# [RFC PATCH] Make :/ accept a regex rather than a fixed pattern 2 messages from 2010-05-13 to 2010-06-18. Participants: Linus Torvalds, Junio C Hamano. Thread: https://gitlist.dev/t/23802 ## Linus Torvalds, 2010-05-13 22:08 Subject: [RFC PATCH] Make :/ accept a regex rather than a fixed pattern Message-ID: URL: https://gitlist.dev/e/alpine.LFD.2.00.1005131457350.3711%40i5.linux-foundation.org ``` From: Linus Torvalds Date: Fri, 23 Apr 2010 08:20:20 -0700 Subject: [PATCH] Make :/ accept a regex rather than a fixed pattern This also makes it trigger anywhere in the commit message, rather than just at the beginning. Which tends to be a lot more useful. Signed-off-by: Linus Torvalds --- I'm re-sending, because after having this in my tree for several weeks now, I actually end up still using it. I agree that it's still not a wonderful thing, and it's entirely possible that we should strive to use just HEAD in :/ commit finding rather than all refs we can find, but for at least the kernel, I don't care (since my tree tends to have just one main branch anyway, apart from some random testing stuff I keep around). But even if we want to limit it to HEAD, that would be a separate patch. I've personally used it for things like git show :/mqueue # did I apply that 'mqueue' patch? git show :/akpm # what was the last patch I got from Andrew? and while in all cases I could admittedly have done exactly the same thing with something like "git log -p -1 --grep=mqueue" instead, I've used it as a nice shortcut. Of course, I'm still not convinced that :/ was a good feature to begin with. If we had a short form of "--grep" ("-/"?), I would probably happily have written "git log -/akpm" instead. So take this patch with a large pinch of salt. It's just not very important. sha1_name.c | 12 +++++++++--- 1 files changed, 9 insertions(+), 3 deletions(-) diff --git a/sha1_name.c b/sha1_name.c index bf92417..8cf635a 100644 --- a/sha1_name.c +++ b/sha1_name.c @@ -679,8 +679,8 @@ static int handle_one_ref(const char *path, /* * This interprets names like ':/Initial revision of "git"' by searching - * through history and returning the first commit whose message starts - * with the given string. + * through history and returning the first commit whose message matches + * the given regular expression. * * For future extension, ':/!' is reserved. If you want to match a message * beginning with a '!', you have to repeat the exclamation mark. @@ -692,12 +692,17 @@ static int get_sha1_oneline(const char *prefix, unsigned char *sha1) struct commit_list *list = NULL, *backup = NULL, *l; int retval = -1; char *temp_commit_buffer = NULL; + regex_t regex; if (prefix[0] == '!') { if (prefix[1] != '!') die ("Invalid search pattern: %s", prefix); prefix++; } + + if (regcomp(®ex, prefix, REG_EXTENDED)) + die("Invalid search pattern: %s", prefix); + for_each_ref(handle_one_ref, &list); for (l = list; l; l = l->next) commit_list_insert(l->item, &backup); @@ -721,12 +726,13 @@ static int get_sha1_oneline(const char *prefix, unsigned char *sha1) } if (!(p = strstr(p, "\n\n"))) continue; - if (!prefixcmp(p + 2, prefix)) { + if (!regexec(®ex, p + 2, 0, NULL, 0)) { hashcpy(sha1, commit->object.sha1); retval = 0; break; } } + regfree(®ex); free(temp_commit_buffer); free_commit_list(list); for (l = backup; l; l = l->next) -- 1.7.1.83.g9d02 ``` ## Junio C Hamano, 2010-06-18 17:03 Subject: Re: [RFC PATCH] Make :/ accept a regex rather than a fixed pattern Message-ID: <7vhbl0xoh2.fsf@alter.siamese.dyndns.org> URL: https://gitlist.dev/e/7vhbl0xoh2.fsf%40alter.siamese.dyndns.org In-Reply-To: ``` Linus Torvalds writes: > From: Linus Torvalds > Date: Fri, 23 Apr 2010 08:20:20 -0700 > Subject: [PATCH] Make :/ accept a regex rather than a fixed pattern > > This also makes it trigger anywhere in the commit message, rather than > just at the beginning. Which tends to be a lot more useful. > > Signed-off-by: Linus Torvalds > --- > > I'm re-sending, because after having this in my tree for several weeks > now, I actually end up still using it. > > I agree that it's still not a wonderful thing, and it's entirely possible > that we should strive to use just HEAD in :/ commit finding rather than > all refs we can find, but for at least the kernel, I don't care (since my > tree tends to have just one main branch anyway, apart from some random > testing stuff I keep around). > > But even if we want to limit it to HEAD, that would be a separate patch. > > I've personally used it for things like > > git show :/mqueue # did I apply that 'mqueue' patch? > git show :/akpm # what was the last patch I got from Andrew? > > and while in all cases I could admittedly have done exactly the same thing > with something like "git log -p -1 --grep=mqueue" instead, I've used it as > a nice shortcut. I already took this but I actually started hating it. As I have merge.summary on, git show ':/diff: add configuration option' almost never hits the commit I am looking for---instead it ends up finding a merge that merges it. I however suspect that I would hate it much less after retraining my fingers to anchor the pattern at the beginning of the line. But not yet... ```