On Fri, Apr 03, 2009 at 04:52:46AM +0200, Johannes Schindelin wrote: > In case it was not clear yet: if I have to chose between following a > standard and reality that just took over, I will _always_ choose the > latter. The problem with "reality" here is that it is not 100% case-insensitive (and I'm not sure we have anything resembling actual data), and worse, that the person using git and the person whose reality is case-sensitive are _different_ people. So you could be breaking email addresses and have no idea, because the case-sensitivity is implemented on some random contributor's mail server. That being said, this is actually restoring behavior broken only about 6 weeks ago (and looks like it was broken in 1.6.2, but no other release). So I am inclined to say this patch should be applied regardless of standards and reality if only because it is making git's behavior consistent across versions. And it looks like Junio has already applied it. Now what git _should_ do is another question. Mailmap has been lower-casing addresses since 5496523 (shortlog: handle email addresses case-insensitively, 2006-11-19), and I don't think we've had a single complaint or bug report. Which implies that either reality _is_ primarily case-sensitive, or that people are not actually feeding mailmap'd addresses to mail servers. So if there is an issue, it is certainly not pressing, and while I can see and support a concerned somebody making a patch to address it, I am not that somebody. Such a patch would have to take into account the very real use case Dscho is handling, but also address some other real use case (like cutting and pasting from shortlog output into a MUA). I assume the lower-casing is primarily used for aggregating similar addresses; we can do better to _preserve_ case while matching insensitively (and if there are multiple cases, then we pick one arbitrarily -- presumably that address _is_ case insensitive if it was used both ways). Doing my best to turn a flame war into an actual discussion of merit, -Peff -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
