{"thread":{"id":"6956","subject":"[PATCH/RFC 0/8] use \"git diff\" in the test suite","startedAt":"2007-02-25T22:34:18Z","lastAt":"2007-02-25T23:19:52Z","messageCount":4,"participants":["Johannes Schindelin","Junio C Hamano"],"isPatch":true,"patchVersion":1,"patchTotal":8},"messages":[{"id":"35462","messageId":"Pine.LNX.4.63.0702252329550.22628@wbgn013.biozentrum.uni-wuerzburg.de","threadId":"6956","inReplyTo":null,"subject":"[PATCH/RFC 0/8] use \"git diff\" in the test suite","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2007-02-25T22:34:18Z","receivedAt":"2007-02-25T22:34:18Z","isPatch":true,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nNow that --no-index is supported by git's diff, we can use much of the \ntest suite to do the job, instead of relying on GNU diff.\n\nThis is a bit work-in-progress, since the handling of \"-\" as a filename is \nnot fleshed out yet, and we really should have the \"--quiet\" option. But I \ngot side-tracked by the edit-patch-series script, and did not want to hold \nthese patches off any longer.\n\nCiao,\nDscho\n"},{"id":"35470","messageId":"Pine.LNX.4.63.0702252349390.22628@wbgn013.biozentrum.uni-wuerzburg.de","threadId":"6956","inReplyTo":"Pine.LNX.4.63.0702252329550.22628@wbgn013.biozentrum.uni-wuerzburg.de","subject":"Re: [PATCH/RFC 0/8] use \"git diff\" in the test suite","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2007-02-25T22:50:36Z","receivedAt":"2007-02-25T22:50:36Z","isPatch":true,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nsomehow, patch 7/8 did not come through, I guess, so here is how to do it \nyourself:\n\n$ perl -pi.bak -e \"s/(^\\s*|'|&&\\s*)cmp -s /\\1git diff /\" *.sh\n\nWe should probably add a \"--quiet\" flag for the diff family first, and use \nthat...\n\nCiao,\nDscho\n"},{"id":"35473","messageId":"7vps7xrfxa.fsf@assigned-by-dhcp.cox.net","threadId":"6956","inReplyTo":"Pine.LNX.4.63.0702252329550.22628@wbgn013.biozentrum.uni-wuerzburg.de","subject":"Re: [PATCH/RFC 0/8] use \"git diff\" in the test suite","fromName":"Junio C Hamano","fromEmail":"junkio@cox.net","sentAt":"2007-02-25T23:11:29Z","receivedAt":"2007-02-25T23:11:29Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Johannes Schindelin writes:\n\n> Now that --no-index is supported by git's diff, we can use much of the \n> test suite to do the job, instead of relying on GNU diff.\n>\n> This is a bit work-in-progress, since the handling of \"-\" as a filename is \n> not fleshed out yet, and we really should have the \"--quiet\" option. But I \n> got side-tracked by the edit-patch-series script, and did not want to hold \n> these patches off any longer.\n\nAs you said yourself [6/8], [7/8], and [8/8] are mechanical\nconversions, and I do not think it is too much effort involved\non your side to \"hold these patches off\".\n\nMy preference is to apply [1/8] now, cook the result in 'next'\nfor a few days longer to see if anybody complains or finds\nglitches in the diff-ni series so far, push it out to 'master'\nand then apply [2/8] and [3/8] while on 'master'. The remainder\nwould be forked into a topic branch and cooked in 'pu' and then\n'next' as usual.\n\nAlso I do not think we would necessarily want [7/8] and [8/8].\nIt would help with tests on broken distributions whose 'cmp'\nmisbehaves when the output is redirected to /dev/null, but I do\nnot think it is a good enough reason to make the tests slower.\n"},{"id":"35475","messageId":"Pine.LNX.4.63.0702260015560.22628@wbgn013.biozentrum.uni-wuerzburg.de","threadId":"6956","inReplyTo":"7vps7xrfxa.fsf@assigned-by-dhcp.cox.net","subject":"Re: [PATCH/RFC 0/8] use \"git diff\" in the test suite","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2007-02-25T23:19:52Z","receivedAt":"2007-02-25T23:19:52Z","isPatch":true,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Sun, 25 Feb 2007, Junio C Hamano wrote:\n\n> Johannes Schindelin writes:\n> \n> > Now that --no-index is supported by git's diff, we can use much of the \n> > test suite to do the job, instead of relying on GNU diff.\n> >\n> > This is a bit work-in-progress, since the handling of \"-\" as a \n> > filename is not fleshed out yet, and we really should have the \n> > \"--quiet\" option. But I got side-tracked by the edit-patch-series \n> > script, and did not want to hold these patches off any longer.\n> \n> As you said yourself [6/8], [7/8], and [8/8] are mechanical conversions, \n> and I do not think it is too much effort involved on your side to \"hold \n> these patches off\".\n> \n> My preference is to apply [1/8] now, cook the result in 'next' for a few \n> days longer to see if anybody complains or finds glitches in the diff-ni \n> series so far, push it out to 'master' and then apply [2/8] and [3/8] \n> while on 'master'. The remainder would be forked into a topic branch \n> and cooked in 'pu' and then 'next' as usual.\n\nI fully agree.\n\nBTW I think that using 020600 as mode for \"-\" when it means stdin would \nfix [4/8]...\n\n> Also I do not think we would necessarily want [7/8] and [8/8]. It would \n> help with tests on broken distributions whose 'cmp' misbehaves when the \n> output is redirected to /dev/null, but I do not think it is a good \n> enough reason to make the tests slower.\n\nI think with the upcoming \"--quiet\", which exits early on error, \nperformance should be about par. Actually, I went off in this direction \nbecause I still use an OpenSUSE 10.1 install DVD, which has this broken \ncmp...\n\nBut I do not care deeply. (It was a good test case for diff-ni, though...)\n\nCiao,\nDscho\n"}]}