PC Media 3

home *** CD-ROM | disk | FTP | other *** search

/ PC Media 3 / PC MEDIA CD03.iso / share / os2 / geog0615 / note0614.doc < prev next >

Wrap

Text File | 1994-06-13 | 56.9 KB | 1,243 lines

---------------------------------------------------------------------- --- TECHNICAL NOTES --- This file contains detailed and often highly technical notes about the news/fixes/changes of each version. Symbols used: - Bugfix. + New feature. ! Changed feature. * Information/clarification. ---------------------------------------------------------------------- Notes for GoldED 2.42.A0614, GAMMA release 14. june 1994. ---------------------------------------------------------------------- * This gamma release is mainly a bugfix and update release of the GoldED 2.42.G0214 version, which was getting a bit old. Most of the fixes in this release should have been out long ago, but time flew, and development of the next version (2.50) went way ahead. Much development has been made on the next version. It is currently in alpha test (latest is A0531), and looks very good. It was planned to have been released as a beta on 6. june, but I lost a weeks worth of development in a nasty harddisk crash that day. The setback means that it will be at least a week or two before the 2.50 beta can be released. Several features and fixes of 2.50 have been included in this 2.42 gamma, most notably the capability to have up to 65500 nodes in the nodelist index. Many AREAFILE's have been updated to support the latest versions of various other software. * The name of the DOS extended 32-bit version has been changed from "GoldED/P32" to "GoldED/386". This is easier to understand and similar to the costum of most other programs that are compiled for the 386 processor. The GoldED/386 EXE file is now named GED386.EXE and GoldNODE/386 is named GN386.EXE. + New or updated AREAFILE's: Ezycom 1.10, FastEcho 1.41, FMail 0.98, IMAIL 1.50, InterMail 2.25/2.27, LoraBBS 2.33, QuickBBS (Goldbase). The new include feature in Squish 1.10 is not supported yet, sorry. ! The GoldED nodelist index now supports up to 65500 nodes in all versions, including the regular DOS version! Hopefully it will take a few years before anyone hits that limit! :-) The new index is not 100% compatible with the old indexes, so I have changed the index file extensions once again. These are the new nodelist index files: GOLDNODE.GXA Address index (index into .GXN). GOLDNODE.GXL List of nodelist files + timestamps. GOLDNODE.GXN Name index. The old index files are obsolete and should be deleted to save space. This NOT done automatically. These are the files that should be deleted: GOLDNODE.?IX (.AIX .LIX .NIX) Old 16-bit indexes. GOLDNODE.IX? (.IXA .IXL .IXN) Old 32-bit indexes. The new index is a hybrid of these two. It's a 16-bit index (.GXA contains index values in the range 0-65499 which fit into a 16-bit unsigned integer), but uses the same name index (.GXN) and list of nodelists (.GXL) as the old 32-bit index. The advantage of staying with a 16-bit index is that it uses less memory in the regular DOS version. The price to pay for the larger index capability is speed, at least in the DOS version. Due to the 64k memory segments of DOS, I had to use so-called "huge" pointer to handle memory allocations larger than 64k. The use of huge pointers causes a significant overhead, resulting in more than doubling the compile time of the DOS version :-( Try using the 386 version instead if you can. Regarding V7 and FD nodelists: I am sorry to disappoint those who want GoldED to support V7 and FD nodelist indexes directly. I am researching the possibilities for supporting these indexes. It is likely that I will support them in a future version, but they both suffer from design flaws that makes it difficult to support full nodelist browsing on both name and address. I'm investigating ways to work around that. We'll see what the future brings. + Ezycom 1.10 msgbase changes implemented. NOTE: You MUST have the EZY environment variable set, otherwise GoldED cannot detect the version. GoldED defaults to version 1.02 unless it can find CONFIG.EZY and detect the new version. - The MSGID address interpreter code assumed that the address was a FidoNet-style address if it started with a non-zero number. Unfortunately some internet-style addresses also start with a non-zero number. The code is now changed so that it assumes a FidoNet-style address if it starts with a non-zero number _and_ the first character after the number is a ':' (indicates a FidoNet-style zone). ! Default for EDITCRLFTERM keyword changed. GoldED will now by default only terminate lines (paragraphs) with a single CR instead of a CR+LF combination. - Fixed bug in reading of Ezycom msg attributes. The Netmail Pending attribute was not read correctly, which could cause an unsent msg to be marked as if was sent, except the Snt attribute was missing. - Internal editor: If a line was almost as wide as the margin and Ctrl-Right (WordRight) was pressed at the last word on the line, the editor would report a Range Check error. Fixed. - Internal editor: The overwrite mode did not work correctly. It allowed overwrite past the end of the line, which could cause a crash or protection violation. - Made adjustments to the quote reflow function, so that it will not scramble tables nearly as much as before. It will attempt to detect when lines begin with the same two characters (for tables, typically a line char and a space). This will also help with indented text that is hard terminated on each line. ! GoldED now tests sharing/locking capability by creating a temporary file in the GOLDPATH and attempting to lock it. If the lock fails, locking capability is assumed to be non-existing, and GoldED will handle msgbases as single-user, no multitask/network. - JAM areas no longer need SHARE loaded or "SHAREMODE 0" in single-user setups. + Speeded up JAM scan a bit more. The .JDT file is no longer opened when just scanning areas. - JAM areas were double-scanned, thereby slowing down scanning a whole lot. Not so noticable with a good cache, but measurable. Watch it zip through 'em now! :-) ! The soundcard (SB, SBPRO etc.) support has been removed. I cannot support and fix it well enough at this time, so I prefer to just remove it. Multimedia features will return some day, when more important and relevant features have been done first... ! The ANSI screen output mode has been removed. Too slow to be useful. Too buggy. Wasted bytes in the exe file. + It is now possible to carbon copy to more than one person with the same address. Useful for a moderator who wants to notify both a BBS sysop and a user on the BBS about something. - If you had AREAAUTONEXT disabled, and hit Alt-J to jump to next area, the auto-next feature would be enabled until you exit GoldED. Fixed. - The quotebuffer feature (ctrl-w) did not work with JAM areas. - Fixed the cursor hide function to work properly under DESQview, when running in a DV window with a 50 lines text display. - Fixed the setting of the border (overscan) color. + GoldED now understands definitions of JAM areas in an AREAS.BBS file. JAM areas in an AREAS.BBS file are marked with a '!' in front of the msgbase filename. ! GoldED now gives an error message if it finds a random item outside a GROUP, while compiling GOLDRAND.CFG. - If a random item was defined in GOLDRAND.CFG _before_ any GROUP statement, GoldED could crash or cause a protection violation. Fixed. + When commandline parameter -VV is used, GoldED now also reports what it finds in GOLDRAND.CFG. ! GoldED now uses the defined FIDOMSGTYPE for *.MSG areas in AREAFILE GEcho, except if the GEcho version is 1.02 or newer. FIDOMSGTYPE defaults to Opus format. By using the Opus format, the "funny" zone/point in the msg headers is eliminated. If you use GEcho 1.02 or newer, and see "funny" zone/point in msg headers, you should enter GSETUP and change Miscellaneous, GEcho options, MSG compatibility to Opus. The Opus format stores date/time information where the FTS-1 format stores zone/point information. That is what causes the "funny" numbers. Most software (notably FrontDoor) uses the Opus de-facto standard, but some software (notable GEcho) is set to use FTS-1 by default. This mismatch causes problems. NOTE: If GEcho is set to use a *.MSG format other than Opus, GoldED will now warn you when compiling the configuration. - If you tried to ROT13 in an empty area (zero msgs), GoldED could crash. Fixed. ! For registered users, the main address is no longer displayed together with the name and serial number in the startup screen. - If an AREAS.BBS file was used, and the configuration compiled, GoldED could hang the system or cause an exception error upon exit (DOS), or cause an access violation (OS/2). Fixed. - Fixed some inconsistencies with the JAM scanning files: The NET/ECHOMAIL.JAM files were updated, even when only the Rcv attribute was set. The NETMAIL.JAM file was updated for every msg read. The SEMAPHORE EXPORTLIST was updated for JAM areas. The NET/ECHOMAIL.JAM files are now only updated for a msg, if the msg is marked "local but not sent". NOTE: GoldED does NOT remove entries in the JAM scanning files when deleting msgs! If it should do so, deletion of msgs would be slowed down a lot. In my opinion, the mail processor should quietly ignore invalid entries. - GoldED is now a bit more tolerant of errors in the JAM subfields. + Added new parameter to the IF/ELSEIF conditional directive. The new parameter is "INOS2" and it is always true for GoldED/2 and true for GoldED/DOS when OS/2 is detected. I use it to include a special set of colors when running under OS/2. - A slight change in the behavior of the MsgAPI in Squish 1.10 caused GoldED to exit with errors in certain circumstances. + Greatly improved Hudson/Goldbase scanning speed when QBBSSCAN Rescan is used. It was very slow before. It is recommended to use this option instead of the default, especially with a large msgbase, because it does not require nearly as much memory. - Fixed several bugs in the Goldbase support: Mistake in the MSGHDR.DAT structure. Due to an oversight while converting the Pascal structures, my implementation was two bytes too short. This caused GoldED to report index errors when used on a correct Goldbase msgbase. Board numbers above 255 were truncated internally to a number below 256, when writing a msg to the msgbase. The lastread number was alway reset to zero when leaving an area. ---------------------------------------------------------------------- Notes for GoldED 2.42.G0214, GAMMA release 14. february 1994. ---------------------------------------------------------------------- ! This release of GoldED/P32 uses a different "stub", which does not suppress the DOS extenders' loading banner. To suppress the banner, issue the following command in DOS (put it in your AUTOEXEC.BAT): SET DOS4G=quiet See the DOS4GW.DOC for further information about configuring the DOS extender. - Fixed nasty bug in crossposting to JAM areas. The crossposted msgs got empty msg bodies. + For JAM areas, the NETMAIL/ECHOMAIL.JAM files are now updated when new messages are written or changed in JAM netmail/echomail areas. The files are written/updated in the HUDSONPATH. If you don't use a Hudson msgbase and haven't defined a HUDSONPATH, the HUDSONPATH defaults to the GOLDPATH. At the time of writing, the NETMAIL/ECHOMAIL.JAM files are not a part of the official JAM specs, but they are used in RA2 and most JAM compatible mail processors to specify the msgs that need to be exported from the JAM msgbase files. + In the nodelist browser, the bottom line of the window now shows the base filename of the nodelist where the current node entry was found. If the nodelist is missing, or the node was found in a USERLIST, nothing is shown. + Added support for the Goldbase msgbase format. Goldbase is basically the Hudson format with some size limits increased to allow virtually unlimited number of msgs, up to 500 areas instead of 200, and breaks the 16MB barrier for the MSGTXT file. The extension of the msgbase files is ".DAT" instead of ".BBS". If you want to use Goldbase areas, you must specify the path to the MSG*.DAT files with the new keyword GOLDBASEPATH. If you do not specify a GOLDBASEPATH, it will default to the HUDSONPATH (see also below). Currently GoldED does not support any AREAFILE to get Goldbase area configuration, so each Goldbase area must be defined manually by using the AREADEF or AREA keywords. With AREADEF, use "Gold" or "Goldbase" as the msgbase format parameter. With AREA, use 'G' as the msgbase format parameter. Examples: AREADEF QUICKBBS "Support" I Echo Goldbase 456 2:234/159 AREA QUICKBBS "Support" GE 456 See the docs for details about the general syntax of the AREADEF and AREA keywords. The implementation of Goldbase in GoldED was done by formatting my existing Hudson sourcecode for easy conversion to Goldbase with a simple AWK script. Therefore the Goldbase code should work just as well as the Hudson code. I had to make a number of changes to the Hudson source and various other places in the GoldED source to allow this easy conversion and the addition of Goldbase. Therefore it is possible that bugs could have crept in. However I have not seen any problems with the Hudson code to date, and I'm using several Hudson areas daily. The Goldbase code is not tested much, mainly because I don't have access to/have not installed other programs that use Goldbase. Therefore you should treat the Goldbase code as "beta stuff". If you experience problems with the Goldbase or Hudson implementation in GoldED, please send me a bug report as soon as possible. NOTE: If you use both Hudson *and* Goldbase msgbases in the same GoldED configuration, you may have to increase the FILES setting in your CONFIG.SYS. When scanning mail, GoldED keeps complete sets of msgbase files open (7 files per msgbase) for the Hudson and Goldbase msgbases. So therefore I recommend adding 10 to your FILES in CONFIG.SYS to be on the safe side, at least if you currently only have FILES=30. I have FILES=50, which works for me. + Updated AREAFILE IMAIL support for IMAIL 1.40G3. Now also reads JAM area configuration. + Updated AREAFILE GEcho support for GEcho 1.02. Now also reads JAM area configuration. The *.MSG areas are now set to the type specified in the GEcho 1.02 setup: Main->Miscellaneous->GEcho options->MSG compatibility (Fido or Opus). In earlier versions of GEcho, the default is Fido (FTS-1 compatible with Zone/Point information instead of Opus date/time stamps). + To improve area scanning speed, I have moved the scanning message down into the statusline instead of the window in the middle of the screen. The constantly changing size of that window was quite a bottleneck, as you will notice. In the window there is now instead the message that was in the statusline before ("Scanning Messagebase (Press ESC to skip)"). - The AREAFILEs for IMAIL (reported) and perhaps FastEcho and FMail were broken in the 32-bit versions GoldED/2 and GoldED/P32. The original C structures used for these AREAFILEs contains so-called bitfields, which may be implemented differently by different compiler makers. Apparently the Watcom C++32 compilers which are used for GoldED/2 and GoldED/P32, implement bitfields slightly differently than the Borland C++ OS/2 32-bit compiler (used for the earlier versions of GoldED/2). I have now fixed the all the C structures I could find with potential problems, so they compile correctly with the compilers I currently use. - The JAM support needed SHARE or network loaded in order to be able to lock the msgbase. I have now changed the code so that it will not attempt to lock the msgbase if the SHAREMODE keyword is set to zero (0). - Removed the automatic setting of intense colors (if enabled), because it caused snow in OS/2 fullscreen. If you have problems with blinking colors after a windowed/fullscreen mode switch, try a shell and exit from the shell immediately. TIP: You can use the EXTERNUTIL feature to call a non-existing program, and thereby assign a key to enable intense colors. (Not tested). - Fixed bugs in GoldED/2 and GoldED/P32 where the zone in msgs could get reset to zero if the msg contained FMPT/TOPT kludges. - The Enter key on the numeric keyboard did not work as expected. In the IE it would make a note character (actually a CR character). ! Added keyword HUDSONPATH to specify the path to the Hudson msgbase files. This keyword is simply an alias for the old QBBSPATH keyword. If not specified, the HUDSONPATH/QBBSPATH will default to the GOLDPATH. - Fixed the code that reads AREAS.BBS. It should no longer crash GoldED when you have many areas in the file. NOTE: This fix is not tested intensively. Let me know if it works okay with many areas. ---------------------------------------------------------------------- Notes for GoldED 2.42.G1219, GAMMA patch release 19. december 1993. ---------------------------------------------------------------------- - Fixed nasty bug in the JAM support. If you read a message to you in a JAM area, and GoldED set the Received attribute, the message text would be lost if you later returned to the message :-( What happens is that the text offset field in the header gets reset to the end of the .JDT file. It is possible to recover the lost text, although it is probably not really worth the bother. It requires that you have intimate knowledge of the JAM header format, know how to use a binary hex editor, and can remember enough of the the lost message text to find the location of it in the .JDT file. Then all you have to do is change the text offset field back to the correct value (the offset in the .JDT file). In addition you should also subtract one from the text length. GoldED output a terminating NUL char that was not necessary. - Fixed a bug in GoldNODE/2 (and maybe GoldNODE/P32), which caused several zones to be missing in the nodelist index, to be replaced with a large number of duplicate nodes. I haven't the faintest idea what went wrong, or even how I fixed it - the code I have changed to make it work had very little to do with what seemed to be the problem. But it seems to work now :-? <puzzled> To be on the safe side, I have included BDF's for all versions of GoldNODE. - Due to a mistake in my setup of the Watcom compiler for OS/2, the wrong version of the OS/2 toolkit was used. This could cause odd behaviour with some operations. I haven't actually seen any problems with GoldED/2 myself, but I discovered some problems with getting and setting the date/time of files when compiling another program with Watcom. I have recompiled GoldED/2 and GoldNODE/2 with the correct version of the toolkit this time. ---------------------------------------------------------------------- Notes for GoldED 2.42.G1218, GAMMA release 18. december 1993. ---------------------------------------------------------------------- + Added JAM msgbase support at last!!! Example cfg definitions: AREA ECHO.JTEST1 "Test 1" JE M:\JAM\JTEST1 AREADEF ECHO.JTEST2 "Test 2" 0 Echo Jam M:\JAM\JTEST2 . (Loc) As you can see, the syntax is very similar to Squish definitions. Please read the document GOLD_JAM.DOC for important information about the implementation of the JAM msgbase support in GoldED. + New GoldED/P32 replaces GoldED/P16. GoldED/P32 is a 32-bit DOS extended version, compiled with Watcom C++32 9.5a. It has been tested successfully in a DOS box under OS/2 2.1 and under DesqView 2.32 with QEMM 6.03. GoldED/P32 needs the DOS4GW DOS extender version 1.92. DOS4GW 1.92 is distributed separately in the archive D4GW_192.ZIP. If you already have a copy of DOS4GW, it may work with GoldED/P32, even if isn't version 1.92. In fact I had an old version 1.0 in my path before 1.92 and it worked fairly well before I discovered the mistake. GoldED/P32 is able to use the new GoldNODE 32-bit index files (.IX*) which can contain more than 32700 nodes. GoldNODE/P32 and GoldNODE/2 generates this new index. GoldED/P32 can also use the old 16-bit index files, but defaults to use the new index files, if present. GoldED/P32 needs about 1.5 to 2 megabytes of free RAM to run. * GoldED/P16 has been dropped completely. Most of the problems reported about P16 is due to a buggy DOS extender and compiler, and the makers of these don't seem to be interested in releasing better versions. I have better things to do than worry about buggy compilers that doesn't get fixed, so this is the end of GoldED/P16 as we know it today. * GoldED/2 and GoldNODE/2 are now compiled with Watcom C++ 9.5a instead of Borland C++ 1.0 for OS/2. + Added support for the new xMail 1.0 mail processor's area configuration file. Use "AREAFILE xMail [areapath]". The optional [areapath] points to where AREAS.XM can be found. Only the AREAS.XM file is read, not any other file. Currently only tested with JAM areas. Note that auto-added areas are often stored with long filenames that are invalid with the FAT file system. To overcome this problem, GoldED internally truncates the base filename to 8 chars, but it does not check for duplicate filename that may result from the truncations. + Updated FastEcho support to version 1.3x. Now also gets JAM areas. Untested! + Updated RemoteAccess 2.x support to get JAM areas. Untested! + Changed behaviour when switching to "read marked" mode. Previously GoldED would go to the next marked msg. Now it will stay on the current msg if it is marked, or go to the next marked msg if not. + Added a new border type, number 7. If border type 7 is used, all border/linedraw characters are pure 7-bit ASCII instead of highbit IBM linedraw characters. This border type was added on request from Asian users who have a character set where the IBM linedraw characters are national characters instead. It is unlikely to be useful for others. Here is how to set all border types to type 7: COLOR AREA BTYPE 7 COLOR ASK BTYPE 7 COLOR BACKGROUND BTYPE 7 COLOR BRAG BTYPE 7 COLOR HEADER BTYPE 7 COLOR HELP BTYPE 7 COLOR INFO BTYPE 7 COLOR MENU BTYPE 7 COLOR READER BTYPE 7 COLOR STATUS BTYPE 7 Add the above to your color setup in GOLDED.CFG. Note that the help file contains IBM linedraw characters. These are NOT converted automatically. You have to change them yourself in the GOLDHELP.CFG file, if necessary. - There have been reports of the cursor remaining visible after saving a msg. I haven't seen this myself, but I have now added a cursor-hide in a strategic place to ensure that the cursor goes away when expected. + If intense colors are enabled, GoldED now sets intense colors every time the statusline is updated. This is a help in multitasking environments that doesn't save the intensity state when switching between tasks. - In the msg browser, the date is now cut to fit if for some reason it is too long. + GoldED now recognizes the FAX flag in the FLAGS kludge and shows it in the header display if found. * TIP: You can use the EXTERNUTIL feature to setup keys to view or print an attached fax. Use something like this in GOLDED.CFG: EXTERNUTIL 11 c:\zfax\zfax.com 2 @subject ; print fax EXTERNUTIL 12 c:\zfax\zfax.com 4 @subject ; view fax And this in GOLDKEYS.CFG: @F11 ExternUtil11 ; Press Alt-F11 to print the fax @F12 ExternUtil12 ; Press Alt-F12 to view the fax Or choose your own key assignments and fax view/print utils. Note that this assumes that the fax file is listed in the subject line like an attached file. + Added temporary new keyword: EDITORVERSION <version>. Due to the large number of bugreports about range check errors in the new internal editor, I have decided to put in the old editor as an option in addition to the new editor. The old editor is the default. If you want to take your chances with the new editor, use this in your GOLDED.CFG: "EDITORVERSION 2". When the new editor is properly bugfixed, the old editor and this keyword will be removed again. + Enhanced keyboard is now auto-detected by checking a bit in low memory (DOS versions only of course). Use "KEYBEXT No" if your keyboard seems to lock up. - Detection of a real, non-emulated, Hercules card was buggy. Fixed. - The msg size display could disappear under some circumstances, after editing a msg. Fixed. - GoldED/P16 paused for 10 seconds when testing if locking was possible and SHARE or network was not loaded. Fixed. Obsolete, since P16 is no longer released. - When trying to generate a userlist (Ctrl-L) with GoldED/2, it would dump with a protection violation SYS3175. Fixed. + Added -Q (for Quiet) switch to GoldNODE. If -Q is used, GoldNODE will not display anything on the screen. This can improve compile speed a lot, especially on systems with slow text mode (such as a textmode windowed session under Windows or OS/2). + Changed all highbit IBM chars to pure ASCII chars in GoldNODE. Too bad it looks ugly :-( - GoldNODE/2 could not compile more than 32767 as advertised. Fixed. This fix is now tested successfully with a nodelist collection which contains more than 32767 nodes. - Fixed Sound Blaster support to work with the CT-VOICE driver in Sound Blaster 16. If you have a SB-compatible non-SB soundcard, but have been unable to make a CT-VOICE driver work with GoldED, try again now. Note: Some users still have problems getting GoldED to work with Sound Blaster 16. The fix was tested successfully with a borrowed SB16 in two of my own machines. The same card would not work with GoldED when it was returned and used in the lenders own machine. I have no idea why it doesn't work :-( ---------------------------------------------------------------------- Notes for GoldED 2.42.G1121, GAMMA Patch Release 21. november 1993. ---------------------------------------------------------------------- - Fixed bug in AREAFILE FastEcho. It caused protection violations in GoldED/2. It did not affect the DOS versions. - Fixed some bugs in the msg marking functions. For example, Alt-S,N didn't always mark all new msgs, and could cause crashes. - While writing a msg in the internal editor, GoldED would lock the Hudson msgbase unnecessarily. This prevented mail tossers from tossing incoming mail in multitask/network environments. Fixed. - GoldED/P16 had problems sharing the Ezycom msgbase files properly. It opened the files in Read/Write mode, where it should have been ReadOnly. Fixed. The regular DOS and OS/2 versions did not have this problem, which was caused by a bug (or undocumented behavior) in the TopSpeed C library. - Fixed minor bug in the file request function for FRQEXT's. The first word of the description would be skipped if there was only one space between filename and description. + The DISPATTACHSIZE and EDITAUTOATTACH features now recognize network drives beginning with "\\" (without a drive: in front) as valid drives. Untested, but should work. In general GoldED currently doesn't support the "\\" netdrives very well. Better support is planned. + Added support for calling of external utilities. This new feature can be used to "filter" msgs, for example calling PGP or other crypting utilities. A new keyword has been implemented for the definition of external utilities: EXTERNUTIL <utilno> [-options] <commandline> The <utilno> must be in the range 1 to 24 for utilities that will be called with key definitions (see below). Higher numbers can be used for utilities that are called from the EDITSAVEMENU (if enabled). See the EDITSAVEUTIL keyword for details. The default options are those specified with the EXTERNOPTIONS keyword. With [-options], you can change the those options locally for specific utils. See the EXTERNOPTIONS keyword for a list of valid options. The <commandline> specfies the DOS or OS/2 commandline you want to execute. This works in the same way as for the external editor and spellchecker. See the manual for details about this, especially if you want to execute batchfiles. In the <commandline> you can use @tokens to transfer information from the msg to the commandline. The @tokens are the same as for templates (see the manual), and in addition you can use two other @tokens: @path The GOLDPATH, including a trailing backslash. @file The full filename of the message file (GOLDED.MSG) that will be written to disk before the utility is called. Examples: EXTERNUTIL 1 c:\4dos\4dos.com /c c:\crypt\encrypt.bat @file EXTERNUTIL 2 c:\4dos\4dos.com /c c:\crypt\decrypt.bat @file To call a defined external utility, you must assign a key to it. There are 24 new keyboard commands you can use in GOLDKEYS.CFG for this purpose: ExternUtilNN Where NN is in the range 01 to 24 _with_ leading zero. Examples: F11 ExternUtil01 F12 ExternUtil02 Then if you press F11, you would call external utility number 1 and so on. The ExternUtilNN keywords ONLY work in reader mode, not in the internal editor or any other place. HOW IT WORKS: 1. Just before the external utility is called, GoldED writes the current message text to the GOLDED.MSG file in the GOLDPATH. The file is written as a textfile with each line CR-LF terminated. The content is exactly as you see it on the screen, which means that kludges are only included if you have enabled kludge viewing. 2. GoldED clears the screen and then calls the utility after swapping itself out of memory (if swapping is enabled or relevant for the version). 3. The utility can now load and process the @file, or do anything else you want. It doesn't have to have anything to do with the current msg. You could call a spreadsheet, a game, whatever. But I think this feature will mainly be used for utilities that process the @file. If the utility processes the @file, it could write the changes back to the @file. 4. After returning from the utility, GoldED reloads the @file and displays it just as if it was the current message. For example, if your utility was a decrypter and the msg was encrypted, you would now see the decrypted msg. Neat eh? :-) 5. If you want to make the reloaded text permanent (save it in the msg), you can use the Change Msg function and immediately choose "Save Message" from the Editing menu (if enabled with EDITMENU Yes). This is in fact the method you could use if you wanted to EN-crypt a msg (however, it is easier to use the EDITSAVEUTIL method, see below). The QUOTESPACING feature can interfere with encoded msgs that include the '>' character at the beginning of lines, by automatically inserting blank lines before and after the lines with '>'. I have therefore changed the quotespacing default from YES to NO. The file GOLD_PGP.DOC contains batchfile examples for using this new feature with PGP. Many thanks to Thomas Horsten 2:234/150 for showing me how to setup PGP and for alpha testing the feature. !!! IMPORTANT !!! From FidoNet Policy 4.07 (section 2.1.4): "[..] Therefore, encrypted and/or commercial traffic that is routed without the express permission of all the links in the delivery system constitutes annoying behavior." So be careful with this feature! + New keyword EXTERNOPTIONS <-options>. Defines the default options for the EXTERNUTIL's (see above). Valid options are: -Cls * Clear screen. -Swap * Swap GoldED out of memory before shelling. -Cursor * Cursor in shell. -Reload * Reload the message file (@file). -Pause Pause for keypress before returning to GoldED. -PauseOnError * Pause only if utility errorlevel is nonzero. Plus the reverse options with a "No" prefix, for example -NoCls. The default options are marked with an asterisk (*). + New keyword EDITSAVEUTIL <utilno> <"L menu text">. Defines the external utilities that will be added to the EDITSAVEMENU (if enabled). The menu text is inserted in the menu. The first two characters of the menu text are the "hotkey" letter that will be highlighted in the menu text, plus a space. Example: EDITSAVEUTIL 1 "S PGP Sign the msg" EDITSAVEUTIL 2 "E PGP Encode the msg" EDITSAVEUTIL 3 "D PGP Decode the msg" If you have changed the default language in this menu (in GOLDLANG.CFG), then make sure the highlight letters don't clash. With EDITSAVEUTIL definitions and EDITSAVEMENU enabled, you can directly call an external utility to do things like encoding msgs before saving them. + Added conditional directives to the GOLDED.CFG reader. You can now use the new keywords IF, ELSEIF and ENDIF to control or exclude parts of your GOLDED.CFG under certain conditions. The IF and ELSEIF directives are identical. ELSEIF is provided for cosmetic reasons. Valid parameters for IF/ELSEIF are: Yes, No, True, False, DOS and OS/2. Example: IF DOS EXTERNUTIL 1 c:\4dos\4dos.com c:\crypt\encode.bat @file EXTERNUTIL 2 c:\4dos\4dos.com c:\crypt\decode.bat @file ELSEIF OS/2 EXTERNUTIL 1 c:\4os2\4os2.exe c:\crypt\encode.cmd @file EXTERNUTIL 2 c:\4os2\4os2.exe c:\crypt\decode.cmd @file ENDIF The example speaks for itself, I think... + Added enhancement of the Random System feature. It is now possible to specify an external file as the source of random items in groups. This helps to avoid duplicating items such as origins that are should be share between several otherwise different groups. The random items that can use external files are: NETNAME, ORIGIN, PLAY, TEARLINE, TEMPLATE, USERNAME, WHOTO. An external file is specified by a '@' character in front of the filename, where the item text would normally be placed. More than one item with an external file can be specified. Example: GROUP FidoNet: TEARLINE @tearline.lst ORIGIN @origins1.lst ORIGIN @origins2.lst The contents of external files must be one item on each line. Empty lines and comments are not allowed in an external file. Items must be specified exactly as they would appear in GOLDRAND.CFG, but without the leading keyword. Example: Pointless point... Don't point at me! etc... With this feature, it is now easy to create large collections of things like origins, and then share them between any number of groups without duplication, or even share them between several users with different setups. NOTE: The default path for the external files is the GOLDPATH, unless you manually specify a path. - Detection of color/monochrome mode VGA could, and did, fail in some circumstances. On some EGA adapters, GoldED would only detect 24 lines. Fixed. - When using BIOS output mode (SCREENUSEBIOS Yes), the screen would scroll one line every time the statusline was updated. Fixed. - The DISPMSGSIZE feature handled parameters "no" and "bytes" wrongly. Fixed. - Fixed a couple of bugs concerning long lines in the new internal editor. Changed the default value of QUOTEMARGIN from being screenwidth dependent (width-5) to 65. ---------------------------------------------------------------------- Notes for GoldED 2.42.G1115, GAMMA Patch Release 15. november 1993. ---------------------------------------------------------------------- - The file import function in the internal editor would use the wrong drive when using a drive letter and wildcards (like A:\*.* if you wrote C:\*.*). Fixed. ---------------------------------------------------------------------- Notes for GoldED 2.42.G1114, GAMMA release 14. november 1993. ---------------------------------------------------------------------- * THIS RELEASE DOES _NOT_ HAVE JAM OR GOLDBASE SUPPORT! JAM and Goldbase support will be provided in a later beta release. Stay tuned to the GOLDED and/or GOLDED.BETA echoes for the latest status reports and release announcements. + The file request function has been much improved: 1. It will now handle filenames such as ">FILE.ZIP" correctly instead of including the invalid '>' character in the filename. 2. If there were more than one known FRQEXT on a line, it would use the first of the FRQEXT's found, which was often the wrong one of course. Now it will take the first filename on the line which has a known FRQEXT. 3. It will now detect the file announcement format used by the Ticket utility even if Ticket's tearline has been removed or overwritten. The detection is based on the position of certain elements on the line. 4. When extracting the description of files detected with the FRQEXT's, it will now skip the first words after the filenames if they begin with a digit or the '[' or '-' characters. It will skips words until the first occurrence of a word that does not match these criteria. This produces a clean description in most cases, such as this one: SOMEFILE.ZIP 123.456 93/11/13 [00] - Some description Which will produce "Some description", instead of "123.456 93/11/13 [00] - Some description" as previous versions would. Obviously this cuts the FILES.BBS maintenance to a minimum when requesting files for your BBS. There is one thing the file request function can't do: It cannot produce more than one message. This is on the TODO, of course. + The internal editor has finally been rewritten! All the known bugs are gone, and the new editor works much better. Editing of quotes have been improved - you can now both split and merge quoted lines with "intelligent" handling of the quote initials string. NOTE: At this time, the new internal editor is not 100% finished and debugged. There are still some inconsistencies and minor bugs, but the editor is fully usable and safe - I use it myself. The block functions (anchor, cut, paste) are currently disabled, because they are still too buggy. + New keyword: EDITCHARPARA [']<char>[']. Defaults to ' ' (space). This keyword defines the character GoldED displays at the end of paragraphs in the internal editor. This is where the CR character will be placed once the msg is saved. See below. + New keyword: EDITCHARSPACE [']<char>[']. Defaults to ' ' (space). This keyword defines the character GoldED displays when it should display a space character in the internal editor. See below. * The new keywords EDITCHARPARA and EDITCHARSPACE were added to aid me while rewriting and debugging the new internal editor. By redefining them to visible characters instead of spaces, I could see if strange things were happening while inserting or deleting characters. This was a great help. Personally I now always use redefinitions to ASCII 20 (a paragraph sign) and CP437 250 (a small dot). ! The file extensions for compiled configuration files have changed. GoldED will now use the extension .G16 for 16-bit versions (DOS and DOS/P16) and extension .G32 for 32-bit versions (OS/2 and later DOS/P32). These extensions were previously named .GED (DOS) and .GEO (OS/2). Remember to delete the obsolete *.GED and/or *.GEO files from your GOLDED directory, so that they wont take up disk space. + New keyword: DISPSOFTCR <yes/no>. Defaults to "no". If enabled, GoldED will treat the so-called Soft-CR character (141d, 8Dh) just like any other displayable character, instead of ignoring it like linefeed chars (LF). Note that by enabling this feature, you _disable_ the character translation feature that uses the Soft-CR as an escape character. This feature was added to help users in countries which use the Soft-CR character for other purposes like 2-byte characters in Japan. NOTE: The Ezycom msgbase format requires the Soft-CR to terminate each line. Therefore this feature is unlikely to be useful to Ezycom sysops. - When using AREALISTSORT to sort by descriptions, the sort was case sensitive. The sort is now case IN-sensitive. + In the Random System, the AREAFILEGROUPS setting (if set to Yes) no longer prevents you from making groups for specific echoes. This means that if you have for example a "GROUP X", corresponding to group X in your GEcho setup, but wish to override the origin or other items for some of the group X echoes, you can now put the GROUP's for these specific echoes before the GROUP X definition. Example: GROUP SOMEECHO ORIGIN "Some origin" GROUP X ORIGIN "X origin" Here you will get the origin "Some origin" in the SOMEECHO echo, even if SOMEECHO is included in the X group in your setup. This was not possible in previous versions - the GROUP X definition would always override any specific definition. + New keyword: "DISPMSGSIZE <bytes/kbytes/no>". Defaults to "bytes". When enabled, this keyword displays the msgbody size in the lower left side of the header. The location of the size display is not necessarily final. The size displayed is for the message body text only, the header and nul-terminator (and anything that may lurk beyond it) is excluded from the calculation. NOTE: This feature currently only works when _reading_ msgs. While editing a msg in the internal editor, this feature is disabled - however, the size will be displayed when you are in the Save msg menu (if EDITSAVEMENU is enabled). + New keyword: "REPLYLINK <chain/direct>". Defaults to "chain". If set to "direct", GoldED will link your reply directly to the original message. If set to "chain", it will link to the last message in the reply chain. The default ("chain") is how GoldED has done it in all previous versions. The advantage of the "direct" linking method is that you can easily find the the original message the reply was for. Unless of course you have later re-linked using a chain-linking replylinker utility. I can recommend the utility SQLINK by David L. Nugent. SQLINK links Squish areas using the MSGID/REPLY kludges and makes direct links instead of chain-linking on the subject line like most other replylinkers do. There are probably also similar replylinkers for other msgbase formats, I just don't know them. NOTE: Direct linking is not always useful or practical. It can be very hard to follow a discussion with more than one participants, because the links become branched trees instead of simple chains. Currently GoldED is not designed to follow the branched direct links, but such a feature is on the TODO, although with low priority. + The SCREENBLANKER keyword now accepts an additional parameter: If you add the word "AllBlank" after the seconds parameter, GoldED will not display the moving box, but leave the screen completely blank. Example: SCREENBLANKER 180 AllBlank "AllBlank" must be placed _after_ the seconds, or it won't work. NOTE: The "Midnight Bug" has not yet been found or fixed :-( + The 32-bit versions of GoldED and GoldNODE (currently only the OS/2 versions) are no longer restricted to a maximum of 32700 nodes in the nodelist index. A slightly different nodelist index is produced by the 32-bit GoldNODE. The index is NOT backward compatible. The new index files have extensions .IXA, .IXN and .IXL. For indexes with less than 32700 nodes, the .IXN file is identical to the old .NIX file. The 32-bit GoldED can use both the new and the old index files, but will use the new ones if both are found. If you no longer use the DOS version of GoldED, you should delete the .AIX, .NIX and .LIX files to save disk space. - When points were using GoldNODE to compile USERLIST's, their point number would be appended to all non-point addresses. Fixed. Note that this fix is *not* tested. Let me know if it worked. - Fixed the bug that caused mangled CHARSET kludges. Under some circumstances, previous versions would put multiple "CHARSET:"'s in the CHARSET kludge, and if the combined length became more than 40 chars, GoldED would lockup or crash when trying to read the msg. - Intense and border/overscan colors now work again in the Borland C++ compiled version of GoldED/2. Borland had made a stupid little bug, which was fortunately easily fixable once located. - The new TEMPLATEPATH keyword did not work for templates defined in the Random System (GOLDRAND.CFG). It does now. + Added a "twirly" feature to GoldNODE. While GoldNODE is reading, sorting and writing it's index files, it displays a rotating "twirly" (the characters |/-\ in quick succession) to show that it is still alive and working. Because of this, GoldNODE now turns off the cursor while working. It is turned on again when exiting. The cursor turnoff currently only works with the Borland compiled versions. - When trying to write marked msgs to an existing file and hitting ESC instead of entering another filename, GoldED would mess up the screen, crash or lockup. Fixed. - Fixed a long-standing bug in GoldNODE. It was not possible to specify the configfile to use on the commandline using the -C option. + When displaying the size of an attached file, GoldED now looks for the file in the INBOUNDPATH if there is no drive specified for the attached file. This is useful when files have been attached _to_ you. If you don't have an INBOUNDPATH in your GOLDED.CFG, the GOLDPATH is assumed instead. + Added support of Binkley-style "^filename" in the subject line for deleting attached files after they have been sent. - GoldED/2 did not update the Hudson NETMAIL/ECHOMAIL.BBS files correctly, and sometimes caused a trap in the OS/2 session while trying. Fixed. The regular DOS version did not have this problem. - Mangled CHARSET kludges were not handled well. When finding such a mangled CHARSET kludge, GoldED/2 would trap and GoldED/DOS would crash, hang or sometimes go on apparently unaffected. Fixed. - Fixed hang in GoldED/2 when using it without a GOLDRAND.CFG or with all keywords in it commented out. Could possibly also explain similar hangs in previous DOS versions. + New feature: You can now change the color palette used in GoldED. The palette has 16 color registers, corresponding to the 16 colors from black (0) to intense white (15). By changing the values in the palette registers, it is possible to make any of the 16 colors a completely different color. You can even make the background colors intense, without using the intense color feature. There are 64 different colors to chose from. To configure the palette colors in GoldED, the SCREENPALETTE keyword has been added. There are two different syntaxes: SCREENPALETTE <reg> <value> SCREENPALETTE <reg> (red green blue) So you can either compose the color value using separate red, green, blue components, or directly use a precalculated value. The red/green/blue values can only be in the range 0-3. These are the original palette values: SCREENPALETTE 0 (0 0 0) SCREENPALETTE 1 (0 0 2) SCREENPALETTE 2 (0 2 0) SCREENPALETTE 3 (0 2 2) SCREENPALETTE 4 (2 0 0) SCREENPALETTE 5 (2 0 2) SCREENPALETTE 6 (2 2 0) SCREENPALETTE 7 (2 2 2) SCREENPALETTE 8 (0 1 0) SCREENPALETTE 9 (1 1 3) SCREENPALETTE 10 (1 3 1) SCREENPALETTE 11 (1 3 3) SCREENPALETTE 12 (3 1 1) SCREENPALETTE 13 (3 1 3) SCREENPALETTE 14 (3 3 2) SCREENPALETTE 15 (3 3 3) Copy these lines into your GOLDED.CFG and start experimenting! :-) NOTE: This feature has not been tested much. There may be bugs. - Known bug: If you start GoldED/2 (The Borland compiled version) as the _first_ program in a full-screen OS/2 session, the screen will be filled with garbage when you exit GoldED/2. Harmless but annoying. The garbage is put on the screen when GoldED/2 tries to restore the original screen (registered only feature). My tests seem to indicate that the original screen read by GoldED/2 at startup is _already_ garbage, but that can't be right. I think this is a Borland compiler bug, because when the problem does not seem to exist when I compile GoldED/2 with Watcom C++. So why don't I release the Watcom compiled version instead? Because I'm still having problems getting it to work correctly. The Borland compiled version looks more stable at this time. ! Changed GoldED/2 back to non-multithreaded mode. ---------------------------------------------------------------------- Notes for GoldED 2.42.A0710, ALPHA release saturday 10. july 1993. ---------------------------------------------------------------------- * This was a limited alpha. It was only sent to a few specific testers for debugging. + Added TZUTC and TID as "known" kludges. They are only recognized, not handled. Full support of TZUTC will probably be added in a future release. ! GoldED/2 is now compiled in multithreaded mode, and runs a separate thread to update the statusline (for the clock). ---------------------------------------------------------------------- Notes for GoldED 2.42.A0701, ALPHA release thurday 1. july 1993. ---------------------------------------------------------------------- + New keyword: "FIDOMSGTYPE <Fido/FTSC>". Defaults to Fido. This keyword defines the default format of Fido *.MSG files. It is only used by some AREAFILE's when reading external area configuration files, where the preferred format is unknown. The FTSC (FTS-0001.012 and later) format uses zone/point fields, where the Fido format uses date/time stamps. The Fido format is also sometimes referrered to as the "Opus" format. When set to Fido format, GoldED interprets the date/time stamps as DOS-style bitmapped date/time. + New keyword: "RA2USERSBBS <yes/no>". Defaults to no. Use this keyword to *force* GoldED to use a RemoteAccess 2.xx (RA2) compatible USERS.BBS file, even if RA2 is not detected. If used, this keyword should be placed in the configuration file _after_ any AREAFILE keyword. + New "AREAFILE InterMail". Supports the InterMail mailer. This is untested! Currently the support provided is similar to the FrontDoor 1.99b (and later) support. NOTE: You still have to define the correct semaphore filenames by yourself using the SEMAPHORE keyword. + AREAFILE D'Bridge now supports the new Beta 2011 DBRIDGE.ADF. This is untested! Normally GoldED tries to detect the DBRIDGE.ADF version based on the known record sizes of the of file. You can now override the autodetection by specifying one of these: AREAFILE D'Bridge -2011 AREAFILE D'Bridge -1047A27 AREAFILE D'Bridge -1047A22 AREAFILE D'Bridge -1046 The numbers refer to the versions which had new/changed DBRIDGE.ADF formats (that I know of). + New "AREAFILE Portal". Supports the Portal of Power mailer (v0.61gamma). Currently supported are the net/bad/dupemail areas and limited support of the Conferences (all conferences are treated as Fido *.MSG echomail areas). Planned for later implementation is closer integration to Portal (reading of non-area information too) and more support of Conferences (when Portal is released with a Mailscanner). + AREAFILE IMAIL now supports version 1.4x. - Fixed long-standing problem with GoldED sometimes crashing when shelling to an external spell checker from the internal editor. + When replying in another area using Alt-N or Alt-B, GoldED now looks for the AREA:<echoid> kludge and automatically positions the arealist selection bar on the area indicated by the kludge. This is particularly useful with systems such as GEcho when the personal mail feature is enabled. + It is now possible to use wildcards in the FRQEXT keyword. Example: FRQEXT .A?? With this definition, GoldED will recognize filenames such as "NODEDIFF.A12". + Improved screen update speed in many places. Most noticable when using GoldED/2 in a window, compared to GoldED/2 2.41.B0108. In some places such as the arealist, the screen update is still a bit slow. I'll continue working on more screen update speed. + New keyword: "SCREENUSEANSI <yes/no>". Defaults to no. This is an EXPERIMENTAL feature only - There is no guarantee that it will be supported in future releases! If enabled, GoldED will write all screen output via DOS (stdout) using ANSI codes for colors, cursor movement, screen clearing etc. SLOW!!! This feature may be useful to people who read echoes where ANSI codes are allowed in the msg text. Should work okay as long as only color codes are used... Another use of ANSI output may be for those who use GoldED remotely via modem in sysop drop-to-dos shells. Make sure you have ANSI.SYS or another ANSI compatible screen driver loaded before testing this feature!!! - High-bit (above ascii 127) chars are no longer converted to (.) chars in the header when printing msgs. - Fixed GoldED crash when using echoid's longer than 29 chars. Changed the max echoid size limit from 49 to 50. - The file request function now stops at the citation char ("). - Changed AREAFILE GEcho to use FTS-1 format *.MSG files instead of Opus format. - Fixed a problem with msg #0 error exits in certain circumstances. + Long filenames are now supported in GoldED/2. It seems to work well (tested with very long Squish filenames and writes to files), but I cannot guarantee that all valid HPFS filenames are accepted anywhere in GoldED. In particular, you probably cannot use spaces in filenames, and you should be careful with characters which are not supported in MS-DOS filenames. There may also be problems when using the import file picker, if there are filenames longer than the screen width (not tested). ! I have not been able to get intense colors and border (overscan) color to work in the 32-bit OS/2 version, so I had to disable these features in GoldED/2 for now. This is not a problem if you only run GoldED/2 in a window, but if you run it full-screen, the colors that should be intense will blink, and the border color will not be set. ----------------------------------------------------------------------