[fossil-users] visual studio solution and project file for fossil
Hi I am not familiar enough with the fossil build system so I hope somebody can answer these questions. Is there a way to generate a visual studio solution and project file from the fossil source which I can then load up in visual studio 2015 community edition? Or an alternative: is there a way to generate a debug build with pdbs which I can debug through visual studio? I used to using visual studio for debugging and it would help me to finish my gmerge patch. Thank you Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] unversioned files - converting files from versioned to unversioned and limiting the number of versions
Hi, I have some binary files in my repository which are currently versioned. I see there is a new unversioned file feature and I would like to convert them to unversioned. Is there a way to "convert" a file from versioned to unversioned? In Perforce there is a way to limit the number of versions (I can set this through P4V), is there anything similar in fossil? Is there a way to define and keep the last 'n' revisions? Thank you Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] visual studio solution and project file for fossil
> Fossil's primary use is to keep track of changes in your files, it is not a > build system perse. If you want to use Visual Studio, why not use something > like CMake: Hi Arjun, Thankyou for your reply. Sorry but after reading back my question I don't think I worded it well and I didn't make it clear about what I wanted. I want to build fossil itself because I want to patch it and hopefully submit my patches for review. I see that fossil can be built with a batch file "buildmsvc.bat" which uses the command-line visual studio compiler. It uses some kind of custom build where the headers which are needed are added to the files and output into a separate directory. I don't really understand fully how this works and is a guess from what I have seen so far. I want to make a solution for fossil so I can build fossil under visual studio itself from the IDE itself. I can then use the IDE's debugger. I tried to add all the source files from inside the src directory but that didn't work. So I then ran the batch file, waited for it to build and then used the source files from the msvcbld and that didn't seem to work either. So I have failed so far and wondered if anyone has made a solution and if they did, how did they do it? The second question was: - Is there a way I can modify (or pass a parameter) to the buildmsvc.bat to make a "debug" build without the compiler doing optimisations and with it generating a debug pdb. I could also debug fossil from within visual studio this way. Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] unexpected result with fossil mv on a directory
Hi, I think I may have found a bug with fossil mv related to moving directories. I am using fossil version 1.35 3aa86af6aa 2016-06-14 11:10:39 UTC. I wanted to use fossil mv to move a directory tree. The tree looks like this: test/drivetest and I wanted to move it to test/disc/drivetest If I use: cd test fossil mv drivetest disc/drivetest if disc, drivetest or neither exist on my machine then the move doesn't work as I expected. fossil reports: RENAME test/drivetest test/disc/drivetest then "fossil changes" lists nothing. if both disc and drivetest exist then fossil reports RENAME for every file as I expected and fossil submit works as expected. Is this by design or is it a bug? Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] unexpected result with fossil mv on a directory
Hi, I re-read my message back and realised I may have caused confusion: disc/drivetest exists - it is the directory with files I want to move, but when I say when neither "disc" or "drivetest" exists, I meant "test/disc" and "test/disc/drivetest" sorry if I caused confusion. Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] unexpected result with fossil mv on a directory
LOL one more try: test/drivetest exists and has files. test/disc/drivetest or test/disc doesn't exist. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] unable to revert fossil mv
Hi, I have a directory called test it contains a directory called disc and a directory called disk_sizes. test/disk_sizes contains a few files with extension "dsk". I wanted to do this move: test/disk_sizes -> test/disc/disk_sizes I worked around the other problem I reported by creating test/disc/disk_sizes. but I typed: fossil mv test/disk_sizes test/disc this moved all files in test/disk_sizes to test/disc. (test/disc is now filled with files with extension dsk). I realised my mistake. I did: fossil revert test/disc/*.dsk fossil reported the UNMANAGE. i then did "fossil changes". fossil then listed each file with dsk extension with "MISSING". I can't revert the change. i tried to do the mv again and fossil did nothing :( (I don't know if this is important but I had other files and directories in my fossil changes at the same time). How can I fix this? I tried to do the move again (but correct this time) and although fossil reported it had renamed the files and to the correct location, fossil changes didn't have those changes and had the same MISSING. I can't revert and I can repeat the operation. :( Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] unable to revert fossil mv
Update: I am seeing this: kevin@kevin:~/arnold/test/disc$ fossil changes MISSING disk_sizes/data.dsk MISSING disk_sizes/data2sides.dsk MISSING disk_sizes/ramdosd1.dsk MISSING disk_sizes/system.dsk MISSING disk_sizes/system42.dsk MISSING disk_sizes/system80t1s.dsk MISSING disk_sizes/test.dsk MISSING disk_sizes/vortex.dsk if I do touch to make these missing files and then do a revert. Fossil shows me kevin@kevin:~/arnold/test/disc/disk_sizes$ fossil revert DELETE test/disc/disk_sizes/data.dsk DELETE test/disc/disk_sizes/data2sides.dsk DELETE test/disc/disk_sizes/ramdosd1.dsk DELETE test/disc/disk_sizes/system.dsk DELETE test/disc/disk_sizes/system42.dsk DELETE test/disc/disk_sizes/system80t1s.dsk DELETE test/disc/disk_sizes/test.dsk DELETE test/disc/disk_sizes/vortex.dsk REVERT test/disk_sizes/data.dsk REVERT test/disk_sizes/data2sides.dsk REVERT test/disk_sizes/ramdosd1.dsk REVERT test/disk_sizes/system.dsk REVERT test/disk_sizes/system42.dsk REVERT test/disk_sizes/system80t1s.dsk REVERT test/disk_sizes/test.dsk REVERT test/disk_sizes/vortex.dsk "fossil undo" is available to undo changes to the working checkout. and then: kevin@kevin:~/arnold/test/disc/disk_sizes$ fossil changes kevin@kevin:~/arnold/test/disc/disk_sizes$ cd .. i try again: kevin@kevin:~/arnold/test$ fossil mv disk_sizes disc/disk_sizes RENAME test/disk_sizes/data.dsk test/disc/disk_sizes/data.dsk RENAME test/disk_sizes/data2sides.dsk test/disc/disk_sizes/data2sides.dsk RENAME test/disk_sizes/ramdosd1.dsk test/disc/disk_sizes/ramdosd1.dsk RENAME test/disk_sizes/system.dsk test/disc/disk_sizes/system.dsk RENAME test/disk_sizes/system42.dsk test/disc/disk_sizes/system42.dsk RENAME test/disk_sizes/system80t1s.dsk test/disc/disk_sizes/system80t1s.dsk RENAME test/disk_sizes/test.dsk test/disc/disk_sizes/test.dsk RENAME test/disk_sizes/vortex.dsk test/disc/disk_sizes/vortex.dsk kevin@kevin:~/arnold/test$ fossil changes MISSING disc/disk_sizes/data.dsk MISSING disc/disk_sizes/data2sides.dsk MISSING disc/disk_sizes/ramdosd1.dsk MISSING disc/disk_sizes/system.dsk MISSING disc/disk_sizes/system42.dsk MISSING disc/disk_sizes/system80t1s.dsk MISSING disc/disk_sizes/test.dsk MISSING disc/disk_sizes/vortex.dsk :( No way to recover :( ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] windows build patch to use vs2015 static analysis
Hi, Here is a patch that sets the /analyze compiler switch which enables the vs2015 static analysis output. I have tested it with vs2015 community edition and the latest source distribution of fossil 1.36. START --- C:/Users/kev/Downloads/fossil-src-1.361/fossil-1.36/win/Makefile.msc Mon Oct 24 15:59:33 2016 +++ C:/Users/kev/Downloads/fossil-src-1.36/fossil-1.36/win/Makefile.msc Sat Oct 29 14:26:46 2016 @@ -34,6 +34,11 @@ DEBUG = 0 !endif +# Enable analysis? +!ifndef ANALYZE +ANALYZE = 0 +!endif + # Build the OpenSSL libraries? !ifndef FOSSIL_BUILD_SSL FOSSIL_BUILD_SSL = 0 @@ -241,6 +246,10 @@ LDFLAGS = $(LDFLAGS) /DEBUG !else CFLAGS = $(CFLAGS) $(CRTFLAGS) /O2 +!endif + +!if $(ANALYZE)!=0 +CFLAGS = $(CFLAGS) /analyze !endif BCC = $(CC) $(CFLAGS) END use it like this: buildmsvc ANALYZE=1 The piped output is too large to attach here. I think it's worth checking these. My experience with other projects has shown that sometimes it's a bit over sensitive but it has also found some real problems. A few lines taken from the output: c:\users\kev\downloads\fossil-src-1.36\fossil-1.36\src\timeline.c(762) : warning C6297: Arithmetic overflow: 32-bit value is shifted, then cast to 64-bit value. Results might not be an expected value. c:\users\kev\downloads\fossil-src-1.36\fossil-1.36\src\allrepo.c(323) : warning C6246: Local declaration of 'rc' hides declaration of the same name in outer scope. For additional information, see previous declaration at line '178' of 'c:\users\kev\downloads\fossil-src-1.36\fossil-1.36\src\allrepo.c'.: Lines: 178 c:\users\kev\downloads\fossil-src-1.36\fossil-1.36\src\delta.c(465) : warning C6385: Reading invalid data from 'collide': the readable size is '_Old_5`nHash*sizeof(int)' bytes, but '8' bytes may be read.: Lines: 350, 351, 352, 353, 354, 355, 356, 360, 361, 367, 380, 381, 382, 383, 384, 385, 386, 387, 388, 385, 386, 387, 388, 385, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 405, 407, 408, 424, 425, 426, 427, 431, 432, 433, 434, 435, 437, 441, 444, 447, 448, 449, 454, 455, 465 c:\users\kev\downloads\fossil-src-1.36\fossil-1.36\src\dispatch.c(505) : warning C6011: Dereferencing NULL pointer 'z'. : Lines: 448, 449, 450, 451, 452, 453, 454, 464, 468, 472, 476, 480, 481, 485, 486, 488, 489, 494, 500, 501, 502, 505 Best Wishes, Kevin___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] unable to revert fossil mv
Steve, I agree with you it is confusing! I have used fossil for over 2 years now (and I continue to use it because I like it very much) but I haven't used the move operation much in that time and each time I do use it the two step move is confusing. I am used to other source control systems which move the files. I use them much more than I use fossil. I use fossil for my own personal code and the other scms for work. What confused me this time is that although technically it is true that the files are classed as MISSING after a move because fossil thinks they are in the new location , and the files have not been moved there, their status as MISSING should not be reported. If it remains as two step then saying their MISSING makes it sounds like something went wrong. Fossil should know it has moved/renamed the files as part of it's operation so if I do a mv command and then list changes, fossil should list them as RENAMED rather than MISSING. This would help. I think fossil should also remind the user to move the files if it's operating in two step mode. This would help me to remember each time ;) It's as simple as putting that in the changes and after the mv operation has been done-> "please don't forget to move your files". For changes, it could have 1 status for "renamed but file doesn't exist there yet" and "renamed and file exists" as a way to confirm to the user that they have done both operations. Also, if it is two step, then in theory I should be able to: 1. submit without moving files 2. do update to get the files in the new location. I am then free to delete the files in the original location the same way that I am free to copy the files or not. Also, what is confusing is what happens if there are changes to the files at the time of merge..? Of course not moving them is confusing for somebody like me who 1) doesn't use this method often 2) is used to the methods used by other scm makes it confusing each time. What also confused me is that when I did revert, fossil appeared to revert TWO operations (it listed the delete and revert). But changes only listed ONE operation. It seemed like the changes wasn't telling me all the information :( ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] windows build patch to use vs2015 static analysis
BTW, There are around 400 warnings. I posted just a few of them. So please compile with this patch to the makefile and visual studio 2015 with the analyze option enabled to see them all. A few more: c:\users\kev\downloads\fossil-src-1.36\fossil-1.36\src\sqlite3.c(113673) : warning C6239: ( && ) always evaluates to the result of . Did you intend to use the bitwise-and operator? c:\users\kev\downloads\fossil-src-1.36\fossil-1.36\src\vfile.c(528) : warning C6287: Redundant code: the left and right sub-expressions are identical. c:\users\kev\downloads\fossil-src-1.36\fossil-1.36\src\diff.c(979) : warning C6001: Using uninitialized memory 'aNext'.: Lines: 944, 945, 946, 947, 948, 949, 950, 951, 952, 953, 954, 956, 957, 958, 959, 960, 961, 962, 963, 964, 965, 966, 967, 968, 969, 970, 971, 976, 977, 978, 979, 980, 981, 982, 979 ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] fossil mv with wildcards inconsistent operation (and more two step merge frustrations)
Hi, I found an unexpected inconsistency with fossil mv on files when using wildcards. I want to move 2 files into another directory: itest.asm and itest.bin Both files exist in the repository. If both exist locally: fossil mv itest.* newdir then fossil will rename *both*. changes show both have been renamed. If only itest.bin exists in the starting directory: fossil mv itest.* newdir will only rename itest.bin and not itest.asm I can still do this (including when itest.asm doesn't exist): fossil mv itest.asm newdir then fossil will rename the file regardless of it existing or not because fossil knows it exists in the repository. So wildcards appear to apply to the files locally rather than those in the repository. Should wildcards apply to both locally and in the repository? And this highlights another issue with the two step move: I can use fossil mv to move files but if I don't then move them locally then I can't submit and fossil reports it can't find the file. The process relies on both local moving of files and moving them in the repository. The operations have to be done in a specific order and are not independent. So my argument is that it should not be two step. It should be single step. Fossil already handles file operations so why is this different and why can't it be done automatically? (Sorry about the number of e-mails relating to mv, I'm re-organising files in my repository and I've e-mailed as I found these issues). Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] revert of missing files using wildcard not possible.
Hi, This is probably related to move and missing files in general. It's not possible to do a revert using a wildcard to revert multiple missing files. You have to specify them one by one. For example: I used mv to move some files: intlace.asm intlace.bat intlace.bin intlace.sh I did not move them yet in the filesystem and with changes they showed up as missing. So I tried this: fossil revert intlace* UNMANAGE intlace* fossil changes showed no changes to the files they were still listed as MISSING. fossil revert intlace.asm this showed DELETE of file from new location and REVERT of old location and changes showed this operation was successful. Please support wildcards for missing files :) Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] revert of missing files using wildcard not possible.
Ok, I understand that wildcards are expanded by the shell but it would make it *SO* much easier to use if fossil could do some kind of pattern matching or use wildcards to allow me to revert a group of files. It feels a little inconsistent to me that I can't use wildcards throughout. I can use them to add files (because the shell does the work), but as a user it seems a bit odd I can't use it to revert missing files? As a user it feels like a bug - a user doesn't really know or care what the shell is doing or not doing. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] revert of missing files using wildcard not possible.
I tested the automatic expansion of wildcards on windows and linux.
Here is my program:
#include
#include
int main(int argc, char **argv)
{
int i;
for (i=0; ihttp://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] revert of missing files using wildcard not possible.
Hi Andy, Try this: kevin@kevin:~$ mkdir fossil kevin@kevin:~$ cd fossil kevin@kevin:~/fossil$ fossil new test project-id: 72090bfca4c8d0618530e7b7b524056a25bf4052 server-id: f38e157351c762cbdb09d581ca17e1d70cb68f0c admin-user: kevin (initial password is "c821e4") kevin@kevin:~/fossil$ fossil open test project-name: repository: /home/kevin/fossil/test local-root: /home/kevin/fossil/ config-db: /home/kevin/.fossil project-code: 72090bfca4c8d0618530e7b7b524056a25bf4052 checkout: a0a2d816aee3e2bc1cb0212e9869abc17c6691d5 2016-10-31 18:50:20 UTC tags: trunk comment: initial empty check-in (user: kevin) check-ins: 1 kevin@kevin:~/fossil$ touch intlace.asm kevin@kevin:~/fossil$ touch intlace.bin kevin@kevin:~/fossil$ touch intlace.dsk kevin@kevin:~/fossil$ fossil add intlace* ADDED intlace.asm ADDED intlace.bin ADDED intlace.dsk kevin@kevin:~/fossil$ fossil commit -m "add files" New_Version: 828f91088a96ba00371148edb59da15ae42ac5d4 kevin@kevin:~/fossil$ ls intlace.asm intlace.bin intlace.dsk subdir test kevin@kevin:~/fossil$ fossil mv intlace* subdir RENAME intlace.asm subdir/intlace.asm RENAME intlace.bin subdir/intlace.bin RENAME intlace.dsk subdir/intlace.dsk kevin@kevin:~/fossil$ fossil changes MISSING subdir/intlace.asm MISSING subdir/intlace.bin MISSING subdir/intlace.dsk kevin@kevin:~/fossil$ fossil revert subdir/intlace* UNMANAGE subdir/intlace* kevin@kevin:~/fossil$ fossil changes MISSING subdir/intlace.asm MISSING subdir/intlace.bin MISSING subdir/intlace.dsk kevin@kevin:~/fossil$ fossil revert "subdir/intlace*" UNMANAGE subdir/intlace* kevin@kevin:~/fossil$ fossil changes MISSING subdir/intlace.asm MISSING subdir/intlace.bin MISSING subdir/intlace.dsk kevin@kevin:~/fossil$ fossil revert subdir/intlace.asm DELETE subdir/intlace.asm REVERT intlace.asm "fossil undo" is available to undo changes to the working checkout. kevin@kevin:~/fossil$ fossil changes MISSING subdir/intlace.bin MISSING subdir/intlace.dsk kevin@kevin:~/fossil$ You reverted the files in their original location. To me that doesn't make sense. I consider the operation to have succeeded and fossil now thinks of them in their new location so I revert them in their new location to go back to their old location. I can revert them one at a time in their new location and that works. OR I could copy or create files in their new location and revert those. But still it's not possible to revert missing files using a wildcard. I guess because the shell doesn't know about them and doesn't expand them and fossil considers the * to be a valid character in a filename. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] revert of missing files using wildcard not possible.
>I suppose this is so obvious to people who grew up with Unix that we >don't even think about it. Are you coming from a windows background? Yes. I have been programming on windows since 1995 and from the beginning I used Visual C++ and later Visual Studio (before then it was mostly Z80). I have written lots of tools and utilities and command-line tools too. I have been using Linux in my spare time for about 4 years. I have used GUIs and IDEs much more than command-line tools and for source control I've used Visual SourceSafe, svn, cvs, Alienbrain and Perforce and for all of those I use shell extensions or GUI tools to make it much easier. I have used perforce through the command-line quite a lot and I am used to it's way of working. I use fossil on windows and linux and it does a great job with my own projects. :) So yes, I don't know how shells handle things because I'm not exposed to it that much and I guess the GUIs I use must "iron-out" some of these command-line "wrinkles" which I see. I am also not that good at replying properly to messages in this mailing list. ;) ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Re: debugging fossil with VisualStudio debugger
Hi Artur, Thank you for the information. I never knew an exe could be debugged this way because I've never needed to do it that way. What I would like to point out is that for most visual studio and windows users this is a strange way to work. Windows users are used to working almost exclusively through the gui and using devenv or msbuild is normally restricted to running in a batch file or on a build machines as some kind of build for end user or for a verification process. For working with visual studio, I used to have to reboot my machine every day, but now I leave it on because it's much more stable. I find the visual studio icon (normally pinned to the taskbar), start up visual studio, find the solution, build the code, press f5 to launch the debugger, step through and debug it via the watch windows, local registers, callstack etc. Debugging is easy because I can hover over a variable and it tells me it's value, or put it into the watch window. I can also go through structures easily by expanding and contracting the + mark next to the structure in the watch window. It's just a different way of working compared to other OS's. The solution file I posted was not the normal one visual studio uses. Normally a project type is defined and files are added and visual studio performs the build and link using that information. In this case I setup a project in makefile mode and configured it to make debug or release builds. I also included the original source since the other files are not there before a build is initiated. I was still able to place a breakpoint anywhere in the code, step through the code and view the variables and their values. For a normal visual studio user this is closer to what they know. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] request for change of error message (autosync fail and pressing ctrl+c).
Hi, I attempted to commit to the fossil server I use but it was inaccessible (I think the connection was down) and the autosync failed. If I press ctrl+c I get a different error compared to answering 'n'. The ctrl+c error seems more scary. Here is the output: C:\Users\kev\Documents\arnold>fossil commit -m "x" Autosync: http://x...@xx.xxx.xx.xx:x/arnold cannot connect to host xx.xxx.xx.xx:x Pull done, sent: 0 received: 0 ip: Autosync failed. continue in spite of sync failure (y/N)? SQLITE_MISUSE: API call with invalid database connection pointer SQLITE_MISUSE: misuse at line 136682 of [fc49f556e4] ^C C:\Users\kev\Documents\arnold>fossil update Autosync: http://k...@xx.xxx.xx.xx:x/arnold cannot connect to host xx.xxx.xx.xx:x Pull done, sent: 0 received: 0 ip: Autosync failed. continue in spite of sync failure (y/N)? n update abandoned due to sync failure It would be great if the error was consistent and always gave "update abandoned due to sync failure". The second error is less scary and nicer especially if non-programmers used fossil as part of a team. I have experience with non-programmers using perforce and if something wrong happens they always call over the person who they believe is the "perforce expert" to help them out when anything unexpected happens. I think also the less scary messages the better because feel less worried something is going to be lost (of course it will not be but that is what they think). The first error message is ok if verbose logging is enabled :) Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Follow-up to reverting missing files
Hi Andy, That is useful. What would the windows equivalent command-line? ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] fossil revert is not recursive?
Hi, I am using windows and it's standard command-line. I can use "fossil add " to add a directory and it's sub-directories and the files within. However, I can't seem to revert a directory in the same way even though that directory exists. I tried "fossil revert " and "fossil revert /*.*" and it doesn't do it. It will revert the files in the immediate directory, but if there are files added in sub-directories then these are not reverted. I think I read that the directories are implicit in fossil? Their knowledge is derived from the file path's. So I am confused. If I can add recursive, why can't I do a revert recursive? Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] fossil reports erroneous revert?
Hi, If I do the following: fossil new test fossil open test echo >1.txt fossil add 1.txt fossil commit -m "add file" fossil changes (doesn't report changes) fossil revert 1.txt REVERT 1.txt fossil changes (doesn't report changes) I am reverting a file that has been added to the repository but there are no local changes. Fossil reports "revert" but this is confusing because there is nothing to revert. Is this expected? ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] UNMANAGE status
Hi If I do a revert of a file that doesn't exist locally or in the repository I get a status like this: fossil revert 2.txt UNMANAGE 2.txt I think the UNMANAGE status can be read in more than one way. I often read it as "fossil used to manage this file but it's not managing it now". But it can also be read as: "fossil is not managing this file. It is only a local file". Does anyone else read it the first way? And a final question which is related: is there documentation which describes the possible responses and what they mean? If not would anyone mind if I wrote some documentation and submitted it to this list? Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] src/info.c use of zUuid after it has been free.
Hi, I want to ask a question about some lines in src/info.c In this function: show_common_info on this line: if( zUuid && showComment ) zUuid is potentially pointing to freed memory (freed in the previous function). It is ok because the code doesn't use it. Should the code check zUUid? If it requires zUUid, then could it be moved into the block above? Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] fossil amend comment
Hi, I submitted a change and made a mistake in the comment. I wanted to change the comment and fix it. I did this: fossil amend ba67fcd67c -m "fixed snapshot" command reported the comment, tags and uuid. fossil changes showed nothing. I looked at the server and there was no changes. I forced a submit with fossil commit --allow-empty and it appears to have worked. I wish "fossil changes" would list this change (and changes like changing file type) because it's confusing that some changes are listed and some are not. I should then just be able to submit without forcing. The way I see it is I did make a change, so it should be listed and the commit is not empty. Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] crash - fossil 2.1
Hi, I got a crash when using fossil 2.1 downloaded from www.fossil-scm.org Fossil server is using: Fossil version [5ed8477b6c] 2017-02-27 14:46:01 I am using windows 10. The change had 4 edited files and 1 added file. Added file is a 4mb binary which is mostly zeros others are edited text files. The output looks like this: C:\Users\kev\Documents\arnold\test\snapshot>..\..\fossil commit -m "added javacpc 4mb snapshot. updated snapshot code to not crash if the snapshot defines more ram than supported. indicate to user the amount of ram wanted and if we can support that" Autosync: x Round-trips: 1 Artifacts sent: 0 received: 0 *** time skew *** server is slow by 20.2 seconds Pull done, sent: 372 received: 506 ip: 80.229.30.31 continue in spite of time skew (y/N)? y ../../src/arngui/SnapshotLoadDialog.cpp contains mixed line endings. Use --no-warnings or the "crlf-glob" setting to disable this warning. Commit anyhow (a=all/c=convert/y/N)? a The server is for private use so I changed the ip and url with xx here. Each time I tried this fossil.exe crashed. I didn't have any symbols so the stack wasn't meaningful. The event viewer doesn't give useful information, but looking at the dmp it appears to be a null pointer access. I downloaded the fossil source and built a debug version using visual studio 2015 which worked and the submit was successful. I don't have a crash now. Sorry I can't provide more information. Is it possible to get the pdb from the 2.1 release? I could use this with the dmp file to hopefully give a good callstack or does anyone want the callstack e-mailing to them. It's 6MB uncompressed. Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] fossil 1.2->2.1 - hash policy "auto" doesn't seem to work and gives SQL error.
Hi All, i downloaded windows version of fossil 2.1 - after initial problems crashing when I ran it (see other thread) and rebuilding it was fine. I used fossil hash-policy to set it to sha3 and forced a commit to the server because I didn't have files in my changes at that time. From then on commits were using sha3 fine on windows. So I went to my linux machine. It was using fossil 1.2. I got this: kevin@kevin:~/arnold$ fossil update Autosync: http://kev@xxx/arnold Round-trips: 1 Artifacts sent: 0 received: 0 unknown command: [igot] Round-trips: 1 Artifacts sent: 0 received: 0 Pull done, sent: 353 received: 2161 ip: x Autosync failed. continue in spite of sync failure (y/N)? n update abandoned due to sync failure I then downloaded linux x64 pre-compiled 2.1 kevin@kevin:~/arnold$ fossil version This is fossil version 2.1 [83e3445f67] 2017-03-10 17:07:08 UTC kevin@kevin:~/arnold$ fossil hash-policy auto kevin@kevin:~/arnold$ fossil update Autosync: http://kev@xxx/arnold Round-trips: 1 Artifacts sent: 0 received: 0 SQLITE_CONSTRAINT: abort at 14 in [INSERT INTO blob(rcvid,size,uuid,content)VALUES(0,-1,:uuid,NULL)]: CHECK constraint failed: blob fossil: SQL error: CHECK constraint failed: blob kevin@kevin:~/arnold$ fossil hash-policy sha3 sha3 kevin@kevin:~/arnold$ fossil update Autosync: http://kev@xx/arnold Round-trips: 3 Artifacts sent: 0 received: 202 Pull done, sent: 8806 received: 157107 ip: xxx UPDATE changes.txt ... So it seems that hash-policy of auto is not working? This is a rough sequence of events: * fossil 1.2 exe on windows and on linux * fossil server updated to 2.1 * I updated windows binary to 2.1 * on windows I used hash-policy to set sha3 and used force because there were no more changes * committed some files on windows * I tried to update with linux 1.2 binary * I then updated to linux 2.1 binary * I then had to set sha3 policy (auto didn't work - see error) * update succeeded. Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] fossil 1.2->2.1 - hash policy "auto" doesn't seem to work and gives SQL error.
Hi Richard, >The "auto" hash-policy automatically flips to "sha3" as soon as Fossil >sees one or more SHA3 artifacts in the repo. Since your repo contains >SHA3 artifacts, you won't be able to set the policy to "auto" because >it will immediately and automatically flip to "sha3". > >That is the way "auto" is designed to work. I expected it to automatically do that when I did a "fossil update" and get the files provided I had fossil 2.1 installed. I have autosync set and I thought that would then sync from the server and automatically switch to sha3 and get the new files. >How were you expecting it to work? How can the documentation be improved? Setting the hash policy was fine that worked without confusion. When I then accessed the remote repository from another machine I expected it to automatically see the change and automatically switch the client to the new policy. If the client needed to be updated then ideally fossil client should tell me. It would have been great if this is what had happened: >fossil update Autosync: http:///arnold Autosync failed. This fossil repository is not compatible with the version of your client. You need version 2.1 or later. Please visit http://www.fossil-scm.org and download a new client. Then after installing the new client and with the hash-policy set to auto: >fossil update Autosync: http:///arnold Hash-policy on server has been updated to sha3. Switching to sha3 for client. ... <- normal sync here with cards and files listed. If I had set the policy to sha1 or other: >fossil update Autosync: http:///arnold Autosync failed. The fossil repository is using sha3 hash policy. Please set hash-policy to sha3 or use auto. Maybe for now, instead of (or in addition) to the errors I saw it could give a command to try to resolve the issue such as one that causes the database to update. Thank you Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] painful merge
Hi, I use fossil. I think it's great. I always find merging painful and hard work. For my workflow I have a trunk branch which is stable. I then have branches, each branch is where I implement a new feature. I often jump between branches with a few active at one time. I will also add fixes to the trunk. So all have commits. While I work on a branch I don't merge from trunk. I never need to because the feature can be done without that. When I am happy with the branch I then merge back into trunk. The merge is always painful - that's why I avoid it as much as possible. I use linux. meld is my merge/diff tool of choice. I like it because I can see two files side by side, I can transfer the changes and I can make edits to resolve them. The user interface is similar to winmerge which is my tool of choice on windows. Having two files side by side is nice and easy to manage. Fossil likes to do 3-way merges. With kdiff3 I see 4 windows. Two of them are from the head versions of trunk and the branch. One is "baseline", I don't know what this and to be honest I don't care. The other window is the edit window where it shows the result. I don't care because I am only interested in the head versions. kdiff3 makes it hard work, I must go and resolve every little thing including whitespace and empty lines. Merging multiple lines is hard work. So often I've given up and choose a different approach. I use perforce at work and it has automatic merge. This takes away a lot of pain and I must only suffer if there are merge conflicts. I always diff before commiting and always build to make sure it's good. My alternative approach for merge is: 1. get head on trunk and copy the files to another directory 2. get head on branch and copy the files to another directory 3. get head on trunk, perform merge (fossil now records the fact of the merge) 4. copy files from copied trunk directory 5. diff against trunk using meld checking each file one by one. Is there a way to make merges much less painful? Is there any form of automatic resolution in fossil or a tool that has this and works similar to perforce in this respect? Thanks ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] painful merge
Thank you and sorry I didn't respond furthur. The problem is not with fossil. I think the real problem is that I am used to a different merge program and I can't get along with kdiff3. It is true I am using Windows and Linux. I never had a problem with line endings with fossil. It just works! On perforce, at work, when I get latest it automatically converts line endings, but I never set it up on fossil. Never needed to! :) All my fossil settings are empty except: autosync (local) on gdiff-command (local) meld gmerge-command (local) kdiff3 "%baseline" "%original" "%merge" -o "%output" I use scite to edit my code on Linux and Microsoft Developer Studio 2013 (Community edition) on Windows. scite doesn't care about line endings. I use command-line to access fossil on both Linux and Windows: "./arngui/arnguiApp.cpp contains CR/NL line endings. Use --no-warnings or the "crnl-glob" setting to disable this warning. Commit anyhow (a=all/c=convert/y/N)? ". I answer "a" for all. If I gdiff with meld the window shows me 2 panels and it's obvious which is fossil file and which is local. meld is like WinMerge and similar to p4merge and that is why I use it on linux. I like meld because of it's simplicity, it's visual display of the diffs and ability to edit in-line. The way it shows changes is exactly what I am used to with other tools on windows. I tried using kdiff3 and it works but I just can't get along with it's interface! The number of buttons, the tagging in the code [Merge Conflict] [No src line] that kind of thing. That is the real problem, not fossil. The tags in the merge output and how each is handled. It's horrible, and all the buttons!!! single arrow, double arrow, triple arrow... I used this command for kdiff3: fossil settings gmerge-command 'kdiff3 "%baseline" "%original" "%merge" -o "%output"' I tried the suggested meld command: fossil settings gmerge-command 'meld "%baseline" "%original" "%merge" "%output"' That doesn't work anymore :( It says it can't read the "-output" file :( I tried p4merge (64-bit linux version downloaded from www.perforce.com). I used it today at work to do a 3-way merge ;) I used this syntax: fossil settings gmerge-command 'p4merge "%baseline" "%original" "%merge" "%output"' This is it's syntax for 3-way merge. It also says it can't find the "output" file. With p4v, which we use at work, I try and do automatic resolve. That solves many problems, but I had some merge conflicts today, I notice that there are 4-windows. I ignore the middle one which is baseline - it's almost like it's not there so that is why I thought there were fewer windows. Left and right are the two head versions, one of trunk one of the branch. At the bottom is the local file the merges go into. The changes are pre-merged here. Where there is a conflict it shows the 3 changes next to each other and I choose them by clicking on the icon of the appropiate change. So for me it's the choice of merge program. Please please keep the way the gmerge command is defined, it's great and adds flexibility to specify the command-line exactly! My workflow is to do the merge, using a graphical merge tool to help me, then do a gdiff against the fossil repository then commit. So my questions are now: * What is the default/server line ending? * Could gmerge-command support additional '%' variables which can be given to merge tools which support it which can be used to name the windows in the merge tool and the title in the tool? p4merge allows naming of left/right/merge etc. Would be great if we had access to fossil change number, or even "source"/"target"/"merge" or something like that? It has -nb -nl -nr -nm so fossil using fossil gmerge-command I could put this in the command-line: e.g. -nb "%baseversion" -nl "%originalversion" * Can fossil support other merge programs and their syntax? e.g. to support p4merge as it is, or meld with it's 3 way merge? Ones where they don't take 4 parameters? Perhaps instead of %output you have %local for the local file such as AboutDialog.cpp? p4merge's bottom window with the merge result is AboutDialog.cpp instead of fossil using AboutDialog.cpp-output. * Why is there a seperate "%output" file? Why doesn't it put the output direct into the local file? * What does convert do? Convert from native (e.g. \n on linux and \r\n on windows) to... what line ending??? Thanks. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] painful merge
>That's a painful way to do it. Unless you have conflicts, merging should be >painless. It sounds to me like you are using 'fossil gdiff' to resolve your >merges manually instead of using 'fossil merge'? This was only a work around because I couldn't use meld to resolve my merge. I do always use fossil gdiff before I commit after a merge to double check the results. If I can't use meld, can anyone suggest a merge tool which shows it's differences like meld/p4merge do, and allow me to edit and works with fossil's gmerge-command as it currently stands? Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] quick poll: do you generally use add/rm or mv
I would use mv. Having had bad experiences with other source control systems where you loose history with add/rm type commands, I would use mv to retain that. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] painful merge
I never got any problems with line endings and I do want fossil to be my merge
tool.
I got the trunk source of fossil and modified it locally to work with p4merge.
p4merge requires the output file to exist.
in merge_3way function I changed it with the following lines:
zCmd = string_subst(zGMerge, 8, azSubst);
printf("%s\n", zCmd); fflush(stdout);
file_touch(zOut);
fossil_system(zCmd);
if( file_wd_size(zOut)>0 ){
blob_read_from_file(pOut, zOut);
file_touch is implemented as:
void file_touch(const char *zFilename){
FILE *fh = fopen(zFilename,"w");
fclose(fh);
}
This makes p4merge work :) I think this will also support other merge tools
where they require the output file to exist.
Would this or a better patch be considered for fossil?
This however does change functionality with other tools and I didn't test
those; more specifically the size of the file being 0 was accepted before so
you could blank the content of the file and it would accept it.
p4merge will leave the output file untouched if a merge is cancelled (leaves it
as a 0 byte file) otherwise it contains content. So I changed the size check to
be greater than zero.
I did try to look at the return code of fossil_system to see what p4merge was
returning. It always seemed to return 0.
If it had returned an error code I would have used that too in the check.
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] painful merge
diff
Index: src/file.c
==
--- src/file.c
+++ src/file.c
@@ -118,40 +118,45 @@
}else{
if( fossil_stat(zFilename, &fileStat, isWd)!=0 ){
fileStatValid = 0;
rc = 1;
}else{
fileStatValid = 1;
rc = 0;
}
}
return rc;
}
/*
** Return the size of a file in bytes. Return -1 if the file does not
** exist. If zFilename is NULL, return the size of the most recently
** stat-ed file.
*/
i64 file_size(const char *zFilename){
return getStat(zFilename, 0) ? -1 : fileStat.st_size;
}
+
+void file_touch(const char *zFilename){
+ FILE *fh = fopen(zFilename,"w");
+ fclose(fh);
+}
/*
** Same as file_size(), but takes into account symlinks.
*/
i64 file_wd_size(const char *zFilename){
return getStat(zFilename, 1) ? -1 : fileStat.st_size;
}
/*
** Return the modification time for a file. Return -1 if the file
** does not exist. If zFilename is NULL return the size of the most
** recently stat-ed file.
*/
i64 file_mtime(const char *zFilename){
return getStat(zFilename, 0) ? -1 : fileStat.st_mtime;
}
/*
** Same as file_mtime(), but takes into account symlinks.
*/
Index: src/merge3.c
==
--- src/merge3.c
+++ src/merge3.c
@@ -16,40 +16,42 @@
***
**
** This module implements a 3-way merge
*/
#include "config.h"
#include "merge3.h"
#if 0
#define DEBUG(X) X
#define ISDEBUG 1
#else
#define DEBUG(X)
#define ISDEBUG 0
#endif
/* The minimum of two integers */
#ifndef min
# define min(A,B) (A0 ){
const char *zGMerge; /* Name of the gmerge command */
zGMerge = db_get("gmerge-command", 0);
if( zGMerge && zGMerge[0] ){
char *zOut; /* Temporary output file */
char *zCmd; /* Command to invoke */
const char *azSubst[8]; /* Strings to be substituted */
-
+ int status;
+
zOut = file_newname(zV1, "output", 1);
azSubst[0] = "%baseline"; azSubst[1] = zPivot;
azSubst[2] = "%original"; azSubst[3] = zOrig;
azSubst[4] = "%merge"; azSubst[5] = zOther;
azSubst[6] = "%output"; azSubst[7] = zOut;
zCmd = string_subst(zGMerge, 8, azSubst);
printf("%s\n", zCmd); fflush(stdout);
+ file_touch(zOut);
fossil_system(zCmd);
- if( file_wd_size(zOut)>=0 ){
+ if( file_wd_size(zOut)>0 ){
blob_read_from_file(pOut, zOut);
file_delete(zPivot);
file_delete(zOrig);
file_delete(zOther);
file_delete(zOut);
}
fossil_free(zCmd);
fossil_free(zOut);
}
}
fossil_free(zPivot);
fossil_free(zOrig);
fossil_free(zOther);
}
blob_reset(&v1);
return rc;
}
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] bring back a deleted file; unsure of command syntax
Hi, I did a merge, merging trunk into my "sound" branch. I submitted it and then I noticed that merge had caused fossil to delete some files (the files were in the "sound" branch but not in "trunk"). How can I go back to the previous revision of a specific file to restore it? on sound branch [3fc31bf75c] is previous and [66c97725ba] is new. in the changes for [66c97725ba] I see this in the fossil ui: Deleted src/arngui/BitmapFont.cpp version [128443f62f]. How do I bring this file back? Thanks ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] bring back a deleted file; unsure of command syntax
I did that and it worked, but not as I expected. :( I had to re-add them and then commit. I was hoping that fossil would have them marked as add so that when I commited they were re-added. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] fossil reporting MISSING for existing file
Hi, I have a strange problem with fossil on my linux machine. Somehow I broke it and I don't know how. Whenever I make an edit, "fossil changes" will report those edits but also display: MISSING build_notes.txt That file exists locally and exists in the "main" fossil repository too. My local copy of the repository seems to think it is missing. The main repository is hosted on another machine on the internet. My local repository has autosync on and my changes are pushed up when I submit. Doing: fossil revert build_notes.txt Will then allow me to submit. Is there a way to find out what I have done? Thanks. Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Name of ZIP file for "ZIP archive"
Hi, In fossil I can select a change, click on "ZIP archive" to download a zip. This is nice. I have been using it frequently and I have been asking people who are not using fossil to get a snapshot of the code. However the file is always named "zip" without extension. When extracted it has a folder which has the name of the repository and a short artifact number. Please would the zip be named the same as this and have the zip extension? It makes it easier on Windows because shell extensions will recognise it and the shell will give it a nice icon. It takes away an extra unncessary renaming step. Thanks Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Name of ZIP file for "ZIP archive"
Hi Stephen, Can the fossil zip command be used to configure the name for the UI? >If that's happening when you click the zip button in the UI, then that's a >bug. i cannot reproduce that on the main fossil site. Yes it happens via the UI. Click on artifact Click on Download ZIP Login with anonymous user See the display: ZIP Archive named Arnold-97bb1e1236.zip holding the content of check-in 97bb1e12363f3865: Click on the download button. File is named "zip". Happens with firefox. Thanks Kev ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Name of ZIP file for "ZIP archive"
I put Stephen and not Stephan. Sorry. My mistake. Kev ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Name of ZIP file for "ZIP archive"
Hi Stephan, The bug doesn't happen if you're logged in. You have to be logged out. Click the download zip link login as anonymous then a page appears with a download button clicking this button gives the wrong name. I have e-mailed you the link to the fossil repository I am using. I didn't post it here because I don't want it to be public. Thanks Kev ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Name of ZIP file for "ZIP archive"
Thankyou Andy and Stephan! :) Kev ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] characters outside of commit comment are included in comment
Hi, I found a small bug. If I type: fossil commit -m "This is a test"£ Then the commit comment shows as "This is a test£". The character after the quote is included. I don't know if this expected behaviour. I saw this when using fossil on linux. Kevin ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] characters outside of commit comment are included in comment
>Your shell (bash) is doing this, not Fossil. Bash hands Fossil the >string "This is a test£" with no indication of whether or not it was >originally quoted nor where the quotes ended. Ok. :) ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
