This is patch Elm 2.4 PL25 -> Elm 2.4ME+ PL50 (25) Warning: Elm 2.4ME+ is UNOFFICIAL! - K E H ------------------------------------------------------------------------------- Index: elm2.4.ME+.50/hdrs/patchlevel.h *** elm2.4.25/hdrs/patchlevel.h Mon Dec 4 17:50:56 1995 --- elm2.4.ME+.50/hdrs/patchlevel.h Tue Nov 17 21:14:08 1998 *************** *** 1 **** ! #define PATCHLEVEL "25" --- 1,6 ---- ! #define PATCHLEVEL "50 (25)" ! #define LAST_REPORT_TIME 942862610 ! #define BUG_REPORT_ADDR "Kari Hurtta " ! #define VERS_DATE "Nov, 1998" /* for elm -v option */ ! #define WHAT_STRING \ ! "@(#) Version 2.4(ME+), USENET supported version, released Nov, 1998" Index: elm2.4.ME+.50/README.ME+ *** /tmp/8255-very-long-file-name-NULL-comes-in-here Sun Nov 22 09:14:43 1998 --- elm2.4.ME+.50/README.ME+ Tue Nov 17 21:14:08 1998 *************** *** 0 **** --- 1,1962 ---- + $Header: /data/cvsroot/mail/elmme+/README.ME+,v 1.11 1998/11/17 18:20:24 hurtta Exp $ + + Version Elm2.4ME+ PL0 (25) is based to version Elm2.4 PL24 ME8b+. + Version Elm2.4 PL24 ME8b+ is based to version Elm2.4 PL24 ME8b. + + Version Elm2.4 PL24 ME8b is done by Michael Elkins . + For details, look file ANNOUNCE.ME + + [ Michael Elkins posted equivalent of MIME code in Elm2.4 PL24 ME8b + to Elm Development Cordinator. Later changes are not posted to + Elm Development Cordinator. ] + + Version Elm2.4 PL24 ME8b is based to version Elm2.4 PL24. + Version Elm2.4ME+ PLx (25) includes patch of version Elm2.4 PL25. + + Changes of Elm2.4ME+ PL50 (25) compared to Elm2.4ME+ PL49 (25) + -------------------------------------------------------------- + + - elm -vvvv now print names of some configuration files + - elm -vvvvv prints some configuration data (mainly + locale dependent data) + + - "When I use the keys '+' or '-' in an empty attachment + list elm exits with a segmentation violation. This seems + to occur because the variable 'ptrs' points to NULL (and + 'ptr_max' is equal to 0) which is not checked. + + The following patch fixes this bug." + From: Henrik Pletat + + - Added locale_init() to lib/read_rc.c and moved to + catopen and setlocale to it. + + - Added mapping from LC_CTYPE locale to display_charset. + Mapping is done on file $lib/elm.mimecharsets or on + .elm/mime.charsets + + - Special value DISPLAY on elmrc variable charset + means that use display_charset. + + - Moved variable substitution from Configure to + ConfTool/Substitute (called automatically) + + - Ctrl-U and Ctrl-D on builtin++ pager was working + inconsistently. + Problem reported by: Andrew J. Piziali + + Changes of Elm2.4ME+ PL49 (25) compared to Elm2.4ME+ PL48 (25) + -------------------------------------------------------------- + + - Put ME+ to under of CVS. Because revision numbers + changed there was need of changing header text of files. + + - "It fixes a small problem with the date representation + in the upper right corner. Currently, times like + "09:08:02" are displayed as " 9: 8: 2". This is because + src/date.c uses the format "%2.2d:%2.2d:%2.2d" which is + not supported by elm_vmessage() (gets interpreted as + "%2d:%2d:%2d"). The patch changes this to "%02d:%02d:%02d" + (this is used everywhere else anyway), which fixes the + problem." + From: Gertjan van Oosten + + - If To: is empty and I have sender, show sender of mail + (DisplayAddrress (lib/dispaddr.c) was not correctly + detectection empty To: -header.) + + Changes of Elm2.4ME+ PL48 (25) compared to Elm2.4ME+ PL47 (25) + -------------------------------------------------------------- + + - Changed to and cc field on header_rec to include + result of break_down_address + - Added to_matches and cc_matches to pattern.c + + - moved and reimplemented forwarded from utils/newmail.c + and src/addr_util.c to lib/forwarded.c + + - shift_lower needs const for argument + + - Add buffer overflow check to stip_parens() and get_parens() + (lib/striparens.c) + + - Moved main part of parse_arpa_date() to new routine + parse_date_time (lib_/parsarpdat.c) and let parse_arpa_date() + to call that routine. + + - Reimplemented process_expiration_date() (src/expires.c). + Original code did not handled 4 number years and was broken + on year 2000. New code uses just parse_date_time() -- + However it accepts expiration times where date part + is only given from RFC 822 date-time format. + + - Now real_from() (lib/realfrom.c) assumes local timezone, + if timezone is not given, because From mailbox separator + is normally inserted locally. + + - There was "multi-character character constant" -warning + coming from src/mkhdrs.c + Reported by: Witold Witkowski + + - "If you switch to the option editor screen, change + "O)utbound mail saved" from "=sent" to "=sent" (i.e. + enter the displayed character sequence), return to the + index, and then mail a message, the outgoing message + will be saved in a file named ~/=sent rather than in + ~/Mail/sent. The equal sign is not getting expanded + to the user's mail folder directory." + Reported by: Andrew J. Piziali + + - "I suggest checking and accepting (extending) existing PATH" + From: Witold Witkowski + > Moved $PATH to beginning of initial PATH of Configure + + Changes of Elm2.4ME+ PL47 (25) compared to Elm2.4ME+ PL46 (25) + -------------------------------------------------------------- + + - Do not treate characters > 127 as controls (allow + they on words of phrases) + + - Fix name of "Forwarded message from" -text + + Changes of Elm2.4ME+ PL46 (25) compared to Elm2.4ME+ PL45 (25) + -------------------------------------------------------------- + + - Split break_down_address to look_special_tokens(), + scanned_to_phrase(), scanned_to_str() routines + and reimplement break_down_address as state machine + (4 states, 6 terminals). + > Now phrase from group -syntax is ignored and + route from address is dropped. + + - strtok(), strpbrk(), strstr() was declared in both + of elmlib.h and defs.h (remove later declaration). + Noted by: John Goerzen , + Emmanuel Bigler + + - Changed from field on header_rec to include + result of break_down_address (instead of just fullname). + - "From " address is stored to new env_from field + > Sorting based on sender uses first "From " address + and then parsed From: -header (fullname and then address) + > Limit with 'from' now works both of full name and address + and it uses now MIME-decoded full name. Also "From " + address is considered. + - get_return (on returnaddr.c) is simplified. Now it uses + stored (and parsed) From: -header from from field of header_rec + + - Assume INTERNET (style addresses). Remove question from + Configure and all ifdefs. + + - Assume USE_EMBEDDED_ADDRESS (From: and Reply-Top:). Remove + question from Configure and all ifdefs. + + - Call user_init(), init_defaults() and read_rc_file() on + newmail. + + Changes of Elm2.4ME+ PL45 (25) compared to Elm2.4ME+ PL44 (25) + -------------------------------------------------------------- + + - skip_envelope on melib/parse_util.c was not skipping + ">From " -lines. + + - Save composition filename to global variable + (so it does not need to be genrated in signal handler + for removal) + + Changes of Elm2.4ME+ PL44 (25) compared to Elm2.4ME+ PL43 (25) + -------------------------------------------------------------- + + - Added lib/getaddr.c (break_down_address, + free_addr_list) + - Moved rfc822_tokenize, remove_space_tokenized, + free_rfc822tokenized from melib/parse_util.c + to lib/getaddr.c + - Reimplemented parse_arpa_who (lib/parsarpwho.c) + and get_address_from (lib/getaddrfrm.c) with + break_down_address + - Added size argument to (get_address_from) + + - replace break_down_to_list (src/reply.c) with + break_down_address + + - "patch-hmo-1 extends the '#' debug menu, showing + all defined message flags, in the order present + in hdrs/defs.h." + From: Helge Oldach + + Changes of Elm2.4ME+ PL43 (25) compared to Elm2.4ME+ PL42 (25) + -------------------------------------------------------------- + + - print attachment instead of just printing [skipping...] + message when printing from atttachment -menu + + - Changes calling sequence of Write_to_screen + - Added some messages to message catalog + - Added PutLineX to curses.c + - Dropped Putline1, PutLine2, PutLine3 to favor + PutLineX + + - Remove hostname initialization from fastmail, + fastmail now calls read_rc -routines + + Changes of Elm2.4ME+ PL42 (25) compared to Elm2.4ME+ PL41 (25) + -------------------------------------------------------------- + + - Tweak little (add missing space) to build_header_line + + - "When I'm in the attach menu before sending a mail and + I use the delete function on the LAST attachment, the + cursor diappears. Trying then to move up the cursor + elm exits with a sigmentation violation. + + This error appears in several versions of elm." + From: Henrik Pletat + + - Fix call of parse_arpa_who in figure_out_addressee, which + caused that To -address in 'frm' was truncated. + + - Change days_ahead to return year in 4 numbers and simplify it. + - Some year 2000 fixes for src/date.c + - Make elm_date_str to return year in 4 numbers + + [ Undo Elm2.4ME+ PL41h patch before installing this patch. ] + + Changes of Elm2.4ME+ PL41h (25) compared to Elm2.4ME+ PL41 (25) + -------------------------------------------------------------- + + - Tweak little (add missing space) to build_header_line + + Changes of Elm2.4ME+ PL41 (25) compared to Elm2.4ME+ PL40 (25) + -------------------------------------------------------------- + + [ Warning: This is so much changes (and some are + incomplete) so this this quite sure more buggy than + PL40. ] + + - "it's not "a X-Windows" but rather "an X-Windows"; and + BTW, it's not "X-Windows" at all cause that term simply + does not exist (cf. man X). Simply replay "an X-Windows" + by "an X11". Thanks." + From: (sender's name lost) + + ( I managed to delete hurtta+elm Mailbox, so resend + your mails if they are sent between 1998-03-04 and + 1998-04-04. ) + + - Added some messages to message catalog + + - iprintf.c dropped (so far) -- it is not used (at least yet) + + - elm_sfprintf, elm_message + > Global change: Most of sprintf calls replaced with elm_sfrpintf + + - Do not longer test existence of mkdir() + + - "this is another very simple fix that repairs this: + posixsig.c:89: warning: preprocessing directive not + recognized within macro arg" + From: Helge Oldach + + - Dropped error1, error2, error3 to favor lib_error + + - Changed how menu2.c works (and therefore changed + DSN and remailer menus.) + + - Tweak parsearpawho + > Now From: is not assument to be part of buffer + > from in header struct is now mime decoded + > That is now mime encoded when generating comment to + reply address. + + - Handles now phrases on From: -header which uses + continuation lines + + - Dropped get_real_name -- it is it is at least + equal broken than parse_arpa_who + + - ' added to set of characters which need quotation because + sendmail otherwise will quote words which includes ' on + phrases -- even that ' is not special character -- and that + way destroy MIME encoding. + + - MIME decoding is now done when showing values in header menu + (for address headers) + + - MIME encoding also full name got from gecos field of passwd + + - MIME encode also name entered to comments in header edit screen + (for address headers) + + Changes of Elm2.4ME+ PL40 (25) compared to Elm2.4ME+ PL39 (25) + -------------------------------------------------------------- + + - Enumerated options + userlevel: beginner intermediate expert + noencoding: pass-7bit pass-8bit pass-binary + + - Move reading of elmrc to elmlib and put elm tools to read + elmrc. This may make to some time in future to possible + to do binary distributions of elm. + > pmalloc.c moved from src/ to lib/ + + - Added option elm -vvv + + - Add routine lib_error (lib/output.c) + + - Added some messages to message catalog + + - "I have a nice patch to elm-me+-2.4pl25ME+39: + If you press (l)imit and then type "thread", all messages + with the same subject are shown. A leading "Re: " is stripped. + + That is usefull for (large) mailing lists, where you don't read + all messages. (It's like the thread command in most net news + readers)" + From: Christian Hammers + + - "I have determined what the problem is. When you `bounce' a + message elm fails to strip the `From ' line from the top of + the message. This causes Smail to think that the message is + an incoming UUCP transmission. Smail uses the `From ' line + to build the new return path using a !, getting a wrong answer. + + Elm should strip the `From ' from bounced messages." + From: Ian Jackson + > Added CM_REMOVE_ENVELOPE flag to call of copy_message + in remail.c in case of INTERNET is defined + (From comments it looks that passing of "From " is done + by purpose: "For those cases when you want to have a + message continue along to another person in such a way + as they end up receiving it with the return address the + person YOU received the mail from") + + - Elm -v now prints warning "Elm is not intended to be + distributed in binary form. If that is done, then it + is very likely that compile time configuration does not + match to your system." + + Changes of Elm2.4ME+ PL39 (25) compared to Elm2.4ME+ PL38 (25) + -------------------------------------------------------------- + + - "The filter included in elm-2.4ME+37 also appears to + be vulnerable to the "save_embedded_address()" attack, + but not to the "get_filter_rules()" attack." + From: KSR[T] + > I didn't used supplied patch. Instead I added bound + check to one loop in save_embedded_address (I was + added it to one loop but was missed other loop. ) + > Note that Makefile will not compile ot install filter + -- that is intentionally (it was dropped because of + security problems) + + - "here is another small fix relative to elm-2.4ME+37. + This is relevant for empty folders with keepempty=OFF. + Previously an empty folder was deleted *only* when this + folder was left with all (and >= 1) messages marked + 'deleted'. Now an already empty folder is also deleted + upon quit or change to another folder." + From: Helge Oldach + > I didn't used supplied patch. Instead I modified leave_mbox. + + - Added elmrc -option "dsn-success" + + - "it looks like that the help function in change folders + of this used version (2.4me+38) doesn't work." + Reported by: Roeland Th. Jansen + > init_helpmsg was called with wrong number of parameters. + + Changes of Elm2.4ME+ PL38 (25) compared to Elm2.4ME+ PL37 (25) + -------------------------------------------------------------- + + - getarpdate generated date wrongly in year 2000 + + - "Here's another small fix which allows elm to deal more + reasonably with empty last name/first name/comment fields + in the alias file. This is intended for situations where + people use mail addresses which already include the full + name (like: Helge.Oldach@sep.hamburg.com), so that + duplicating the last name and first name on the alias list + seems not reasonable. If one just leaves the last name/first + name/comment fields empty like in: + + hmo = = Helge.Oldach@sep.hamburg.com + + then the alias screen will display nothing but the alias, + which doesn't tell anything. This patch makes elm display + the actual mail address instead of an empty last name/first + name/comment field. + + It also fixes the display of such empty fields when displaying + the actual address while sending a mail ("Message addressed as: + Helge.Oldach@sep.hamburg.com ()") - the empty brackets won't + appear in this case." + From: Helge Oldach + + - "background_wait_time should be background-wait-time in + doc/elmrc-info, otherwise the elmrc update function (elm -w) + won't copy the comment." + From: Helge Oldach + + - "It wasn't properly truncating the subject line if + the rest of the header was longer than 44 characters" + From: Guy Harris + + - If msgcat.c will be compiled, don't include + In that situation always include "elm_nl_types.h" + + - rfc822_tokenize, remove_space_tokenized, free_rfc822tokenized, + strmcat + + Changes of Elm2.4ME+ PL37 (25) compared to Elm2.4ME+ PL36 (25) + -------------------------------------------------------------- + + - include in defs.h + + - include in lib/posixsig.c on Ultrix V4.4 + From: Bernhard Simon + + - "The latest elm.me has a nl_types.h include which on + Digital UNIX is conflicting with a system include having + the same name, which is of course needed for proper + compilation :(." + From: Christophe Wolfhugel + > Rename nl_types.h to elm_nl_types.h + > There was wrong number of argument in catgets + call in iprintf.c [ not currently used ] + From: Christophe Wolfhugel + + - Changes for configure to run in Dec Unix (OSF1) + "- cc -std (to have prototypes) + - nm_opts='' for DEC Unix or it gives wrong output with -p + and Configure won't see some functions (like strspn) + - signal hardcoded, because cc -E or cpp or DEC Unix does not + pass the standard defines (the ones put by cc -std) and we + will miss the signal prototype anyway :(." + From: Christophe Wolfhugel + + - Some changes that Elm compiles cleanly on Digital Unix (OSF1) + [ It was not defining __STDC__, but was complaining + if char * argument was given to unsigned char * argument + of function (and opposite) ] + + - Digital Unix (OSF1) was defineing macto BSD (sig!) + Renamed BSD macro to BSD_TYPE + + Changes of Elm2.4ME+ PL36 (25) compared to Elm2.4ME+ PL35 (25) + -------------------------------------------------------------- + + - "If the mail transport agent (MTA) is missing, elm + silently fails. No message is delivered and no warning or + error message is displayed." + From: Jan Prikryl , + Andrew J. Piziali + > Return status of process was decoded incorrectly. + > Also errno status was written to wrong end of report -pipe. + + - Redraw screen also if old position was on screen + when new mail arrives. + + - Dropped return_value_of from curses.c + + - Support arrow cursor on attachment menu + + - "I recently tried to run the Configure script on AIXv4.2.1 which + failed trying to extract the function names out of libc.a. + I updated the script so it will work on AIX." + From: Will Fiveash + + - "Also, I am including my file locking patch so that + elm will lock a folder when trying to save or copy to it." + From: Will Fiveash + > Added elmrc option lock-in-copy, that locking is done if + this option is TRUE (default). + * NOTICE: This does not work with dot-locking + * NOTICE: This is usefull only with folders which elm + treats as mailboxes, because folder are not + locked (when visited). + + - Changed can_open to use my_wait and prepared for EINTR + + - Dropped move_old_files_to_new() + + Changes of Elm2.4ME+ PL35 (25) compared to Elm2.4ME+ PL34 (25) + -------------------------------------------------------------- + + - "I've built elm-2.4.ME+3[34] under SunOS 4.1.2 using + libc 1.1.7 and gcc-2.7.2 and can't actually open a mail + file with them. " + From: Dave Rynne + - wait in can_access was not prepared for EINTR + - can_access was not using my_wait + [ And why can_access just not use access system call? + -- it used real uid and gid anyway!!!! ] + + - "The run_state structure declared in me.h has a member + called "errno". However, it is unsafe to use errno as an + identifier under some Unices, e.g. when errno is a macro. + Here is a patch against PL34 which renames the struct member." + From: Jesse Thilo + [ It turns out that culprit is Linux libc 6 ] + + - Add support for multipart/alternative. Elmrc + variable pagealternative is added. If that is TRUE (default), + and at least one subpart of multipart/alternative is displayable + metamail is not called. If that is FALSE, metamail is not called + if all subparts of multipart/alternative are displayable. + + Changes of Elm2.4ME+ PL34 (25) compared to Elm2.4ME+ PL33 (25) + --------------------------------------------------------------- + + - Print "Elm: Unsupported locale (check $LANG)" if setlocale + fails. + Suggested by: Michael Rudolf + + - "I've found a bug in the Configure script for elm2.4.ME+.33: + when Configure asks for the metamail path and the path doesn't + exist elm asks you: "File ... doesn't exist. Use that name + anyway?", if you answer 'y' then metamail_path is set + to 'y'. I think that's not the desired action... :)" + From: Jose A. Rodriguez + + - Add elmrc -varaible pgp. This gives path to PGP. + If "none", pgp is not called. Also checked that + pgp is executable with given path. + Suggested by: John Goerzen + + - check with can_open that debug file can be opened. + + - -A (attach) option didn't used "~/.elm/mime.types" file. + Reported by: Andrew J. Piziali + + Changes of Elm2.4ME+ PL33 (25) compared to Elm2.4ME+ PL32 (25) + -------------------------------------------------------------- + + - Content-Disposition: inline; filename = "filename" + was not parsed correctly. + Reported by: Jussi Kaurola + + - Remove NIS tests from Configure. $passcat was not used + ( Next section "set up shell script to do ~ expansion" + may perhaps need it, but actually it does not use it. ) + Hint from: Jerzy Sobczyk + > Now sun is always added to from of $libswanted + + - Added elmrc option background-wait-time (default 2 seconds) + If = 0 wait mailer to completion + > 0 wait background-wait-time seconds for completion + if not completed in this time, let it go to background + + (completion of mailing is still reported if user have + not exited from elm in that time when mailer have completed) + Suggested by: Heiko W.Rupp + Daniel C. Sobral + * THIS option is supported only in POSIX systems + > Added my_wait to lib/posigsig.c + + - Added VOLATILE to defs.h + > to resize_screen, redraw_screen (elm.h, headers.h) + + - "When I tried to compile elm2.4.ME+.32 for Digital Unix 4.0 I got + a few compiler errors. The problem is that ANSIC/not ANSIC headers + are not in sync... + + They are fixed with this patch (for lib/iprintf.c file)" + From: Jose A. Rodriguez + * Actually this (ELMCHAR, iprintf) code isn't yet in use + + - "The bug was that filter truncated the "Saved folder" to + only 3 chars (out of the full path). To fix it, just change + the following line in 'utils.c' in the filter directory" + From: From: Yuval Shamir + [ Suggested fix was incorrect. ] + + > Notice: filter code is not tested (and compilation + of it is commented out from Makefile in all ME+) + + Changes of Elm2.4ME+ PL32 (25) compared to Elm2.4ME+ PL31 (25) + --------------------------------------------------------------- + + - Incorrect Content-length: -header was causing corruption + of folders. + From: Guy Harris + + - argv_from_to was not handled ',' in comments correctly. + Detected from report of Gary Casterline + + > Use rfc822_toklen instead of len_next_part + + - SECURITY: strcpy -> strfcpy changes of Elm2.4ME+ PL29 (25) + was not done for curses.c in this source tree. + Bug report on List . + From: John Goerzen + + - Some changes on curses.c (bl -- bell) + + - Change output of option -v + + - "If you use 'answer -u', then every user name is truncated to + three letters." ... "Change line 232 to: " ... + From: Jean-Pierre Radley + [ Suggested fix was incorrect. ] + + - "Configure always has had a problem on Linux systems, + extracting names in a usuable format (from /usr/lib/libc.so). + The following small change fixes this." + From: Bauke Jan Douma + + - "In Solaris enviroment NIS+ is now used instead of Yellow + Pages so I have modified Configure to reflect this and + allow to use niscat in the same fashion as ypcat was." + From: Jerzy Sobczyk + [ Undone in PL33 ] + + - Compilation of filter (actions.c) fail. + Reported by: Arnout Boer + + - Typo mismatch in lib/strftime.c + From: Yuval Shamir + - [wordwrap.c, bultin editor] + "Wordwrap and delete at the + beginning of the line does not work in PL31. Fix is below." + From: "Zoltan T. Hidvegi" + [ I didn't used supplied patch. ] + + - [builtin editor] There was 'sizeof buffer' instead of + 'buffer_size' in get_with_expansion() + + - Add same fflush -fix to remail.c which is in mailmsg2.c + + Changes of Elm2.4ME+ PL31H (25) compared to Elm2.4ME+ PL31 (25) + --------------------------------------------------------------- + + [ This is out of branch patch. ] + + - Incorrect Content-length: -header was causing corruption + of folders. + From: Guy Harris + + - argv_from_to was not handled ',' in comments correctly. + Detected from report of Gary Casterline + + > Use rfc822_toklen instead of len_next_part + + Changes of Elm2.4ME+ PL31 (25) compared to Elm2.4ME+ PL30 (25) + -------------------------------------------------------------- + + - Print "Warning: BINARY data? Check Content-Type!" + when trying send binary (*) data as text in attachment + menu. + + (*) binary = not fill requirements of "7bit" and "8bit". + + - Convert some system_call() to use + start_run() ... wait_end() instead. + + - optionally_enter() in pgp_decode.c was called without + size -argument + + - "The problem appears to be that when a "FILE *" is rewound, + the underlying file is not, causing mail messages to be empty. " + Reported by: Jon Harder + [ Now input file descriptor is fflush'ed -- this should + synzronize it. + ] + + + - "I often mail PGP-encrypted attachments to people with the + appropriate content type. If I want to review the attachment + before sending it, I go the to a)ttachments screen, select the + attachment of interest, and press . Elm prompts me for a + pass phrase because I always encrypt using my public key along + with the recipient's. If I mistype my pass phrase, I cannot use + control-F (^F) to cause Elm to forget the cached pass phrase from + either the attachments screen or the send screen. + + I wish ^F were recognized on those two screen also." + Suggested by: Andrew J. Piziali + * Added ^F to verify_transmission, attachment menu, and to + showmsg_c.c + + Changes of Elm2.4ME+ PL30 (25) compared to Elm2.4ME+ PL29 (25) + -------------------------------------------------------------- + + - mailerflags -array (argument vector) was not NULL + terminated in remail.c. + + - Elm was not recognized SortBy (and aliassortby) keywords + [ strfcpy was got size of pointer instead of array size + in read_rc.c ] + Reported by: "Ralf S. Engelschall" + + - Fixed some error messages in read_rc.c + + - "The problem is that 'case:' statement use int range, so + 0xFC02 expanded to 0x0000FC02, not to 0xFFFFFC02, + but xxx->magic value is short and expanded to 0xFFFFxxxx, + so comparation is unsuccessful. + Fix is simple: direct cast to short." + From: "Andrey A. Chernov" + [ I don't actually used that fix. Instead I changed magic + values to 'unsigned short' in elmlib.h. Now they + are used same way as in melib.h. ] + * Actually this (ELMCHAR) code isn't yet in use + + Changes of Elm2.4ME+ PL29 (25) compared to Elm2.4ME+ PL28 (25) + -------------------------------------------------------------- + + [ Warning: This is so much changes (and some are + incomplete) so this this quite sure more buggy than + PL28. ] + + - SECURITY: Elm was passing `shell command` without + quotation to shell when invoking sendmail. + Reported by: Wolfgang Ley + from: "Computer Emergency Response Team, Germany" + + (Originally from: Dirk Nimmich ) + > Don't use shell when calling sendmail (or other mailers) + - quote_args() dropped from strings.c + - now elm waits until sendmail returns + (and simple not put it background without checking + if mailing succees) + > Now -V option of Elm works better + + - Small change in displaying of multipart messages + + - (mime_decode.c) Use in_state_seekable instead of + testing magic agaist STATE_in_file + + - Move state.c from src/ to melib/ + + - New header files melib.h and elmlib.h + + - Add routine strfcat to strfcpy.c + + - get_folder_type is litle more smarter to detect files + in mail directory + + - d)omainze removed from header config -menu + + - Don't ask shell in configure -- take shell from passwd + If passwd is used, default value for elmrc is null string + If $SHELL env variable is used, defaualt value for elmrc is + $SHELL string + + - If $EDITOR env varaible is used, default value for elmrc is + $EDITOR string + + - Default to p)gp if replying to pgp encoded message + (in verify_transmission -menu) + + - Remove bounceback variable and bouncebk.c + + - [defs.h] In IRIX define 'int32' by using 'int32_t' + Warning: If you have compiled Elm with ABI=-64 + (default on IRIX64), this changes on disk + format of aliases database + * Actual use inttypes.h + (does not work if -cckr is used, + (also can't be used in IRIX 5.3 because of conflicts + with sys/types.h -- ARGH!) + + - Removed references to VMS and EUNICE from Configure + + - Removed I_STDARG -- + now if ANSI_C is defined stdarg.h is used, + otherwise varargs.h is used + + - Add do { .... } while(0) to dprint -macro + + - Global change: Use strfcpy instead of strcpy + and: Use strfcat (new routine) instead of strcat + (but sprintf -calls are still untouched!) + + [ Undo Elm2.4ME+ PL28s patch before installing this patch. ] + + Changes of Elm2.4ME+ PL28s (25) compared to Elm2.4ME+ PL28 (25) + -------------------------------------------------------------- + + - SECURITY: Elm was passing `shell command` without + quotation to shell when invoking sendmail. + Reported by: Wolfgang Ley + from: "Computer Emergency Response Team, Germany" + + (Originally from: Dirk Nimmich ) + > Quote ` -character + + Changes of Elm2.4ME+ PL28 (25) compared to Elm2.4ME+ PL27 (25) + -------------------------------------------------------------- + + - mailerflags was too short for DSN flags + + Changes of Elm2.4ME+ PL27 (25) compared to Elm2.4ME+ PL26 (25) + -------------------------------------------------------------- + + - 'hdr -> magic = NULL;' should be 'hdr -> magic = 0;' + in parse_util.c + Noted by: "Andrew J. Piziali" + + - "Change Elm to exit after printing the error message, rather than + continuing and erasing the screen message and all, if it + encounters an error when reading your "elmrc" file." + From: Guy Harris + * But do not do this if in batch mode! + * Don't exit if there is errors in global elmrc! + * Let user save new .elm/elmrc with option -w + + - state_read_headers was building next_header chain + incorrectly (and possible other chains also). + + - add_paramater_t routine (wrapper for add_parameter) + to mime_encode.c + + - Let user attach files with option -A + To get this work correctly, user needs to have + mapping file extension -> mime/type in + his .elm/elmrc + + - Added field magic to mime_t and routine mime_panic. + Now it is possible to get error message MIME_PANIC + if in mime structure in memory is something wrong + (analogy to STATE PANIC and HEADER PANIC). + + - Size of forwarded mail in message/rfc822 was not set + correctly in Elm2.4ME+ PL23. + + - Added DSN (Delivery Status Notification) Configure + menu to verify_tranmission. You need sendmail 8.8 + to get this. + + - In menu of verify_transmission, show + a)attachments option also in user level = 0, if there + is attachments. + - In menu of verify_transmission, show + p)gp option also in user level = 0, if original message + was pgp -encoded (when replying). + - In menu of verify_transmission, show + D)SN option also in user level = 0 if there is dsn settings + (does not apply now). + + Changes of Elm2.4ME+ PL26 (25) compared to Elm2.4ME+ PL25 (25) + -------------------------------------------------------------- + + - Adding final \n\n was badly broken in copy_message() + Reported by: Eilon Gishri + + - Try fixing handling of null bodys in read_headers() + { newmbox.c } in some cases. Perhaps this brokes + nothing? + + - rfc1522_encode_text failed to \0 terminate Subject + From: Claus Assmann + + - PGP fails to check signature if KEEPBINARY=ON + From: Heiko W. Rupp + + - Created new directory melib and made archive libme.a. + Some code is moved to that directory. You need to + create that directory before applying path! + + Changes of Elm2.4ME+ PL25 (25) compared to Elm2.4ME+ PL24 (25) + -------------------------------------------------------------- + + - Problem with screen size (elm_LINES and elm_COLUMNS + are inconsistent). + [ I was commented out line what was comment + /* kludge for HP Window system? ... */ + in PL23. VERY misleading........................ + + I HATE this. - K E H + ] + + + - Fix release month + + Changes of Elm2.4ME+ PL24 (25) compared to Elm2.4ME+ PL23 (25) + -------------------------------------------------------------- + + - ks/ke capabilities are optional. They are not present + in ANSI -terminal. + From: "Andrey A. Chernov" + + - X-ELM-OSV was not correctly parsed + + Changes of Elm2.4ME+ PL23 (25) compared to Elm2.4ME+ PL22 (25) + -------------------------------------------------------------- + + - Add a new config variable, called askpgpsig. + When set to ON, Elm will ask for a username to + sign pgp messages with. + From: Guido van Rooij + + - menu2.c: ch must be int, or comparations with -1 (EOF) + and 257 (REDRAW_MARK) produce wrong results. + From: Andrey A. Chernov + From: Guy Harris + + - In "do_pipe()", "redraw" was both a variable and a label. + From: Guy Harris + + - If you've set characters such as the word-erase or + reprint character to non-standard values, use your values, + not the standard ones. + From: Guy Harris + [ Added also !escaped test, because now these characters can + be printable. ] + + - Moved processing of cursor keys to ReadCh + (old cursor key handling was buggy). + > HP's shoftkeys are not supported: + No options: -k -K + No elmrc variables: softkeys keypad + > Recognized keys are: + Key tercap name terminfo name + ----- ----------- ------------- + UP arrow ku kcuu1 + DOWN arrow kd kcud1 + LEFT arrow kl kcub1 + RIGHT arrow kr kcuf1 + PAGE UP kP kpp + PAGE DOWN kN knp + HOME kh khome + HELP %1 khlp + FIND @0 kfnd + > To get keys to use, your $TERM variable must point + termcap/terminfo defination what includes these keys. + Notice that 'vt100' does not include other than cursor + keys. You must perhaps make new defination, for example + terminal type 'vt100-plus' :-) For terminfo use 'tic' and + 'infocmp' -commands. + + - state operations are moved outside of #ifdef MIME ... #endif + - Added new routines state_write_headers, state_read_headers, + delete_headers, locate_header and find_header to + parse_util.c. read_header_line is now implemented with routine + state_read_hdr_line. file_read_headers is implemented with routine + state_read_headers. + > Now included RFC 822 parts have printed text + -- Start of included mail From: xxxxx + instead of + -- Start of included mail. + - Rewritten read_mime_header to use file_read_headers, mainly by + dividing it to new routine parse_mime_headers + > Rewritten rfc822_parse. Now non-mime messages inside of + message/rfc822 should be handled more correctly. + - Setting elmrc variable + require-mime-version-for-body-encoding to FALSE causes + that almost all messages are treated as MIME -messages and + MIME-Version is not required for these (in top level + or inside of message/rfc822). + + - Add flag CM_REMOVE_ENVELOPE for copy_message + > make message/RFC822 attachment when forwarding with copy_message + + - Make wait() in matapager restart in case of EINTR error + + - Do MIME Part 2 (RFC 1522) decoding even when not MIME + even when not Mime-Version -header is present + (If you want olf behauviour, set elmrc variable + require-mime-version-for-hdr-encoding to TRUE). + Header decoding is not done if sending elm was + disabled to do MIME Part 2 (RFC 1522) encoding + (as set in elmrc variable nohdrencoding). That + information is carried in header X-ELM-OSV: + (Our standard violations). + + - quoted-printable-decode was not decoding zero length part + correctly. + > Indicate with length == -1 as not set instead of length == 0 + + - Added symbolics values RHL_MARK_FOLDING and RHL_CHECK_HEADER + for flags of read_header_line. Now read_header_line check + header for carefully. Side effect: If there is spaces before + ':' in header, them are deleted (assuming RHL_CHECK_HEADER). + + - Add to mime_types[] following non-standard main content types: + x-world, model + + Changes of Elm2.4ME+ PL22 (25) compared to Elm2.4ME+ PL21 (25) + -------------------------------------------------------------- + + [ Warning: There is quite much changes in + user interface, so perhaps there is bugs. ] + + > Changes in Makefile.SH -- remember to run + 'Configure' or 'Configure -d' or 'Configure -S' + + - 'ch' need to be 'int' in alias() + - 'cmd' need to be 'int' in MenuLoop() + - 'ch' need to be 'int' in generic_menu() + Problem noted by: Claus Assmann + + - There was unwanted trigraph in run_decoder() + Problem noted by: Claus Assmann + + - There was missing 'default:' in state_put() + Problem noted by: Claus Assmann + Problem noted by: Guy Harris + + - rename 'timeout' to 'elm_timeout' to get rid out linker warning + - rename 'filter' to 'elm_filter' to get rid out linker warning + - rename 'LINES' to 'elm_LINES' to get rid out linker warning + (These are defined also in libtermlib.so) + + - rename 'COLUMNS' to 'elm_COLUMNS' to match 'elm_LINES' + + - If there is Raw(ON) in parent in metapager(), then + also Raw(OFF) must be in parent + Noted by (?): Jurgen Fluk + + - Replace some getchar()s with ReadCh(REDRAW_MARK) + in syscall.c + + - Replace some getchar()s with ReadCh(REDRAW_MARK) + in read_rc.c + + - Replace some getchar()s with ReadCh(REDRAW_MARK) + in alias.c + + - Replace some getchar() with ReadCh(REDRAW_MARK) + in utils.c + + - Use type application/X-ELM-encode with + [enclode] ... [clear] if message is multipart + (Elm 2.4ME+ PL6 should know this type, but there + was error is some versions of ME+ in mime_notplain). + + - mime_notplain was returning TRUE (should be FALSE) + for application/X-ELM-encode. + + - The builtin++ pager clears the screen before displaying the + first screen of the message; however, the metapager wasn't + clearing the screen before running the external pager, which + left some junk on the screen. + From: Guy Harris + + - Convert optionally_enter to use flag bits and + now it returns REDRAW_MARK in most places when redraw + is needed + + - LOCK_DIR removed from Configure / config.h.SH + + - Rememmber if last Configure run there was answer + 'none' for additional librariers + + - Some new checks in Makefile (or actually in Makefile.SH), + Configure is run lately (that confih.sh or Makefile or + config.h are not stale) + + + Changes of Elm2.4ME+ PL21 (25) compared to Elm2.4ME+ PL20 (25) + -------------------------------------------------------------- + + - More Configure fixes for IRIX 6.2 + + - PGP signed messages (with type text/plain) was not handled + correctly. + Noted by: Volker Paulsen + + - Just a short one-liner to fix the display: + When you leave the key selection menu in + pgp_mail_public_key() with `q' (i.e., don't mail a key), + this routine currently returns 0. This means that the main list + won't be redrawn and seems to be a bug. + From: Thomas Roessler + + - read_number should create buffer of size SLEN, not NLEN + since it uses optionally_enter() which assumes that the + buffer has length SLEN + From: Thomas Roessler + + - Additionally, one might want to have the pgp functions + in the help menu. + From: Thomas Roessler + + Changes of Elm2.4ME+ PL20 (25) compared to Elm2.4ME+ PL19 (25) + -------------------------------------------------------------- + + [ There is changes of error handling of leave_mbox -- + so report, if you lost mail... ] + + - Look also termlib.so and curses.so in Configure + + - Remove PGP test from need_meta in showmsg.c + (It was causing that metamail was not called for + multipart/signed what isn't supported.) + + - Move metamail flags before PGP flags in show_message + in screen.c + + - lib/safeopen.c need to be added to MANIFEST + Noted by: Keith Manley + + - Elm as acting quite oddly if ending boundary of multipart + structure was missing. + Detected from report of: Rolf Larsson + + - Add sleep_message() to out_utils.c + + - Detect if there is read error in newmbox! + + - Check read errors also in some other places.... + + - Changes in leave_mbox -- now it returns with value -2 if + it fails to leave folder + + - SIGUSR1 and SIGUSR2 is anly handled when in prompt + (because otherwise we can be in wrong state -- + for example in middle in resync.) + + - Check in ReadCh if it was called from GetPrompt + (GetPrompt want to see EINTR error!) + + - ReadCh takes now redraw flag: Either 0 or REDRAW_MARK. + Now elm is able in many places automatically redraw screen + when elm resumes after suspending. + + - skip_envelope() to parse_util.c + + - Use skip_envelope() in get_and_expand_everybody() and + handle_reply_to() -- now elm should not be confused + with MMDF envelope of mail + Reported by: { Unknown -- mail lost } + + Changes of Elm2.4ME+ PL19 (25) compared to Elm2.4ME+ PL18 (25) + -------------------------------------------------------------- + + - Fix SIGSEGV (with application/pgp; format=mime) + + - With application/pgp; format=mime in included mime type + nested structures was not parsed (actually that was causing + SIGSEGV) + + - recognize application/pgp; x-action=signclear + + - Add "-- Start of included mail." and "-- End of included mail." + texts. + + Changes of Elm2.4ME+ PL18 (25) compared to Elm2.4ME+ PL17 (25) + -------------------------------------------------------------- + + [ This patch includes very many changes, so perhaps there + are bugs ... ] + + - Clean up some of the questions in the "Configure" script. + From: Guy Harris + + - Added a '-S' option to the Configure script to just redo the + "variable substitutions on .SH files" that the script performs + at the end. "I found this extremely useful, as I often made + changes to the config.sh (for instance, the 'optimize' flags), + and I wanted to re-create the Makefiles without running the + entire Configure script all over again." + - Also added '-h' option to report usage and available options to + Configure script. + From: Will Day + + - Add support for a "mailpermissions" variable in your "elmrc" file + that can be set to specify the permissions to be used when creating + a new mail folder. The syntax to use when specifying the permissions + is *not* numeric, it's an "ls"-style string such as "rw-rw----", and + the default is "rw-------". + From: Guy Harris + + - (Elm)HelpViewAttach was missing from s_elm.m + + - Have the NLS stuff make "s_elmalias.h" from "s_elmalias.m". + From: Guy Harris + + - Fix up "gencat" to compile on systems that don't do ANSI C + (e.g., SunOS 4.1[.x] when using the bundled C compiler). + From: Guy Harris + + - Some additions to help text of Elm + + - Reconstruct mime_send_t. There was wrong type when multipart + message was sent and pgp was used. + Noted by: J. Porter Clark + + - Added elmrc option: "quoteforward", which will let the user + restore the behavior of original elm when forwarding messages + - that is, to quote them like replies. + "I wanted this available, because usually when I forward a + message, I will comment various lines, and if it's not quoted + like a reply, my comments can be confused with the forwarded + message (and I didn't want to have to manually comment the + message every time)." + From: Will Day + + - convert mail() and send_msg() to use binary flags. + Get rid of global variable "forwarding". + + - Added safeopen_rdwr() call yet some places. + "While browsing through elm me+ xyz kh's source code, + I stumbled across an invocation of tempnam() in a + somewhat insecure environment: In mailmsg2.c, it's + output is used as argument to fopen(...,"w+"). This is + dangerous, since it may allow other users to access those + temporary files." + From: Thomas Roessler + > Also use open_end_update() call, then code is cleaner + [ In suplied patch there was possibility that elm destroyes + =sent folder... ] + + - Modification for Configure: On a Linux system which uses + ELF-format shared libaries, the library symbols do not have + an underscore prepended to them. + "I simply removed the [._] part from the sed expression. This + makes everything compile and run happily on my machine, + but if you want to correct this problem in your distribution, + you would probably want to keep the old rule as well, to keep + it from breaking elsewhere." + Noted by: Matthew Hunt + + - "The second patch (re-)enables the use of the `T' key in the menu + at the end of a message (tag + next). This used to work in + earlier releases of elm." + From: J Wunsch + + - "I think there is a small bug in metapager.c which has the + consequence that the pager menu is sometimes displayed on the + wrong screen page. The patch below fixes this behaviour." + From: Thomas Roessler + + [ This patch little questionable -- 'ti' is "string to begin + brograms that use cup" (cursor addressing) and 'te' is + "string to end programs rhat use cup" -- problem is + that xterm uses ti/te to swap screen page... ] + + - "Additionally, there is still an ugly timezone bug in elm-me+; + the old patch for elm 2.4 applies cleanly to the current me+ + version." + Noted by: Thomas Roessler + > Fix timezone: + "OK...here is the fix for the wierd timezone shifting. + Besides the botched month parameter, the year also was wrong. + + This patch better documents the input parameters to the + lib/date_util.c procedures, and fixes get_tz_mins() to avoid + a wierd offset shifting problem. Thanks to Thomas Roessler + for chasing down the cause of the + problem." + From: Chip Rosenthal + + - Fix dangling pointer bug in attach_viewer! + + - Recognize text/x-pgp. + From: Guido van Rooij + [ + #ifdef USE_PGP added ] + > If type was text/x-pgp, don't consider case + that PGP file is binary + + - Don't convert application/x-pgp* to application/pgp + in newmbox.c -- instead recognize it in mime_decode. + + - Don't reset type_opts when converting text/plain to + application/pgp. + [ That change was buggy -- fixed in PL21 ] + + - Move "Decrypting message..." and "Checking signature..." + messages to another place. + + Changes of Elm2.4ME+ PL17 (25) compared to Elm2.4ME+ PL16 (25) + -------------------------------------------------------------- + + - Some systems use OXTABS bit instead of TABDLY + [ Various paches for *BSD ] + From: Ollivier Robert + From: Jeff Thieleke + From: Andrey A. Chernov + + - Undone HP-UX change for ANSI_C + "This is mostly incorrect. We figured out that + the problem was the -traditional flag that was included + by default when compiling with gcc." + Noted by: Jeff Thieleke + + - Removed -traditional from gcc flags in Configure + "The real fix would be to remove the default CFLAGS when + compiling with gcc, since they probably aren't needed any more." + Suggested by: Jeff Thieleke + + - Don't allow change a filename of forwarded mail in attachment + menu. + + - If content-transfer-encoding is unknown when saving attachment, + treate it as binary. + + - Don't allow change filename in attachment menu to non-existent + + Changes of Elm2.4ME+ PL16 (25) compared to Elm2.4ME+ PL15 (25) + -------------------------------------------------------------- + + [ Warning: state routines are rewritten -- so look below + what to do if you get STATE PANIC message ... ] + + - Add #include "headers.h" to safeopen.c to get rid + out one stupid compiler warning + Urged by: Andrew J. Piziali + + - struct state is divided to two structs: struct in_state + and struct out_state. Large modifications to state + routines (state.c). + + If you get message STATE PANIC: Bad magic number + compile elm with debugging -g and -DDEBUG + and run elm with -d11 and try get it repeated. + Now STATE PANIC should cause producing of core file. + + Look with debugger where that STACE PANIC accured (ask + stack trace from debugger) and look also file ~/ELM:debug.info + + - If terminal driver is expanding tabs, do not print + them -- instead jump to correct location + + - "HP-UX apparently doesn't declare __STDC__ (like AIX), + so compiling with cc or gcc fails in src/init.c at the getgid(), + getuid() prototypes." + From: Jeff Thieleke + [ Undone in ME+ PL17 ] + + - If you sign 8-bit mail, it will be later encoded with + quoted-printable or base64 (by elm or by sendmail). + Thus 'invalidating' because pgp_decode does not undo contet- + transfer encoding before calling PGP. + Noted by: Zoltan Hidvegi + Noted by: Hans-Christoph Wirth + * Now content-transfer-encoding is decoded before bodypart + is passed to pgp (it also now work, when *.pgp file + is encoded with base64). + > That decoding is done via temporary file -- perhaps I + some time stack state routines so temporary file is not needed + and decoding of content-ttransfer-encodings are done during + reading. + > If I some day get implemented MIME/PGP then 8-bit mail is always + encoded with quoted printable and then passed to PGP for signing. + + - Display text before PGP armor if elmrc variable showpgppreamble + is set (default). Binary PGP files are detected by looking first + byte (after decoding content-transfer-encoding). + + - Fix to PL15 fix: "Elm fail in resyncing if mailfile is created + after initial entering to mailbox." -- previous fix can cuase + elm exiting with message + "What's this? The temp folder already exists??" and then + "Ahhhh... I give up." + Perhaps noted by: Andrew J. Piziali + + - SECURITY: Elm was running PGP as setgid mail (by using popen) + when searching matching keys -- converting to use + pipe - fork - setgid (groupid) - execl + + - Need pass +language=en to PGP when searching matching keys + Noted by: Boris Staeblow + + - Initialize last (extra) element also to NULL in DynamicArray + so that DestroyDynamicArray works (lib/dynarray.c) + -- was causing SIGSEGV in GetPGPKey + + [ Undo Elm2.4ME+ PL15s patch before installing this patch. ] + + Changes of Elm2.4ME+ PL15s (25) compared to Elm2.4ME+ PL15 (25) + --------------------------------------------------------------- + + - SECURITY: Elm was running PGP as setgid mail (by using popen) + when searching matching keys -- converting to use + pipe - fork - setgid (groupid) - execl + + Changes of Elm2.4ME+ PL15 (25) compared to Elm2.4ME+ PL14 (25) + -------------------------------------------------------------- + + [ Warning: This is perhaps more buggy than PL14. ] + + - Some changes in Makefiles, Configure + - Change of temporary mbox name, put temporary mbox to + back to temp directory (from home directory) + - lib/safeopen.c + - can_open(filename,"sw") checks if creating of _new_ file + is possible (O_EXCL) -- and use that in leavembox.c + Side-effect: New file is created + - use isprint((unsigned char)ch) in src/curses.c + - "I removed all fopen(file, "w") where file was a filename in /tmp + and replaced these with a safeopen(file) call. These changes + should be also applied to the vanilla elm source." + - "Also there are some minor bugfixes for PL11. + Elm did not asked my PGP password when I wanted to + read a PGP encrypted mail since it wrongly thought + that the mail is only signed but not encrypted. + That's the most serious bug." + - "I add a ClearScreen() before the fork in pgp_decrypt_init() + to make the screen look clearer." + From: Zoltan Hidvegi + + - Routine safeopen_rdwr added to safeopen.c + + - Test agaist special modes was WRONG in leavembox.c + (it was 0x7000 and it should be 07000) + - Treate files with special modes as SPOOL files (newmbox.c) + - Handle 'intelligent' way also SPOOL files what haven't in mailhome + - Restore also special modes of file if file is not executable + - Don't chown file if owner is already correct -- it resets + special modes + * With these changes it is possible to redirect sendmail aliases + to files (mode 04600) and cause that elm does not reset setgid + bit (what is required to be set, before sendmail accepts forwarding + to no world writable files). + + - Multiple To: -headers was cauisng of trashing of memory + and SIGSEGV + Noted by: Richard Allen + + - Elm fail in resyncing if mailfile is created after initial + entering to mailbox. + Noted by: Miroslav Ruda + { I didn't used suplied patch. } + + - Elm fails to create received -folder + Noted by: Jussi Kaurola + - Added function open_end_update to file_util.c + + - Replace iscntrl((unsigned char)*pending) with + !isprint((unsigned char)*pending) in builtin++.c + because Configure asks that is isprint + working with 8-bit values, but it does not ask about iscntrl + + * Notice: Seems that someones does not agreed with interpretation + of ASCII_CTYPE defination. My interpretation is: + + ASCII_CTYPE defined: isprint works only with arguments + 0-127 + + Conclusion: We can't test that is characters in + range 128-255 printable or not, so + don't print them + + ASCII_CTYPE is defined when you answer "no" to following question: + "Should Elm assume that isprint() won't break on + non-ASCII characters?" + + To get 8-bit characters to work you need: + 1) 8-bit clean ctype functions and macros + 2) ASCII_CTYPE is not defined + 3) setlocale exists (Configure have found locale.h; + that is: I_LOCALE is defined) + 4) Correct locale defination (environment variable + $LANG or $LC_CTYPE usually) + Look manual page of setlocale + - If you are not using builtin(++) pager in Elm, thinks + can go more complicated -- all pagers does not use + locale definations + + - Use kill(getpid(), SIGSTOP) instead of kill(0, SIGSTOP) + so that delivering of mail does not stop when user suspends + Elm. + Reported by: Richard J. Auletta + + Changes of Elm2.4ME+ PL14 (25) compared to Elm2.4ME+ PL13 (25) + -------------------------------------------------------------- + + - PGP signed (or encrypted) MIME mail was not quoted when + replying. Quick fix for pgp_decode in pgp.c. It still + perhaps have broken... + + Changes of Elm2.4ME+ PL13 (25) compared to Elm2.4ME+ PL12 (25) + -------------------------------------------------------------- + + - Compiler was giving warning that 'char *' and + 'unsigned char *' are not assignment-compatible + (for code in mime_encode.c) + Noted by: Arkady Altman + + - Include headers of forwarded message to editor + buffer even when you have noheader = ON + (that is in situation mimeforward = OFF). + That is now controlled with elmrc variable "noheaderfwd" + (default: OFF) + + - Elm was getting SIGSEGV in remailer_parse_opts in remailer.c + Noted by: V Menayang + [ Notice that this REMAILER code is still broken! ] + + - remailer.c was calling finger without resetting egid! + + - FreeBSD changes to utils/newmail.c also pertain to NetBSD + From: Jeff Thieleke + + Changes of Elm2.4ME+ PL12 (25) compared to Elm2.4ME+ PL11 (25) + -------------------------------------------------------------- + + Warning: This version is probably much more buggy + than PL11 !! + + - The blank line between two messages in a folder + was missing. + Noted by: Juergen Obermann + + - If nohdrencoding = TRUE and there is 8bit characters in subject, + treat as there was 8bit characters in body. + From: Andrey A. Chernov + [ This does not work very well... ie. It works only when + Content-Type is Text/plain. ] + " Before RFC 1522 issued, too many users (f.e. Russian users) use + 8bit chars in Subject: header field. There is a lots of 8bit + capable MUA & MTA exists now, much more then RFC1522 compatible + ones. In existent variant when you enter 8bit subject with ASCII + body result content charset stays as ASCII, it makes impossible to + recognize Subject: charset. I UNDERSTAND that my method violates + MIME slightly (no 8bit characters must be in header field) but it + is adoptation of real life practic." + + - Add /usr/sbin to default path, there is sendmail on FreeBSD + - Disable national extensions of 'date' + - Add FreeBSD lock directory + - Add KOI8-R charset to user prompt, it is officially registered, + RFC1700 + - Fix utime_buffer declaration for FreeBSD + - Allow mailbox editing for Expert user level only. Total disabling + of this option makes life harder for experienced users and total + enabling cause accidental 'e' hitting damage novice users mailboxes. + It seems that enabling it for Expert user leve is the best solution. + From: Andrey A. Chernov + + - "I also discovered another small 'bug'... Michael Elkins changed + someday in pgp.c a get_address_from() call to a simple strcpy() + which caused my pgp not to find my secret key because of a + leading blank: + " Volker Paulsen " + ^ + After I changed it back to get_address_from() the leading blank + becomes removed and pgp matches the key correctly against its key + ring ." + From: Volker Paulsen + + - Add binary -flag for header_rec. No longer hack for remail.c. + + - Make more clear what part of message was PGP signed or encrypted. + (Why pgp returns exit status 0 even when signature doesn not + match???) + + - Some massive rewriting of copy_message. Now should decoding of + RFC 1522 encoding work correctly when replying (and forwarding). + Side effects: + - Headers which have longer than 32 KB after unfolding + are truncated + - Exact folding is not preserved + + Changes of Elm2.4ME+ PL11 (25) compared to Elm2.4ME+ PL10 (25) + -------------------------------------------------------------- + + - Elm wrote incorrect Content-Length: -header when + updating Status: -header + Noted by: Juergen Obermann + [ That content-length is lightly too big and is not fixed + if it is last mail in folder (it is not big enough that + newmbox.c notices) also incorrent Content-Length: -header + causes in other situations that it is ignored. This causes + problems when "From ":s are not escaped. ] + + Changes of Elm2.4ME+ PL10 (25) compared to Elm2.4ME+ PL9 (25) + ------------------------------------------------------------- + + - CM_DISPLAYING was wrong value in hdrs/defs.h + + - Elm was corrupting folder by adding number to end of mail... + And therefore causes that next mail does not start + with "From " in folder. + Noted by: Volker Paulsen + + - REMAILER code does not compile. + From: Andrey A. Chernov + From: Arkady Altman + + Changes of Elm2.4ME+ PL9 (25) compared to Elm2.4ME+ PL8 (25) + ------------------------------------------------------------ + + - Attach viewer does not completely fill header_rec + so "lines" was uninitiazed, which may cause + "metapager()" to choose randomly whether to use the + built-in pager or not, if you haven't set "pager" to + "builtin" or "internal". Now structure is pre-initialized + completely. "lines" is set to number of bytes / 60 + (not saticfactory, but actually metapager should use + number of lines in decoded message -- + not in original anyway.) + Prompted by: Guy Harris + + Changes of Elm2.4ME+ PL8 (25) compared to Elm2.4ME+ PL7 (25) + ------------------------------------------------------------ + + - In syscall.c, around line 180, the space for p + was allocated but the actual string was never built. + From: Mark Johnson + + - At the end of a message with the buildin pager I + sometimes get the prompt "MORE (you've seen 101%):" + and I can not read the next message with + pressing the space bar. + From: Juergen Obermann + + - "rfc822_reap_comments()" wasn't properly handling + Content-Type: headers like + + Content-Type: Multipart/Mixed; + boundary="--Next_Part(Wed_Feb_14_09:54:49_1996)--" + + because it didn't realize that the ")" in the boundary + was inside a quote, and therefore should be copied rather + than being stripped out. + From: Guy Harris + + - It wasn't RFC 1522-decoding the name in the "From:" header + when it put in the "Forwarded message from" and + "End of forwarded message from" lines. + From: Guy Harris + + - Put defualt disposition for attachment to be 'attachment' + when sending. + + - Check that attachment is not deleted before mail is sent. + + - Because copy_message is now called from metapager, it + needs two more arguments (mailfile and current_header). + I hope that this change breaks nothing! + Detected from report of: Mark Johnson + + Changes of Elm2.4ME+ PL7 (25) compared to Elm2.4ME+ PL6 (25) + ------------------------------------------------------------ + + - Incorrect subject + + Changes of Elm2.4ME+ PL6 (25) compared to Elm2.4ME+ PL5 (25) + ------------------------------------------------------------ + + - Sometimes extra newline was added when message or + body of it was copied. + + > copy_message in fileio.c was gone too hairy. + Tried to reorganize it. I hope that nothing + breaks. + + > Semantic of flag CM_DECODE is now changed. + + > Now message is decoded before replying and forwarding + (excpet when mime_forward is true -- forwarded message + as attachment is not decoded). + + > Well, now [encode] [clear] does not work when metamail + is called. + + > [encode] ... [clear] is now internally handled as + application/X-ELM-encode (in MIME -message) + + - Do RFC 1522 decoding only in MIME messages. + + - builtin++ code in metapager was broken [encode] ... [clear] + + > I simplified metapager. I hope that nothing + breaks. + + > Notice that [encode] ... [clear] is broken when sending + multipart messages. + + - Now also headers of message/rfc822 are prefixed when replying + + - Nicer printing of folded headers (in metapager and rfc822_decode) + + - Letter 'e' (= uses Elm's [encode] ... [clear]) added to status + letter of mailbox menu. + + Changes of Elm2.4ME+ PL5 (25) compared to Elm2.4ME+ PL4 (25) + ------------------------------------------------------------ + + - "nohdrencoding" was in wrong place in list + + - To instead of From in headers was always displayed: + From: Petr Lampa + + - There was some confusion in (char *) versus (unsigned char *) + in rfc1522_encode_text. + Pointed by: Petr Lampa + + - Replaced check of R_OK with ECCESS_EXISTS in can_save + in attach_menu.c + + - Reordering of attach_viewer so metamail is not called + when it isn't needed (new defination of 'pagemultipart' + was broken it.) + + - Message "Parsing MIME structure..." is printed + + - Added to some unlink and elm_chown calls to attach_menu.c + + - Now prints "Executing metamail..." if metamail is called for + display message. + + - Tried fix MIME decoding of headers in copy_message + (well, it still can't work correctly.) + + - MIME (RFC 1522) encoding of In-Reply-To -header + + > Address headers and headers from .elm/elmheaders are still + not encoded. + + - Don't try print To: address in menu if it does not exists. + + Changes of Elm2.4ME+ PL4 (25) compared to Elm2.4ME+ PL3 (25) + ------------------------------------------------------------ + + [ Warning: This is perhaps more buggy than PL 3 (25). ] + + * Sending of simple Text/Plain messages with quoted-printable encoding + was broken when remailer -code was added. + Noted by: Petr Lampa + + - In the index screen Tag all, Delete all and Undelete all + didn't show the new status of the messages. + From: Juergen Obermann + + - In readmsg the Cc-header line is always shown. + From: Juergen Obermann + [ This should be also posted to Elm Development Cordinator ] + + - Add a "Status: RO" line to all outgoing messages (when the "copy" + option is turned on), so that they do not show up as "New" when you + look at a folder. + From: Pete Wenzel + + - The macro X_OK is undefined in some systems. + Use EXECUTE_ACCESS instead. + From: Gunnar Ekolin + + > Failing to include . + Test agaist I_UNISTD instead of ANSI_C. + Noted by: Petr Lampa + + > Now EXECUTE_ACCESS, WRITE_ACCESS, READ_ACCESS, ACCESS_EXISTS, + EDIT_ACCESS is defined in terms of X_OK, W_OK, R_OK, F_OK + from unistd.h (if available) + + - Fix defination in hdrs/mcprt.h to match with lib/mcprt.c + From: Claus Assmann + + - DisplayAddress not work (well, I really not used that patch). + That needs still reconsideration. + From: Claus Assmann + + - Mime (RFC 1522) encoding of following headers: + - From (fullname) + - Subject + - Content-Description + + Encoding of In-Reply-To, To, CC or headers in + .elm/elmheaders is NOT done. + + - New elmrc variable "nohdrencoding". If this variable is TRUE, + then MIME encoding of headers isn't done. + + - Wrong tests with allow_no_encoding in src/mailmsg2.c. + Noted by: Petr Lampa + + - Use strincmp for lib/strincmp.c instead of strcasecmp. + Noted by: Petr Lampa + + - Added check for write errors in mime_encode.c + + Changes of Elm2.4ME+ PL3 (25) compared to Elm2.4ME+ PL2 (25) + ------------------------------------------------------------ + + - Fix to builtin++.c SIGSEGV fix + + Changes of Elm2.4ME+ PL2 (25) compared to Elm2.4ME+ PL1 (25) + ------------------------------------------------------------ + + - Really fix In-Reply-To -header + [ This is also posted to Elm Development Cordinator ] + + Changes of Elm2.4ME+ PL1 (25) compared to Elm2.4ME+ PL0 (25) + ------------------------------------------------------------ + + - SIGSEGV in builtin++.c when looking full headers of mail + and Ctrl-L was hit. + + - Typo in remailer.c (causes that it don't link) + From: Ollivier Robert + + - Ultrix patch from comp.mail.elm + From: John Simonson , + Sydney Weinstein , + Bill Gianopoulos + + - Now Elm's Configure search for metamail -program + + - DisplayAddress: Don't munge sender with get_real_name + + - Elm was generating syntactically incorrect In-Reply-To -headers + + - Because some peoples seems have trashed src/state.c that patch + includes file src/state.c.PL1. + If src/state.c.PL1 and src/state.c don't have identical, + use command: mv src/state.c.PL1 src/state.c + otherwise: rm src/state.c.PL1 + + - SIGSEGV when hitting 'G' in builtin++.c + + + Changes of Elm2.4ME+ PL0 (25) compared to version Elm2.4 PL24 ME8b+ + ------------------------------------------------------------------- + + - Added (security) patch PL24 -> PL25 and + added same kind (security) fixes for MIME code + + - Version number is now formed with new style :-) + + - elmrc -variable 'displaycharset' is propgated to + environment variable $MM_CHARSET when calling metamail. + Environment variable $MM_CHARSET overwrides default value + of 'displaycharset'. + + - New elmrc -variable 'metamail'. So now you not need define + environment variable $NOMETAMAIL, if you want compile Elm + with MIME support, and you don't have metamail. Special values: + none Don't call metamail + metamail Call metamail, if environment variable + $NOMETAMAIL is not defined + /path Call tiss, if this is executable + + - Meaning of elmrc -variable 'pagemultipart' is changed. If + this is FALSE (default), then call metamail if elm can't + handle all subpart of multipart -type or if type is other + than multipart/mixed, multipart/digest or multipart/report. + If 'pagemultipart' is TRUE, then always handle multipart type + without metamail (and treate unknown subtypes as multipart/mixed). + + Decision can Elm handle that type, is done when message is + displayed. So there is now 'M' -letter (= metamail needed) in + multipart mails until message is displayed even when Elm can + show it (that is because that defination is recursive and + requires parsing of structure of message.) + + - Multipart/digest now works correctly when viewing attachment + what we have sending and we have put type for it to be + Multipart/mixed. + + - Configure now tries test if sendmail supports -B8BITMIME and + -BBINARYMIME (unlikely). + + - Makefile now don't compile or install 'filter' because security + problems in it. + + Changes of Elm2.4 PL24 ME8b+ compared to version Elm2.4 PL24 ME8b + ----------------------------------------------------------------- + + - New version of my Reply-To patch included + [ Old version of that patch is posted to Elm Development Cordinator ] + + - Made MIME part 2 (RFC 1522) decoding more accoriding of RFC. + + - Some bugfixes + + Elm mailbox menu's status letters + --------------------------------- + - First letter: D Deleted + E Expired + N New + O Unread + r replied + + - Second letter: C Confidental + U Urgent + P Private (or ...) + A Action + F Form letter + e Uses Elm's [encode] ... [clear] + P PGP message (or ...) + S PGP signed message + K PGP public key + ? MIME message with unknown + MIME version + M MIME message wich requires metamail + or isn't yet parsed + m Pre-MIME message wich requires metamail + + New elmrc variables for Elm 2.4ME+ (compared to Elm 2.4) + -------------------------------------------------------- + + askpgpsig PGP + background-wait-time - + dsn-success MIME / DSN + keeppassfor PGP + lock-in-copy - + metamail MIME + mimeforward MIME + noencoding MIME + nohdrencoding MIME + noheaderfwd - + pagealternative MIME + pagemultipart MIME + pgp PGP + quoteforward - + require-mime-version-for-body-encoding MIME + require-mime-version-for-hdr-encoding MIME + showpgppreamble PGP + usepgppass PGP + + Obsolete elmrc variables + ------------------------- + hpkeypad (alias to keypad) + hpsoftkeys (alias to softkeys) + keypad - + softkeys - + textencoding MIME + bounce (alias to bounceback) + bounceback UUCP + + New elm options for Elm 2.4ME+ (compared to Elm 2.4) + ---------------------------------------------------- + + -w Write .elm/elmrc on startup + -A Attach file + + Obsolete elm options + -------------------- + + -k Enable HP 2622 terminal keyboard\n\r + -K Keypad&softkeys - enable use of softkeys + "-k" + + Options which now takes enumerated values + ----------------------------------------- + + userlevel beginner, intermediate, expert + noencoding pass-7bit, pass-8bit, pass-binary + + Warnings + -------- + + - Not compiled with MMDF or REMAILER support defined, so + probably they don't work (or even compile). + + - As far I know REMAILER code have newer worked. + Specially: remailers are incompatible with MIME + + - This is not tested very much so there can be bad bugs. Bug + Reports to: Kari E. Hurtta + + - Decoding of MIME is not done when searching content of mail + + - RFC 1522: Mime part 2 support is incomplete + + - There is not support for multipart/signed and multipart/encrypted + (specially: no support for PGP/MIME) + + > But anyway, Elm will newer be MIME Conformant + + - HP's shoftkeys are no longer supported + + - DECNET addresses like host::user may be interpreted as phrase + of group syntax or as route of source routed address. + + Collector of these patches + -------------------------- + + Kari E. Hurtta Index: elm2.4.ME+.50/ANNOUNCE.ME *** /tmp/8255-very-long-file-name-NULL-comes-in-here Sun Nov 22 09:14:43 1998 --- elm2.4.ME+.50/ANNOUNCE.ME Tue Nov 17 21:12:53 1998 *************** *** 0 **** --- 1,287 ---- + + This ANNOUNCE.ME is for ME patches of the 2.4 version. + + Look file README.ME+ for the 2.4ME+ version. + + -------------------------------------------------------------------- + + ELM 2.4 PL24 ME8 + + This is an announcement for Patchlevel ME8 for Elm 2.4pl24. + + The most current version of this document is available via the web: + http://www.cs.hmc.edu/~me/elm/announce.html. You can also visit the "ME" home + page at http://www.cs.hmc.edu/~me/elm/me.html. + NOTE THAT THIS ADDRESS HAS CHANGED FROM THE PREVIOUS ANNOUNCEMENT! + + What's new: + + * Anonymous remailer support (plus chaining, PGP encryption & MIME support) + * mime.types file when guessing the content-type of an attachment. + * Forwarded messages can be attached as MESSAGE/RFC822 data + * Supports extraction of PGP public keys directly + * Support for application/pgp type inside multipart/mixed and message/rfc822 + messages + + Changes from ME7a: + + * Replying to MULTIPART messages now works correctly. + * MESSAGE/RFC822 is now properly supported. + * BINARY encoding is better supported. + * Fixed bugs with BASE64 and QUOTED-PRINTABLE encodings. + * Enhanced attachment menu + * Improved interface for PGP encrypting messages + + Where to get it: + + * ftp://cs.hmc.edu/pub/me/elm-2.4pl24me8.tar.Z + * ftp://cs.hmc.edu/pub/me/elm-2.4pl24me8.tar.gz + * ftp://cs.hmc.edu/pub/me/elm-2.4pl24me8.diffs.Z + * ftp://cs.hmc.edu/pub/me/elm-2.4pl24me7a-8.diffs.Z + + -OR- + + * ftp://uiarchive.cso.uiuc.edu/pub/packages/mail/elm/me/elm-2.4pl24me8.tar.Z + * ftp://uiarchive.cso.uiuc.edu/pub/packages/mail/elm/me/elm-2.4pl24me8.tar.gz + * ftp://uiarchive.cso.uiuc.edu/pub/packages/mail/elm/me/elm-2.4pl24me8.diffs.Z + * ftp://uiarchive.cso.uiuc.edu/pub/packages/mail/elm/me/elm-2.4pl24me7a-8.diffs.Z + + ------------------------------------------------------------------------------- + + STATUS + + These patches are not an official release of Elm, and are not supported by the + Elm Development Group. Features included in this patch may or may not be + included in the next official release of Elm. I can say for certain that the + PGP portion will not. I am hoping that my MIME patches will be included in the + 2.5 release, but that is still an open question. Should the 2.5 release appear + without these patches, I plan on "porting" them and using 2.5 as the base. + ------------------------------------------------------------------------------- + And now on to the interesting stuff... + + PGP SUPPORT + + [This work was inspired by pgpelm which was started by Mic the Missing and + furthur developed by Ken Cox (kenc@viewlogic.com). This version is a complete + rewrite of that system but looks very similar.] + + Features include: mark on folder menu showing PGP encrypted messages and public + key blocks, encryption/signing of outgoing messages, decrypting/signature + check, mailing public key blocks, and extracting public key blocks from + messages. Special feature allows you to have ELM keep track of your PGP + passphrase so you only need to enter it a minimal number of times. + + I have written a primer for using PGP with this version of ELM which can be + found at http://www.cs.hmc.edu/~me/elm/pgp.html. It contains everything you + need to know about how to use PGP with ELM. + + MIME SUPPORT + + With this version of Elm, the need for metamail is much less since it can now + handle the MIME encodings and multipart/* messages internally. The only case in + which metamail is called is when there is a content-type which Elm does not + understand (eg. image/*, audio/*). In most cases you will probably only receive + this as attachments anyway, so you might still not require metamail. + + * Incoming Attachments + + Access to individual portions of a MIME mulitpart message can be accessed + via the "Attachments" menu (type 'v' in the index menu). This brings up a + menu listing the attachments for the current message. You then have the + option of viewing or saving the attachment. + + * Outgoing Attachments + + In addition to using the "[include ..." construct when editing your + message, there is now an "Attachments" menu accesible from the "Send" menu + (it is abbreviated to "a)tt" because of space problems). Simply type "a" + to get into the Attachments menu. You may then add/delete/edit/modify + attachments. When possible, Elm will attempt to guess the correct + Content-Type (based on file extension, see the "mime.types" section below) + and Content-Transfer-Encoding. + + * Automatic encoding for outgoing messages. + + If Elm finds 8bit or control characters in text files to be attached, it + will automatically encode them QUOTED-PRINTABLE if Elm has been compiled + without the 8BITMIME option (meaning that your MTA isn't 8bit clean). This + behavior can be overriden with the "noencoding" elmrc option (see below) + in cases where your sendmail is 8bit clean, but doesn't support the + -B8BITMIME option. + + * mime.types file + + Elm now mime.types file which maps filename extensions to Content-Type + field when sending attachments via the new attachment menu. The format of + these files is as follows: + + + + A hash character ('#') may be used as a comment, but it must be in the + first column of a line. + + Here is an example of mime.types file: + + # This is a commented line + html text/html + tgz application/octet-stream; type=tar; conversions="x-gzip" + + Note that the content-type field is not limited to just the type/subtype + part; in can also include suplementary information. + + Elm first looks to see if the user has a ~/.elm/mime.types file. If there + is no match, Elm will then try to read the system types file (typically + /usr/local/lib/elm.mimetypes). Failing that, Elm has a few builtin types + that it understands. + + You might want to pick up a copy of my mime.types file as a starting point + for either your personal or system file at + http://www.cs.hmc.edu/~me/elm/mime.types. + + REMAILER SUPPORT + + This version of Elm has the ability to prepare messages for use with anonymous + remailers (if you don't know what that is, read alt.privacy.anon-server for + more information). + + Elm gets its information about anonymous servers by either fingering + "remailer-list@kiwi.cs.berkeley.edu" or by reading ~/.elm/remailers (if this + file exists, it overrides the "finger" so you can still use this feature even + if you are not directly net connected.) + + To send an anonymous message, first compose your message as you normally would, + entering the name of the person who should ultimately receive the message as + the To: address. Then, at the "send" menu, you can type "r" to get into the + "remailer" menu. + + There are several options in this menu: remail, chain, path and encrypt. + The "remail" option indicates whether or not the current message will be + remailed. + + "chain" indicates how many servers this message should be sent through (the + higher chain is, the more secure, but the tradeoff is how long until the + receiver gets the message). + + "path" is where you can specify which servers you would like this message to go + through (this overrides "chain"). You can specify a semi-colon (";") separated + list of server "nicknames" here. + + "encrypt" means that Elm should encrypt your messages with the PGP public key + of server you are remailing through (note that if chain > 1 or there is more + that one server in "path", Elm will do the "right" thing and call PGP multiple + times for each hop). + + Once the desired options are set, you can quit the menu and send the message. + What you see next will vary... If you are a PGP user and set encrypt=TRUE in + the remailer menu, you will see the call(s) to pgp flash across your screen. + Otherwise you will most likely just see the "Mail sent." message. + + If you have MIME defined, it should be possible for you to include attachments + in your anonymous mail in the normal manner. Most servers these days support + the "hash" option which Elm makes use of to pass the proper MIME headers. + + NEW ELMRC OPTIONS + + * usepgppass = [ ON | OFF ] + Default: usepgppass = OFF + o OFF: Elm does not keep your passphrase in memory. You must enter it + each time you want to decrypt a pgp mail. + o ON: Elm keep your passphrase in memory for "keeppassfor" seconds. + * keeppassfor = + Default: keeppassfor = 0 + o keeppassfor>=0: Elm will keep your passphrase in memory for seconds + after entering the passphrase. + o keeppassfor<0: Elm will keep your passphrase in memory without + clearing it from memory. + * noencoding = + Default: noencoding = 0 + + This control makes it possible to send raw 8bit or binary data when the + MTA doesn't support 8BITMIME and the -B8BITMIME option or BINARYMIME and + the -BBINARYMIME option. + + Possible values: + o 0 : Always convert 8BIT and BINARY data to QUOTED-PRINTABLE if the + MTA doesn't support the -B8BITMIME and -BBINARYMIME options. + o 1 : Allow 8bit without -B8BITMIME, but binary data is encoded + o 2+: Allow binary without -BBINARYMIME and and 8bit without + -B8BITMIME + It is STRONGLY recommended that it be left set to 0 unless you know what + you are doing!! + * showto = [ ON | OFF ] + Default: showto = OFF + o ON: Elm should show you who the message is addressed to rather than + who it is from if the message is not addressed to you. + o OFF: normal behavior + * pagemultipart = [ ON | OFF ] + Default: pagemultipart = ON + o ON: Use the internal MIME support to display MULTIPART messages + rather than calling metamail + o OFF: display MULTIPART messages with metamail + * mimeforward = [ ON | OFF ] + Default: mimeforward = OFF + o ON: When forwarding a message, Elm will attach as a MIME body part of + type MESSAGE/RFC822. + o OFF: Elm will include the forwarded message in your message, just + like it always has. + + BUILTIN++ PAGER + + This pager looks a lot like more, but has some of the movement features of + less, most notably, paging backwards through your messages. Press '?' inside of + the pager for a list of valid keystroke commands. + + TAGS + + Use the keyword "tagged" in the "Delete messages with pattern" (invoked by ^D + in the index menu) to delete all tagged messages. + + You can now use the keyword "all" to tag all visible messages. + + LIMIT FUNCTIONALITY + + * Simple compound statements can be made using '|' (OR) and '&' (AND). + * The "Cc:" field is now availible for search. + * New keyword "tagged" for narrowing to only the currently tagged messages. + + MISCELLANEOUS + + '*' may be used as a wildcard in the "alternatives" list in the elmrc file. For + instance, "me@*.hmc.edu" would me to match the user "me" in all hosts from the + "hmc.edu" domain. + + Elm will keep track of the messages which you have replied to with an 'r' mark + in the index menu. + + Forwarded messages are encapsulated instead of quoted like replies: + + ----- Forwarded message from Joe User ----- + [...] + ----- End of forwarded message from Joe User ----- + + You can effect an automatic Bcc: line by setting the address(es) in the "BCC" + environment variable. Note that this is non-standard, but provided as a means + of accomplishing the task. + + BUGS + + A list of known bugs can be found at + http://www.cs.hmc.edu/~elkins/elm/bugs.html. + + AUTHOR + + Michael Elkins + Home Page: http://www.cs.hmc.edu/~me/index.html + + If you are interested in knowing what my future plans are, you can browse my + "TO DO" list at http://www.cs.hmc.edu/~me/elm/todo.html + + THANKS + + Kari Hurrta has contributed a great deal to the MIME + support (and fixed lots of my stupid bugs ;-) + + I would also like to thank everyone who took the time to report bugs or suggest + improvements! + ------------------------------------------------------------------------------- + Last updated Sept 21, 1995. Index: elm2.4.ME+.50/ChangeLog.ME *** /tmp/8255-very-long-file-name-NULL-comes-in-here Sun Nov 22 09:14:43 1998 --- elm2.4.ME+.50/ChangeLog.ME Tue Nov 17 21:12:53 1998 *************** *** 0 **** --- 1,812 ---- + + This ChangeLog.ME is for ME patches of the 2.4 version. + + Look file README.ME+ for changes of the 2.4ME+ version. + + ------------------------------------------------------------------------- + + ----------------- + Changes from ME8a + ----------------- + When checking to see if the subject line should be centered when displaying + a mail message, the weedlist check should be for "subject:", not "subject". + This case is more general. + + control-F in the main menu now causes Elm to forget your passphrase. Useful + for when you screw up so you don't have to exit Elm before using PGP + again. + + Added a "metoo" argument in pgp_encrypt to decide whether or not the + +encrypttoself=on flag should be passed to pgp. This was needed so that + the value of auto_cc can be overidden when remailing. + + Added support for decoding of RFC1522 style encoded headers. This should + work in the index menu, the pager, and when replying (the "attribution"). + + metapager will now recover if pgp fails by displaying the message "as is" + in the case where the message is only signed. + + Simplified "pager" check in metapager since there are two pagers now. + + Fixed problem in copy_message where the rfc822 newline at the end of the + headers was missing. + + Fixed bug in menu2.c:generic_menu (used by the remailer) which caused + Elm to segfault when not compiled with gcc. + From: Andrey A. Chernov + + Fixed problem where trailing MMDF separator was still being printed. + + The default filename for the Content-Disposition header was not getting + set if the included file did not have an extension. + + text/plain => application/pgp conversion is now NOT done for public key + blocks. + + Configure now figures out where "finger" is instead of hardcoding a path + in remailer.c + + Hopefully fixed definition for sleep() in defs.h. + + Fixed bug where elm would segfault when typing ^E when there are no messages + in the mailbox. + + Fixed bug where a leading space could be inserted before the real name when + looking up a PGP public key from the return address. + + Fixed some of the movement commands in builtin++ so that they work now. + + Convert application/x-pgp-* => application/pgp to decode messages generated + by pgpelm. + + Fixed problem where if the signature was added after editing the message, + it would be written over the message right after the content length. Fixed + by seeking to the end of the file after writing the content length field. + + ---------------- + Changes from ME8 + ---------------- + Fixed bug in conversion of text/plain => application/pgp in newmbox.c which + was causing Elm to segfault. (Thanks to Kari.Hurtta@fmi.fi) + + Fixed bug where builtin++ segfaults when viewing zero length messages. + + ----------------------- + Changes from ME8alpha19 + ----------------------- + Fixed bug where external pager did not work with non-MIME messages when + USE_PGP is defined. + + Cleaned up mime_pgp_decode so that it does not need to look at the + header_rec for the current message any longer. + + Do text/plain => application/pgp conversion in newmbox.c if PGP type data + is discovered in the message. This supports those who are sending PGP + messages but are not compliant with what Elm expects. + + Added the control-E (extract public key) command to showmsg_c.c. + + Security fix from Olaf Kirch which forces Elm to write + the temp spool mailbox in the user's home directory instead of /tmp. This + prevents exploitation of symlinks to create a .rhosts in the user's + directory. + + Fixed bug in show_status where flags from the previous message were not + being properly cleared with USE_PGP was defined and MIME was not. + + ----------------------- + Changes from ME8alpha18 + ----------------------- + added application_decode for support of APPLICATION/* types. Calls + null_decode if the subtype is unknown. + + rfc822_decode now matches the headers against the weedlist when printing + out. You can view all the headers for a message by pressing 'h' while + viewing a message just as you normally would. + + added mime_pgp_decode for inline decoding of application/pgp parts. + application/pgp at the top level is still handled by metapager since some + people will be configured without MIME. I might eventually require that + to use PGP, you must also define MIME, but not yet! + + Status header was printed in the wrong place with BINARY encoding. + From: Kari.Hurtta@fmi.fi + + Added the DONT_ADD_FROM and NEEDS_LONE_PERIOD_ESCAPE checks to write_encoded, + and other MIME text parts. + From: Kari.Hurtta@fmi.fi + + Added a "type_opts" field in mime_send_t. + + mime_write_header now just writes the MIME headers. The code to add the + boundary for multipart messages and the charset for text/plain was moved into + mailmsg2.c. + + Changed pgp code to set MIME fields to use add_parameter and type_opts instead + of directly writing to subtype. + + Made some changes in the Content-Type for PGP messages to be more compliant + with the unofficial standard. + + ----------------------- + Changes from ME8alpha17 + ----------------------- + MIME header code in write_header_info() was moved to + mime_write_header in mime_encode.c. This was done to greatly simplify the + code, and to make it possible for me to send MIME messages with the + remailer code. This will also make it possible for me to have PGP + encrypted MIME messages in the future. + + Content-Length headers are generated regardless of whether or not MIME + is defined. + + Lots of changes to the remailer code to make it work with MIME. + + ----------------------- + Changes from ME8alpha16 + ----------------------- + Removed Content-Length header from attachments included by the "[include ...]" + statement. + + Made the existing global variables that set the toplevel Content-Length header + a part of mime_send_t. This has the side effect that if MIME is not + defined, no Content-Length header will be added on outgoing messages. + + Commented out the code which reads the content-length header from attachments + in mime_parse.c. It should now ignore that header and figure out the length + in multipart_parse(). + + ----------------------- + Changes from ME8alpha15 + ----------------------- + New option "mimeforward" controls whether or not forwarded messages are + sent as MESSAGE/RFC822 MIME messages, or are sent in the old non-MIME + way. + + - Fixes for my mistakes in alpha15. + - text_unsupported_decode() added for printing warning message about + unsupported text subtypes. + From: Kari.Hurtta@fmi.fi + + Changed disposition handing such that a message saying that the user should + type 'v' to view an attachment which is declared something other than + "inline". + + ----------------------- + Changes from ME8alpha14 + ----------------------- + Make sure that arguments to fseek(), etc., are type long. + From: Kari.Hurtta@fmi.fi + + Added some general access commands for attachments in the attach_menu.c + routines. + From: Kari.Hurtta@fmi.fi + + The disposition is now displayed in the little mini-header. This looks nicer + than the previous message about "[Content-Disposition: attachment, skipping]". + + select_CT_decoder now returns text_decode for all TEXT types. My reasoning + here is that since the MIME spec says that all unknown subtypes of TEXT + should be treated as TEXT/PLAIN so long as the charset is understood, then + all TEXT can be treated in this manner. + + Made sure that the pgp field of header_rec gets set to zero when initializing + in attach_viewer and also in newmbox.c. + + Removed lots of unused variables. + + Added some parentheses in places to make "gcc -Wparentheses" happy. + + ----------------------- + Changes from ME8alpha13 + ----------------------- + Forgot to initialize usepass=FALSE in pgp_decrypt_init. This caused elm + to hang when trying to decode a signed message while pgp was waiting for + the passphrase... + + The attachment menu should now work properly for viewing message/rfc822 + messages with the builtin pager. + + When attaching a message/rfc822 with the "forward" command, you should + now be able to perform all the actions in attach_menu() as for a "normal" + attachment. + + ----------------------- + Changes from ME8alpha12 + ----------------------- + Tried to fix MIME_NOTPLAIN code again. This time MESSAGE/RFC822 will not + have it set, which MULTIPART will if pagemultipart=OFF. + + Fixed attach_menu so that Elm does not segfault when you try to do an + operation on an attachment when there are no attachments defined. + + Removed the MIME_NOTPLAIN flag and made it a one-bit flag in mime_t instead. + Removed PGP flags from defs.h and use a 3-bit field in header_rec instead. + These two now ensure that the "status" field of header_rec is always less + than 65536 so that 16bit integer machines will not have troubles. + + Elm now pays attention to what the MIME-Version header says and passes any + messages with an unknown MIME-Version header value to metamail, or displays + an error if $NOMETAMAIL is set. + From: Kari.Hurtta@fmi.fi + + ----------------------- + Changes from ME8alpha11 + ----------------------- + pgp_call() now uses the expanded_* address lists so that aliases are + expanded before trying to look up the ids with PGP. + + Fixed calls in mailmsg2.c with wrong number of arguments (botched job + including Kari Hurtta's patch). + From: Kari.Hurtta@fmi.fi + + Reordered the preMIME code in newmbox.c to make it a little more efficient. + + Fixed bug in write_header_info() where a spurious newline was being printed + after the Content-Type header for TEXT. This was causing the remaining + fields not to be a part of the message header. + + Removed check for old-style RFCXXX Mime-Version header. I don't believe + anybody uses that anymore since RFC1341 has been out so long. + + pgp_status needed to be reset when the user aborts sending a message, + otherwise you can't encrypt another message until you successfully send + another message. + + MIME_NOTPLAIN is not set for multipart messages if "pagemultipart" is TRUE. + + Cleaned up code for deciding if metamail mail should be called to display + a message in showmsg.c. Created a need routine, need_meta(), for this + purpose. + + ----------------------- + Changes from ME8alpha10 + ----------------------- + Fixes to remove some global MIME variables used for composing. + From: Kari.Hurtta@fmi.fi + + Added a redraw command to the attachment menu. + + Fixed the code in newmbox.c where MIME_NOTPLAIN was not set for non-TEXT + content-types. Also, check for "plain" should be "text" for old-style + pre-MIME content-type headers. + + Added a print command to the attachment menu. + + Made the remailer support a Configure option. + + ---------------------- + Changes from ME8alpha9 + ---------------------- + Use Elm's error routines in the attachment menu. + Protected iscntrl() more calling isascii() before it in builtin++. + From: Kari Hurtta + + Added bit of code lost from old showmsg.c which centers the subject line + after the title line in the case where "subject" is in the weedlist. + + Viewing attachments with metamail while in send mode now works. It required + different arguments to be passed to metamail. + + ----------------------- + Changes from ME8alpha8a + ----------------------- + mime_t_clear() now sets the defaults of TEXT/PLAIN, ENCODING_7BIT, DISP_INLINE, + when called. + + More fixes from Kari Hurtta + + Removed padding bitfield from struct mimeinfo since it isn't needed. + + Added +clearsig=ON when adding a PGP signature to a message. + + Added prototypes for the functions defined in pgp.c. + + Elm will now allow you to edit the list of usernames to encrypt to when + using PGP. This fixes the problem where sometimes the return address + doesn't match the address on the key. However, it still requires that + the user know this. + + Implemented Kari Hurtta's suggestion of using a "state" variable for the + MIME output when decoding messages. This will make it easier to add new + extensions for output. The state information is defined in hdrs/state.h + + The user interface for the attachment menu now is more consistent with + the mailbox menu. (Kari.Hurtta@fmi.fi) + + User is now asked for confirmation upon deleting an attachment from the + attachment menu if "ask=ON" in the elmrc. + + Fixed replying to messages with PGP signatures. They were getting sucked + into the new MIME code and getting skipped as an unknown type. + + Added new procedure to allow for extraction of PGP public keys directly + instead of relying on metamail. + + If Elm can't find a PGP key that it is looking for, it now tells the user + which key instead of just that there was an error. + + The user is now allowed to re-edit the filename in attach_modify() if the + file was not found. This also allows a user to change the file associated + with an attachment. + + Fixed very old bug of MMDF separators being printed when displaying a + message. + + ---------------------- + Changes from ME8alpha8 + ---------------------- + Fixed problem where you couldn't edit the default filename in attach_save(). + + ---------------------- + Changes from ME8alpha7 + ---------------------- + Use safe_malloc() and safe_realloc() instead. (Kari Hurtta) + + Fixed bug where no EOLN token was printed after the Content-Type header + when type_opts was NULL in attach_generate_message(). + + Reorganized struct mimeinfo so that bitfields are next to each other and + added padding to make them 1 byte long total. + + ----------------------- + Changes from ME8alpha6+ + ----------------------- + Reduced memory usage in "struct mimeinfo" by using bitfields or pointers + instead of predefined strings when possible. Also removed the "comments" + fields. + + New routine lib/strmcpy.c for doing malloc/realloc and strcpy in one shot. + Helps out with the above objective. + + The attachment menu now defaults to disposition "inline" when sending + mail. + + Reorganized the screen in the attachment "configuration" menu. + + Removed code for adding "name=" field to APPLICATION/OCTET-STREAM types. + Uses Kari's code for Content-Disposition "filename" instead. attach_menu() + still supports both for getting default filename when saving an + attachment. + + Removed MIME_MULTIPART flag. Use MIME_TYPE_MULTIPART with "type" instead. + + Removed "encoding" field from "header_rec". Use "encoding" in "mimeinfo" + instead (this was already mostly true, but there were a few cases left + over from the early support for encoding...) + + ---------------------- + Changes from ME8alpha6 + ---------------------- + Incorporated changes from Kari Hurtta . + Big changes are the use of Content-Disposition and routines to better parse + rfc822 headers. Also fixed multi-line MIME header problem in newmbox.c + + ---------------------- + Changes from ME8alpha5 + ---------------------- + Fixed places where "/tmp" was hardcoded for temporary files instead of using + "temp_dir" in the ME extensions. + + Fixed more isspace() problems with "signed char". + + Made some formatting improvements in multipart_decode(). + + Fixed bug where multi-line Content-Type headers were not properly parsed in + newmbox.c. + + Added builtin++2.c which contains a tempfile based pager which may become + part of the "official" distribution. For right now it is "experimental". + + ---------------------- + Changes from ME8alpha4 + ---------------------- + Fixed attach_modify() to use ->type and ->subtype like the rest of the + attachment routines. + + Viewing an attachment while sending didn't work because it read from + mailfile... Now it can view external files as well. + + Fixed problem with signed char and iscntrl() in AddLineToCache(). + This fixes the problem with umlaut characters in some sets. + From: guy@netapp.com. + + Modified mime_generate_boundary() to use mime_count as a counter so that + subsequent calls will generate different values. + + Elm will now read ~/.elm/mime.types and $libdir/elm.mimetypes (system + mime.types file), if they exist, to map filename extensions to + Content-Type values when sending attachments. The format of this file + is: + + There must be no line breaks in the Content-Type field. + Example: + tar application/octet-stream; type=tar + for .tar files. + NOTE: data which is declared "application/octet-stream" will automatically + get the option "name=" added to it, so there is no + need to put it in the definition. + + Modified attach_info() so that it uses BASE64 encoding for "application" + types that have 8bit or control chars in them. + Also, "image" automatically defaults to BASE64 without reading the file. + + attach_save() now uses the "name=" field of "application" + types as the default filename. + + Added '+' as a synonym for NEXT_PAGE in builtin++ per popular request. + + attach_generate_msg() no longer puts Content-Length headers on parts that + it generates. + + ---------------------- + Changes from ME8alpha3 + ---------------------- + Added support for multi-line Content-Type: headers in read_headers() and + mime_read_headers(). + + me.h now contains prototypes for functions defined by this set of patches. + + Fixed some problems with using PGP and remailers. + + Fixed parsing of options in the Content-Type header so that semicolons + inside quotes are protected. + + If MIME is defined, forwarded messages will be an attachment of type + MESSAGE/RFC822. + + Fixed problem with viewing MULTIPART or MESSAGE/RFC822 attachments from + the attachment menu. + + Replaced MIME_BOUNDARY with a function (mime_generate_boundary()) that + sets the global variable "mime_boundary" to generate unique multipart + boundaries. This is important now that Elm will be sending out + MESSAGE/RFC822 data which Elm might have originally generated... + + ---------------------- + Changes from ME8alpha2 + ---------------------- + Added a second line for menu items in verify_transmission() since adding + the "r)emailer" option didn't fit on the first line. + + Modified the PGP code a little for more flexibility. You can now specify + a string of "ids" to encrypt with in pgp_call(). + + --------------------- + Changes from ME8alpha + --------------------- + The menus from attach.c and attach_send.c were combined into a single + menu in attach_menu.c which servers both purposes. The difference is + whether or not the user is allowed to modify the given attachments. + + The rest of attach_send.c was renamed as mime_encode.c. + + ----------------- + Changes from ME7a + ----------------- + + Added a check in quoted_printable_decode() to make sure that the index + into outbuf[] does not run off the end and cause problems. + + attach_edit() didn't return TRUE after editing in one spot, which causes + problems on some platforms, namely Solaris. + From: Martin Walter + + base64_decode() was broken because it would fread() a data chunk, and + wasn't always a multiple of 4 bytes, causing the decoding to break. + + Changed char => unsigned char in most places in *_encode() to fix problems + with ISO characters >127. + + Added message about PGP support for the 'elm -v' command in args.c. + + Support for anonymous remailers. [ALPHA] + + Major rewrite of the MIME handling code. MULTIPART and MESSAGE/RFC822 + messages are parsed the first time they are accessed (eg. viewed/replied) + and then are ready to use without parsing after that. See mime_parse.c + and mime_decode.c for details. + + From: Kari Hurtta + Put a loop around verify_transmission() so that Elm will no longer just + be lame when an error occurs in [include ...] statements. (Way cool! -Ed) + + Don't allow user read or writes files as attachmentds which he don't heve + access. Notice that elm is (usually) setgid mail. + + Consider if MTA supports 8BITMIME also when encoding attachments. + + Now mail is encoded with quoted-printable if it includes control characters. + 8BITMIME doesn't help in this case. + + Attachments are checked for 8bit characters, then content-transfer-encoding + 7bit is not allowed. + + If any of subpart (also attachments) have content-tarnsfer-encoding 8bit + or binary it propagates to content-transfer-encoding of top multipart -type. + + Now taken note of MIME's canonical encoding model: + - End of line in text -subtypes when they have encoded with BASE64 + must be encoded as CR LF. + - Also CR and LF must be encoded in quoted-printable-encoding if + type have binary data (ie. not line oriented). + - Done mapping CR LF -> LF when decoding text -subtypes. + + These content-transfer-encoding checks for attachments are also done + for [include ...] keywords when sending. + + Now [include ...] also don't need metamail's help (they use same routines + than attachments menu). + + Added configuration file option 'noencoding' (see elmrc-info for details + on what this does). + + ---------------- + Changes from ME7 + ---------------- + + Forgot to kill the newline from the PGP output before putting it in to the + builtin++ pager. This resulted in messages being displayed double-spaced. + Ick! + + ---------------- + Changes from ME6 + ---------------- + + attach_send.c needs on some machines. + + Fixed initialization of arrays in attach_send.c because CC isn't quite + ANSI compliant. You can only initialize static arrays. + From: Guy Harris + + Added support for sendmail 8.7 which has support for 8BIT content + encoding. Uses the -B8BITMIME option. + From: Kari Hurtta + + Fixed problem where QUOTED-PRINTABLE messages were not decoded when + replying if USE_PGP was not defined. + From: Kari Hurtta + + [keyword ...] constructs are no longer broken when using MIME and the + message is to be encoded QUOTED-PRINTABLE. + From: Kari Hurtta + + Made the QUOTED-PRINTABLE encoding routine more compliant with the + standard. + From: Kari Hurtta + + Added patch to mark messages NEW in all folders (not just the SPOOL) + if there is no status: line. + NOTE: This will probably go away when 2.5 is released. Instead, they + will use some sort of "magic" flag to accomplish the same thing. + + Added check to make sure that we don't run off the end of outbuf[] in + base64_decode() in mime.c. + From: Kurt.Jaeger@rus.uni-stuttgart.de + + Fixed type casting problem with BSD 4.4 derivatives in mime.c. + From: Andrey A. Chernov + + KOI8-R added as a valid compat charset. + From: Andrey A. Chernov + + builtin++ now wraps the line at the number of columns on the screen, not + one character before. NewLine() apparently "does the right thing" and + doesn't print an extra line if the last character is in the last column. + From: Kari Hurtta + + The arrow keys now do what they did before instead of controlling the + bultin++ pager. + + builtin++ always clears the screen before displaying the next (or + previous) page. + + The length of a line in the cache for builtin++ is increased to 150 + chars because the old value of 128 didn't work with hpterms with a + width of 134 chars. + + Fixed problem where Elm would SIGBUS if you tried to use the + attachments menu in an empty folder. + From: Kari Hurtta + + buf[STRING] was too short in AttachmentMenu() and caused certain Elm to + bomb out on certain messages which had header lines which were too + long. + + AddLineToCache() can now handle multiline input. This meant that + calls to it from metapager had to remove the trailing newlines. + + Changed the MIME mini-header format in metapager(). + + attach_edit() looks more like header_edit() now. Makes the interface + more consistent. + + lib/getrealname.c needed on older BSD systems. + + Fixed nasty bug where the blank line in between the headers and the body + of a NEW message would get deleted when resync'ing in the spool folder. + + Fixed infinite loop in Configure when run with the -d option. + + ---------------- + Changes from ME5 + ---------------- + + Added the DoAll() call so that the keyword "all" can be used when tagging + or deleting. + + Decoupled PGP and MIME support such that you don't need MIME to use PGP. + + Added lots of error checking code in metapager.c. + From: guy@netapp.com (Guy Harris). + + Fixed bug in quoted_printable_decode() where I did not convert read + characters to base16. + + Added hex() function to protect index_hex from bad values. + + metapager() prompts for a command if necessary after running the external + pager, just as Elm used to do. + + Fixed the From: line in write_header_info() where I broke it before. + + Fixed bug where the MIME multipart boundary would be written after the + MMDF separator when copy=ON and the message was being copied to the user's + desired folder. + + Added an "Attachments" menu in verify_transmission() for including + attachments. This works in addition to the current method of using + "[include ..." + + Elm now directly supports writing of BASE64 and QUOTED-PRINTABLE + encodings. + + Changed the return value of match_address() to be zero, since it + returns an int, and NULL is not always an int. + From: Michael Shields + + AttachViewer() now uses notplain() to determine whether or not + metamail is needed to view the message. + From: Guy Harris + + needs_mmdecode() now takes the integer version of the encoding rather + than a string. This avoids parsing the string twice. + From: Guy Harris + + Elm now checks outgoing messages to see if there are any non-printable + characters, and if so encodes the message QUOTED-PRINTABLE. + + Cleaned up PGP code by moving more code to pgp.c so that other files + require less modification (ie. the diffs will be simpler). + + Fixed showmsg.c so that it passes anything that doesn't require metamail + to the metapager. This fixed the problem where messages with no + Mime-Version header but had a Content-Type header wouldn't get displayed. + From: Guy Harris + + Fixed bug in dispaddr.c so that if a message is from the user, it will + always use the TO address to display. + + Moved get_real_name() to its own file in lib so it can be used by + dispaddr.c. + + Modified get_real_name() so that if it sees an address like + "" it will return "elkins@aero.org" instead of a zero + length string. + + ---------------- + Changes from ME4 + ---------------- + + Introduced the "metapager". It replaces most of the code in showmsg.c. + It also takes care of all the MIME and PGP functions the bultin++ pager + used to perform, so that the user doesn't have to use builtin++ in those + cases. + + PGP decrypting function was rewritten using pipes so no temp files are + required! + + The attachment handling code was modified to handle multiline content-type + headers generate by such mailers as zmailer. It also looks through each + modifier in the line intead of just assuming that the boundary field will + be the first. + From: guy@netapp.com (Guy Harris) + + The "command" status line in the builtin++ pager now floats around instead + of always being on the bottom. This was done to facilitate the ^L in + the pager so it didn't look stupid. + + Moved the MIME QUOTED-PRINTABLE and BASE64 decoding code to mime.c so it + can be shared. + + Fixed copy_message() so that it will automatically decode MIME B64 and QP + encodings. This means that replies to these types of messages do what the + user will expect. + + ---------------- + Changes from ME3 + ---------------- + + Fixed problems in builtin++.c and pgp.c where I used the return value + from sprintf() as a char* when in SYSV systems it returns an int. + This caused Elm to seg fault while decrypting messages. + + Configure now asks about PGP support and automatically sets up + config.h. Thanks to Ken Cox (kenc@viewlogic.com) for additional help. + + PGP messages are now automatically decrypted when replying if + usepgppass = ON. + + Back to using the application/pgp format content-type. Use + format=keys-only for send a public key block. + + PGPCall() now uses the pgp option +encrypttoself=on instead of tacking + the username on the end of the TO list. + + Fixed problem where any messages after sending a PGP public key had + the key text included before the included message text. + + Use rename() if availible instead of copying the file in PGPCall(). + Otherwise use link(), unlink(). This should be a little more + efficient. + + pgp is now called with the -w (wipe) option when encrypting. + + builtin++ can now handle text which is QUOTED-PRINTABLE or BASE64 + encoded. + + Fixed bug where the last line of messages was not being printed. + + Elm will no longer seg fault if you hit return without entering your + PGP passphrase. + + Discovered the +batchmode option for pgp. This makes dealing with + incorrect passphrases much easier! + + Fixed bug in builtin++ where the last line of pgp messages would be + printed more than once while at the end of the message under certain + circumstances. + + Changed notplain() in mime.c to support message/* content-type. + + Cleaned up the multipart handling by removing the boundary string from + header_rec to conserve memory. This is now handled by builtin++ and + AttachMenu(). + + builtin++ can now handle multipart/digest correctly. + + SaveAttach() no longer needs to call metamail to decode QUOTED-PRINTABLE + or BASE64 encodings. + + ----------------- + Changes from ME2a + ----------------- + + Elm will now keep track of your PGP passphrase for your RSA secret key. + To use this feature, add the line "usepgppass = ON" to your .elmrc. + There is also a timeout feature which will cause Elm to "forget" your + passphrase after so many seconds if you don't use it. The default is + 300 seconds, but this can be changes by setting the variable + "keeppassfor" in your .elmrc. + + The builtin++ pager was completely rewritten (again) so that it can now + handle normal messages, MIME multipart messages, automatic decryption + and automatic signature checking for PGP messages. It also seems to be + lacking some of the memory allocation bugs that the previous pager had. + + Fixed a bug in DynamicArray() which causes DestroyDynamicArray() to + seg fault if exactly max+n elements were allocated. (weo@recco.de) + + ---------------- + Changes from ME2 + ---------------- + + It is now possible to delete all "tagged" messages by entering "tagged" + as the pattern to match after typing -d (^D). + + fixed a bug in the limit command where you could not "un"limit the + display of headers. Thanks to guy@netapp.com (Guy Harris) for + pointing it out! + + fixed a bug in is_multipart() where the "boundary" token match was + case-sensitive. Index: elm2.4.ME+.50/Changes *** elm2.4.25/Changes Sun Dec 20 07:19:11 1992 --- elm2.4.ME+.50/Changes Tue Nov 17 21:12:53 1998 *************** *** 1,3 **** --- 1,9 ---- + + This Changes is for the 2.4 version. + + Look file README.ME+ for changes of the 2.4ME+ version. + + -------------------------------------------------------------------------- This file describes major changes made to the Elm source distribution between Releases 2.3 and 2.4. Please remember that Elm is a volunteer project, we did what we had time for, and perhaps, what interested us. Index: elm2.4.ME+.50/Configure *** elm2.4.25/Configure Mon May 30 19:55:55 1994 --- elm2.4.ME+.50/Configure Tue Nov 17 21:14:08 1998 *************** *** 8,21 **** # and edit it to reflect your system. Some packages may include samples # of config.h for certain machines, so you might look for one of those.) # ! # $Header: /home/syd/elm.rel/RCS/Configure,v 5.38 1994/05/30 16:55:46 syd Exp $ # - # Yes, you may rip this off to use in other distribution packages. - # (Note: this Configure script was generated automatically. Rather than - # working with this copy of Configure, you may wish to get metaconfig.) : sanity checks ! PATH=".:/bin:/usr/bin:/usr/local/bin:/usr/ucb:/usr/local:/usr/lbin:/etc:/usr/new:/usr/new/bin:/usr/nbin:$PATH" export PATH || (echo "OOPS, this isn't sh. Desperation time. I will feed myself to sh."; sh $0; kill $$) if test ! -t 0; then --- 8,18 ---- # and edit it to reflect your system. Some packages may include samples # of config.h for certain machines, so you might look for one of those.) # ! # $Header: /data/cvsroot/mail/elmme+/Configure,v 1.10 1998/11/17 18:55:48 hurtta Exp $ # : sanity checks ! PATH="$PATH:.:/bin:/usr/bin:/usr/sbin:/usr/local/bin:/usr/ucb:/usr/local:/usr/lbin:/etc:/usr/new:/usr/new/bin:/usr/nbin" export PATH || (echo "OOPS, this isn't sh. Desperation time. I will feed myself to sh."; sh $0; kill $$) if test ! -t 0; then *************** *** 34,43 **** cd UU fi ! case "$1" in ! -d) shift; fastread='yes';; esac Log='' Header='' bin='' --- 31,70 ---- cd UU fi ! fastread='' ! extractsh='' ! ! while test $# -gt 0; do ! case "$1" in ! -d) shift; fastread='yes';; ! -S) shift; extractsh=true;; ! -h) shift; error=true;; ! -*) echo "Configure: unknown option $1" >&2; shift; error=true;; ! *) break;; ! esac ! done ! ! case "$error" in ! true) ! cat >&2 << EOM ! Usage: Configure [-dS] ! -d : use defaults for all answers. ! -h : print this help message and exit (with an error status). ! -S : perform variable substitutions on all .SH files ! EOM ! exit 1 ! ;; ! esac ! ! case "$extractsh" in ! true) ! cd .. ! . ConfTool/Substitute ! exit 0 ! ;; esac + Log='' Header='' bin='' *************** *** 46,55 **** cppminus='' d_getopt='' d_memcpy='' - d_mkdir='' d_rename='' d_symlink='' d_whoami='' n='' c='' orderlib='' --- 73,84 ---- cppminus='' d_getopt='' d_memcpy='' d_rename='' d_symlink='' d_whoami='' + d_poll='' + d_inttypes='' + d_nointtypes='' n='' c='' orderlib='' *************** *** 56,66 **** ranlib='' package='' pager='' - prefshell='' startsh='' - d_eunice='' define='' - eunicefix='' loclist='' expr='' sed='' --- 85,92 ---- *************** *** 71,77 **** cp='' tail='' tr='' - mkdir='' sort='' uniq='' grep='' --- 97,102 ---- *************** *** 138,144 **** d_flock='' d_dotlock='' d_fcntlock='' - lock_dir='' has_flock='' has_fcntl='' d_ftruncate='' --- 163,168 ---- *************** *** 160,166 **** d_mallocvoid='' d_mboxedit='' d_mime='' ! defencoding='' defcharset='' defdispcharset='' d_mmdf='' --- 184,194 ---- d_mallocvoid='' d_mboxedit='' d_mime='' ! metamail_path='' ! metamail='' ! d_8bitmime='' ! d_binarymime='' ! d_dsn='' defcharset='' defdispcharset='' d_mmdf='' *************** *** 190,195 **** --- 218,224 ---- d_strftime='' d_strings='' d_pwdinsys='' + d_waitpid='' strings='' includepath='' d_strstr='' *************** *** 200,206 **** tempnamc='' d_termio='' d_termios='' - d_useembed='' d_utimbuf='' d_vfork='' defbatsub='' --- 229,234 ---- *************** *** 211,217 **** mydomain='' autohostname='' i_memory='' - i_stdarg='' i_stdlib='' i_time='' i_systime='' --- 239,244 ---- *************** *** 248,258 **** ldflags='' cc='' libs='' nametype='' d_passnames='' d_berknames='' d_usgnames='' ! passcat='' rmttape='' roff='' roffopts='' --- 275,286 ---- ldflags='' cc='' libs='' + noaddlib='' nametype='' d_passnames='' d_berknames='' d_usgnames='' ! # passcat='' rmttape='' roff='' roffopts='' *************** *** 267,272 **** --- 295,306 ---- xenlf='' d_xenix='' d_bsd='' + d_pgp='' + pgp='' + pgp_path='' + d_remailer='' + finger='' + finger_path='' CONFIG='' : set package name package=elm2 *************** *** 273,279 **** echo " " echo "Beginning of configuration questions for $package kit." - : Eunice requires " " instead of "", can you believe it echo " " define='define' --- 307,312 ---- *************** *** 285,299 **** rmlist='kit[1-9]isdone kit[1-9][0-9]isdone' trap 'echo " "; rm -f $rmlist; exit 1' 1 2 3 - : We must find out about Eunice early - eunicefix=':' - if test -f /etc/unixtovms; then - eunicefix=/etc/unixtovms - fi - if test -f /etc/unixtovms.exe; then - eunicefix=/etc/unixtovms.exe - fi - : Now test for existence of everything in MANIFEST echo "First let's make sure your kit is complete. Checking..." --- 318,323 ---- *************** *** 339,345 **** echo "Found bin directory" fi ! libswanted="intl nls" attrlist="$attrlist sgi" : some greps do not return status, grrr. --- 363,370 ---- echo "Found bin directory" fi ! # libswanted="intl nls" ! libswanted="sun intl nls c_s" attrlist="$attrlist sgi" : some greps do not return status, grrr. *************** *** 374,380 **** echo "Okay, let's see if #! works on this system..." e=/bin/echo; test -f $e || e=/usr/bin/echo echo "#!$e hi" > try - $eunicefix try chmod +x try ./try > today if $contains hi today >/dev/null 2>&1; then --- 399,404 ---- *************** *** 382,388 **** sharpbang='#!' else echo "#! $e hi" > try - $eunicefix try chmod +x try ./try > today if test -s today; then --- 406,411 ---- *************** *** 398,404 **** shsharp=false echo "exec grep -v '^#'" >spitshell chmod +x spitshell - $eunicefix spitshell spitshell=`pwd`/spitshell echo "I presume that if # doesn't work, #! won't work either!" sharpbang=': use ' --- 421,426 ---- *************** *** 416,422 **** EOSS chmod +x try - $eunicefix try if ./try; then echo "Yup, it does." else --- 438,443 ---- *************** *** 553,562 **** if test -f \$dir/\$thing; then echo \$dir/\$thing exit 0 - elif test -f \$dir/\$thing.exe; then - : on Eunice apparently - echo \$dir/\$thing - exit 0 fi ;; esac --- 574,579 ---- *************** *** 565,571 **** exit 1 EOSC chmod +x loc - $eunicefix loc loclist=" cat chgrp --- 582,587 ---- *************** *** 591,596 **** --- 607,613 ---- date emacs execmail + finger ispell line lint *************** *** 597,607 **** --- 614,626 ---- lp lpr mailx + metamail mips more nroff pack pg + pgp pmake pr rmail *************** *** 705,711 **** : see if date exists, if so, init c_date if $test ! "x$date" = "x" ; then ! c_date=`LANGUAGE= $date` else c_date='date was unavailable at configuration' fi --- 724,730 ---- : see if date exists, if so, init c_date if $test ! "x$date" = "x" ; then ! c_date=`LANG= LC_TIME= $date` else c_date='date was unavailable at configuration' fi *************** *** 814,819 **** --- 833,844 ---- echo exit 1 >bsd echo exit 0 >usg echo exit 1 >v7 + elif $test "$uname_os" = "IRIX" ; then + : test `echo abc | tr a-z A-Z` = Abc does not work in IRIX 6.2 + echo "Looks like an IRIX system, but we'll see.." + echo exit 1 >bsd + echo exit 0 >usg + echo exit 1 >v7 else echo "Looks kind of like a BSD system, but we'll see..." echo exit 0 >bsd *************** *** 826,847 **** echo exit 1 >usg echo exit 0 >v7 fi ! case "$eunicefix" in ! *unixtovms*) ! cat <<'EOI' ! There is, however, a strange, musty smell in the air that reminds me of ! something...hmm...yes...I've got it...there's a VMS nearby, or I'm a Blit. ! EOI ! echo "exit 0" >eunice ! d_eunice="$define" ! ;; ! *) ! echo " " ! echo "Congratulations. You aren't running Eunice." ! d_eunice="$undef" ! echo "exit 1" >eunice ! ;; ! esac if test -f /xenix; then echo "Actually, this looks more like a XENIX system..." echo "exit 0" >xenix --- 851,857 ---- echo exit 1 >usg echo exit 0 >v7 fi ! if test -f /xenix; then echo "Actually, this looks more like a XENIX system..." echo "exit 0" >xenix *************** *** 851,857 **** echo "exit 1" >xenix fi chmod +x xenix - $eunicefix xenix if test -f /venix; then echo "Actually, this looks more like a VENIX system..." echo "exit 0" >venix --- 861,866 ---- *************** *** 864,873 **** fi echo "exit 1" >venix fi ! chmod +x bsd usg v7 eunice venix ! $eunicefix bsd usg v7 eunice venix rm -rf foo ! rmlist="$rmlist bsd usg v7 eunice venix xenix" : find out how to find out full name echo " " --- 873,881 ---- fi echo "exit 1" >venix fi ! chmod +x bsd usg v7 venix rm -rf foo ! rmlist="$rmlist bsd usg v7 venix xenix" : find out how to find out full name echo " " *************** *** 881,886 **** --- 889,896 ---- dflt=y elif xenix; then dflt=y + elif $test "$uname_os" = "IRIX"; then + dflt=y else dflt=n fi *************** *** 934,967 **** ;; esac - : see if we have to deal with yellow pages, if so, put sun - : library first, as the yp password routines must override - : the c library ones - if $test -d /usr/etc/yp || $test -d /etc/yp; then - if $test "$passcat" = "ypcat passwd"; then - dflt=y - elif $contains '^\+:' /etc/passwd; then - dflt=y - else - dflt=n - fi - rp="Are you getting the passwd file via yellow pages? [$dflt]" - $echo $n "$rp $c" - . myread - case "$ans" in - y*) passcat='ypcat passwd' - case "$libswanted" in - '') libswanted='sun c_s';; - *) libswanted=`$echo "$libswanted" | $sed -e 's/sun //g'` - libswanted="sun $libswanted";; - esac - ;; - *) passcat='cat /etc/passwd';; - esac - else - passcat='cat /etc/passwd' - fi - : set up shell script to do ~ expansion cat >filexp </dev/null >libc.tmp ! $sed -n -e 's/^.* [ATDS] *[_.]*//p' -e 's/^.* [ATDS] //p' libc.list if $contains '^printf$' libc.list >/dev/null 2>&1; then echo done elif $sed -n -e 's/^__*//' -e 's/^\([a-zA-Z_0-9$]*\).*xtern.*/\1/p' \ --- 1939,1958 ---- set `echo $libc $libnames | tr ' ' '\012' | sort | uniq` $echo $n "Extracting names from $* for later perusal...$c" nm $nm_opts $* 2>/dev/null >libc.tmp ! if $test `uname` = "AIX" ! then ! cat libc.tmp |egrep '^\..* *T'|egrep -v '^\._|Srestf|Ssavef|^\.\$'|\ ! $sed -e 's/\.\(.* \)T.*/\1/' -e 's/ *//g' >libc.list ! ! if $test -f /lib/syscalls.exp; then ! echo " " ! echo "Also extracting names from /lib/syscalls.exp for good ole AIX..." ! $sed -n 's/^\([^ ]*\)[ ]*syscall$/\1/p' /lib/syscalls.exp >>libc.list ! fi ! else ! $sed -n -e 's/^.* [ATDSUW] *[_.]*//p' -e 's/^.* [ATDSUW] //p' libc.list ! fi ! if $contains '^printf$' libc.list >/dev/null 2>&1; then echo done elif $sed -n -e 's/^__*//' -e 's/^\([a-zA-Z_0-9$]*\).*xtern.*/\1/p' \ *************** *** 1966,1971 **** --- 1977,1986 ---- libc.list; \ $contains '^printf$' libc.list >/dev/null 2>&1; then echo done + elif $sed -n -e 's/^.* [ATDSUW] *//p' -e 's/^.* [ATDSUW] //p' \ + libc.list; \ + $contains '^printf$' libc.list >/dev/null 2>&1; then + echo done else nm -p $* 2>/dev/null >libc.tmp $sed -n -e 's/^.* [AT] *_[_.]*//p' -e 's/^.* [AT] //p' libc.list *************** *** 2175,2181 **** exit \$status EOSS chmod +x Cppsym ! $eunicefix Cppsym echo "Your C preprocessor defines the following symbols:" Cppsym -l $attrlist >Cppsym.true cat Cppsym.true --- 2190,2196 ---- exit \$status EOSS chmod +x Cppsym ! echo "Your C preprocessor defines the following symbols:" Cppsym -l $attrlist >Cppsym.true cat Cppsym.true *************** *** 2247,2256 **** $cat << EOM ! Some modern Mail Transport Agents (mailers) support the Content-Length: header. ! In doing so, they do not wish to have messages escaped to protect 'From ' lines ! in the body of the message, among other strings. Does the Mail Transport Agent ! in use on this system honor the Content-Length: header? EOM echo " " --- 2262,2273 ---- $cat << EOM ! Some Local Delivery Agents (mailers) support the Content-Length: header. In ! doing so, they do not wish to have messages escaped to protect 'From ' lines ! in the body of the message, among other strings. Does the Local Delivery Agent ! in use on this system use the Content-Length: header when writing mails to ! (incoming) mailbox? ! EOM echo " " *************** *** 2459,2479 **** echo 'Dot lock style locking will be used' fi - if $test -d /usr/spool/locks ; then - lock_dir=/usr/spool/locks - elif $test -d /usr/spool/uucp ; then - lock_dir=/usr/spool/uucp - elif $test -d /var/spool/locks ; then - lock_dir=/var/spool/locks - elif $test -d /var/spool/uucp ; then - lock_dir=/var/spool/uucp - else - lock_dir=/tmp - fi - echo " " - echo "Non-mailbox locks will use the $lock_dir directory" - echo " " - : now get the host name if $test "$autohostname" != "$undef" -o "$hostname" = "" ; then autohostname="$define" --- 2476,2481 ---- *************** *** 2642,2647 **** --- 2644,2692 ---- d_whoami="$undef" fi + + : see if there is a poll file + echo " " + if $test -r /usr/include/poll.h ; then + d_poll="$define" + echo "poll.h found." + else + d_poll="$undef" + fi + + : see if there is a poll file + echo " " + if $test -r /usr/include/inttypes.h ; then + d_inttypes="$define" + echo "inttypes.h found." + + case "$d_nointtypes" in + "$define") dflt=y;; + "$undef") dflt=n;; + *) dflt=n;; + esac + + echo "In some systems (such as IRIX 5.3, but not in IRIX 6.2) " + echo "including of inttypes.h causes error (types defined also " + echo "in sys/types.h)" + echo " " + rp="Avoid including inttypes.h [$dflt]" + $echo $n "$rp $c" + . myread + case "$ans" in + [Yy]*) + d_nointtypes="$define" + ;; + *) + d_nointtypes="$undef" + esac + + else + d_inttypes="$undef" + d_nointtypes="$undef" + fi + + : see how we will look up host name echo " " d_douname="$undef" *************** *** 2909,2914 **** --- 2954,2971 ---- *) d_mmdf="$undef";; esac + # BSD Sendmail installs itself either /usr/lib/sendmail, /usr/sbin/sendmail + # or /usr/ucblib/sendmail, prefer that over other locations + if [ -x /usr/lib/sendmail ] ; then + sendmail=/usr/lib/sendmail + echo "BSD(?) sendmail is on $sendmail" + elif [ -x /usr/sbin/sendmail ] ; then + sendmail=/usr/sbin/sendmail + echo "BSD(?) sendmail is on $sendmail" + elif [ -x /usr/ucblib/sendmail ] ; then + sendmail=/usr/ucblib/sendmail + echo "BSD(?) sendmail is on $sendmail" + fi : determine mail delivery agent for Elm to use case "$mailer" in '') *************** *** 2964,2998 **** execmail='' fi - : check for internet mailer - case "$d_internet" in - "$define") dflt=y;; - "$undef") dflt=n;; - *) if $test -f "$sendmail"; then - dflt=y - elif $test -f "$submit"; then - dflt=y - elif $test -f "$smail"; then - dflt=y - else - dflt=n - fi - ;; - esac - $cat </dev/null < /dev/null; then ! echo "Yup. Option -B8BITMIME perhaps works." ! echo "(Sendmail 8.6 will accept but ignore that option.)" ! dflt=y ! else ! echo "Seems that option -B8BITMIME does not work." ! dflt=n ! fi ! else ! echo "Can't figure how to test if your mailer supports -B8BITMIME ..." ! dflt=n ! fi ! ;; ! esac ! ! cat </dev/null /dev/null /dev/null 2>&1; then + echo "waitpid() found." + d_waitpid="$define" + else + d_waitpid="$undef" + fi + : see if strspn exists set strspn d_strspn eval $inlibc *************** *** 3820,3856 **** fi fi - : check for valid reply/to fields - case "$d_useembed" in - "$define") dflt=y;; - "$undef") dflt=n;; - *) if $test "$d_internet" = "$define" ; then - dflt=y - else - dflt=n - fi - ;; - esac - $cat < /dev/null 2>&1; then echo "getdomainname() found" --- 4133,4138 ---- *************** *** 4001,4007 **** #include #include #define far /* to keep Xenix from complaining */ ! #if (defined(BSD) || !defined(apollo)) # include #endif #ifdef I_TIME --- 4283,4289 ---- #include #include #define far /* to keep Xenix from complaining */ ! #if (defined(BSD_TYPE) || !defined(apollo)) # include #endif #ifdef I_TIME *************** *** 4010,4016 **** #ifdef I_SYSTIME # include #endif ! #ifdef BSD # include #endif #ifdef I_UTIME --- 4292,4298 ---- #ifdef I_SYSTIME # include #endif ! #ifdef BSD_TYPE # include #endif #ifdef I_UTIME *************** *** 4041,4047 **** cflags="$cflags -DI_SYSTIME" fi if $test "$d_bsd" = "$define"; then ! cflags="$cflags -DBSD" fi if $cc $cflags try.c -o try >/dev/null 2>&1 ; then d_utimbuf="$define" --- 4323,4329 ---- cflags="$cflags -DI_SYSTIME" fi if $test "$d_bsd" = "$define"; then ! cflags="$cflags -DBSD_TYPE" fi if $cc $cflags try.c -o try >/dev/null 2>&1 ; then d_utimbuf="$define" *************** *** 4075,4091 **** i_memory="$undef" fi - : see if stdarg.h is in includepath - echo " " - i_stdarg=`loc stdarg.h "" /usr/include $includepath` - if $test -n "$i_stdarg"; then - echo "Found " - i_stdarg="$define" - else - echo "Did not find " - i_stdarg="$undef" - fi - : see if unistd.h is in includepath echo " " i_unistd=`loc unistd.h "" /usr/include $includepath` --- 4357,4362 ---- *************** *** 4100,4106 **** : see if signal is declared as pointer to function returning int or void echo " " $cppstdin $cppflags < /usr/include/signal.h >$$.tmp ! if $contains 'void.*[^s]signal' $$.tmp >/dev/null 2>&1 ; then echo "You have void (*signal())() instead of int." sigtype="void" elif $contains 'int.*[^s]signal' $$.tmp >/dev/null 2>&1 ; then --- 4371,4380 ---- : see if signal is declared as pointer to function returning int or void echo " " $cppstdin $cppflags < /usr/include/signal.h >$$.tmp ! if test "$uname_os" = "OSF1" ; then ! echo "You have void (*signal())()." ! sigtype="void" ! elif $contains 'void.*[^s]signal' $$.tmp >/dev/null 2>&1 ; then echo "You have void (*signal())() instead of int." sigtype="void" elif $contains 'int.*[^s]signal' $$.tmp >/dev/null 2>&1 ; then *************** *** 4287,4292 **** --- 4561,4635 ---- fi done + if [ "$d_locale" = "$define" ] ; then + $cat <bin/elm.mimecharsets <<'EOM' + # Mapping from LC_CTYPE locale to MIME's charset values + # Format is: + # LC_CTYPE charset + # LC_CTYPE is on form + # LL_cc.encoding + # then also following formats are available: + # LL_cc charset + # LL charset + # *.encoding charset + EOM + fi + + values="`locale -a 2>/dev/null`" + if $test -z "$values" ; then + values="`ls /usr/lib/locale`" + fi + + for lc_ctype in $values ; do + if grep "^$lc_ctype" bin/elm.mimecharsets > /dev/null; then + echo "$lc_ctype ok" + else + echo $n "Adding $lc_ctype $c" + if $test "$lc_ctype" = C -o "$defdispcharset" = "" + then + echo "$lc_ctype US-ASCII" >> bin/elm.mimecharsets + echo "(as US-ASCII)" + elif grep "^$lc_ctype" doc/mime.charsets >> bin/elm.mimecharsets + then + echo " (from doc/mime.charsets)" + else + echo "$lc_ctype $defdispcharset" >> bin/elm.mimecharsets + echo "(as $defdispcharset)" + fi + fi + done + + if $test -f /usr/lib/locale/locale.alias ; then + echo "# /usr/lib/locale/locale.alias gives:" >> bin/elm.mimecharsets + sed 's/^/#/g' < /usr/lib/locale/locale.alias >> bin/elm.mimecharsets + echo "Copying /usr/lib/locale/locale.alias to elm.mimecharsets as comment" + fi + + + echo + echo "Resulting elm.mimecharsets is following:" + echo + cat bin/elm.mimecharsets + echo + + rp="If you need to edit bin/elm.mimecharsets, do it as a shell escape here:" + $echo $n "$rp $c" + . UU/myread + $echo + ) + fi + : ask the preferred line printer and options for this system. case "$linepr" in '') *************** *** 4348,4354 **** : ask the preferred pager for this system. case "$pager" in ! '') dflt="builtin+";; *) dflt="$pager";; esac --- 4691,4697 ---- : ask the preferred pager for this system. case "$pager" in ! '') dflt="builtin++";; *) dflt="$pager";; esac *************** *** 4367,4372 **** --- 4710,4718 ---- builtin+) cont='' ;; + builtin++) + cont='' + ;; /*) if $test -f "$ans"; then cont='' *************** *** 4390,4434 **** esac done - : find out which shell people like to use most - case "$prefshell" in - '') - if $test -f /bin/ksh; then - dflt='/bin/ksh' - elif $test -f /bin/csh; then - dflt='/bin/csh' - else - dflt='/bin/sh' - fi - ;; - *) dflt="$prefshell";; - esac - cont=true - while $test "$cont" ; do - echo " " - echo "Give the full path name of the shell most people like to use on your" - $echo $n "system: [$dflt] $c" - rp="Preferred shell: [$dflt]" - . myread - prefshell=$ans - if test -f $ans; then - cont='' - else - if $test "$fastread" = yes; then - dflt=y - else - dflt=n - fi - rp="File $ans doesn't exist. Use that name anyway? [$dflt]" - $echo $n "$rp $c" - . myread - dflt='' - case "$ans" in - y*) cont='';; - esac - fi - done - : see if ar generates random libraries by itself echo " " echo "Checking how to generate random libraries on your machine..." --- 4736,4741 ---- *************** *** 4476,4481 **** --- 4783,4791 ---- if $test "$ans" = x; then ans=`loc Slibtermlib.a x $libpth` fi + if $test "$ans" = x; then + ans=`loc libtermlib.so x $libpth` + fi case "$ans" in /usr/lib*|/usr/ccs/lib*|/lib*) termlib='-ltermlib' *************** *** 4492,4497 **** --- 4802,4810 ---- if $test "$ans" = x; then ans=`loc Slibcurses.a x $libpth` fi + if $test "$ans" = x; then + ans=`loc libcurses.so x $libpth` + fi case "$ans" in /*) ar t $ans >grimble *************** *** 4626,4631 **** --- 4939,4960 ---- #$i_systime I_SYSTIME #include #include + + + #ifdef I_TIME + # include + #endif + #ifdef I_SYSTIME + # include + #endif + + #ifndef _POSIX_SOURCE + extern struct tm *localtime(); + extern time_t time(); + #endif + EOF + cat >elmlib.h </dev/null 2>&1 ! if $test $? -eq 0 && ./try >try.out 2>&1 ; then $echo "Alright! That seems to have worked." break fi --- 4976,4983 ---- $echo "Trying -DTZNAME_USE_$tzname_handling" $cc -D_CONFIGURE -DTZNAME_USE_$tzname_handling $ccflags \ try.c -o try >/dev/null 2>&1 ! if $test $? -eq 0 && ./try >try.out 2>&1 ! then $echo "Alright! That seems to have worked." break fi *************** *** 4663,4669 **** $echo "You will have to set \"tzname_handling\" manually." fi fi ! $rm -f defs.h try.c try.o try try.out : the "config.over" file can be used to patch configuration changes --- 4993,4999 ---- $echo "You will have to set \"tzname_handling\" manually." fi fi ! $rm -f defs.h elmlib.h try.c try.o try try.out : the "config.over" file can be used to patch configuration changes *************** *** 4714,4723 **** cppminus='$cppminus' d_getopt='$d_getopt' d_memcpy='$d_memcpy' - d_mkdir='$d_mkdir' d_rename='$d_rename' d_symlink='$d_symlink' d_whoami='$d_whoami' n='$n' c='$c' orderlib='$orderlib' --- 5044,5055 ---- cppminus='$cppminus' d_getopt='$d_getopt' d_memcpy='$d_memcpy' d_rename='$d_rename' d_symlink='$d_symlink' d_whoami='$d_whoami' + d_poll='$d_poll' + d_inttypes='$d_inttypes' + d_nointtypes='$d_nointtypes' n='$n' c='$c' orderlib='$orderlib' *************** *** 4724,4734 **** ranlib='$ranlib' package='$package' pager='$pager' - prefshell='$prefshell' startsh='$startsh' - d_eunice='$d_eunice' define='$define' - eunicefix='$eunicefix' loclist='$loclist' expr='$expr' sed='$sed' --- 5056,5063 ---- *************** *** 4739,4745 **** cp='$cp' tail='$tail' tr='$tr' - mkdir='$mkdir' sort='$sort' uniq='$uniq' grep='$grep' --- 5068,5073 ---- *************** *** 4789,4794 **** --- 5117,5123 ---- compress='$compress' execmail='$execmail' libswanted='$libswanted' + noaddlib='$noaddlib' c_date='$c_date' d_ascii='$d_ascii' d_broke_ctype='$d_broke_ctype' *************** *** 4806,4812 **** d_flock='$d_flock' d_dotlock='$d_dotlock' d_fcntlock='$d_fcntlock' - lock_dir='$lock_dir' has_flock='$has_flock' has_fcntl='$has_fcntl' d_ftruncate='$d_ftruncate' --- 5135,5140 ---- *************** *** 4828,4834 **** --- 5156,5166 ---- d_mallocvoid='$d_mallocvoid' d_mboxedit='$d_mboxedit' d_mime='$d_mime' + metamail_path='$metamail_path' defencoding='$defencoding' + d_8bitmime='$d_8bitmime' + d_binarymime='$d_binarymime' + d_dsn='$d_dsn' defcharset='$defcharset' defdispcharset='$defdispcharset' d_mmdf='$d_mmdf' *************** *** 4850,4855 **** --- 5182,5188 ---- d_sighold='$d_sighold' d_sigprocmask='$d_sigprocmask' d_sigblock='$d_sigblock' + d_waitpid='$d_waitpid' d_sigaction='$d_sigaction' d_strcspn='$d_strcspn' d_strspn='$d_strspn' *************** *** 4868,4874 **** tempnamc='$tempnamc' d_termio='$d_termio' d_termios='$d_termios' - d_useembed='$d_useembed' d_utimbuf='$d_utimbuf' d_vfork='$d_vfork' defbatsub='$defbatsub' --- 5201,5206 ---- *************** *** 4920,4926 **** d_passnames='$d_passnames' d_berknames='$d_berknames' d_usgnames='$d_usgnames' ! passcat='$passcat' rmttape='$rmttape' roff='$roff' roffopts='$roffopts' --- 5252,5258 ---- d_passnames='$d_passnames' d_berknames='$d_berknames' d_usgnames='$d_usgnames' ! # passcat='$passcat' rmttape='$rmttape' roff='$roff' roffopts='$roffopts' *************** *** 4935,4940 **** --- 5267,5278 ---- xenlf='$xenlf' d_xenix='$d_xenix' d_bsd='$d_bsd' + d_pgp='$d_pgp' + pgp='$pgp' + pgp_path='$pgp_path' + d_remailer='$d_remailer' + finger='$finger' + finger_path='$finger_path' CONFIG=true EOT *************** *** 4953,5014 **** eval $ans;; esac : if this fails, just run all the .SH files by hand ! . ./config.sh ! ! echo " " ! echo "Doing variable substitutions on .SH files..." ! set x `awk '{print $1}' /dev/null 2>&1; then ! dflt=n ! $cat < makedepend.out &" ! It can take a while, so you might not want to run it right now. ! ! EOM ! rp="Run make depend now? [$dflt]" ! $echo $n "$rp $c" ! . UU/myread ! case "$ans" in ! y*) make depend && echo "Now you must run a make." ! ;; ! *) echo "You must run 'make depend' then 'make'." ! ;; ! esac ! elif test -f Makefile; then ! echo " " ! echo "Now you must run a make." ! else ! echo "Done." ! fi ! $rm -f kit*isdone : the following is currently useless cd UU && $rm -f $rmlist --- 5291,5297 ---- eval $ans;; esac : if this fails, just run all the .SH files by hand ! . ConfTool/Substitute $rm -f kit*isdone : the following is currently useless cd UU && $rm -f $rmlist Index: elm2.4.ME+.50/Instruct *** elm2.4.25/Instruct Thu Mar 10 19:20:28 1994 --- elm2.4.ME+.50/Instruct Tue Nov 17 21:12:53 1998 *************** *** 1,7 **** Instructions ------------ ! Last Update: $Date: 1994/03/10 17:20:23 $ This file contains instructions on how to create and install --- 1,7 ---- Instructions ------------ ! Last Update: $Date: 1998/10/15 15:36:38 $ This file contains instructions on how to create and install Index: elm2.4.ME+.50/MANIFEST *** elm2.4.25/MANIFEST Thu Mar 10 19:14:46 1994 --- elm2.4.ME+.50/MANIFEST Tue Nov 17 21:14:08 1998 *************** *** 1,7 **** --- 1,11 ---- + README.ME+ + ANNOUNCE.ME + ChangeLog.ME Changes Configure Instruct MANIFEST + ConfTool/Substitute Makefile.SH NOTICE Overview *************** *** 31,36 **** --- 35,42 ---- doc/frm.1 doc/listalias.1 doc/messages.1 + doc/mime.types + doc/mime.charsets doc/newalias.1 doc/newmail.1 doc/printmail.1 *************** *** 55,64 **** hdrs/headers.h hdrs/mcprt.h hdrs/mcprtlib.h hdrs/mime.h hdrs/msgcat.h hdrs/ndbz.h ! hdrs/nl_types.h hdrs/patchlevel.h hdrs/regexp.h hdrs/s_aliases.h --- 61,73 ---- hdrs/headers.h hdrs/mcprt.h hdrs/mcprtlib.h + hdrs/me.h + hdrs/menu.h + hdrs/menu2.h hdrs/mime.h hdrs/msgcat.h hdrs/ndbz.h ! hdrs/elm_nl_types.h hdrs/patchlevel.h hdrs/regexp.h hdrs/s_aliases.h *************** *** 84,89 **** --- 93,100 ---- lib/can_open.c lib/chloc.c lib/date_util.c + lib/dispaddr.c + lib/dynarray.c lib/errno.c lib/expand.c lib/figadrssee.c *************** *** 124,133 **** --- 135,146 ---- lib/strftime.c lib/strincmp.c lib/striparens.c + lib/strmcpy.c lib/strstr.c lib/strtokq.c lib/tail_of.c lib/validname.c + lib/safeopen.c nls/C/C/C/s_aliases.m nls/C/C/C/s_answer.m nls/C/C/C/s_elm.m *************** *** 159,166 **** src/alias.c src/aliaslib.c src/args.c ! src/bouncebk.c ! src/builtin.c src/calendar.c src/curses.c src/date.c --- 172,179 ---- src/alias.c src/aliaslib.c src/args.c ! src/attach_menu.c ! src/builtin++.c src/calendar.c src/curses.c src/date.c *************** *** 185,200 **** src/lock.c src/mailmsg1.c src/mailmsg2.c src/mime.c src/mkhdrs.c src/newmbox.c src/options.c src/out_utils.c src/pattern.c ! src/pmalloc.c src/quit.c src/read_rc.c src/remail.c src/reply.c src/returnadd.c src/save_opts.c --- 198,218 ---- src/lock.c src/mailmsg1.c src/mailmsg2.c + src/menu.c + src/menu2.c + src/metapager.c src/mime.c + src/mime_encode.c src/mkhdrs.c src/newmbox.c src/options.c src/out_utils.c src/pattern.c ! src/pgp.c src/quit.c src/read_rc.c src/remail.c + src/remailer.c src/reply.c src/returnadd.c src/save_opts.c *************** *** 227,229 **** --- 245,263 ---- utils/printmail.SH utils/prlong.c utils/readmsg.c + melib/Makefile.SH + melib/mime.c + melib/mime_decode.c + melib/mime_parse.c + melib/parse_util.c + melib/pgp_decode.c + melib/state.c + hdrs/melib.h + hdrs/elmlib.h + lib/read_rc.c + lib/output.c + lib/pmalloc.c + nls/C/C/C/s_me.m + hdrs/s_me.h + lib/getaddr.c + lib/forwarded.c Index: elm2.4.ME+.50/Makefile.SH *** elm2.4.25/Makefile.SH Wed Apr 21 04:53:22 1993 --- elm2.4.ME+.50/Makefile.SH Tue Nov 17 21:12:53 1998 *************** *** 21,44 **** echo "Extracting Makefile (with variable substitutions)" cat >Makefile <Makefile <. If this is";\ $(ECHO) "not the case, interrupt this process quickly!";\ $(SLEEP) 10;) cd lib; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ cd src; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ ! cd filter; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ cd utils; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ cd doc; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ --- 87,96 ---- $(ECHO) "ELM on the remote machine <$(REMOTE)>. If this is";\ $(ECHO) "not the case, interrupt this process quickly!";\ $(SLEEP) 10;) + cd melib; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ cd lib; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ cd src; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ ! # cd filter; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ cd utils; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ cd doc; $(MAKE) $(MJ) -$(MAKEFLAGS) REMOTE=$(REMOTE) $@ *************** *** 111,116 **** --- 115,121 ---- cd doc; $(MAKE) $(MJ) -$(MAKEFLAGS) $@ elm: + cd melib; $(MAKE) $(MJ) -$(MAKEFLAGS) all cd lib; $(MAKE) $(MJ) -$(MAKEFLAGS) all cd src; $(MAKE) $(MJ) -$(MAKEFLAGS) all *************** *** 120,126 **** filter: _filter _filter: ! cd filter; $(MAKE) $(MJ) -$(MAKEFLAGS) all # The dummy dependency here prevents make from thinking the target is the # utils directory that sits in the current directory, rather than --- 125,131 ---- filter: _filter _filter: ! # cd filter; $(MAKE) $(MJ) -$(MAKEFLAGS) all # The dummy dependency here prevents make from thinking the target is the # utils directory that sits in the current directory, rather than Index: elm2.4.ME+.50/README *** elm2.4.25/README Tue May 11 06:19:23 1993 --- elm2.4.ME+.50/README Tue Nov 17 21:12:53 1998 *************** *** 1,3 **** --- 1,7 ---- + This README is for the 2.4 version. + Look file README.ME+ for the 2.4ME+ version. + + ------------------------------------------------------------------ This is the 2.4 (USENET) version of the Elm Mail System See the NOTICE and Instruct files for further details. Index: elm2.4.ME+.50/config.h.SH *** elm2.4.25/config.h.SH Mon May 30 19:55:57 1994 --- elm2.4.ME+.50/config.h.SH Tue Nov 17 21:12:54 1998 *************** *** 56,68 **** */ #$d_memcpy MEMCPY /**/ - /* MKDIR: - * This symbol, if defined, indicates that the mkdir routine is available - * to create directories. Otherwise you should fork off a new process to - * exec /bin/mkdir. - */ - #$d_mkdir MKDIR /**/ - /* RENAME: * This symbol, if defined, indicates that the rename routine is available * to rename files. Otherwise you should do the unlink(), link(), unlink() --- 56,61 ---- *************** *** 82,107 **** */ #$d_whoami WHOAMI /**/ ! /* PREFSHELL: ! * This symbol contains the full name of the preferred user shell on this ! * system. Usual values are /bin/csh, /bin/ksh, /bin/sh. ! */ ! #define PREFSHELL "$prefshell" /**/ ! ! /* EUNICE: ! * This symbol, if defined, indicates that the program is being compiled ! * under the EUNICE package under VMS. The program will need to handle ! * things like files that don't go away the first time you unlink them, ! * due to version numbering. It will also need to compensate for lack ! * of a respectable link() command. ! */ ! /* VMS: ! * This symbol, if defined, indicates that the program is running under ! * VMS. It is currently only set in conjunction with the EUNICE symbol. */ ! #$d_eunice EUNICE /**/ ! #$d_eunice VMS /**/ /* CONFIGURE_DATE * This symbol contains the last date that configure was run for elm -v output. */ --- 75,95 ---- */ #$d_whoami WHOAMI /**/ ! /* POLL: ! * This symbol, if defined, indicates that the program may include ! * poll.h. ! */ ! #$d_poll POLL /**/ ! ! /* I_INTTYPES: ! * This symbol, if defined, indicates that the program may include ! * inttypes.h. */ ! #$d_inttypes I_INTTYPES /**/ ! #$d_nointtypes NO_INTTYPES /**/ + + /* CONFIGURE_DATE * This symbol contains the last date that configure was run for elm -v output. */ *************** *** 185,194 **** /* USE_FCNTL_LOCKING * This symbol, if defined, indicates that SYSV style fcntl file locking should be used. */ - /* LOCK_DIR - * This symbol is the name of the lock directory for access (not mailbox) locks. - * It will be /usr/spool/locks or /usr/spool/uucp - */ #$d_flock USE_FLOCK_LOCKING /**/ #$d_dotlock USE_DOTLOCK_LOCKING /**/ --- 173,178 ---- *************** *** 195,202 **** #$d_fcntlock USE_FCNTL_LOCKING /**/ - #define LOCK_DIR "$lock_dir" /**/ - /* FTRUNCATE: * This symbol, if defined, indicates that the ftruncate() routine exists. */ --- 179,184 ---- *************** *** 235,246 **** #$d_index index strchr /* cultural */ #$d_index rindex strrchr /* differences? */ - /* INTERNET: - * This symbol, if defined, indicates that there is a mailer available - * which supports internet-style addresses (user@site.domain). - */ - #$d_internet INTERNET /**/ - /* ISPELL: * This symbol, if defined, indicates that the GNU ispell * spelling checker that is available to Elm. --- 217,222 ---- *************** *** 292,297 **** --- 268,314 ---- */ #$d_mime MIME /**/ + /* MIME: + * This symbol defines default path to metamail. + * Special values: none Don't call metamail + * metamail Call metamail if environment + * variable NOMETAMAIL is not defined + */ + #define METAMAIL_PATH "$metamail_path" + + /* USE_8BITMIME + * Mailer accepts -B8BITMIME option + * -B == BODYTYPE + */ + #$d_8bitmime USE_8BITMIME /**/ + + /* USE_BINARYMIME + * Mailer accepts -BBINARYMIME option + * -B == BODYTYPE + * + * Elm don't yet fully support BINARYMIME: it requires that lines are + * terminated with \r\n, Unix's \n isn't sufficient! + * That is only partially implemented in this version of Elm. + * + * Quote from RFC 1830 (Experimental Protocol): + * SMTP Service Extensions for Transmission of Large and Binary MIME Messages + * + * It is important to note that when using BINARYMIME, it is + * especially important to ensure that the MIME message itself is + * properly formed. In particular, it is essential that text be + * canonically encoded with each line properly terminated with + * . Any transformation of text into non-canonical MIME to + * observe local storage conventions must be reversed before sending + * as BINARYMIME. The usual line-oriented shortcuts will break if + * used with BINARYMIME. + */ + #$d_binarymime USE_BINARYMIME /**/ + + /* + * If mailer support DSN via options -N -V -R + */ + #$d_dsn USE_DSN /**/ + /* MMDF: * This symbol, if defined, indicates that mailboxes are in * the MMDF format. *************** *** 398,403 **** --- 415,422 ---- #$d_sigprocmask HASSIGPROCMASK /**/ + #$d_waitpid HASWAITPID /**/ + /* STRSPN: * This symbol, if defined, indicates that the strspn() routine exists. */ *************** *** 469,479 **** #$d_termio TERMIO /**/ - /* USE_EMBEDDED_ADDRESSES: - * This symbol, if defined, indicates that replyto: and from: - * headers can be trusted. - */ - #$d_useembed USE_EMBEDDED_ADDRESSES /**/ /* UTIMBUF: * This symbol is defined if this system defines struct utimbuf. --- 488,493 ---- *************** *** 517,528 **** */ #$i_memory I_MEMORY /**/ - /* I_STDARG: - * This symbol, if defined, indicates that the file stdarg.h - * should be included instead of varargs - */ - #$i_stdarg I_STDARG /**/ - /* I_STDLIB: * This symbol, if defined, indicates that the file stdlib.h * should be included instead of declaring the stdlib routines. --- 531,536 ---- *************** *** 596,605 **** * This symbol, if defined, indicates this is a Xenix system, * for knocking out the far keyword in selected places. */ ! /* BSD: * This symbol, if defined, indicates this is a BSD type system, */ #$d_xenix XENIX /**/ ! #$d_bsd BSD /**/ !GROK!THIS! --- 604,635 ---- * This symbol, if defined, indicates this is a Xenix system, * for knocking out the far keyword in selected places. */ ! /* BSD_TYPE: * This symbol, if defined, indicates this is a BSD type system, */ #$d_xenix XENIX /**/ ! #$d_bsd BSD_TYPE /**/ + /* USE_PGP: + * This symbol, if defined, means that PGP support should be compiled in. + */ + /* PGP_PATH: + * The path to the PGP binary. + */ + #$d_pgp USE_PGP /**/ + #ifdef USE_PGP + #define PGP_PATH "$pgp_path" + #endif + + /* USE_REMAILER: + * If defined, includes support for anonymous remailing, chaining, and + * if USE_PGP is defined, encryption. + * + * FINGER_PATH: + * Path to the finger binary. + */ + #$d_remailer USE_REMAILER /**/ + #ifdef USE_REMAILER + #define FINGER_PATH "$finger_path" + #endif !GROK!THIS! Index: elm2.4.ME+.50/doc/Alias.guide *** elm2.4.25/doc/Alias.guide Tue Jul 20 06:02:25 1993 --- elm2.4.ME+.50/doc/Alias.guide Tue Nov 17 21:12:54 1998 *************** *** 1,4 **** ! .\" @(#)$Id: Alias.guide,v 5.7 1993/07/20 03:02:11 syd Exp $ .\" .\" A guide to the ELM alias system and so on. .\" format with: --- 1,4 ---- ! .\" @(#)$Id: Alias.guide,v 1.2 1998/10/20 19:57:19 hurtta Exp $ .\" .\" A guide to the ELM alias system and so on. .\" format with: *************** *** 6,43 **** .\" .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust ! .\" Elm is now in the public trust. Bug reports, comments, suggestions, flames ! .\" etc. should go to: ! .\" Syd Weinstein elm@DSI.COM (dsinc!elm) ! .\" ! .\" $Log: Alias.guide,v $ ! .\" Revision 5.7 1993/07/20 03:02:11 syd ! .\" drop useless reference to Elm address validation ! .\" From: Syd ! .\" ! .\" Revision 5.6 1993/07/20 02:16:26 syd ! .\" Explain that newalias for global takes -g argument ! .\" From: Syd via request from Manfred Ebery ! .\" ! .\" Revision 5.5 1993/04/12 01:45:22 syd ! .\" Update for quotes ! .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) ! .\" ! .\" Revision 5.4 1992/12/24 23:48:05 syd ! .\" make Nroff not use co symbol, just the word ! .\" From: Syd via request from Tim Pierce ! .\" ! .\" Revision 5.3 1992/12/20 05:06:19 syd ! .\" Here are the doc changes for my additions to the alias system and ! .\" the new switches to listalias. ! .\" From: "Robert L. Howard" ! .\" ! .\" Revision 5.2 1992/10/17 22:01:01 syd ! .\" corrected typos ! .\" From: David.W.Tamkin@gagme.chi.il.us (David W. Tamkin) ! .\" ! .\" Revision 5.1 1992/10/03 20:51:50 syd ! .\" Initial checkin as of 2.4 Release at PL0 .\" .\" .\" --- 6,12 ---- .\" .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust ! .\" Elm is now in the public trust. .\" .\" .\" Index: elm2.4.ME+.50/doc/Config.guid *** elm2.4.25/doc/Config.guid Fri Jul 30 05:23:26 1993 --- elm2.4.ME+.50/doc/Config.guid Tue Nov 17 21:12:54 1998 *************** *** 1,4 **** ! .\" @(#)$Id: Config.guid,v 5.9 1993/07/30 02:23:21 syd Exp $ .\" .\" A guide to the ELM alias system and so on. .\" format with: --- 1,4 ---- ! .\" @(#)$Id: Config.guid,v 1.1.1.1 1998/10/15 15:36:39 hurtta Exp $ .\" .\" A guide to the ELM alias system and so on. .\" format with: *************** *** 6,52 **** .\" .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust - .\" Elm is now in the public trust. Bug reports, comments, suggestions, flames - .\" etc. should go to: - .\" Syd Weinstein elm@DSI.COM (dsinc!elm) .\" - .\" $Log: Config.guid,v $ - .\" Revision 5.9 1993/07/30 02:23:21 syd - .\" fix -q to -g in newalias call .\" - .\" Revision 5.8 1993/04/12 04:00:55 syd - .\" remove unused ALIAS parameters .\" - .\" Revision 5.7 1993/04/12 01:47:45 syd - .\" Update for quotes - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) .\" - .\" Revision 5.6 1992/12/24 23:48:05 syd - .\" make Nroff not use co symbol, just the word - .\" From: Syd via request from Tim Pierce - .\" - .\" Revision 5.5 1992/11/27 02:51:18 syd - .\" Make the docs catch up with the code, config.changes -> config.over - .\" From: Syd - .\" - .\" Revision 5.4 1992/11/22 00:49:31 syd - .\" Add support for a "config.over" file to override Configure decisions. - .\" From: chip@chinacat.unicom.com (Chip Rosenthal) - .\" - .\" Revision 5.3 1992/10/24 13:51:44 syd - .\" More typos - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.2 1992/10/11 01:15:53 syd - .\" A cleaned up Config.guid follows. Corrected some spelling errors, - .\" and played with some formatting: - .\" From: David Barr - .\" - .\" Revision 5.1 1992/10/03 20:51:50 syd - .\" Initial checkin as of 2.4 Release at PL0 - .\" - .\" - .\" .tm Have we been run through "tbl" first?? I hope so! .po 1i .ds h0 --- 6,15 ---- *************** *** 248,263 **** a lone period into a period and a blank to avoid this problem. This symbol is ignored if the \*QHonors Content-Length: header?\*U question is answered \f2yes\f1. - .sp - .ne 4 - \f3Does your mailer understand INTERNET addresses?\f1 - .sp - Elm works with systems that can process the \*q@\*u character of - Internet format addresses or with the \*q!\*u format of \f2uucp\f1 - addresses. If your MTA understands the \*q@\*u format addresses, they - should be used and this question answered \f2yes\f1. If messages - bounce when you send them with \*q@\*u format addresses (such as - \*Qelm@dsi.com\*U), then answer this question \f2no\f1. .sp .ne 4 \f3Am I going to be running as a setgid program?\f1 --- 211,216 ---- Index: elm2.4.ME+.50/doc/Elm.cover *** elm2.4.25/doc/Elm.cover Sat May 8 22:58:27 1993 --- elm2.4.ME+.50/doc/Elm.cover Tue Nov 17 21:12:54 1998 *************** *** 1,31 **** ! .\" @(#)$Id: Elm.cover,v 5.4 1993/05/08 19:58:20 syd Exp $ .\" .\" Cover sheet for the ELM mail system... .\" format with .\" troff tmac.n Elm.cover > Coversheet.fmtd' .\" ! .\" Elm is now in the public trust. Bug reports, comments, suggestions, flames ! .\" etc. should go to: ! .\" Syd Weinstein elm@DSI.COM (dsinc!elm) .\" .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust - .\" - .\" $Log: Elm.cover,v $ - .\" Revision 5.4 1993/05/08 19:58:20 syd - .\" use new quote macros - .\" From: Syd - .\" - .\" Revision 5.3 1992/12/24 23:48:05 syd - .\" make Nroff not use co symbol, just the word - .\" From: Syd via request from Tim Pierce - .\" - .\" Revision 5.2 1992/10/24 14:13:39 syd - .\" Add notice file - .\" - .\" Revision 5.1 1992/10/03 20:51:50 syd - .\" Initial checkin as of 2.4 Release at PL0 - .\" .\" .\" .po 1i --- 1,13 ---- ! .\" @(#)$Id: Elm.cover,v 1.2 1998/10/20 19:57:19 hurtta Exp $ .\" .\" Cover sheet for the ELM mail system... .\" format with .\" troff tmac.n Elm.cover > Coversheet.fmtd' .\" ! .\" Elm is now in the public trust. .\" .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust .\" .\" .po 1i Index: elm2.4.ME+.50/doc/Filter.guid *** elm2.4.25/doc/Filter.guid Fri Apr 16 07:11:54 1993 --- elm2.4.ME+.50/doc/Filter.guid Tue Nov 17 21:12:54 1998 *************** *** 1,4 **** ! .\" @(#)$Id: Filter.guid,v 5.7 1993/04/16 04:11:29 syd Exp $ .\" .\" A guide to the Elm Filter program .\" format with: --- 1,4 ---- ! .\" @(#)$Id: Filter.guid,v 1.2 1998/10/20 19:57:19 hurtta Exp $ .\" .\" A guide to the Elm Filter program .\" format with: *************** *** 7,48 **** .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust .\" ! .\" Elm is now in the public trust. Bug reports, comments, suggestions, flames ! .\" etc. should go to: ! .\" Syd Weinstein elm@DSI.COM (dsinc!elm) ! .\" ! .\" $Log: Filter.guid,v $ ! .\" Revision 5.7 1993/04/16 04:11:29 syd ! .\" Fix wrong variable used in shell script ! .\" From: Bruce Tindall ! .\" ! .\" Revision 5.6 1993/04/12 01:46:12 syd ! .\" Update for quotes ! .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) ! .\" ! .\" Revision 5.5 1993/01/27 19:45:15 syd ! .\" Filter turns spaces in quoted strings into _ and then back again. This destroys ! .\" any _ that where there in the first place. This patch removes that. ! .\" Also fixed a minor bug where 'filter -r' wrote out the wrong thing if the ! .\" relation in a rule was '~'. ! .\" From: Jan Djarv ! .\" ! .\" Revision 5.4 1993/01/27 19:40:01 syd ! .\" I implemented a change to filter's default verbose message format ! .\" including %x %X style date and time along with username ! .\" From: mark@drd.com (Mark Lawrence) ! .\" ! .\" Revision 5.3 1992/12/24 23:48:05 syd ! .\" make Nroff not use co symbol, just the word ! .\" From: Syd via request from Tim Pierce ! .\" ! .\" Revision 5.2 1992/11/15 01:41:50 syd ! .\" Add regexp processing to filter. ! .\" Add execc operator ! .\" From: Jan Djarv ! .\" ! .\" Revision 5.1 1992/10/03 20:51:50 syd ! .\" Initial checkin as of 2.4 Release at PL0 .\" .\" .po 1i --- 7,13 ---- .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust .\" ! .\" Elm is now in the public trust. .\" .\" .po 1i Index: elm2.4.ME+.50/doc/Form.guide *** elm2.4.25/doc/Form.guide Mon Apr 12 04:47:13 1993 --- elm2.4.ME+.50/doc/Form.guide Tue Nov 17 21:12:54 1998 *************** *** 1,39 **** ! .\" @(#)$Id: Form.guide,v 5.4 1993/04/12 01:47:13 syd Exp $ .\" .\" A guide to using the Elm Forms mode .\" format with: .\" 'troff tmac.n - > Forms.format' .\" ! .\" Elm is now in the public trust. Bug reports, comments, suggestions, flames ! .\" etc. should go to: ! .\" Syd Weinstein elm@DSI.COM (dsinc!elm) .\" .\" Copyright (c) 1988-1992 USENET Community Trust .\" Copyright (c) 1986,1987 Dave Taylor - .\" - .\" $Log: Form.guide,v $ - .\" Revision 5.4 1993/04/12 01:47:13 syd - .\" Update for quotes - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.3 1992/12/24 23:48:05 syd - .\" make Nroff not use co symbol, just the word - .\" From: Syd via request from Tim Pierce - .\" - .\" Revision 5.2 1992/10/11 01:33:36 syd - .\" fixed a bug in the .lp macro in doc/tmac.n. The lp macros is always - .\" called with a unit specified for the second argument, however, the - .\" macro added a "n" unit specifier to that argument. The following patch - .\" fixes this problem. - .\" - .\" The various .guide books use the troff .tl macro liberally, but often the - .\" extra ' characters needed at the end of the lines to explicitly specify - .\" the trailing nulls fields were left off. The following patch fixes the - .\" erroneous lines. - .\" From: Larry Philps - .\" - .\" Revision 5.1 1992/10/03 20:51:50 syd - .\" Initial checkin as of 2.4 Release at PL0 .\" .\" .\" --- 1,13 ---- ! .\" @(#)$Id: Form.guide,v 1.2 1998/10/20 19:57:19 hurtta Exp $ .\" .\" A guide to using the Elm Forms mode .\" format with: .\" 'troff tmac.n - > Forms.format' .\" ! .\" Elm is now in the public trust. .\" .\" Copyright (c) 1988-1992 USENET Community Trust .\" Copyright (c) 1986,1987 Dave Taylor .\" .\" .\" Index: elm2.4.ME+.50/doc/Makefile.SH *** elm2.4.25/doc/Makefile.SH Thu Mar 10 19:20:29 1994 --- elm2.4.ME+.50/doc/Makefile.SH Tue Nov 17 21:12:54 1998 *************** *** 16,82 **** echo "Extracting doc/Makefile (with variable substitutions)" cat >Makefile < - # - # Revision 5.9 1992/12/13 18:00:05 syd - # Add one more missing rm of a previously packed man page - # From: Bo Asbjorn Muldbak - # - # Revision 5.8 1992/12/12 01:28:50 syd - # Remove the old packed manual page before trying to pack the new one. - # From: Tom Moore - # - # Revision 5.7 1992/12/07 05:05:27 syd - # fix ; placement - # From: Syd - # - # Revision 5.6 1992/12/07 03:28:35 syd - # add else clause with true to TEST lines as some Makefiles exit if - # the if construct doesnt execute a clause as the if returns an error - # status. - # From: Syd via requests from several people - # - # Revision 5.5 1992/12/07 02:17:14 syd - # Add missing -n flag to doc makefile. - # From: gwh@dogmatix.inmos.co.uk - # - # Revision 5.4 1992/11/22 00:47:02 syd - # Change the way Configure and doc/Makefile.SH deal with installing - # man pages. - # From: Tom Moore - # - # Revision 5.3 1992/11/07 16:13:00 syd - # Limit test arguments to -f instead of -x as several shells and - # their resulting test commands dont have -x - # From: Syd via request from Tim Pierce - # - # Revision 5.2 1992/10/30 21:06:09 syd - # add missing MAKE variable definition - # From: syd via note from Keith Petersen w8sdz@tacom-emh1.army.mil - # - # Revision 5.1 1992/10/03 20:51:50 syd - # Initial checkin as of 2.4 Release at PL0 - # - # - # # Variables # Variables established by Configure CHMOD = $chmod --- 16,28 ---- echo "Extracting doc/Makefile (with variable substitutions)" cat >Makefile < Ref.format .\" ! .\" Elm is now in the public trust. Bug reports, comments, suggestions, flames ! .\" etc. should go to: ! .\" Syd Weinstein elm@DSI.COM (dsinc!elm) .\" .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust - .\" - .\" $Log: Ref.guide,v $ - .\" Revision 5.30 1993/09/19 23:43:35 syd - .\" Conform textencoding addition to style standard. - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.29 1993/08/23 02:53:52 syd - .\" Added some documentation for textencoding variable. - .\" From: Jan.Djarv@sa.erisoft.se (Jan Djarv) - .\" - .\" Revision 5.28 1993/08/10 21:25:09 syd - .\" Correct typo and use hard spaces within quotes in *clear-weed-list* note. - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.27 1993/08/03 18:56:24 syd - .\" Added text about how the weedout list works w.r.t From: and From_ - .\" From: Jan.Djarv@sa.erisoft.se (Jan Djarv) - .\" - .\" Revision 5.26 1993/07/20 02:43:05 syd - .\" [ Patch description ] - .\" Update the example elmrc and move a couple out of order variable - .\" descriptions. - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.25 1993/05/14 03:56:39 syd - .\" Moved the new numeric variables to the Numeric Variables section and - .\" applied the quoting/highlight policy to them. - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.24 1993/05/08 20:08:43 syd - .\" add sleepmsg parameter - .\" - .\" Revision 5.23 1993/05/08 20:00:22 syd - .\" fix mail added >From lines - .\" From: Syd - .\" - .\" Revision 5.22 1993/05/08 18:56:16 syd - .\" created a new elmrc variable named "readmsginc". This specifies an - .\" increment by which the message count is updated. If this variable is - .\" set to, say, 25, then the message count will only be updated every 25 - .\" messages, displaying 0, 25, 50, 75, and so forth. The default value - .\" of 1 will cause Elm to behave exactly as it currently does in PL21. - .\" From: Eric Peterson - .\" - .\" Revision 5.21 1993/04/12 03:15:41 syd - .\" These patches makes 'T' (since it was free) do a Tag and Move command in the - .\" index and alias page, and in the builtin pager. - .\" In the alias help in src/alias.c, there is a tolower done on the character - .\" one wants help for. This is clearly wrong. - .\" From: Jan Djarv - .\" - .\" Revision 5.20 1993/04/12 01:44:20 syd - .\" Update for quotes - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.19 1993/01/27 20:57:47 syd - .\" Corrections to PL21 Alpha docs and consistent punctuation use in help files. - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.18 1993/01/20 04:01:07 syd - .\" Adds a new integer parameter builtinlines. - .\" if (builtinlines < 0) and (the length of the message < LINES on - .\" screen + builtinlines) use internal. - .\" if (builtinlines > 0) and (length of message < builtinlines) - .\" use internal pager. - .\" if (builtinlines = 0) or none of the above conditions hold, use the - .\" external pager if defined. - .\" From: "John P. Rouillard" - .\" - .\" Revision 5.17 1993/01/19 04:52:19 syd - .\" add c)hange alias command to alias helpfile - .\" if a deleted alias is changed, undelete it. Also added the 'N' - .\" flag to changed aliases to help remind the user. Documented it. - .\" Note: if they mark the alias for deletion AFTER making the change it - .\" WILL be deleted. (and marked accordingly) - .\" modified alias mode title string to indicate when a resync was - .\" needed. - .\" allow editing alias file when none exist. - .\" Now aliases are check for illegal characters (and WS) and - .\" addresses are check for illegal WS when they are being entered. If - .\" anything illegal is found and message is printed and they keep entering - .\" the item until they get it right. - .\" I fixed a couple of places where int should be long to match - .\" the declared type of alias_rec.length - .\" From: "Robert L. Howard" - .\" - .\" Revision 5.16 1992/12/24 23:48:05 syd - .\" make Nroff not use co symbol, just the word - .\" From: Syd via request from Tim Pierce - .\" - .\" Revision 5.15 1992/12/20 05:22:13 syd - .\" Fix formatting and typograpical problems - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.14 1992/12/20 05:06:19 syd - .\" Here are the doc changes for my additions to the alias system and - .\" the new switches to listalias. - .\" From: "Robert L. Howard" - .\" - .\" Revision 5.13 1992/12/12 01:27:41 syd - .\" Make verb and pronoun agree. - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.12 1992/12/07 02:41:21 syd - .\" This implements the use of SIGUSR1 and SIGUSR2 as discussed on the - .\" mailing list recently, and adds them to the documentation. - .\" From: scs@lokkur.dexter.mi.us (Steve Simmons) - .\" - .\" Revision 5.11 1992/11/07 19:37:21 syd - .\" Enhanced printing support. Added "-I" to readmsg to - .\" suppress spurious diagnostic messages. - .\" From: chip@chinacat.unicom.com (Chip Rosenthal) - .\" - .\" Revision 5.10 1992/11/07 16:31:15 syd - .\" another typo (2.4.6 Ref.guide) - .\" From: David.W.Tamkin@gagme.chi.il.us (David W. Tamkin) - .\" - .\" Revision 5.9 1992/10/30 20:56:37 syd - .\" Things I found in doc/Ref.guide looking over patches 4 & 5: - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.8 1992/10/24 13:53:16 syd - .\" I think all these are typos - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.7 1992/10/24 13:44:41 syd - .\" There is now an additional elmrc option "displaycharset", which - .\" sets the charset supported on your terminal. This is to prevent - .\" elm from calling out to metamail too often. - .\" Plus a slight documentation update for MIME composition (added examples) - .\" From: Klaus Steinberger - .\" - .\" Revision 5.6 1992/10/24 13:25:41 syd - .\" In our global elm.rc I keep the four options as below - .\" - .\" confirmappend = OFF Don't confirm every append to any file. - .\" confirmcreate = ON Confirm creation of every new file. - .\" confirmfiles = ON Confirm append to non folder files though. - .\" confirmfolders = ON In case someone does not want to be asked - .\" every time when creating a new file try - .\" to confirm creation of folders though. - .\" From: Jukka Ukkonen - .\" - .\" Revision 5.5 1992/10/17 22:58:57 syd - .\" patch to make elm use (or in my case, not use) termcap/terminfo ti/te. - .\" From: Graham Hudspith - .\" - .\" Revision 5.4 1992/10/17 22:01:01 syd - .\" corrected typos - .\" From: David.W.Tamkin@gagme.chi.il.us (David W. Tamkin) - .\" - .\" Revision 5.3 1992/10/11 01:33:36 syd - .\" fixed a bug in the .lp macro in doc/tmac.n. The lp macros is always - .\" called with a unit specified for the second argument, however, the - .\" macro added a "n" unit specifier to that argument. The following patch - .\" fixes this problem. - .\" - .\" The various .guide books use the troff .tl macro liberally, but often the - .\" extra ' characters needed at the end of the lines to explicitly specify - .\" the trailing nulls fields were left off. The following patch fixes the - .\" erroneous lines. - .\" From: Larry Philps - .\" - .\" Revision 5.2 1992/10/04 01:15:43 syd - .\" Emphasize that hostdomain, hostname and hostfullname must all be specified - .\" if one is specified in elmrc file. - .\" - .\" Revision 5.1 1992/10/03 20:51:50 syd - .\" Initial checkin as of 2.4 Release at PL0 .\" .\" .\" --- 1,13 ---- ! .\" @(#)$Id: Ref.guide,v 1.2 1998/10/20 19:57:19 hurtta Exp $ .\" .\" Reference guide to the Elm mail system. .\" format with .\" troff tmac.n Ref.guide > Ref.format .\" ! .\" Elm is now in the public trust. .\" .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust .\" .\" .\" Index: elm2.4.ME+.50/doc/Users.guide *** elm2.4.25/doc/Users.guide Thu Mar 10 19:21:24 1994 --- elm2.4.ME+.50/doc/Users.guide Tue Nov 17 21:12:54 1998 *************** *** 1,59 **** ! .\" @(#)$Id: Users.guide,v 5.9 1994/03/10 17:21:21 syd Exp $ .\" .\" Users guide to the ELM mail system. .\" format with: .\" 'tbl Users.guide | troff tmac.n - > Users.fmtd' .\" ! .\" Elm is now in the public trust. Bug reports, comments, suggestions, flames ! .\" etc. should go to: ! .\" Syd Weinstein elm@DSI.COM (dsinc!elm) .\" .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust - .\" - .\" $Log: Users.guide,v $ - .\" Revision 5.9 1994/03/10 17:21:21 syd - .\" remove autoreply - .\" - .\" Revision 5.8 1993/05/08 20:00:22 syd - .\" fix mail added >From lines - .\" From: Syd - .\" - .\" Revision 5.7 1993/04/12 02:12:07 syd - .\" Added elmalias.1 man page. Updated Users.guide to mention elmalias. - .\" From: chip@chinacat.unicom.com (Chip Rosenthal) - .\" - .\" Revision 5.6 1993/04/12 01:42:16 syd - .\" Update for quotes - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.5 1992/12/24 23:48:05 syd - .\" make Nroff not use co symbol, just the word - .\" From: Syd via request from Tim Pierce - .\" - .\" Revision 5.4 1992/12/11 02:01:13 syd - .\" fix Obvious typos. - .\" From: dwolfe@pffft.sps.mot.com (Dave Wolfe) - .\" - .\" Revision 5.3 1992/10/17 21:59:56 syd - .\" Fix spelling of name - .\" From: Larry Philps - .\" - .\" Revision 5.2 1992/10/11 01:33:36 syd - .\" fixed a bug in the .lp macro in doc/tmac.n. The lp macros is always - .\" called with a unit specified for the second argument, however, the - .\" macro added a "n" unit specifier to that argument. The following patch - .\" fixes this problem. - .\" - .\" The various .guide books use the troff .tl macro liberally, but often the - .\" extra ' characters needed at the end of the lines to explicitly specify - .\" the trailing nulls fields were left off. The following patch fixes the - .\" erroneous lines. - .\" From: Larry Philps - .\" - .\" Revision 5.1 1992/10/03 20:51:50 syd - .\" Initial checkin as of 2.4 Release at PL0 - .\" .\" .\" .tm I trust we've run this through "tbl" first... --- 1,13 ---- ! .\" @(#)$Id: Users.guide,v 1.2 1998/10/20 19:57:20 hurtta Exp $ .\" .\" Users guide to the ELM mail system. .\" format with: .\" 'tbl Users.guide | troff tmac.n - > Users.fmtd' .\" ! .\" Elm is now in the public trust. .\" .\" (C) Copyright 1986,1987 Dave Taylor .\" (C) Copyright 1988-1992 Usenet Community Trust .\" .\" .tm I trust we've run this through "tbl" first... Index: elm2.4.ME+.50/doc/elm-help.0 *** elm2.4.25/doc/elm-help.0 Mon Apr 12 06:13:09 1993 --- elm2.4.ME+.50/doc/elm-help.0 Tue Nov 17 21:12:54 1998 *************** *** 1,5 **** ! Command Elm 2.4 Action , Display current message | Pipe current message or tagged messages to --- 1,5 ---- ! Command Elm 2.4ME+ Action , Display current message | Pipe current message or tagged messages to *************** *** 24,35 **** --- 24,38 ---- c Change to another folder d Delete current message ^D Delete messages with a specified pattern + ^E Extract PGP public keys from current message e Edit current folder f Forward current message + ^F Forget PGP passphrase g Group (all recipients) reply to current message h Headers displayed with message J Increment current message by one j, Advance to next undeleted message + ^K Mail PGP public key(s) K Decrement current message by one k, Advance to previous undeleted message l Limit messages by specified criteria *************** *** 50,55 **** --- 53,59 ---- ^T Tag messages with a specified pattern u Undelete current message ^U Undelete messages with a specified pattern + v View attachments in current message. x, ^Q Exit leaving folder untouched, ask permission if folder changed X Exit leaving folder untouched, unconditionally Index: elm2.4.ME+.50/doc/elm-help.3 *** elm2.4.25/doc/elm-help.3 Mon Apr 12 06:13:10 1993 --- elm2.4.ME+.50/doc/elm-help.3 Tue Nov 17 21:12:54 1998 *************** *** 1,5 **** ! Command Elm 2.4 Action Display the current message, --- 1,5 ---- ! Command Elm 2.4ME+ Action Display the current message, *************** *** 38,43 **** --- 38,44 ---- t Tag current message for further operations T Tag current message and go to next message u Undelete current message + v View attachments in current message. x Exit leaving folder untouched, ask permission if folder changed X Exit leaving folder untouched, unconditionally Index: elm2.4.ME+.50/doc/elm.1 *** elm2.4.25/doc/elm.1 Fri Dec 25 01:46:52 1992 --- elm2.4.ME+.50/doc/elm.1 Tue Nov 17 21:12:54 1998 *************** *** 74,79 **** --- 74,84 ---- .B " -a" Arrow - force the arrow cursor (instead of the inverse bar) .TP + .B " -A \fIfile\fR " + Attach file. Look also option -i. + To get this to work correctly, in file "$HOME/.elm/mime.types" + must be mapping from extension of filename to mime type. + .TP .B " -c" Checkalias - expand the following aliases and return. .TP *************** *** 88,102 **** --- 93,112 ---- .TP .B " -i \fIfile\fR " Include - include prepared file in edit buffer for sending. + Look also option -A. .TP .B " -?" Synonymous with the "-h" option. .TP .B " -k" + Ignored. + Keypad - force knowledge of HP terminal keyboard, to allow the use of the NEXT, PREV and HOME/SHIFT-HOME keys. .TP .B " -K" + Ignored. + Keypad + softkeys - enable use of softkeys on HP terminals only. .TP .B " -m" *************** *** 110,115 **** --- 120,129 ---- .TP .B " -v" Version - print version and configuration information. + .TP + .B " -w" + Write .elm/elmrc in startup. Useful if your .elm/elmrc + have errors. .TP .B " -z" Zero - don't enter \fBElm\fR if no mail is pending. Index: elm2.4.ME+.50/doc/elmrc-info *** elm2.4.25/doc/elmrc-info Fri Mar 11 23:19:26 1994 --- elm2.4.ME+.50/doc/elmrc-info Tue Nov 17 21:14:08 1998 *************** *** 1,5 **** ! #@(#)$Id: elmrc-info,v 5.11 1994/03/11 21:19:26 syd Exp $ ! # Elm Version 2.4 # #$HDR - magic line to mark comments to put at top of user's elmrc file. # --- 1,5 ---- ! #@(#)$Id: elmrc-info,v 1.2 1998/11/08 18:20:15 hurtta Exp $ ! # Elm Version 2.4ME+ # #$HDR - magic line to mark comments to put at top of user's elmrc file. # *************** *** 15,20 **** --- 15,21 ---- shell # The shell to use for shell escapes + # If not set value from passwd is used fullname # The full user name for outbound mail *************** *** 22,27 **** --- 23,31 ---- maildir # where to save my mail to, default directory is "Mail" + mailpermissions + # permissions to be given to newly-created saved mail files, default rw------- + tmpdir # where to place temporary files, default directory is "/tmp" *************** *** 88,96 **** --- 92,157 ---- weed # enable the weedout list to be read? + metamail + # This tells path of metamail program + # + # That version (Elm 2.4ME+) does not use Metamail's companion programs + # (suchs as mmencode) directly, but metamail package supposes that + # companion programs are installed to search path. + # + # Special values: none Don't call metamail + # metamail Don't call metamail if environment + # variable $NOMETAMAIL is defined + + noencoding + # This control makes it possible to send raw 8bit or binary data when the MTA + # doesn't support 8BITMIME and the -B8BITMIME option or + # BINARYMIME and the -BBINARYMIME option + # + # Possible values: + # pass-7bit (0) : Always convert 8BIT and BINARY data to QUOTED-PRINTABLE + # if the MTA doesn't support the -B8BITMIME and + # -BBINARYMIME options. + # pass-8bit (1) : Allow 8bit without -B8BITMIME, but binary data is + # encoded + # pass-binary (2) : Allow binary without -BBINARYMIME and and 8bit without + # -B8BITMIME + # + # It is STRONGLY recommended that it be left set to pass-7bit unless you know + # what you are doing!! + # + # Elm doesn't yet fully support BINARYMIME: it requires that text lines are + # terminated with \r\n, Unix's \n isn't sufficient! + # That is only partially implemented in this version of Elm. + # + # Quote from RFC 1830 (Experimental Protocol): + # SMTP Service Extensions for Transmission of Large and Binary MIME Messages + # + # It is important to note that when using BINARYMIME, it is + # especially important to ensure that the MIME message itself is + # properly formed. In particular, it is essential that text be + # canonically encoded with each line properly terminated with + # . Any transformation of text into non-canonical MIME to + # observe local storage conventions must be reversed before sending + # as BINARYMIME. The usual line-oriented shortcuts will break if + # used with BINARYMIME. + + nohdrencoding + # Don't do MIME part 2 (RFC 1522) encoding for 8-bit headers? + # Because 8-bit data is not allowed in headers, + # it is STRONGLY recommended that this is left set to OFF + noheader # when messages are copied into the outbound buffer, don't include headers? + # when replying + noheaderfwd + # when messages are copied into the outbound buffer, don't include headers? + # when forwarding + # (Notice that messages are not copiend to outbound buffer when + # mimeworward = ON) + + titles # display message title when displaying pages of message? *************** *** 164,170 **** # do we want to be able to mail out AT&T Mail Forms? userlevel ! # are we good at it? 0=beginner, 1=intermediate, 2+ = expert! names # just show the names when expanding aliases? --- 225,231 ---- # do we want to be able to mail out AT&T Mail Forms? userlevel ! # are we good at it? beginner (0), intermediate (1), expert (2) names # just show the names when expanding aliases? *************** *** 193,203 **** charset # name of Character Set used with MIME text/plain Content-type # US-ASCII is default. Be aware that if you use a national charset ! # elm probably needs metamail to display US-ASCII. displaycharset # name of Character Set which the display supports. This is independent ! # of the above "charset". compatcharsets # list of Character Sets, which are more or less a superset of US-ASCII --- 254,273 ---- charset # name of Character Set used with MIME text/plain Content-type # US-ASCII is default. Be aware that if you use a national charset ! # elm probably needs metamail to display US-ASCII. Special value ! # DISPLAY means that Charcter Set used in content-type is same ! # which the display supports (see: displaycharset) displaycharset # name of Character Set which the display supports. This is independent ! # of the above "charset". This is also copied to environment variable ! # MM_CHARSET when metamail is called. This is used only as default ! # on systems which supports locale. In these systems LC_CTYPE locale ! # is mapped to Character Set. Mapping is given in file elm.mimecharsets ! # on elm library directory or on file .elm/mime.charsets ! ! dsn-success ! # If TRUE, success Delivery-Status-Notifications are asked by default. compatcharsets # list of Character Sets, which are more or less a superset of US-ASCII *************** *** 247,250 **** textencoding # Type of encoding to be put into the MIME Content-Transfer-Encoding header. # Usual values are 7bit or 8bit. ! # NOTE: Elm will not encode your message based on this variable. --- 317,391 ---- textencoding # Type of encoding to be put into the MIME Content-Transfer-Encoding header. # Usual values are 7bit or 8bit. ! # ! # -- NOT Used in this version of elm (2.4 PL24 ME7) or ELM 2.4ME+ PLxx (25) ! # Now elm analyzes mail when sending and does that encoding of quoted-printable ! # when necessary. ! ! showpgppreamble ! # if this variable is ON, display text before PGP armor. ! ! showto ! # if this variable is ON, Elm will show who the message is to rather than ! # who it is from in the headers display. ! ! usepgppass ! # If ON, Elm will prompt you for your PGP passphrase and use it whenever ! # necessary to make it more convenient to hand PGP messages. ! ! keeppassfor ! # The number of seconds Elm should remember your passphrase. -1 means never ! # expire it. ! ! askpgpsig ! # If ON, Elm will prompt you for a username with which to sign the pgp ! # message ! ! pagealternative ! # Should Elm use "pager" for display one part of multipart/alternative, ! # if there are found one part which can be displayed. If not ! # set, Elm uses "pager" for display one part of multipart/alternative, ! # if are parts are displayable. ! ! pagemultipart ! # Should Elm use "pager" to display MIME multipart messages ! # with unknown subparts or with unknown subtype? ! ! mimeforward ! # This option controls how Elm will forward messages when MIME is defined. ! # ON: Causes Elm to send the message as a MIME attachment of type ! # MESSAGE/RFC822. Highly recommended when forwarding MIME ! # messages so that the receiver can correctly view the message! ! # OFF: Elm will include the forwarded message in your editor instead ! # of making it an attachment. ! ! quoteforward ! # Forwarded messages are quoted like replies, rather than included ! # with "Forwarded message" and "End of forwarded message" banners ! # above and below. This don't have effect if "mimeforward" is TRUE. ! ! require-mime-version-for-hdr-encoding ! # If set MIME Part 2 (RFC 1522) decoding is only done when MIME-version ! # header is present. Default: False ! ! require-mime-version-for-body-encoding ! # If set MIME Part 1 (RFC 1521) decoding is only done when MIME-version ! # header is present. Default: True ! ! background-wait-time ! # If = 0 wait mailer to completion ! # > 0 wait background_wait_time seconds for completion ! # if not completed in this time, let it go to background ! # ! # (completion of mailing is still reported if user have ! # not exited in that time when mailer have completed) ! ! pgp ! # What is the path for pgp? ! # Possible values include: ! # none Don't call pgp ! # /path Call metamail via that path, if it is executable ! ! lock-in-copy ! # If set, elm will lock a folder when trying to save or copy to it. ! # NOTICE: Does not work with dot-locking Index: elm2.4.ME+.50/doc/elmrc.samp *** elm2.4.25/doc/elmrc.samp Sat May 8 23:08:48 1993 --- elm2.4.ME+.50/doc/elmrc.samp Tue Nov 17 21:14:08 1998 *************** *** 1,193 **** ! #@(#)$Id: elmrc.samp,v 5.5 1993/05/08 20:08:43 syd Exp $ # # .elm/elmrc - options file for the ELM mail system # ! # Saved automatically by ELM 2.4 for Elm Development Group # ! # For yes/no settings with ?, ON means yes, OFF means no ! # where to save calendar entries ! calendar = ~/.Agenda ! # what editor to use ("none" means simulate Berkeley Mail) ! # '%s' can be used as the temporary filename for the outgoing message ! editor = none ! # the character to use in the builtin editor for entering commands ! escape = ~ ! # the full user name for outbound mail ! fullname = Elm Development Group ! # where to save received messages to, default file is "=received" ! receivedmail = $HOME/Mail/received ! # where to save my mail to, default directory is "Mail" ! maildir = /users/elmdev/Mail ! #directory to hold my temporary files to avoid NFS cross mount problems ! tmpdir = /users/elmdev/Mail/tmp ! # program to use for displaying messages ('builtin' is recommended) ! pager = builtin ! # prefix sequence for indenting included message text in outgoing messages... ! prefix = >_ ! # how to print a message ('%s' is the filename) ! print = lpr -Plw2 %s # attribution string for replies ('%s' is the author of original message) ! attribution = According to %s: ! ! # where to save copies of outgoing mail to, default file is "=sent" ! sentmail = /users/elmdev/Mail/mail.sent ! # the shell to use for shell escapes ! shell = /bin/csh ! # local ".signature" file to append to appropriate messages... ! localsignature = localsig ! # remote ".signature" file to append to appropriate messages... ! remotesignature = remotesig ! # do we want dashes above signatures? (News 2.11 compatibility and convention) ! sigdashes = ON ! # how to sort folders, "Reverse Sent" by default ! sortby = Reverse-Received ! # should the default be to delete messages we've marked for deletion? ! alwaysdelete = ON ! # should the default be to store read messages to the "received" folder? ! alwaysstore = ON ! # should the default be to keep unread messages in the incoming mailbox? ! alwayskeep = ON ! # should we use the "->" rather than the inverse video bar? ! arrow = OFF ! # should the message disposition questions be displayed(ON) or ! # auto-answered(OFF) with the default answers when we resync or change folders? ! ask = ON ! # would you like to be asked for Carbon-Copies information each msg? ! askcc = ON ! # automatically copy message being replied to into buffer? ! autocopy = OFF ! # threshold for bouncing copies of remote uucp messages... ! # zero = disable function. ! bounceback = 0 ! # save a copy of all outbound messages? ! copy = ON # do we want to be able to mail out AT&T Mail Forms? ! forms = OFF # should we keep folders from which all messages are deleted? ! keepempty = OFF ! # we're running on an HP terminal and want HOME, PREV, NEXT, etc... ! keypad = OFF # should we display the three-line 'mini' menu? ! menu = ON # would you like a copy of a message you send to an alias you are on? ! metoo = OFF # when using the page commands (+ - ) change the current # message pointer...? movepage = ON # just show the names when expanding aliases? ! names = ON # when messages are copied into the outbound buffer, don't include headers? ! noheader = ON # start up by pointing to the first new message received, if possible? ! pointnew = ON # prompt for a command after the external pager exits? ! promptafter = ON ! # emulate the mailx message increment mode (only increment after something ! # has been 'done' to a message, either saved or deleted, as opposed to ! # simply each time something is touched)? ! resolve = ON ! # save messages, incoming and outbound, by login name of sender/recipient? ! savename = ON ! # save outbound messages by login name of sender/recipient even if the ! # associated folder doesn't already exist? ! forcename = OFF ! # are we running on an HP terminal and want HOME, PREV, NEXT, etc...? ! # (this implies "keypad=ON" too) ! softkeys = OFF ! # set the main prompt timeout for resynching... ! timeout = 60 ! # display message title when displaying pages of message? ! titles = ON ! # are we good at it? 0=beginner, 1=intermediate, 2+ = expert! ! userlevel = 2 ! # enable the weedout list to be read? ! weed = ON ! # what headers I DON'T want to see, ever. ! weedout = "Path:" "Via:" "Sent:" "Date" "Status:" "Original" "Phase" ! "Subject:" "Fruit" "Sun" "Lat" "Buzzword" "Return" "Posted" ! "Telephone" "Postal-Address" "Origin" "X-Sent-By-Nmail-V" "Resent" ! "X-Location" "Source" "Mood" "Neuron" "Libido" "To:" "X-Mailer:" ! "Full-Name:" "X-HPMAIL" "Cc:" "cc:" "Mmdf" "Network-" "Really-" ! "Sender:" "Post" "Message-" "Relay-" "Article-" "Lines:" ! "Approved:" "Xref:" "Organization:" "*end-of-user-headers*" ! # alternative addresses that I could receive mail from (usually a ! # forwarding mailbox) and don't want to have listed... ! alternatives = dsidev!elmdev dsirel!elmdev elmdev@dsidev elmdev%dsidev ! # list of delivery precedences allowed, or empty to allow anything ! # precedence may be followed by optional ":priority" specification ! precedences = special-delivery:urgent air-mail:urgent first-class bulk junk ! # name of Character Set used with MIME text/plain Content-type ! # US-ASCII is default. Be aware that if you use a national charset ! # elm probably needs metamail to display US-ASCII. Elm tries to ! # know if that charset could display US-ASCII too, but the list ! # of us-ascii compatible charsets is not yet complete ! charset=US-ASCII ! # name of Character Set which the display supports. This is independent ! # of the above "charset". ! displaycharset=US-ASCII ! # list of Character Sets, which are more or less a superset of US-ASCII ! # so we know that we can display messages with charset=US-ASCII without ! # help of metamail ! compatcharsets = ISO-8859-1 ISO-8859-2 ISO-8859-3 ISO-8859-4 ISO-8859-5 ISO-8859-6 ISO-8859-7 ISO-8859-8 ISO-8859-9 # would you like to use termcap/terminfo ti/te entries? ! usetite = ON ! # Value by which message count is incremented while reading a new mailbox. ! # Setting this value to a number larger than one will speed up the time it ! # takes to read a large mailbox when using a slow terminal. ! readmsginc = 5 ! # time in seconds which Elm will wait after displaying a transient message ! # and before erasing it. Can be 0 or positive integer. ! sleepmsg = 2 --- 1,371 ---- ! #@(#)$Id: elmrc.samp,v 1.2 1998/11/08 18:20:15 hurtta Exp $ # # .elm/elmrc - options file for the ELM mail system # ! # Saved automatically by ELM 2.4ME+ PL50 (25) for Kari E. Hurtta # ! # ! # For yes/no settings, ON means yes, OFF means no. ! # ! # Lines beginning with "#" (like this one!) are comments. ! # ! # The "###" lines mean that you do not have a value set and the system ! # default will be used. The value shown was the default at the time ! # this file was created. Yes...the default could have changed between ! # then and now. ! # ! # how to sort the alias list, "Name" by default ! aliassortby = Name ! # name of editor to use for replies that have text ! ### alteditor = /bin/vi ! # alternative addresses that I could receive mail from (usually a ! # forwarding mailbox) and don't want to have listed... ! # alternatives = ! # should the default be to delete messages we've marked for deletion? ! ### alwaysdelete = OFF ! # should the default be to keep unread messages in the incoming mailbox? ! ### alwayskeep = ON ! # should the default be to store read messages to the "received" folder? ! ### alwaysstore = OFF ! # should we use the "->" rather than the inverse video bar? ! ### arrow = OFF ! # should the message disposition questions be displayed(ON) or ! # auto-answered(OFF) with the default answers when we resync or ! # change folders? ! ### ask = ON ! # would you like to be asked for Carbon-Copies information each msg? ! ### askcc = ON ! # If ON, Elm will prompt you for a username with which to sign the pgp ! # message ! ### askpgpsig = OFF # attribution string for replies ('%s' is the author of original message) ! attribution = %s: ! # automatically copy message being replied to into buffer? ! autocopy = ON ! # If = 0 wait mailer to completion ! # > 0 wait background_wait_time seconds for completion ! # if not completed in this time, let it go to background ! # ! # (completion of mailing is still reported if user have ! # not exited in that time when mailer have completed) ! ### background-wait-time = 2 ! ! # This is used to determine if the builtin pager should be used on some ! # messages even if you would usually use an external pager program. ! # There are two ways of determining whether the builtin pager should be ! # used. If you want any message that is shorter than "n" lines to use the ! # internal pager, set the parameter to "n". If you want the builtin ! # pager to be used if the message is "m" lines shorter than the number of ! # lines on your screen set the parameter to be "-m". Setting the ! # parameter to zero will result in the message always being sent through ! # your external pager. ! ### builtinlines = -3 ! # where to save calendar entries ! ### calendar = /home/people/hurtta/ ! # name of Character Set used with MIME text/plain Content-type ! # US-ASCII is default. Be aware that if you use a national charset ! # elm probably needs metamail to display US-ASCII. Special value ! # DISPLAY means that Charcter Set used in content-type is same ! # which the display supports (see: displaycharset) ! charset = DISPLAY ! # list of Character Sets, which are more or less a superset of US-ASCII ! # so we know that we can display messages with charset=US-ASCII without ! # help of metamail ! ### compatcharsets = ISO-8859-1 ISO-8859-2 ISO-8859-3 ISO-8859-4 ISO-8859-5 ISO-8859-7 ISO-8859-8 ISO-8859-9 KOI8-R X-ROMAN8 ! # list of options that can be configured at the "o)ptions" screen ! ### configoptions = ^_defspyv_am_un_ho ! # Should elm always ask the user to confirm before it appends ! # messages to any existing file? ! # This is used for both folders in the user's mail directory ! # and ordinary files. ! ### confirmappend = OFF ! ! # Should elm always ask the user to confirm before it creates ! # any new files? ! # This is used for both folders in the user's mail directory ! # and ordinary files. ! ### confirmcreate = OFF ! ! # Should elm ask for confirmation before it appends a message ! # to an existing file that is not a folder in the user's Mail ! # directory? ! # This is used for ordinary files only and does not affect ! # folders in the user's mail directory. ! ### confirmfiles = OFF ! ! # Should elm ask for confirmation before it creates a new folder ! # in the user's Mail directory? ! # This is used only for folders in the user's mail directory and ! # does not affect ordinary files. ! ### confirmfolders = OFF ! # save a copy of all outbound messages? ! copy = ON ! # name of Character Set which the display supports. This is independent ! # of the above "charset". This is also copied to environment variable ! # MM_CHARSET when metamail is called. This is used only as default ! # on systems which supports locale. In these systems LC_CTYPE locale ! # is mapped to Character Set. Mapping is given in file elm.mimecharsets ! # on elm library directory or on file .elm/mime.charsets ! ### displaycharset = ISO-8859-1 ! # If TRUE, success Delivery-Status-Notifications are asked by default. ! dsn-success = ON ! # name of editor for ~e command (when editor="builtin") ! ### easyeditor = emacs ! # what editor to use ("none" means simulate Berkeley Mail) ! ### editor = /bin/vi ! # the character to use in the builtin editor for entering commands ! ### escape = ~ ! # save outbound messages by login name of sender/recipient even if the ! # associated folder doesn't already exist? ! ### forcename = OFF # do we want to be able to mail out AT&T Mail Forms? ! ### forms = OFF + # The full user name for outbound mail + ### fullname = Kari E. Hurtta + # should we keep folders from which all messages are deleted? ! ### keepempty = OFF ! ! # The number of seconds Elm should remember your passphrase. -1 means never ! # expire it. ! ### keeppassfor = 300 ! ! # local ".signature" file to append to appropriate messages... ! ### localsignature = ! ! # If set, elm will lock a folder when trying to save or copy to it. ! # NOTICE: Does not work with dot-locking ! ### lock-in-copy = ON ! ! # where to save my mail to, default directory is "Mail" ! ### maildir = /home/people/hurtta/Mail ! # permissions to be given to newly-created saved mail files, default rw------- ! ### mailpermissions = rw------- # should we display the three-line 'mini' menu? ! ### menu = ON ! ! # This tells path of metamail program ! # ! # That version (Elm 2.4ME+) does not use Metamail's companion programs ! # (suchs as mmencode) directly, but metamail package supposes that ! # companion programs are installed to search path. ! # ! # Special values: none Don't call metamail ! # metamail Don't call metamail if environment ! # variable $NOMETAMAIL is defined ! ### metamail = /usr/freeware/bin/metamail # would you like a copy of a message you send to an alias you are on? ! ### metoo = OFF + # This option controls how Elm will forward messages when MIME is defined. + # ON: Causes Elm to send the message as a MIME attachment of type + # MESSAGE/RFC822. Highly recommended when forwarding MIME + # messages so that the receiver can correctly view the message! + # OFF: Elm will include the forwarded message in your editor instead + # of making it an attachment. + mimeforward = ON + # when using the page commands (+ - ) change the current # message pointer...? movepage = ON # just show the names when expanding aliases? ! ### names = ON ! ! # This control makes it possible to send raw 8bit or binary data when the MTA ! # doesn't support 8BITMIME and the -B8BITMIME option or ! # BINARYMIME and the -BBINARYMIME option ! # ! # Possible values: ! # pass-7bit (0) : Always convert 8BIT and BINARY data to QUOTED-PRINTABLE ! # if the MTA doesn't support the -B8BITMIME and ! # -BBINARYMIME options. ! # pass-8bit (1) : Allow 8bit without -B8BITMIME, but binary data is ! # encoded ! # pass-binary (2) : Allow binary without -BBINARYMIME and and 8bit without ! # -B8BITMIME ! # ! # It is STRONGLY recommended that it be left set to pass-7bit unless you know ! # what you are doing!! ! # ! # Elm doesn't yet fully support BINARYMIME: it requires that text lines are ! # terminated with \r\n, Unix's \n isn't sufficient! ! # That is only partially implemented in this version of Elm. ! # ! # Quote from RFC 1830 (Experimental Protocol): ! # SMTP Service Extensions for Transmission of Large and Binary MIME Messages ! # ! # It is important to note that when using BINARYMIME, it is ! # especially important to ensure that the MIME message itself is ! # properly formed. In particular, it is essential that text be ! # canonically encoded with each line properly terminated with ! # . Any transformation of text into non-canonical MIME to ! # observe local storage conventions must be reversed before sending ! # as BINARYMIME. The usual line-oriented shortcuts will break if ! # used with BINARYMIME. ! ### noencoding = pass-7bit ! ! # Don't do MIME part 2 (RFC 1522) encoding for 8-bit headers? ! # Because 8-bit data is not allowed in headers, ! # it is STRONGLY recommended that this is left set to OFF ! ### nohdrencoding = OFF ! ! # when messages are copied into the outbound buffer, don't include headers? ! # when replying ! ### noheader = ON # when messages are copied into the outbound buffer, don't include headers? ! # when forwarding ! # (Notice that messages are not copiend to outbound buffer when ! # mimeworward = ON) ! ### noheaderfwd = OFF ! ! # Should Elm use "pager" for display one part of multipart/alternative, ! # if there are found one part which can be displayed. If not ! # set, Elm uses "pager" for display one part of multipart/alternative, ! # if are parts are displayable. ! ### pagealternative = ON ! ! # Should Elm use "pager" to display MIME multipart messages ! # with unknown subparts or with unknown subtype? ! ### pagemultipart = OFF ! ! # program to use for displaying messages ('builtin' is recommended) ! pager = builtin + # What is the path for pgp? + # Possible values include: + # none Don't call pgp + # /path Call metamail via that path, if it is executable + pgp = /usr/local/pgp/bin/pgp + # start up by pointing to the first new message received, if possible? ! ### pointnew = ON ! ! # list of delivery precedences allowed, or empty to allow anything ! # precedence may be followed by optional ":priority" specification ! ### precedences = ! ! # prefix sequence for indenting included message text in outgoing messages... ! ### prefix = >_ ! ! # how to print a message ('%s' is the filename) ! print = mpage -o2p -Parth7bw %s # prompt for a command after the external pager exits? ! ### promptafter = ON ! # Forwarded messages are quoted like replies, rather than included ! # with "Forwarded message" and "End of forwarded message" banners ! # above and below. This don't have effect if "mimeforward" is TRUE. ! ### quoteforward = OFF ! # Value by which message count is incremented while reading a new mailbox. ! # Setting this value to a number larger than one will speed up the time it ! # takes to read a large mailbox when using a slow terminal. ! readmsginc = 25 ! # where to save received messages to, default file is "=received" ! ### receivedmail = =received ! # remote ".signature" file to append to appropriate messages... ! ### remotesignature = ! # If set MIME Part 1 (RFC 1521) decoding is only done when MIME-version ! # header is present. Default: True ! ### require-mime-version-for-body-encoding = ON ! ! # If set MIME Part 2 (RFC 1522) decoding is only done when MIME-version ! # header is present. Default: False ! ### require-mime-version-for-hdr-encoding = OFF ! ! # emulate the mailx message increment mode (only increment after ! # something has been 'done' to a message, either saved or deleted, ! # as opposed to simply each time something is touched)? ! ### resolve = ON ! # save messages, incoming and outbound, by login name of sender/recipient? ! ### savename = ON ! # where to save copies of outgoing mail to, default file is "=sent" ! ### sentmail = =sent ! # The shell to use for shell escapes ! # If not set value from passwd is used ! ### shell = $SHELL ! ! # if this variable is ON, display text before PGP armor. ! ### showpgppreamble = ON ! ! # if this variable is ON, Elm will show who the message is to rather than ! # who it is from in the headers display. ! ### showto = OFF ! # do we want dashes above signatures? (News 2.11 compatibility and convention) ! ### sigdashes = ON ! # time in seconds which Elm will wait after displaying a transient message ! # and before erasing it. Can be 0 or positive integer. ! ### sleepmsg = 2 ! # how to sort folders, "Reverse Sent" by default ! sortby = Reverse-Sent ! # Set the main prompt timeout for resynching... ! ### timeout = 600 ! # display message title when displaying pages of message? ! ### titles = ON ! # where to place temporary files, default directory is "/tmp" ! ### tmpdir = /usr/tmp/ + # If ON, Elm will prompt you for your PGP passphrase and use it whenever + # necessary to make it more convenient to hand PGP messages. + usepgppass = ON + + # are we good at it? beginner (0), intermediate (1), expert (2) + userlevel = intermediate + # would you like to use termcap/terminfo ti/te entries? ! ### usetite = ON ! # name of editor for ~v command (when editor="builtin") ! ### visualeditor = /bin/vi ! # enable the weedout list to be read? ! ### weed = ON ! ! # what headers I DON'T want to see, ever. ! ### weedout = "*end-of-user-headers*" Index: elm2.4.ME+.50/doc/mime.types *** /tmp/8255-very-long-file-name-NULL-comes-in-here Sun Nov 22 09:14:43 1998 --- elm2.4.ME+.50/doc/mime.types Tue Nov 17 21:12:54 1998 *************** *** 0 **** --- 1,27 ---- + # mime.types for Elm 2.4PL24 ME8 (or greater) + # + # Format of this file: + # + # + # Elm already has the following "builtin" definitions: + # ps application/postscript + # gif image/gif + # tiff image/tiff + # jpeg image/jpeg + # + # NOTE: data which is declared "application/octet-stream" will automatically + # have the "name=" option added to it, so you should not + # put it in your definitions. + + gz application/octet-stream + pbm image/pbm + pgm image/pgm + ppm image/ppm + taz application/octet-stream; type=tar; conversions=x-compress + tgz application/octet-stream; type=tar; conversions=x-gzip + tar application/octet-stream; type=tar + Z application/octet-stream + zip application/octet-stream; type=zip + + # PGP users might want the following + pgp application/pgp Index: elm2.4.ME+.50/doc/tmac.n *** elm2.4.25/doc/tmac.n Mon Apr 12 04:35:36 1993 --- elm2.4.ME+.50/doc/tmac.n Tue Nov 17 21:12:54 1998 *************** *** 1,4 **** ! .\" @(#)$Id: tmac.n,v 5.5 1993/04/12 01:35:29 syd Exp $ .\" The News macro package .\" .\" This is the macro package that is used to format news documents. It --- 1,4 ---- ! .\" @(#)$Id: tmac.n,v 1.1.1.1 1998/10/15 15:36:39 hurtta Exp $ .\" The News macro package .\" .\" This is the macro package that is used to format news documents. It Index: elm2.4.ME+.50/filter/Makefile.SH *** elm2.4.25/filter/Makefile.SH Sun Jun 6 20:35:08 1993 --- elm2.4.ME+.50/filter/Makefile.SH Tue Nov 17 21:12:54 1998 *************** *** 21,58 **** fi cat >Makefile < - # - # Revision 5.1 1992/10/03 22:18:09 syd - # Initial checkin as of 2.4 Release at PL0 - # - # # Variables # Variables established by Configure CC = $cc --- 21,34 ---- fi cat >Makefile < header file is now included ! * globally through hdrs/defs.h and the BROKE_CTYPE patchup is handled ! * there. Inclusion of was removed from *all* the individual ! * files, and the toupper() and tolower() routines in lib/opt_utils.c ! * were dropped. ! * From: chip@chinacat.unicom.com (Chip Rosenthal) ! * ! * Revision 5.7 1993/08/03 19:07:58 syd ! * Removed bogus string lockfile. ! * From: Jan.Djarv@sa.erisoft.se (Jan Djarv) ! * ! * Revision 5.6 1993/04/21 01:25:45 syd ! * I'm using Elm 2.4.21 under Linux. Linux has no Bourne shell. Each ! * user installs her favorite shell as /bin/sh. I use Bash 1.12. ! * ! * Elm invokes the mail transport (MTA) like so: ! * ! * ( ( MTA destination; rm -f tempfile ) & ) < tempfile & ! * ! * This form of command doesn't work with my Bash, in which any command ! * which is backgrounded ("&") gets its stdin attached to /dev/null. ! * ! * The below patch arranges for Elm to call the MTA thusly: ! * ! * ( MTA destination ! * ! * Revision 5.2 1992/12/11 02:16:08 syd ! * remove unreachable return(0) at end of function ! * From: Syd ! * ! * Revision 5.1 1992/10/03 22:18:09 syd ! * Initial checkin as of 2.4 Release at PL0 ! * ! * ! ******************************************************************************/ /** RESULT oriented routines *chuckle*. These routines implement the --- 1,16 ---- ! static char rcsid[] ="@(#)$Id: actions.c,v 1.2 1998/10/20 19:57:27 hurtta Exp $"; ! /****************************************************************************** ! * The Elm (ME+) Mail System - $Revision: 1.2 $ $State: Exp $ * + * Modified by: Kari Hurtta + ****************************************************************************** + * The Elm Mail System + * * Copyright (c) 1988-1992 USENET Community Trust * Copyright (c) 1986,1987 Dave Taylor ! *****************************************************************************/ /** RESULT oriented routines *chuckle*. These routines implement the *************** *** 78,85 **** --- 21,32 ---- #include #include #include + #include + extern int errno; #include "defs.h" + #include "elmlib.h" + #include "filter.h" #include "s_filter.h" *************** *** 97,103 **** */ FILE *pipefd, *tempfd, *mailfd; ! int in_header = TRUE, line_count = 0, mailunit, pid, statusp; char tempfile[SLEN], mailbox[SLEN], buffer[VERY_LONG_STRING], *cp; --- 44,51 ---- */ FILE *pipefd, *tempfd, *mailfd; ! int in_header = TRUE, line_count = 0, mailunit, pid,sig; ! S__ statusp; char tempfile[SLEN], mailbox[SLEN], buffer[VERY_LONG_STRING], *cp; *************** *** 107,113 **** date_n_user(), address); if (! show_only) { ! sprintf(tempfile, "%s.%d", filter_temp, getpid()); if ((tempfd = fopen(tempfile, "r")) == NULL) { if (outfd != NULL) --- 55,63 ---- date_n_user(), address); if (! show_only) { ! elm_sfprintf(tempfile, sizeof tempfile, ! FRM("%s.%d"), ! filter_temp, getpid()); if ((tempfd = fopen(tempfile, "r")) == NULL) { if (outfd != NULL) *************** *** 134,142 **** } if (strcmp(sendmail, mailer) == 0) ! sprintf(buffer, "%s %s %s", sendmail, smflags, address); else ! sprintf(buffer, "%s %s", mailer, address); /* * we do a fork/exec here to prevent evil people --- 84,96 ---- } if (strcmp(sendmail, mailer) == 0) ! elm_sfprintf(buffer, sizeof buffer, ! FRM("%s -oi -oem %s"), ! sendmail, address); else ! elm_sfprintf(buffer, sizeof buffer, ! FRM("%s %s"), ! mailer, address); /* * we do a fork/exec here to prevent evil people *************** *** 145,163 **** if ( (pid=fork()) > 0 ) /* we're the parent */ { ! wait(&statusp); ! if (statusp!=0) { fprintf(outfd, catgets(elm_msg_cat, FilterSet,FilterBadWaitStat, "filter (%s): Command \"%s\" exited with value %d\n"), ! date_n_user(),buffer,statusp); } } else if (pid==0) /* we're the child */ { ! /* use safe permissions */ setuid(user_uid); setgid(user_gid); --- 99,134 ---- if ( (pid=fork()) > 0 ) /* we're the parent */ { ! int w; ! while (pid != (w = my_wait(pid,&statusp)) && ! -1 != w || EINTR == errno); ! ! if (w != pid) { ! fprintf(outfd, ! "filter (%s): Command \"%s\", wait failed (erro = %d) \n", ! date_n_user(),buffer,errno); ! ! } else { ! int sig, stat; ! sig = convert_status(statusp,&stat); ! if (sig) { ! fprintf(outfd, ! "filter (%s): Command \"%s\" died with signal = %d\n", ! date_n_user(),buffer,sig); ! ! } else if (stat!=0) { fprintf(outfd, catgets(elm_msg_cat, FilterSet,FilterBadWaitStat, "filter (%s): Command \"%s\" exited with value %d\n"), ! date_n_user(),buffer,stat); } + } } else if (pid==0) /* we're the child */ { ! /* use safe permissions */ setuid(user_uid); setgid(user_gid); *************** *** 167,175 **** fprintf(outfd,catgets(elm_msg_cat, FilterSet,FilterPopenFailed, "filter (%s): popen %s failed!\n"), date_n_user(), buffer); ! sprintf(buffer, "(%s %s %s < %s ; %s %s) &", ! sendmail, smflags, address, tempfile, ! remove_cmd, tempfile); exit(system(buffer)); } --- 138,147 ---- fprintf(outfd,catgets(elm_msg_cat, FilterSet,FilterPopenFailed, "filter (%s): popen %s failed!\n"), date_n_user(), buffer); ! elm_sfprintf(buffer, sizeof buffer, ! FRM("(%s -oi -oem %s < %s ; %s %s) &"), ! sendmail, address, tempfile, ! remove_cmd, tempfile); exit(system(buffer)); } *************** *** 218,224 **** /** OTHERWISE it is to the current user... **/ ! sprintf(mailbox, "%s%s", mailhome, username); if (!lock()) { if (outfd != NULL) { --- 190,198 ---- /** OTHERWISE it is to the current user... **/ ! elm_sfprintf(mailbox, sizeof mailbox, ! FRM("%s%s"), ! mailhome, username); if (!lock()) { if (outfd != NULL) { *************** *** 256,263 **** } /* end if show only */ } ! save_message(foldername) ! char *foldername; { /** Save the message in a folder. Use full file buffering to make this work without contention problems **/ --- 230,238 ---- } /* end if show only */ } ! int save_message(foldername, namesize) ! char *foldername; ! int namesize; { /** Save the message in a folder. Use full file buffering to make this work without contention problems **/ *************** *** 269,275 **** statusp, ret; /* allow for ~ file names expansion in rules */ ! (void) expand_filename(foldername); if (verbose && outfd != NULL) fprintf(outfd, catgets(elm_msg_cat,FilterSet,FilterSavedMessage, --- 244,250 ---- statusp, ret; /* allow for ~ file names expansion in rules */ ! (void) expand_filename(foldername, namesize); if (verbose && outfd != NULL) fprintf(outfd, catgets(elm_msg_cat,FilterSet,FilterSavedMessage, *************** *** 292,298 **** if ((pid=fork()) > 0) /* we're the parent */ { ! wait(&statusp); return(statusp); } --- 267,273 ---- if ((pid=fork()) > 0) /* we're the parent */ { ! my_wait(pid,&statusp); return(statusp); } *************** *** 351,357 **** char filename[SLEN], buffer[SLEN]; int fdunit; ! sprintf(filename, "%s.%d", filter_temp, filter_pid); if ((fdunit = open(foldername, O_APPEND | O_WRONLY | O_CREAT, 0600)) < 0) { --- 326,334 ---- char filename[SLEN], buffer[SLEN]; int fdunit; ! elm_sfprintf(filename, sizeof filename, ! FRM("%s.%d"), ! filter_temp, filter_pid); if ((fdunit = open(foldername, O_APPEND | O_WRONLY | O_CREAT, 0600)) < 0) { *************** *** 407,418 **** "filter (%s): Executing %s\n"), date_n_user(), command); if (! show_only) { ! sprintf(buffer, "%s %s.%d | %s", ! cat, filter_temp, getpid(), command); if ( (pid=fork()) > 0) /* we're the parent */ { ! wait(&statusp); if (statusp!=0) { fprintf(outfd, --- 384,396 ---- "filter (%s): Executing %s\n"), date_n_user(), command); if (! show_only) { ! elm_sfprintf(buffer, sizeof buffer, ! FRM("%s < %s.%d"), ! command, filter_temp, getpid()); if ( (pid=fork()) > 0) /* we're the parent */ { ! my_wait(pid,&statusp); if (statusp!=0) { fprintf(outfd, *************** *** 455,461 **** char mailbox[SLEN]; int mailunit; ! sprintf(mailbox, "%s/%s", home, EMERGENCY_MAILBOX); if ((mailunit = open(mailbox, O_APPEND | O_WRONLY | O_CREAT, 0600)) < 0) { if (outfd != NULL) --- 433,441 ---- char mailbox[SLEN]; int mailunit; ! elm_sfprintf(mailbox, sizeof mailbox, ! FRM("%s/%s"), ! home, EMERGENCY_MAILBOX); if ((mailunit = open(mailbox, O_APPEND | O_WRONLY | O_CREAT, 0600)) < 0) { if (outfd != NULL) *************** *** 463,469 **** "filter (%s): Can't open %s either!!\n"), date_n_user(), mailbox); ! sprintf(mailbox,"%s/%s", home, EMERG_MBOX); if ((mailunit = open(mailbox, O_APPEND | O_WRONLY | O_CREAT, 0600)) < 0) { --- 443,451 ---- "filter (%s): Can't open %s either!!\n"), date_n_user(), mailbox); ! elm_sfprintf(mailbox,sizeof mailbox, ! FRM("%s/%s"), ! home, EMERG_MBOX); if ((mailunit = open(mailbox, O_APPEND | O_WRONLY | O_CREAT, 0600)) < 0) { *************** *** 498,500 **** --- 480,483 ---- tempfd = fdopen(mailunit, "a"); return((FILE *) tempfd); } + Index: elm2.4.ME+.50/filter/audit.c *** elm2.4.25/filter/audit.c Tue Feb 9 21:04:48 1993 --- elm2.4.ME+.50/filter/audit.c Tue Nov 17 21:12:54 1998 *************** *** 1,30 **** ! static char rcsid[] ="@(#)$Id: audit.c,v 5.3 1993/02/09 19:04:38 syd Exp $"; ! /******************************************************************************* ! * The Elm Mail System - $Revision: 5.3 $ $State: Exp $ * * Copyright (c) 1988-1992 USENET Community Trust * Copyright (c) 1986,1987 Dave Taylor ! ******************************************************************************* ! * Bug reports, patches, comments, suggestions should be sent to: ! * ! * Syd Weinstein - elm@DSI.COM ! * dsinc!elm ! * ! ******************************************************************************* ! * $Log: audit.c,v $ ! * Revision 5.3 1993/02/09 19:04:38 syd ! * use standard include method on time.h style includes ! * ! * Revision 5.2 1993/01/27 21:30:10 syd ! * fix include syntax ! * ! * Revision 5.1 1993/01/27 19:40:18 syd ! * Initial checkin ! * ! * ! ******************************************************************************/ /** allow date/time to be combined with username in verbose audit --- 1,16 ---- ! static char rcsid[] ="@(#)$Id: audit.c,v 1.2 1998/10/20 19:57:27 hurtta Exp $"; ! /****************************************************************************** ! * The Elm (ME+) Mail System - $Revision: 1.2 $ $State: Exp $ * + * Modified by: Kari Hurtta + ****************************************************************************** + * The Elm Mail System + * * Copyright (c) 1988-1992 USENET Community Trust * Copyright (c) 1986,1987 Dave Taylor ! *****************************************************************************/ /** allow date/time to be combined with username in verbose audit *************** *** 35,48 **** #include #include - #ifdef I_TIME - # include - #endif - #ifdef I_SYSTIME - # include - #endif - #include "defs.h" #include "filter.h" #include "s_filter.h" --- 21,29 ---- #include #include #include "defs.h" + #include "elmlib.h" + #include "filter.h" #include "s_filter.h" *************** *** 54,59 **** --- 35,41 ---- static char date[NLEN]; static char *stuff = NULL; time_t now; + int size; now = time(NULL); if (!strftime(date, NLEN, "%c", localtime(&now))) *************** *** 60,68 **** return (username); if (stuff) free(stuff); ! if (!(stuff = malloc(strlen(username) + strlen(date) + 3))) ! return (username); ! sprintf(stuff, "%s %s", date, username); ! return(stuff); } --- 42,49 ---- return (username); if (stuff) free(stuff); ! ! stuff = elm_message("%s %s", date, username); ! return(stuff); } Index: elm2.4.ME+.50/filter/filter.c *** elm2.4.25/filter/filter.c Mon May 30 19:31:55 1994 --- elm2.4.ME+.50/filter/filter.c Tue Nov 17 21:14:08 1998 *************** *** 1,62 **** ! static char rcsid[] ="@(#)$Id: filter.c,v 5.7 1994/05/30 16:31:40 syd Exp $"; ! /******************************************************************************* ! * The Elm Mail System - $Revision: 5.7 $ $State: Exp $ * * Copyright (c) 1988-1992 USENET Community Trust * Copyright (c) 1986,1987 Dave Taylor ! ******************************************************************************* ! * Bug reports, patches, comments, suggestions should be sent to: ! * ! * Syd Weinstein - elm@DSI.COM ! * dsinc!elm ! * ! ******************************************************************************* ! * $Log: filter.c,v $ ! * Revision 5.7 1994/05/30 16:31:40 syd ! * make getpwuid dependent on ANSI_C not posix flag ! * From: Syd ! * ! * Revision 5.6 1993/08/03 19:28:39 syd ! * Elm tries to replace the system toupper() and tolower() on current ! * BSD systems, which is unnecessary. Even worse, the replacements ! * collide during linking with routines in isctype.o. This patch adds ! * a Configure test to determine whether replacements are really needed ! * (BROKE_CTYPE definition). The header file is now included ! * globally through hdrs/defs.h and the BROKE_CTYPE patchup is handled ! * there. Inclusion of was removed from *all* the individual ! * files, and the toupper() and tolower() routines in lib/opt_utils.c ! * were dropped. ! * From: chip@chinacat.unicom.com (Chip Rosenthal) ! * ! * Revision 5.5 1993/06/06 17:58:20 syd ! * make white space skipping work for blank or tab ! * ! * Revision 5.4 1993/01/27 19:40:01 syd ! * I implemented a change to filter's default verbose message format ! * including %x %X style date and time along with username ! * From: mark@drd.com (Mark Lawrence) ! * ! * Revision 5.3 1992/11/15 01:40:43 syd ! * Add regexp processing to filter. ! * Add execc operator ! * From: Jan Djarv ! * ! * Revision 5.2 1992/11/07 16:20:56 syd ! * The first is that when doing a summary, macros are expanded when printing the ! * rule. IMHO they should be printed as with the -r option (i.e. %t is ! * printed as "