From: Jeff King <peff@peff.net>
To: Nguyen Thai Ngoc Duy <pclouds@gmail.com>
Cc: Johannes Sixt <j.sixt@viscovery.net>,
Git Mailing List <git@vger.kernel.org>
Subject: Re: 'git grep needle rev' attempts to access 'rev:.../.gitattributes' in the worktree
Date: Tue, 9 Oct 2012 08:41:44 -0400 [thread overview]
Message-ID: <20121009124144.GB12465@sigill.intra.peff.net> (raw)
In-Reply-To: <20121009120144.GA8780@do>
On Tue, Oct 09, 2012 at 07:01:44PM +0700, Nguyen Thai Ngoc Duy wrote:
> On Tue, Oct 09, 2012 at 04:38:32PM +0700, Nguyen Thai Ngoc Duy wrote:
> > #5 0x0815e736 in userdiff_find_by_path (path=0x820e7f0
> > "HEAD:Documentation/.gitattributes") at userdiff.c:278
> > #6 0x081058ca in grep_source_load_driver (gs=0xbfffd978) at grep.c:1504
>
> A bandage patch may look like this. But it does not solve the real
> problem:
>
> - we should be able to look up in-tree .gitattributes, not just
> ignore like this patch does
>
> - gs->name seems to be prepared for human consumption, not for
> accessing files. grep_file() with opt->relative is true can put
> quotes in gs->name, for example.
Right. For the second, you would probably want gs->identifier in the
case of GREP_SOURCE_FILE. But that identifier information is not
available at all for GREP_SOURCE_SHA1, which is what is breaking the
first point.
> I feel like we should make this function a callback and let git-grep
> deal with driver loading itself.
I think we just need to have callers of grep_source_init provide us with
the actual pathname (or NULL, in the case of GREP_SOURCE_BUF). That is
where the information is lost.
Like this incomplete and untested patch, which should fix the quoting
problem for GREP_SOURCE_FILE, but leave the sha1 bits broken (see the
in-code comment). I'm traveling this week, so I doubt I'll have time to
look for a few more days. If you want to work on it, please do.
diff --git a/builtin/grep.c b/builtin/grep.c
index 82530a6..be602dd 100644
--- a/builtin/grep.c
+++ b/builtin/grep.c
@@ -86,7 +86,7 @@ static void add_work(struct grep_opt *opt, enum grep_source_type type,
static int skip_first_line;
static void add_work(struct grep_opt *opt, enum grep_source_type type,
- const char *name, const void *id)
+ const char *name, const char *path, const void *id)
{
grep_lock();
@@ -94,7 +94,7 @@ static void add_work(struct grep_opt *opt, enum grep_source_type type,
pthread_cond_wait(&cond_write, &grep_mutex);
}
- grep_source_init(&todo[todo_end].source, type, name, id);
+ grep_source_init(&todo[todo_end].source, type, name, path, id);
if (opt->binary != GREP_BINARY_TEXT)
grep_source_load_driver(&todo[todo_end].source);
todo[todo_end].done = 0;
@@ -378,14 +378,21 @@ static int grep_sha1(struct grep_opt *opt, const unsigned char *sha1,
if (opt->relative && opt->prefix_length) {
quote_path_relative(filename + tree_name_len, -1, &pathbuf,
opt->prefix);
+ /* XXX Why do we insert here instead of just putting it in
+ * first? */
strbuf_insert(&pathbuf, 0, filename, tree_name_len);
} else {
strbuf_addstr(&pathbuf, filename);
}
+ /* XXX We seem to get all kinds of junk via the filename field here,
+ * including partial filenames, sha1:path, etc. We could parse it
+ * ourselves, but that is probably insanity. We should ask the
+ * caller to break it down more for us. For now, just pass NULL. */
+
#ifndef NO_PTHREADS
if (use_threads) {
- add_work(opt, GREP_SOURCE_SHA1, pathbuf.buf, sha1);
+ add_work(opt, GREP_SOURCE_SHA1, pathbuf.buf, NULL, sha1);
strbuf_release(&pathbuf);
return 0;
} else
@@ -394,7 +401,7 @@ static int grep_sha1(struct grep_opt *opt, const unsigned char *sha1,
struct grep_source gs;
int hit;
- grep_source_init(&gs, GREP_SOURCE_SHA1, pathbuf.buf, sha1);
+ grep_source_init(&gs, GREP_SOURCE_SHA1, pathbuf.buf, NULL, sha1);
strbuf_release(&pathbuf);
hit = grep_source(opt, &gs);
@@ -414,7 +421,7 @@ static int grep_file(struct grep_opt *opt, const char *filename)
#ifndef NO_PTHREADS
if (use_threads) {
- add_work(opt, GREP_SOURCE_FILE, buf.buf, filename);
+ add_work(opt, GREP_SOURCE_FILE, buf.buf, filename, filename);
strbuf_release(&buf);
return 0;
} else
@@ -423,7 +430,7 @@ static int grep_file(struct grep_opt *opt, const char *filename)
struct grep_source gs;
int hit;
- grep_source_init(&gs, GREP_SOURCE_FILE, buf.buf, filename);
+ grep_source_init(&gs, GREP_SOURCE_FILE, buf.buf, filename, filename);
strbuf_release(&buf);
hit = grep_source(opt, &gs);
diff --git a/grep.c b/grep.c
index edc7776..06bc1c6 100644
--- a/grep.c
+++ b/grep.c
@@ -1373,7 +1373,7 @@ int grep_buffer(struct grep_opt *opt, char *buf, unsigned long size)
struct grep_source gs;
int r;
- grep_source_init(&gs, GREP_SOURCE_BUF, NULL, NULL);
+ grep_source_init(&gs, GREP_SOURCE_BUF, NULL, NULL, NULL);
gs.buf = buf;
gs.size = size;
@@ -1384,10 +1384,12 @@ void grep_source_init(struct grep_source *gs, enum grep_source_type type,
}
void grep_source_init(struct grep_source *gs, enum grep_source_type type,
- const char *name, const void *identifier)
+ const char *name, const char *path,
+ const void *identifier)
{
gs->type = type;
gs->name = name ? xstrdup(name) : NULL;
+ gs->path = path ? xstrdup(path) : NULL;
gs->buf = NULL;
gs->size = 0;
gs->driver = NULL;
@@ -1409,6 +1411,8 @@ void grep_source_clear(struct grep_source *gs)
{
free(gs->name);
gs->name = NULL;
+ free(gs->path);
+ gs->path = NULL;
free(gs->identifier);
gs->identifier = NULL;
grep_source_clear_data(gs);
diff --git a/grep.h b/grep.h
index c256ac6..c2cf57b 100644
--- a/grep.h
+++ b/grep.h
@@ -158,11 +158,13 @@ void grep_source_init(struct grep_source *gs, enum grep_source_type type,
char *buf;
unsigned long size;
+ char *path; /* for attribute lookups */
struct userdiff_driver *driver;
};
void grep_source_init(struct grep_source *gs, enum grep_source_type type,
- const char *name, const void *identifier);
+ const char *name, const char *path,
+ const void *identifier);
void grep_source_clear_data(struct grep_source *gs);
void grep_source_clear(struct grep_source *gs);
void grep_source_load_driver(struct grep_source *gs);
next prev parent reply other threads:[~2012-10-09 12:41 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-09 9:03 'git grep needle rev' attempts to access 'rev:.../.gitattributes' in the worktree Johannes Sixt
2012-10-09 9:38 ` Nguyen Thai Ngoc Duy
2012-10-09 12:01 ` Nguyen Thai Ngoc Duy
2012-10-09 12:41 ` Jeff King [this message]
2012-10-09 18:59 ` Junio C Hamano
2012-10-10 5:17 ` Nguyen Thai Ngoc Duy
2012-10-10 5:33 ` Junio C Hamano
2012-10-10 5:45 ` Nguyen Thai Ngoc Duy
2012-10-10 11:34 ` Nguyễn Thái Ngọc Duy
2012-10-10 11:34 ` [PATCH 1/3] quote: let caller reset buffer for quote_path_relative() Nguyễn Thái Ngọc Duy
2012-10-10 21:13 ` Junio C Hamano
2012-10-11 13:04 ` Nguyen Thai Ngoc Duy
2012-10-11 16:42 ` Junio C Hamano
2012-10-10 11:34 ` [PATCH 2/3] grep: pass true path name to grep machinery Nguyễn Thái Ngọc Duy
2012-10-10 11:34 ` [PATCH 3/3] grep: stop looking at random places for .gitattributes Nguyễn Thái Ngọc Duy
2012-10-10 11:51 ` Johannes Sixt
2012-10-10 12:03 ` Nguyen Thai Ngoc Duy
2012-10-10 12:12 ` Johannes Sixt
2012-10-10 12:32 ` Nguyen Thai Ngoc Duy
2012-10-10 12:43 ` Johannes Sixt
2012-10-10 12:51 ` Nguyen Thai Ngoc Duy
2012-10-10 19:44 ` Junio C Hamano
2012-10-11 5:55 ` Johannes Sixt
2012-10-11 7:04 ` Michael Haggerty
2012-10-11 8:17 ` Nguyen Thai Ngoc Duy
2012-10-10 13:59 ` [PATCH v2 0/2] Re: 'git grep needle rev' attempts to access 'rev:.../.gitattributes' in the worktree Nguyễn Thái Ngọc Duy
2012-10-10 13:59 ` [PATCH v2 1/2] quote: let caller reset buffer for quote_path_relative() Nguyễn Thái Ngọc Duy
2012-10-10 13:59 ` [PATCH v2 2/2] grep: stop looking at random places for .gitattributes Nguyễn Thái Ngọc Duy
2012-10-10 14:21 ` Johannes Sixt
2012-10-10 19:56 ` Junio C Hamano
2012-10-11 5:45 ` Johannes Sixt
2012-10-11 15:51 ` Junio C Hamano
2012-10-12 7:33 ` Johannes Sixt
2012-10-14 4:29 ` Junio C Hamano
2012-10-15 6:02 ` Johannes Sixt
2012-10-15 16:54 ` Junio C Hamano
2012-10-16 6:39 ` Johannes Sixt
2012-10-17 7:05 ` Johannes Sixt
2012-10-17 7:33 ` Junio C Hamano
2012-10-11 1:49 ` Nguyen Thai Ngoc Duy
2012-10-11 3:15 ` Junio C Hamano
2012-10-12 10:49 ` [PATCH v3] " Nguyễn Thái Ngọc Duy
2012-10-12 16:47 ` Junio C Hamano
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: http://vger.kernel.org/majordomo-info.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20121009124144.GB12465@sigill.intra.peff.net \
--to=peff@peff.net \
--cc=git@vger.kernel.org \
--cc=j.sixt@viscovery.net \
--cc=pclouds@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://80x24.org/mirrors/git.git
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).