patman: Use the Change-Id, version, and prefix in the Message-Id

As per the centithread on ksummit-discuss [1], there are folks who
feel that if a Change-Id is present in a developer's local commit that
said Change-Id could be interesting to include in upstream posts.
Specifically if two commits are posted with the same Change-Id there's
a reasonable chance that they are either the same commit or a newer
version of the same commit.  Specifically this is because that's how
gerrit has trained people to work.

There is much angst about Change-Id in upstream Linux, but one thing
that seems safe and non-controversial is to include the Change-Id as
part of the string of crud that makes up a Message-Id.

Let's give that a try.

In theory (if there is enough adoption) this could help a tool more
reliably find various versions of a commit.  This actually might work
pretty well for U-Boot where (I believe) quite a number of developers
use patman, so there could be critical mass (assuming that enough of
these people also use a git hook that adds Change-Id to their
commits).  I was able to find this git hook by searching for "gerrit
change id git hook" in my favorite search engine.

In theory one could imagine something like this could be integrated
into other tools, possibly even git-send-email.  Getting it into
patman seems like a sane first step, though.

NOTE: this patch is being posted using a patman containing this patch,
so you should be able to see the Message-Id of this patch and see that
it contains my local Change-Id, which ends in 2b9 if you want to
check.

[1] https://lists.linuxfoundation.org/pipermail/ksummit-discuss/2019-August/006739.html

Signed-off-by: Douglas Anderson <dianders@chromium.org>

1 files changed, 62 insertions, 2 deletions

diff --git a/tools/patman/patchstream.py b/tools/patman/patchstream.py
index b6455b0fa3..ef06606297 100644
--- a/tools/patman/patchstream.py
+++ b/tools/patman/patchstream.py
@@ -2,6 +2,7 @@
 # Copyright (c) 2011 The Chromium OS Authors.
 #
 
+import datetime
 import math
 import os
 import re
@@ -14,7 +15,7 @@ import gitutil
 from series import Series
 
 # Tags that we detect and remove
-re_remove = re.compile('^BUG=|^TEST=|^BRANCH=|^Change-Id:|^Review URL:'
+re_remove = re.compile('^BUG=|^TEST=|^BRANCH=|^Review URL:'
     '|Reviewed-on:|Commit-\w*:')
 
 # Lines which are allowed after a TEST= line
@@ -32,6 +33,9 @@ re_cover_cc = re.compile('^Cover-letter-cc: *(.*)')
 # Patch series tag
 re_series_tag = re.compile('^Series-([a-z-]*): *(.*)')
 
+# Change-Id will be used to generate the Message-Id and then be stripped
+re_change_id = re.compile('^Change-Id: *(.*)')
+
 # Commit series tag
 re_commit_tag = re.compile('^Commit-([a-z-]*): *(.*)')
 
@@ -156,6 +160,7 @@ class PatchStream:
 
         # Handle state transition and skipping blank lines
         series_tag_match = re_series_tag.match(line)
+        change_id_match = re_change_id.match(line)
         commit_tag_match = re_commit_tag.match(line)
         cover_match = re_cover.match(line)
         cover_cc_match = re_cover_cc.match(line)
@@ -177,7 +182,7 @@ class PatchStream:
             self.state = STATE_MSG_HEADER
 
         # If a tag is detected, or a new commit starts
-        if series_tag_match or commit_tag_match or \
+        if series_tag_match or commit_tag_match or change_id_match or \
            cover_match or cover_cc_match or signoff_match or \
            self.state == STATE_MSG_HEADER:
             # but we are already in a section, this means 'END' is missing
@@ -275,6 +280,16 @@ class PatchStream:
                 self.AddToSeries(line, name, value)
                 self.skip_blank = True
 
+        # Detect Change-Id tags
+        elif change_id_match:
+            value = change_id_match.group(1)
+            if self.is_log:
+                if self.commit.change_id:
+                    raise ValueError("%s: Two Change-Ids: '%s' vs. '%s'" %
+                        (self.commit.hash, self.commit.change_id, value))
+                self.commit.change_id = value
+            self.skip_blank = True
+
         # Detect Commit-xxx tags
         elif commit_tag_match:
             name = commit_tag_match.group(1)
@@ -345,6 +360,47 @@ class PatchStream:
             self.warn.append('Found %d lines after TEST=' %
                     self.lines_after_test)
 
+    def WriteMessageId(self, outfd):
+        """Write the Message-Id into the output.
+
+        This is based on the Change-Id in the original patch, the version,
+        and the prefix.
+
+        Args:
+            outfd: Output stream file object
+        """
+        if not self.commit.change_id:
+            return
+
+        # If the count is -1 we're testing, so use a fixed time
+        if self.commit.count == -1:
+            time_now = datetime.datetime(1999, 12, 31, 23, 59, 59)
+        else:
+            time_now = datetime.datetime.now()
+
+        # In theory there is email.utils.make_msgid() which would be nice
+        # to use, but it already produces something way too long and thus
+        # will produce ugly commit lines if someone throws this into
+        # a "Link:" tag in the final commit.  So (sigh) roll our own.
+
+        # Start with the time; presumably we wouldn't send the same series
+        # with the same Change-Id at the exact same second.
+        parts = [time_now.strftime("%Y%m%d%H%M%S")]
+
+        # These seem like they would be nice to include.
+        if 'prefix' in self.series:
+            parts.append(self.series['prefix'])
+        if 'version' in self.series:
+            parts.append("v%s" % self.series['version'])
+
+        parts.append(str(self.commit.count + 1))
+
+        # The Change-Id must be last, right before the @
+        parts.append(self.commit.change_id)
+
+        # Join parts together with "." and write it out.
+        outfd.write('Message-Id: <%s@changeid>\n' % '.'.join(parts))
+
     def ProcessStream(self, infd, outfd):
         """Copy a stream from infd to outfd, filtering out unwanting things.
 
@@ -358,6 +414,9 @@ class PatchStream:
         fname = None
         last_fname = None
         re_fname = re.compile('diff --git a/(.*) b/.*')
+
+        self.WriteMessageId(outfd)
+
         while True:
             line = infd.readline()
             if not line:
@@ -481,6 +540,7 @@ def FixPatches(series, fnames):
     for fname in fnames:
         commit = series.commits[count]
         commit.patch = fname
+        commit.count = count
         result = FixPatch(backup_dir, fname, series, commit)
         if result:
             print('%d warnings for %s:' % (len(result), fname))