home *** CD-ROM | disk | FTP | other *** search
-
- SENDMAIL:
-
- .31
- FINALLY -- arbitrary size headers can exist in the article
- without trashing memory (normally a problem only for lists).
-
- .30
- This corrects the invalid address patch to properly deal with
- "`'\"/: <>" as it should (from 0.28). Thanks Ty.
-
- .29
- A couple of fixes to patch 27. The checkuser patch was depending
- on a file to be in a specific array that the "pipes" code used
- to. This caused a conflict. No more.
-
- A pointer to an array that was no longer valid was used, now
- the array is more global.
-
- .28
- sendmail used to check all incoming addressed for use
- of special alias command ("><!") which was OK. However,
- at the same time it checked for ':/' which are used in
- X.400 address. That check prevented X.400 addresses
- from being passed through rmail.
-
- Now, rmail allows those address to pass until it may
- become time to deliver one with a '/' or ':', at which
- point OneMailToUser() doesn't allow it (security fault).
-
- .27
- Add in Ty Sarna's checkuser patches. See the man page.
-
- Change fatal error messages to ulog(-1) from ulog(0).
-
- .26
- Had to special case the optimization for delivering
- local mails without going through UUX.
-
- If rmail is called with a hostname, EVEN THE LOCAL
- HOSTNAME, then UUX has to be called. This is required
- for aliases to be expanded.
-
- This only seems to happen for batch SMTP variants. All
- UUCP's I can see do rmail lines properly (ie, for the
- destination host, the rmail line doesn't contain any
- hostname, just the destination username).
-
- For example, incoming mail to mbs@adastra could be either
- of:
-
- rmail mbs <D....
-
- or
- rmail adastra.cvl.va.us!mbs <D....
-
- The second case has to make an extra path through UUX.
- The first case does not.
-
- Get whoever is responsible to fix your mailer.
-
- .25
- When MaxRMailLen was greater than zero, the calculated
- length of the RMAIL line was wrong. A '+=' was just an
- '='. I hate when that happens.
-
- .24
- Part of patch .22 was to use UUX for creating files in
- UUSpool: to be sent out via UUCP. An optimization part
- of that patch was to deliver all local mails directly --
- without having to go through the UUX/UUXQT process.
-
- Well, based on the way SENDMAIL/RMAIL are designed, that
- just won't work. Alias expansion, when a hostname is
- attached, is only done for RMAIL. To make SENDMAIL
- strip hostnames in the proper places is a non-trivial
- effort.
-
- This patch removes the optimization. It also REQUIRES
- a matching patch in UUX which will allow the local
- NodeName to be recognized without it being in the L.Sys
- file, and the filename built accordingly (a X. file
- instead of a D. file, and no C. file at all).
-
- .23
- Properly check if the localhost is where mail should be
- delivered, without requiring localhost to be in L.Sys.
-
- Call UUX with "-g A" for Mail, instead of allowing it to
- default to "N" (for News).
-
- Only do special "From " processing if the "From " header
- has "remote from" in it. This fixes a postprocessing date/
- year problem.
-
- If "LogMail" is defined in Config, then log a record about
- all outgoing mail in the MAILLOG file (which also may be
- defined in Config, default UUSpool:Maillog). Idea/patch
- from Jeff Van Epps.
-
- .22
- Added in support for wusel's RFC-976 SmartUUCP option.
-
- No longer lose the first header if it isn't "From " and
- properly process mail that starts with a ">From ". SMail 3.1.*
- and sendmail 8.* both do this differently.
-
- .21
- Check malloc() results.
-
- Reformat. Added debugging when PROCDEBUG defined at compile time.
-
- Message-id is now RFC-822 compliant.
-
- If a message-id header exists, a new one is NOT added.
-
- A new config option StripUUCP causes addresses to have ".UUCP"
- stripped prior to resolving domain stuff. Primarily for the
- uunet.ca folks...
-
- Bounces now return up to 65535 lines of the bounce.
-
- Now calls UUX instead of building UUSpool: files internally.
-
- Errors to stderr instead of stdout.
-
- A locally stored message now has a proper "id" in the Received
- header.
-
- Comparisons for headers are no longer case sensitive.
-
- Deal properly with TIMEZONES of the form -0500 and +0400.
- Partial timezones are still not supported, but this would
- be easily added if anyone needs it.
-
- Use standard list processing features instead of the "funky"
- stuff that Matt depended on.
-
- If RealName is empty, then an empty "()" is no longer appended.
-
- Multiple From: and Subject: lines are no longer submitted in
- the final mail. The last one in applies.
-
- If the UserName supplied via "-f" has a "@" or "!" in it, then
- "@node.domain" is no longer appended; the "-f" argument is taken
- as a FQDN for the From: header.
-
- .20
- OwnDevUnit.Library added
-
- Message-ID: fix made (mwm), messages would be given different
- ID's for multiple receipients.
-
- 'From ' line skip stuff removed, don't know why it was put in
- there in the first place!
-
- .19
- recompiled under DICE 2.06.35, which now supports 2.0 local
- variables for getenv() calls. Thus, under 2.0, sendmail
- fully supports a multi-user capability.
-
- .18
- aliased entries disable mail-bouncing. that is, they are
- treated as outgoing mail.
-
- .17
- fixed security hole for incomming mail to specific users.
- Was not properly checking the 'username' for being a path
- instead of a user name.
-
- Added Martin J. Laubach's bouncing code to sendmail. Sendmail
- will now bounce if no domain or L.sys entry exists for a
- machine route. TO USE, MUST BE ENABLED WITH CONFIG ENTRY,
- See UUMAN:SendMail
-
- Added Henning's DomainPath enhancement to sendmail, see
- UUMAN:SendMail (puts fully domained machine names in header
- paths)
-
- .16
- fixed bug in To: line generation, does not put any
- | or > alias entries on the To: line !
-
- .15
- fixed bug in local mail going though the node, i.e. you
- email to yourmachine!fubar instead of just fubar, bug was
- introduced in .14
-
- .14
- fixed Domain bug, MF entries specifying multiple paths,
- such as: '* MF UU able!beta' screwed up be prepending the
- rmail line address with a '!'.
-
- names queue files differently, FINALLY does it more
- intuitively for local names.
-
- .13
- no longer logs 'sending mail to' messages for the contents of
- aliases when said contents are hidden (like huge mailing
- lists).
-
- .12
- fixed U line in X. file
-
- .11 (1.08 release)
- Aliases now expand, unless '::' is used instead of ':' after
- the alias name, in which case they do not. Using :: is useful
- when one is mailing to large groups.
-
- AmigaUUCP now generates a multi-rmail line in the X. file,
- which means that if you have multiple recipients AmigaUUCP
- will generate only one batch file per UUCP node. Thus, if
- you send a message: To: fo!a, fo!b then AmigaUUCP generates
- only one batch for this.
-
- Additional european timezones added.
-
- .10
- now uses the VERSION macro for the mail-header version string
-
- .09
- mail-append bug fixed that caused dmail to screw up some times
- queue file naming fixed (now names remote files after local
- machine name, NOT after remote machine name !!!!!!)
-
- -raw option added for data files that contain no headers at all
-
- .08
- multiple users on rmail line allowed, and bug in MakeNode() fixed.
- Tnx Chris.
-
- .07
- duplicate case in options .. 'r' .. -R realname now.
-
- 1.06 release
-
- .06
- Uses USERNAME and REALNAME ENV: variables if they exist, else
- falls back to Config entries.
-
- .05
- Intermediate
-
- .04
- Runs program specified by 'MailReadyCmd' Config entry when
- local mail sent.
-
- Fixes local mail. sending mail to yourself 'myself!user' instead
- of just 'user' now works properly. Also fixes bug in some
- versions of UUPC (e.g. Mac UUPC) that do not remove the
- destination machine from the rmail line.
-
- Without this fix improperly formatted rmail lines from other
- UUPCs will route through our default route instead of to us.
-
- .03
-
-