----------------------------------------------------------------------
ChangeLog for GoldED.
$Id: ChangeLog,v 1.25 1999/02/06 02:56:31 damware Exp $
----------------------------------------------------------------------

1999-02-05  Dirk A. Mueller <dmuell@gmx.net>

        * gepost.cpp: Applied second charset translation fix. Either
        it works or not, don't have time to investigate.

        * geutil.cpp: don't accept lines as quotes if "-" found in 
        prefix. 

        * various: Extended @oname etc. syntax, patches by
        Alexander S. Aganichev <asa@lan.eed.miee.ru>.

        * geinit.cpp: "Splash screen" is gone. I'm simply fed up
        getting around 5 mails/day with reports about it.

1999-02-03  Leonid Lisovsky <lly@aha.ru>

        * geusrbse.cpp: Borland and Watcom compilers wouldn't accept
        such update_screen() function redefinition after previous
        addressbook updates patch.

1999-02-01  Dirk A. Mueller <dmuell@gmx.net>

        * various files: Integrated patches by Alexander S. Aganichev,
        <asa@lan.eed.miee.ru>. Adding NICKNAME, major addressbook
        updates (file sharing, packing, adding current writer). 

1999-01-31  Dirk A. Mueller <dmuell@gmx.net>

        * geline.cpp: GoldED doesn't use header information or
        INTL/FMPT/TOPT in echomail areas any more. This should fix the
        problems squish-users had. 

1999-01-29  Dirk A. Mueller <dmuell@gmx.net>
        
        * gectrl.cpp: fixed segfault, clean code a bit up. 

	* getpls.cpp: changed back to old code, as the new one seems
	to produce more problems than the old one.
	
1999-01-28  Dirk A. Mueller <dmuell@gmx.net>

        * various: Added INTERNETDOMAIN keyword. Defines the "domain"
        part of to-be-generated Message-IDs in News/Email areas. Done
        to be really compatible to Son-Of-RFC1036. Defaults to
        "username%domain.de".

        * geline.cpp: Again changed the "originating address" guess
        code to the following priority:

          - If available, use INTL/FMPT/TOPT.
          - else, if available, use Origin aka
          - else, if available, use Msgid aka
          - else leave it as default (0:0/0 aka unknown).

        I hope that this code now will work in all cases.
	
1999-01-27  Dirk A. Mueller <dmuell@gmx.net>

        * gesoup.cpp: Probably fixed SOUPEXPORTMARGIN-Bug (broken 
        headerlines during export if headerline longer than margin). 
        Please test.
	
1999-01-22  Dirk A. Mueller <dmuell@gmx.net>

        * (LL): Ged/W32 won't quit when pressing CTRL-BREAK any more.

        * geline.cpp: Added Mailing-List to the list of known RFC
        headerlines. The value of that headerline is ignored. 

        * geinit.cpp: Workaround added for those users who can't read
        the manual. Some of them still declare areas in front of their
        aka setup. GoldED will now check for surch areas "without" 
        default aka and add the main aka (if any akas defined). This 
        won't work in all cases, but I don't care. 

1999-01-16  Dirk A. Mueller <dmuell@gmx.net>

        * Again changed the Message-ID format. This should have been
        the last change for a while. 

        * Fixed a bug in ImportSoup() Routine, which generated a
        broken MSGID/REPLY kludge if there was no such headerline.

1999-01-13  Dirk A. Mueller <dmuell@gmx.net>

        * Added resource/golded.xpm. It's a GoldED-icon, whatelse ;)

        * There was a bug in xlat translation of header fields when
        quoting or forwarding (Pavel Bazanov, banzay@banzay.spb.ru). 

        * Other small things I forgot. 

        * gesoup.cpp: MSGID/REPLY kludges will be now placed (nearly) 
        at the top of the message. Some msgbase linkers won't find 
        them if they are placed below the first 1024 bytes of a 
        message. Anyway, it should be a speedup-change for msgbase
        linkers. 

        * removed the ugly debugging code from gesoup.cpp I added four
        days ago, as the bug is fixed. 

        * gectrl.cpp: Some rearrangements were needed because of the 
        bugfix below, so I also modified the Message-ID-generation.
        In my opinion, the Message-ID is still RFC1036 compatible,
        however I'm not sure. Feedback is welcome. I did this because
        I didn't like the old style at all ;)

        * gesoup.cpp: Major bugfix in ProcessSoupMsg() routine. The
        calculated MSGID/REPLY crc values where almost always com-
        pletely wrong. This is now fixed, however, MAILTOSS now only 
        parses the internet headerlines and not the whole email for
        technical reasons. I think this is still okay. 

        * geline.cpp: KludgeMESSAGE_ID(..) produced a leading space
        in msg->msgids. Fixed.

        * gesoup.cpp: removed CvtMSGIDto...(), as it's never used. 

1999-01-12  Dirk A. Mueller <dmuell@gmx.net>

        * gefile.cpp: Fixed Y2K bug in filebrowser display.
   
        * gepost.cpp: Don't convert to upper case when attaching
        files under UNIX. Patched by Leonid Lisovsky <lly@aha.ru>.

1999-01-09  Dirk A. Mueller <dmuell@gmx.net>
        
        * gesoup.cpp: Added debug-output to stderr to find a Bug in
        the CvtMessageId...() routine. I know it's nerving, sorry. 

        * gemrks.cpp: MARKYOURMAIL didn't work any more with the new
        advanced search function.

        * gearea.cpp: Also the Lnx-version will now show uppercase
        letters when fuzzy-searching for an area. 

1999-01-08  Dirk A. Mueller <dmuell@gmx.net>

        * EXPERIMENTAL: Golded/Lnx is now linked dynamically.

1999-01-07  Odinn Sorensen <odinn@goldware.dk>

	* gesoup.cpp: Added support for Mailing-List header.

1999-01-07  Dirk A. Mueller <dmuell@gmx.net>

        * gesoup.cpp: GoldED/Lnx wil now write correct REPLIES files.
        Previously, it created a uppercase GOLDMAIL.MSG, but wrote
        the lowercase-version of it into the REPLIES file. Now fixed
        to uppercase. 

1999-01-01  Dirk A. Mueller <dmuell@gmx.net>

        * Probably fixed mysterious crashes while editing in the
        addressbook. Transfered ChgAttr() to gehdre.cpp instead of
        doing it in geusrbse.cpp.

1998-12-28  Dirk A. Mueller <dmuell@gmx.net>

        * Notework.txt contained a new entry but Changelog not?!

        * Logo screen wait is disabled of -NOSCAN or if any keys are
        given on commandline (it'll break the soupexport/import
        batches for usenet mail transfer otherwise).

1998-12-27  Dirk A. Mueller <dmuell@gmx.net>

        * Version numbering: Okay, any number as long as it'll be
        released soon. 12/31/98 or 01/01/99 seems to be a good day.

        * Logo screen wait is a bit long for me, anyway it'll need to
        be made configurable... (maybe with the HAPPYBIRTHDAY FRIEND
        switch? ;)).

        * geusrbse.cpp: Added possibility to delete entries in the
        addressbook. It is not yet what users will expect (as they're
        marked only as deleted and get never hidden/removed), but it
        works. You can now permanently forbit GoldED to lookup certain
        users address in addressbook.

1998-11-23  Odinn Sorensen <odinn@goldware.dk>

        * It's version 3.0.0, not 2.9.0. Use PREVERSION "pre-" and
        POSTVERSION "dam-1" if you want. I call mine pre-3.0.0os-1.

        * Added a (to be configurable) wait to keep the logo screen a
        short while.

        * Fixed a shift in the language strings.

1998-11-23  Dirk A. Mueller <dmuell@gmx.net>

        * Didn't have time to convert my changes to the now
        opengoldware tree. I'm pretty much busy with school .. :-(

        * Improved Styleguide a little bit.

        * Added a nice (?) new startup screen layout. Hope you'll find
        it useful. Note I don't like the possibility to change the
        product name by editing the new mygolded.h definitions, so I
        added a "warning" as a comment ;)

        * Added vendor information to mygolded.__h & golded3.cpp. This
        will help to identify "inofficial" compiles done by third
        parties from "official" ones. I'm wondering it isn't better to
        add a small startup delay (configurable), because its quite
        tricky to grab the startup-screen when using the program on
        fast machines.

1998-11-21  Odinn Sorensen <odinn@goldware.dk>

        * Moved the main() function and overlay stuff to a new
        gemain.cpp file.

        * Revamped the product name and version number system. You
        must now add a mygolded.h file of your own. In this file you
        can change some strings to personalize the product name and
        version strings used in GoldED. Added a template for the file
        in mygolded.__h. Rename this to mygolded.h and modify it if
        you need to. Do NOT cvs add a mygolded.h !!!


1998-11-14  Odinn Sorensen <odinn@goldware.dk>

        * Switched to GPL.


1998-11-01  Dirk A. Mueller <damware@gmx.net>

        * READgotouplink and READgotodownlink removed, as they're
        replaced with READgotorepl* since years.

        * Converted gckeys.cpp to the new parsing sheme to enhance
        speed and reduce code size. This is also needed to implement
        the "online-help always shows current key layout" feature.
        It'll probably break the Dosversion, as it enlarges the data
        segment much. Much stupid typing work :-(


1998-10-31  Dirk A. Mueller <damware@gmx.net>

        * geline.cpp, ScanKludges(): Don't detect line as origin if
        already detected a tearline or tagline.

        * geqwks.cpp, gesoup.cpp: If -noscan and soupimport/qwkimport
        used, GoldED didn't update goldlast.lst to disk on exit.
        Previous code faild because startupscan_success is always
        false under such a condition.

        * gcalst.cpp, AL::ReadGoldLast(): If there was no matching
        Area for a given goldlast.lst entry, function lost sync,
        because it didn't skip the variable-length fields (message
        marks). Worked around that problem, pretty quick and dirty
        (need to be adjusted if file format changes). :-/


1998-10-25  Odinn Sorensen <odinn@goldware.dk>

        * gehdre.cpp: A const string was passed to non-const parameter
        for lookup_addressbook(). Work-around by copying to a buffer
        and back.

        * gesoup.cpp: Fixed a bug in the soup import. The Linux
        version core dumped when checking addresses for mailing lists.

        * Started ChangeLog.

----------------------------------------------------------------------

