On Tue, Apr 25, 2023 at 01:52:45AM -0400, Jeff King wrote:
> Here's a v2 of my series. The behavior should be identical, but I've
> incorporated some comment and small code tweaks based on feedback from
> the first round.
>
> I also added a fourth patch which adds a new comment explaining some of
> the cases that were alluded to in the earlier round's patch 3.
>
> [1/4]: t4212: avoid putting git on left-hand side of pipe
> [2/4]: parse_commit(): parse timestamp from end of line
> [3/4]: parse_commit(): handle broken whitespace-only timestamp
> [4/4]: parse_commit(): describe more date-parsing failure modes
>
> commit.c | 47 +++++++++++++++++++++++++++++++++++-------
> t/t4212-log-corrupt.sh | 39 +++++++++++++++++++++++++++++++++--
> 2 files changed, 76 insertions(+), 10 deletions(-)
Whoops, forgot my range-diff (though nothing should be too surprising
based on the round 1 discussion):
1: 07932cf666 = 1: ac38ce133d t4212: avoid putting git on left-hand side of pipe
2: 7ee34c7d5f ! 2: f59e61262d parse_commit(): parse timestamp from end of line
@@ Commit message
parse back to the final ">". In theory we could use split_ident_line()
here, but it's actually a bit more strict. In particular, it requires a
valid time-zone token, too. That should be present, of course, but we
- wouldn't want to break --until for malformed cases that are working
- currently.
+ wouldn't want to break --until for cases that are working currently.
We might want to teach split_ident_line() to become more lenient there,
but it would require checking its many callers (since right now they can
@@ commit.c: static timestamp_t parse_commit_date(const char *buf, const char *tail
- if (buf >= tail)
+
+ /*
-+ * parse to end-of-line and then walk backwards, which
-+ * handles some malformed cases.
++ * Jump to end-of-line so that we can walk backwards to find the
++ * end-of-email ">". This is more forgiving of malformed cases
++ * because unexpected characters tend to be in the name and email
++ * fields.
+ */
+ eol = memchr(buf, '\n', tail - buf);
+ if (!eol)
return 0;
- dateptr = buf;
- while (buf < tail && *buf++ != '\n')
-+ for (dateptr = eol; dateptr > buf && dateptr[-1] != '>'; dateptr--)
- /* nada */;
+- /* nada */;
- if (buf >= tail)
++ dateptr = eol;
++ while (dateptr > buf && dateptr[-1] != '>')
++ dateptr--;
+ if (dateptr == buf || dateptr == eol)
return 0;
- /* dateptr < buf && buf[-1] == '\n', so parsing will stop at buf-1 */
3: e8e94083f5 ! 3: c62fc59bf1 parse_commit(): handle broken whitespace-only timestamp
@@ Commit message
It's not subject to the same bug, because it insists that there be one
or more digits in the timestamp.
- We can use the same logic here. If there's a non-whitespace but
- non-digit value (say "committer name <email> foo"), then
- parse_timestamp() would already have returned 0 anyway. So the only
- change should be for this "whitespace only" case.
-
Signed-off-by: Jeff King <peff@xxxxxxxx>
## commit.c ##
@@ commit.c: static timestamp_t parse_commit_date(const char *buf, const char *tail)
- if (dateptr == buf || dateptr == eol)
+ dateptr = eol;
+ while (dateptr > buf && dateptr[-1] != '>')
+ dateptr--;
+- if (dateptr == buf || dateptr == eol)
++ if (dateptr == buf)
return 0;
+- /* dateptr < eol && *eol == '\n', so parsing will stop at eol */
+ /*
-+ * trim leading whitespace; parse_timestamp() will do this itself, but
-+ * it will walk past the newline at eol while doing so. So we insist
-+ * that there is at least one digit here.
++ * Trim leading whitespace; parse_timestamp() will do this itself, but
++ * if we have _only_ whitespace, it will walk right past the newline
++ * while doing so.
+ */
+ while (dateptr < eol && isspace(*dateptr))
+ dateptr++;
-+ if (!strchr("0123456789", *dateptr))
++ if (dateptr == eol)
+ return 0;
+
- /* dateptr < eol && *eol == '\n', so parsing will stop at eol */
++ /*
++ * We know there is at least one non-whitespace character, so we'll
++ * begin parsing there and stop at worst case at eol.
++ */
return parse_timestamp(dateptr, NULL, 10);
}
+
## t/t4212-log-corrupt.sh ##
@@ t/t4212-log-corrupt.sh: test_expect_success 'absurdly far-in-future date' '
-: ---------- > 4: 28ed51a2ca parse_commit(): describe more date-parsing failure modes
