-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Coburn schrieb am 12.07.2006 14:45: > This is what darcs does when you pull in a patch that conflicts with > one of your local patches. It marks the conflicting variations in the > code. The conflict is then manually resolved (by editing the file) and > a new patch recorded. You will probably find the darcs manual helpful. > See: > http://www.abridgegame.org/darcs/manual/ > more specifically > http://www.abridgegame.org/darcs/manual/ > node6.html#SECTION00635000000000000000 Aaaah, thanks! This is the track i was looking for ;-) > > On Jul 12, 2006, at 4:02 AM, Werner Flamme wrote: > >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Hi, >> >> I just used the darcs plugin to update my installation. Something went >> wrong, in inc/template.php I read: >> >> - ---snip--- >> v v v v v v v >> trigger_event('TPL_CONTENT_DISPLAY',$html_output,'ptln'); > > Code from the incoming conflicting patch. > >> ************* >> trigger_event('TPL_CONTENT_DISPLAY',$html_output,ptln); >> >> return $html_output; > > Code from the conflicting patch in your repository. > >> ^ ^ ^ ^ ^ ^ ^ >> - ---pins--- >> >> and, of course, this is not valid PHP code. The line "return >> $html_output;" >> I added for my PDF export (see >> http://wiki.splitbrain.org/users:wflamme:pdfdownload). After adding >> this >> line, I used "darcs record" so darcs should know about this line (and >> the >> other changes I made for "pdfdownload"). >> > > Both patches conflict (see darcs manual) because they change the same > "trigger_event" line in a preceding patch. The command 'darcs whatsnew' > should clearly show the part of the file that needs to be changed (if > there are no other unrecorded changes). The whole block can be replaced > (without comments): > > // start > trigger_event('TPL_CONTENT_DISPLAY',$html_output,'ptln'); > > return $html_output; > // end > > > Regards, Ben Coburn Hmmm... no, I do not fully agree here. I *added* one line, the line "return $html_output;", I did *not* modify the line "trigger_event(...)" above. The conflict may be caused by by the context - but since patches are applied by line number, the following patches (same file, but lines below) may run wrong too (but I am not sure here). Shouldn't the darcs plugin give at least a hint that there was something not OK with the patch(es) just applied? Or show "Your pending_buggy file has reached a size of $x bytes" (if $x > 4)? I would take this as a hint to look for this file (I see ~18 k now) and to 'revert' all the files mentioned there... BTW, 'darcs revert' gets the version from the file server, I have to reapply my local patches afterwards... Nice to know that my webserver allows ssh login... Thanks to Chris, Andi an Ben for the helpful answers! Regards, Werner - -- Werner Flamme, Abt. WKDV UFZ Umweltforschungszentrum Leipzig-Halle GmbH, Permoserstr. 15 - 04318 Leipzig Tel.: (0341) 235-3921 - Fax (0341) 235-453921 http://www.ufz.de - eMail: werner.flamme@xxxxxx -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org iD8DBQFEtPcDk33Krq8b42MRAmtzAJ4xO9TheK3VqgJPUSP9X8HEkULylgCcDG8V d0WwMYLyFLa7FJn9oNqGMFg= =jh7p -----END PGP SIGNATURE----- -- DokuWiki mailing list - more info at http://wiki.splitbrain.org/wiki:mailinglist