The following python error:
Traceback (most recent call last):
File "./tools/patman/patman", line 144, in <module>
series = patchstream.FixPatches(series, args)
File "./tools/patman/patchstream.py", line 477, in FixPatches
commit = series.commits[count]
IndexError: list index out of range
is seen when:
- 'END' is missing in those tags
- those tags are put in the last part in a commit message
- the commit is not the last commit of the series
Add testing logic to see if a new commit starts.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
'Series-changes' uses blank line to indicate its end. If that is
missing, series internal state variable 'in_change' may be wrong.
Correct its state.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
If 'END' is missing in a 'Cover-letter' section, and that section
happens to show up at the very end of the commit message, and the
commit is the last commit of the series, patman fails to generate
cover letter for us. Handle this in CloseCommit of patchstream.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
'Cover-letter', 'Series-notes' and 'Commit-notes' tags require an
'END' to be put at the end of its section. If we forget to put an
'END' in those sections, and these sections are followed by another
patman tag, patman generates incorrect patches. This adds codes to
handle such scenario.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Like other patman tags, use a new variable cover_match to indicate
a match for 'Cover-letter'.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
At present if you try to use buildman with the branch 'test' it will
complain that it is unsure whether you mean the branch or the directory.
This is a feature of the 'git log' command that buildman uses. Fix it
by resolving the ambiguity.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Joe Hershberger <joe.hershberger@ni.com>
It is convenient to install symlinks to buildman and patman in the search
patch, such as /usr/local/bin. But when this is done, the -H option fails to
work because it looks in the directory containing the symlink instead of its
target. Fix this.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Joe Hershberger <joe.hershberger@ni.com>
A patman series with a 'Series-notes' section causes
buildman to crash with:
self.series.notes += self.section
TypeError: cannot concatenate 'str' and 'list' objects
Fix by initializing series.notes as a one-element array
rather than a scalar.
Signed-off-by: Albert ARIBAUD <albert.u.boot@aribaud.net>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
Correct spelling of "U-Boot" shall be used in all written text
(documentation, comments in source files etc.).
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Minkyu Kang <mk7.kang@samsung.com>
Add option to create threaded series of patches.
With it, it will be possible to create patch threads like this:
[PATCH 0/10] Add support for time travel
[PATCH 1/10] Add Flux Capacitor driver
[PATCH 2/10] Add Mr. Fusion driver
(...)
Internally it will call git send-email with --thread option
Signed-off-by: Mateusz Kulikowski <mateusz.kulikowski@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
This reverts commit 19b4a33698.
Since that commit, patman generates useless patches for file removal;
"git format -D" prints only the header but not the diff when deleting
files, and "git am" always refuses such patches.
The following is the quotation from "man git-format-patch":
-D, --irreversible-delete
Omit the preimage for deletes, i.e. print only the header but
not the diff between the preimage and /dev/null. The resulting
patch is not meant to be applied with patch nor git apply; this
is solely for people who want to just concentrate on reviewing
the text after the change. In addition, the output obviously
lack enough information to apply such a patch in reverse, even
manually, hence the name of the option.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Simon Glass <sjg@chromium.org>
Commit 488d19c (patman: add distutils based installer) has the side effect
of making patman run twice with each invocation. Fix this by checking for
'main program' invocation in patman.py. This is good practice in any case.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Chris Packham <judge.packham@gmail.com>
To make it easier to use patman on other projects add a distutils style
installer. Now patman can be installed with
cd u-boot/tools/patman && python setup.py install
There are also the usual distutils options for creating source/binary
distributions of patman.
Tested-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
For the local project, we may specified format.subjectprefix setting.
Then the patch will be formated as [Project_prefix][PATCH].
But patman will not check this setting. It will remove the
format.subjectprefix.
So This patch will let patman check this setting and add it as a
project prefix.
Signed-off-by: Josh Wu <josh.wu@atmel.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
We should read this file to obtain a set of aliases. This reduces the need
to create them in the ~/.patman file.
This feature did exist in some version of patman, and is mentioned in the
help but it did not find its way upstream.
Reported-by: Graeme Russ <gruss@tss-engineering.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
This causes an error when trying to build a local branch which has a local
branch as its upstream.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Add an explanation for how to set up git so that patman can find the alias
file. Fix up the get_maintainers message too.
Reported-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
True commit lines start at column zero. Anything that is indented
is part of the commit message instead. I noticed this by trying to
run buildman with commit e3a4facdfc
as master, which contained a reference to a Linux commit inside
the commit message. ProcessLine saw that as a genuite commit
line, and thus buildman tried to build it, and died with an
exception because that SHA is not present in the U-Boot tree.
Signed-off-by: Scott Wood <scottwood@freescale.com>
Acked-by: Simon Glass <sjg@chromium.org>
When run with the --dry-run argument patman prints out information
showing what it would do. This information currently doesn't line up
with what patman/git send-email really do. Some basic examples:
- If an email address is addressed via "Series-cc" and "Patch-cc" patman
shows that email address would be CC-ed two times.
- If an email address is addressed via "Series-to" and "Patch-cc" patman
shows that email address would be sent TO and CC-ed.
- If an email address is addressed from a combination of tag aliases,
get_maintainer.pl output, "Series-cc", "Patch-cc", etc patman shows
that the email address would be CC-ed multiple times.
Patman currently does try to send duplicate emails like the --dry-run
output shows, but "git send-email" intelligently removes duplicate
addresses so this patch shouldn't change the non-dry-run functionality.
Change patman's output and email addressing to line up with the
"git send-email" logic. This trims down patman's dry-run output and
prevents confusion about what patman will do when emails are actually
sent.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
Adjust the -b flag to permit a range expression as well as a branch.
Signed-off-by: Simon Glass <sjg@chromium.org>
Suggested-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Tested-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Buildman normally obtains the upstream commit by asking git. Provided that
the branch was created with 'git checkout -b <branch> <some_upstream>' then
this normally works.
When there is no upstream, we can try to guess one, by looking up through
the commits until we find a branch. Add a function to try this and print
a warning if buildman ends up relying on it.
Also update the documentation to match.
Signed-off-by: Simon Glass <sjg@chromium.org>
Suggested-by: Wolfgang Denk <wd@denx.de>
The get_maintainers script is a useful default, but sometimes is copies
too many people, or takes a long time to run.
Add an option to disable it and update the README.
Signed-off-by: Simon Glass <sjg@chromium.org>
For an occasional user of patman some failures are not obvious: for
instance when checkpatch reports warnings, the dry run still reports
that the email would be sent. If it is not dry run, the warnings are
shown on the screen, but it is not clear that the email was not sent.
Add some code to report failure to send email explicitly.
Tested by running the script on a patch with style violations,
observed error messages in the script output.
Signed-off-by: Vadim Bendebury <vbendeb@chromium.org>
Reviewed-by: Doug Anderson <dianders@chromium.org>
Acked-by: Simon Glass <sjg@chromium.org>
Tags like Series-version are normally expected to appear once, and with a
unique value. But buildman doesn't actually look at these tags. So ignore
conflicts.
This allows bulidman to build a branch containing multiple patman series.
Reported-by: Steve Rae <srae@broadcom.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
For reasons that are not well-understood, GetMetaDataForList() can end up
adding to an existing series even when it appears that it should be
starting a new one.
Change from using a default constructor parameter to an explicit one, to
work around this problem.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add a test point for the command module. This allows tests to emulate
the execution of commands. This provides more control (since we can make
the fake 'commands' do whatever we like), makes it faster to write tests
since we don't need to set up as much environment, and speeds up test
execution.
Signed-off-by: Simon Glass <sjg@chromium.org>
RunPipe() currently pipes the output of stdout and stderr to a pty, but
this is not the intended behaviour. Fix it.
Signed-off-by: Simon Glass <sjg@chromium.org>
When running unit tests we don't want output to go to the terminal.
Provide a way of collecting it so that it can be examined by test code
later.
Signed-off-by: Simon Glass <sjg@chromium.org>
patman collects tags that it sees in the commit and places them nicely
sorted at the end of the patch. However, this is not really necessary and
in fact is apparently not desirable.
Suggested-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
In a headless environment the pager can apparently hang. We don't want a
pager anyway so let's request that none be used.
Reported-by: Tom Rini <trini@ti.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
It seems that this is no longer needed, since checkpatch.pl will catch
whitespace problems in patches. Also the option is not widely used, so
it seems safe to just remove it.
Suggested-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
It seems that doctest behaves differently now, and some of the unit tests
do not run. Adjust the tests to work correctly.
./tools/patman/patman --test
<unittest.result.TestResult run=10 errors=0 failures=0>
Signed-off-by: Simon Glass <sjg@chromium.org>
This option is currently not supported, but needs to be, for buildman to
operate as expected.
Reported-by: York Sun <yorksun@freescale.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
"patman [options]" is displayed by default.
Append the rest of help messages to parser.usage
instead of replacing it.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
In Python, sys.exit() function can also take an object other
than an integer.
If an integer is given to the argument, Python exits with the return
code of it. If a non-integer argument is given, Python outputs it
to stderr and exits with the return code of 1.
That means,
print >> sys.stderr, "Blah Blah"
sys.exit(1)
is equivalent to
sys.exit("Blah Blah")
The latter is a useful shorthand.
Note:
Some error messages in Buildman and Patman were output to stdout.
But they should go to stderr. They are also fixed by this commit.
This is a nice side effect.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: Simon Glass <sjg@chromium.org>
Older versions of git (e.g. Ubuntu 10.04) do not support this flag. By
default they do not decorate. So only enable this flag when supported.
Suggested-by: Tom Rini <trini@ti.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
If Series-to tag is missing, Patman exits with a message
"No recipient".
This is just annoying for those who had already added
sendemail.to configuration.
I guess many developers have
[sendemail]
to = u-boot@lists.denx.de
in their .git/config because the 'To: u-boot@lists.denx.de' field
should always be added when sending patches.
That seems more reasonable rather than adding
'Series-to: u-boot@lists.denx.de' to every patch series.
Patman should exit only when both Series-to tag and sendemail.to
configuration are mising.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Cc: Simon Glass <sjg@chromium.org>
Acked-by: Simon Glass <sjg@chromium.org>
When patman applies the patches it checks out a new branch, uses 'git am'
to apply the patches one by one, and then tries to go back to the old
branch. If you try this when the branch is 'undefined', this doesn't work
as patman cannot restore the correct branch after applying the patches.
It seems that 'undefined' is created by git and is persistent after it is
created, so that you can end up on quite an old branch.
Add a check for the 'undefined' branch to avoid this.
Reported-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Signed-off-by: Simon Glass <sjg@chromium.org>
There is an unfortunate bug in the signoff suppression logic. The first
pass is performed with 'git log', and all signoffs are added to the
supression set, such that the second time (when processing the real
patches) we always suppress the signoffs.
Correct this by only suppressing signoffs in the second pass.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Michal Simek <monstr@monstr.eu>
Tested-by: Andreas Bießmann <andreas.devel@googlemail.com>
Add a new Patch-cc: tag which performs the service now provided by
the Cc: tag. The Cc: tag is interpreted by git send-email but
ignored by patman.
So now:
Cc: patman does nothing. (git send-email can cc patches)
Patch-cc: patman Cc's patch and removes this tag from the patch
Signed-off-by: Simon Glass <sjg@chromium.org>
