http://permalink.gmane.org/gmane.comp.version-control.darcs.user hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16291 Nice one! Simon Michael 2008-12-04T03:54:50 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16290 The best I can think of (while waiting for getpref) is parsing the output of show repo and concatenate the values of Root and boringfile Pref (defaulting boringfile Pref to _darcs/prefs/boring if it is not set). ; darcs show repo Type: darcs Format: hashed, darcs-2 Root: /home/tommy/tmp/foo Pristine: HashedPristine Cache: thisrepo:/home/tommy/tmp/foo, cache:/home/tommy/.darcs/cache boringfile Pref: d/f Num Patches: 1 This will give an absolute path to the boringfile no matter where in the repo you are. A quick test shows it works with non-ascii chars in the path too. Hm well, it doesn't work with newlines in the path, but I think that can pass. O my, after changing the boringfile to 'a\nb', and than to something else, I have a line with a 'b' in _darcs/prefs/prefs. :-) Tommy Pettersson 2008-12-03T20:52:18 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16289 In particular, we had a bug in 6.8.3 and earlier with buffer management in the code that parses the module header. On Windows this results in the dreaded "Bad interface file" error that randomly appears, but I imagine it's possible to get segfaults too. If I recall correctly, the bug was aggravated by having long comments before the 'module' keyword in a source file, or lots of {-# OPTIONS #-} pragmas. Cheers, Simon Simon Marlow 2008-12-03T19:30:55 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16288 http://projects.haskell.org/camp/unique Kari Hoijarvi 2008-12-03T18:32:30 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16287 -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 On Tue, Dec 2, 2008 at 6:46 PM, Trent W. Buck wrote: No, fortunately that's not the case. My first patch removed IFS from all of them. This caused the tests which did in fact need IFS to fail. My second patch did it again, but spared the failing tests - which still removed the IFS stuff from a fair number of tests. So all the remaining calls to IFS need it somehow. - -- gwern -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREKAAYFAkk11SoACgkQvpDo5Pfl1oKMaACdEs3Gd9gf9PgiRdpTfmmACHc0 CCYAoI99nrMxglGzxClvqGL1gOb0uve7 =d4HM -----END PGP SIGNATURE----- Gwern Branwen 2008-12-03T00:39:47 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16286 I don't see why not, but if IFS is there to affect read, perhaps it would be better to use bashisms. At the very least, I would expect while IFS='' read foo; do ...; done to be better than the stateful oldIFS="$IFS" IFS='' while read foo; do ...; done IFS="$oldIFS" I thought that was being done already. Trent W. Buck 2008-12-02T23:49:21 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16285 In general, setting IFS can lead to hairy and confusing bugs, and so I'd advocate leaving it visible. But I haven't looked at the tests, so I don't know why IFS is being set -- it might be that many of the tests that currently set IFS don't need to; someone just blindly copied the line from an older test. _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users Trent W. Buck 2008-12-02T23:46:37 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16284 Yes, I have. .From looking at http://github.com/jgm/gitit/tree/master/Gitit%2FGit.hs with my small knowledge of Haskell, it looks like it'd be possible to write a similar layer by calling darcs' binary, as I do for wikiri; with the exception of grep (can be replaced by regular grep over the files resulting from "darcs query manifest") and gitMergeFile (I have no idea if this can be replaced, but maybe it can be avoided). If anyone is interested in writing such code, maybe the wikiri code (which is written in Python) can be useful for the commands equivalence (and maybe parsing). The relevant section of the code is in wikiri.cgi, line 644 (http://blitiri.com.ar/git/?p=wikiri;a=blob;f=wikiri.cgi;hb=HEAD#l644). Obviously if anyone has any questions regarding it, please let me know. Thanks, Alberto Alberto Bertogli 2008-12-02T22:11:44 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16283 _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users Eric Kow 2008-12-02T21:40:28 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16282 _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users Eric Kow 2008-12-02T21:39:02 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16281 _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users Eric Kow 2008-12-02T11:16:20 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16280 _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users Christian Kellermann 2008-12-01T21:02:33 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16278 Folks, For reasons unrelated to darcs I need to remove cygwin gcc from the Windows machine. This should not be a problem for the darcs buildslave, of course, darcs doesn't use the locally-installed gcc when it is built on Windows -- it uses the gcc that comes bundled inside the ghc distribution. ghc is still installed on that machine. However, the autoconf scripts test for the existence of gcc when they start, and discovering that there is no gcc installed, they immediately error out. You can follow the links in the appended buildbot mail to see what I mean. If someone can patch darcs so that it builds on Windows even when gcc is not installed, then that would be great, but if not then I'm afraid I'm going to remove the Windows buildslave from the darcs buildbot farm for now. Regards, Zooko On Dec 1, 2008, at 13:19 PM, buildbot-darcs< at >allmydata.org wrote: zooko 2008-12-01T20:27:04 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16277 -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 On Mon, Dec 1, 2008 at 10:47 AM, Eric Kow wrote: (Sorry about the previous email, I pressed the wrong buttons.) Points: 1) Only lib callers are affected. 2) Should any of our scripts be calling read? 3) Each test should be running in its own shell or subshell for maximum separation. 4) We can conditionally set IFS on Windows, ie. if echo $OS | grep -i windows; then IFS='' fi - -- gwern -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREKAAYFAkk0F00ACgkQvpDo5Pfl1oJELQCeLy/MTVHG/st/sm9KDnz5Od6t DLkAnjZJ5+u3c5N5t6OEk0l314F0ggNP =PTyu -----END PGP SIGNATURE----- Gwern Branwen 2008-12-01T16:56:47 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16276 -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 - -----BEGIN PGP MESSAGE----- Version: GnuPG v1.4.9 (GNU/Linux) owFNVM+LHEUYnZgoWqAYDJKD4AcKk4GZ2d24oGxww/hjYZBkhRXEk1R3fz1dTk9V 21U9nQHB4EnwYI4RD7n4P3gTQTyIOS3i3aCgIngST4Lvq5pEmz50d3396n3vva9u PX52oM5/devXn//+5cePznz/ezaY7z99emzpmrNjeo1z2hvT5d3dl0gH2ts92H+R ZtfG9HprcnrD9UR96wIfqEOaVWOaU689vdf5QDpzXaDgUGAC08Z18tJy41p8roAW 2AdPumXywdQ1Ndp7YxeA0mXgVn5pqTQtwBod8mqqDrH2ClDlHwH0HR5CZTzh1rRw riBTsB5TxrXrx7IjlyXnway53tBKL5l2eM3tZgdQQoA8B5ofnbw8HF6N+G9VTJn2 FYptoxcgpzf+IC6hjHBJxdyCodU1HRmuCzrhRrc6uDZ1BjbUeS6ISnyi3rV49E1t QkgNpiu1SXyj0dYbZ/HFohCkUzFRbSyjOSsyAsRTb0KFPRhC6uIBUNaZGshEuVuB djHdsqSCS93VAFrrusO7ELt38zbu7Z+yweG9m5/hTvLOgWFL9j51Mt96Uzg7DFR2 NVTsbMEwRbgKE9T79zu2OZi6MtoBKDq20Rq7oPlwJexbA0FSKkAjQwNLWXXAaMk6 O4G8k01KBRqY2bgCpIK5QYlAbOTXSKyX1GhE6VJnl9b1gKo3oxg2QU3ZipUF/rai FQSOWcm40muDcKWO345FLiUUhJdXlZqhq66pOQrKNiAAeyO67gLF8KTk5LquPQ1r kw3V5RG92bLvVjqrN2NinVepCPu1nbXCCQYZkHK9JV8xREU2fJfF5yuRq5Lp6ZPu SAVyy0njCFXp1kZjHFoAY2BO6cRJgcxQ5bq6gEsKi9puIvXOmlwHiZbsBQDgrboA Gwern Branwen 2008-12-01T16:54:51 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16275 _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users Eric Kow 2008-12-01T15:47:22 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16274 Mon Dec 1 10:39:32 EST 2008 gwern0< at >gmail.com * tests/*: refactor remaining IFS calls to lib _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users gwern0< at >gmail.com 2008-12-01T15:40:08 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16273 _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users Eric Kow 2008-12-01T09:04:09 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16272 _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users Eric Kow 2008-12-01T09:01:56 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16271 _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users Eric Kow 2008-12-01T08:57:34 http://permalink.gmane.org/gmane.comp.version-control.darcs.user/16270 _______________________________________________ darcs-users mailing list darcs-users< at >darcs.net http://lists.osuosl.org/mailman/listinfo/darcs-users Eric Kow 2008-12-01T08:40:38 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.comp.version-control.darcs.user