From: Sverre Rabbelier <srabbelier@gmail.com>
To: Jonathan Nieder <jrnieder@gmail.com>, "Jeff King" <peff@peff.net>,
Git List <git@vger.kernel.org>,
Daniel Barkalow <barkalow@iabervon.org>,
Ramkumar Ramachandra <artagnon@gmail.com>
Cc: Sverre Rabbelier <srabbelier@gmail.com>
Subject: [PATCH 10/19] fast-import: introduce 'done' command
Date: Wed, 8 Jun 2011 20:48:41 +0200 [thread overview]
Message-ID: <1307558930-16074-11-git-send-email-srabbelier@gmail.com> (raw)
In-Reply-To: <1307558930-16074-1-git-send-email-srabbelier@gmail.com>
Add a 'done' command that causes fast-import to stop reading from the
stream and exit.
If the new --done command line flag was passed on the command line
(or a "feature done" declaration included at the start of the stream),
make the 'done' command mandatory. So "git fast-import --done"'s
input format will be prefix-free, making errors easier to detect when
they show up as early termination at some convenient time of the
upstream of a pipe writing to fast-import.
Another possible application of the 'done' command would to be allow a
fast-import stream that is only a small part of a larger encapsulating
stream to be easily parsed, leaving the file offset after the "done\n"
so the other application can pick up from there. This patch does not
teach fast-import to do that --- fast-import still uses buffered input
(stdio).
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Sverre Rabbelier <srabbelier@gmail.com>
---
This is actually Jonathan's version that makes done required if
feature done was specified. See also:
http://mid.gmane.org/20110213094212.GA25435@elie
Documentation/git-fast-import.txt | 25 ++++++++++++++++++++++
fast-import.c | 8 +++++++
t/t9300-fast-import.sh | 42 +++++++++++++++++++++++++++++++++++++
3 files changed, 75 insertions(+), 0 deletions(-)
diff --git a/Documentation/git-fast-import.txt b/Documentation/git-fast-import.txt
index 249249a..0fc68a9 100644
--- a/Documentation/git-fast-import.txt
+++ b/Documentation/git-fast-import.txt
@@ -101,6 +101,12 @@ OPTIONS
when the `cat-blob` command is encountered in the stream.
The default behaviour is to write to `stdout`.
+--done::
+ Require a `done` command at the end of the stream.
+ This option might be useful for detecting errors that
+ cause the frontend to terminate before it has started to
+ write a stream.
+
--export-pack-edges=<file>::
After creating a packfile, print a line of data to
<file> listing the filename of the packfile and the last
@@ -330,6 +336,11 @@ and control the current import process. More detailed discussion
standard output. This command is optional and is not needed
to perform an import.
+`done`::
+ Marks the end of the stream. This command is optional
+ unless the `done` feature was requested using the
+ `--done` command line option or `feature done` command.
+
`cat-blob`::
Causes fast-import to print a blob in 'cat-file --batch'
format to the file descriptor set with `--cat-blob-fd` or
@@ -1015,6 +1026,11 @@ notes::
Versions of fast-import not supporting notes will exit
with a message indicating so.
+done::
+ Error out if the stream ends without a 'done' command.
+ Without this feature, errors causing the frontend to end
+ abruptly at a convenient point in the stream can go
+ undetected.
`option`
~~~~~~~~
@@ -1044,6 +1060,15 @@ not be passed as option:
* cat-blob-fd
* force
+`done`
+~~~~~~
+If the `done` feature is not in use, treated as if EOF was read.
+This can be used to tell fast-import to finish early.
+
+If the `--done` command line option or `feature done` command is
+in use, the `done` command is mandatory and marks the end of the
+stream.
+
Crash Reports
-------------
If fast-import is supplied invalid input it will terminate with a
diff --git a/fast-import.c b/fast-import.c
index 78d9786..8a8a915 100644
--- a/fast-import.c
+++ b/fast-import.c
@@ -354,6 +354,7 @@ static unsigned int cmd_save = 100;
static uintmax_t next_mark;
static struct strbuf new_data = STRBUF_INIT;
static int seen_data_command;
+static int require_explicit_termination;
/* Signal handling */
static volatile sig_atomic_t checkpoint_requested;
@@ -3139,6 +3140,8 @@ static int parse_one_feature(const char *feature, int from_stream)
relative_marks_paths = 1;
} else if (!strcmp(feature, "no-relative-marks")) {
relative_marks_paths = 0;
+ } else if (!strcmp(feature, "done")) {
+ require_explicit_termination = 1;
} else if (!strcmp(feature, "force")) {
force_update = 1;
} else if (!strcmp(feature, "notes") || !strcmp(feature, "ls")) {
@@ -3288,6 +3291,8 @@ int main(int argc, const char **argv)
parse_reset_branch();
else if (!strcmp("checkpoint", command_buf.buf))
parse_checkpoint();
+ else if (!strcmp("done", command_buf.buf))
+ break;
else if (!prefixcmp(command_buf.buf, "progress "))
parse_progress();
else if (!prefixcmp(command_buf.buf, "feature "))
@@ -3307,6 +3312,9 @@ int main(int argc, const char **argv)
if (!seen_data_command)
parse_argv();
+ if (require_explicit_termination && feof(stdin))
+ die("stream ends early");
+
end_packfile();
dump_branches();
diff --git a/t/t9300-fast-import.sh b/t/t9300-fast-import.sh
index 6b1ba6c..526231b 100755
--- a/t/t9300-fast-import.sh
+++ b/t/t9300-fast-import.sh
@@ -2197,6 +2197,48 @@ test_expect_success 'R: quiet option results in no stats being output' '
test_cmp empty output
'
+test_expect_success 'R: feature done means terminating "done" is mandatory' '
+ echo feature done | test_must_fail git fast-import &&
+ test_must_fail git fast-import --done </dev/null
+'
+
+test_expect_success 'R: terminating "done" with trailing gibberish is ok' '
+ git fast-import <<-\EOF &&
+ feature done
+ done
+ trailing gibberish
+ EOF
+ git fast-import <<-\EOF
+ done
+ more trailing gibberish
+ EOF
+'
+
+test_expect_success 'R: terminating "done" within commit' '
+ cat >expect <<-\EOF &&
+ OBJID
+ :000000 100644 OBJID OBJID A hello.c
+ :000000 100644 OBJID OBJID A hello2.c
+ EOF
+ git fast-import <<-EOF &&
+ commit refs/heads/done-ends
+ committer $GIT_COMMITTER_NAME <$GIT_COMMITTER_EMAIL> $GIT_COMMITTER_DATE
+ data <<EOT
+ Commit terminated by "done" command
+ EOT
+ M 100644 inline hello.c
+ data <<EOT
+ Hello, world.
+ EOT
+ C hello.c hello2.c
+ done
+ EOF
+ git rev-list done-ends |
+ git diff-tree -r --stdin --root --always |
+ sed -e "s/$_x40/OBJID/g" >actual &&
+ test_cmp expect actual
+'
+
cat >input <<EOF
option git non-existing-option
EOF
--
1.7.5.1.292.g728120
next prev parent reply other threads:[~2011-06-08 18:50 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-06-08 18:48 [PATCH 00/19] remote-helper improvements Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 01/19] transport-helper: fix minor leak in push_refs_with_export Sverre Rabbelier
2011-06-08 21:57 ` Jeff King
2011-06-08 22:08 ` Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 02/19] t5800: factor out some ref tests Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 03/19] t5800: document some non-functional parts of remote helpers Sverre Rabbelier
2011-06-08 19:28 ` Jonathan Nieder
2011-06-08 19:36 ` Jonathan Nieder
2011-06-08 19:51 ` Sverre Rabbelier
2011-06-08 21:13 ` Sverre Rabbelier
2011-06-09 12:45 ` Sverre Rabbelier
2011-06-10 1:18 ` Jeff King
2011-06-08 18:48 ` [PATCH 04/19] teach remote-testgit to import non-HEAD refs Sverre Rabbelier
2011-06-08 19:30 ` Jonathan Nieder
2011-06-08 19:47 ` Sverre Rabbelier
2011-06-08 22:15 ` Jeff King
2011-06-08 23:48 ` Junio C Hamano
2011-06-09 6:23 ` Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 05/19] transport-helper: don't feed bogus refs to export push Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 06/19] git_remote_helpers: push all refs during a non-local export Sverre Rabbelier
2011-06-08 19:42 ` Jonathan Nieder
2011-06-08 22:19 ` Jeff King
2011-06-08 22:21 ` Sverre Rabbelier
2011-06-09 8:09 ` Jonathan Nieder
2011-06-09 8:29 ` Sverre Rabbelier
2011-06-09 8:43 ` Jonathan Nieder
2011-06-09 10:26 ` Sverre Rabbelier
2011-06-10 1:40 ` Jeff King
2011-06-08 18:48 ` [PATCH 07/19] remote-curl: accept empty line as terminator Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 08/19] git-remote-testgit: only push for non-local repositories Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 09/19] git-remote-testgit: fix error handling Sverre Rabbelier
2011-06-08 18:48 ` Sverre Rabbelier [this message]
2011-06-08 20:03 ` [PATCH 10/19] fast-import: introduce 'done' command Jonathan Nieder
2011-06-08 20:07 ` Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 11/19] fast-export: support done feature Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 12/19] transport-helper: factor out push_update_refs_status Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 13/19] transport-helper: check status code of finish_command Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 14/19] transport-helper: use the new done feature where possible Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 15/19] transport-helper: update ref status after push with export Sverre Rabbelier
2011-06-09 9:10 ` Jonathan Nieder
2011-06-09 10:23 ` Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 16/19] transport-helper: change import semantics Sverre Rabbelier
2011-06-08 20:47 ` Jonathan Nieder
2011-06-08 20:52 ` Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 17/19] transport-helper: export is no longer always the last command Sverre Rabbelier
2011-06-09 1:07 ` Junio C Hamano
2011-06-09 6:48 ` Sverre Rabbelier
2011-06-09 7:51 ` Jonathan Nieder
2011-06-09 8:28 ` Sverre Rabbelier
2011-06-13 15:24 ` Junio C Hamano
2011-06-08 18:48 ` [PATCH 18/19] transport-helper: Use capname for gitdir capability too Sverre Rabbelier
2011-06-08 20:54 ` Jonathan Nieder
2011-06-08 20:57 ` Sverre Rabbelier
2011-06-08 18:48 ` [PATCH 19/19] transport-helper: implement marks location as capability Sverre Rabbelier
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
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1307558930-16074-11-git-send-email-srabbelier@gmail.com \
--to=srabbelier@gmail.com \
--cc=artagnon@gmail.com \
--cc=barkalow@iabervon.org \
--cc=git@vger.kernel.org \
--cc=jrnieder@gmail.com \
--cc=peff@peff.net \
/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.
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 NNTP newsgroup(s).