Skip to contentSkip to conversation

Update release.sh

PATCH Superseded Andrey Albershteyn 2 participants 17 messages Feb 13, 2025

v4 of 5 Superseded Latest is v5 →

lore
Share & tools
lore
#0 lore raw

v4 cover Update release.sh

To: [email protected]

The release.sh seems to be last updated in 2015. Every one seems to have their own script to release. This patchset updates release.sh to do the basic stuff as updating version files, committing them, creating tag, creating release tarball, uploading tarball (optional), generate for-next announce.

Signature · 45 lines
Andrey

---
Changes in v4:
- Add a few from recently generated files to gitignore
- Drop Cc: trailers in libxfs-apply
- Better handling of trailers with multiple emails/hash marks/not quoted
  names
- Link to v3: https://lore.kernel.org/r/[email protected]

Changes in v3:
- Add -f to generate for-next ANNOUNCE email
- Update for-next when new released is pushed
- Link to v2: https://lore.kernel.org/r/[email protected]

Changes in v2:
- Added git-contributors.py from Darrick
- A bit better/shorter commands for maintainer to run
- Link to v1: https://lore.kernel.org/r/[email protected]

---
Andrey Albershteyn (10):
      release.sh: add signing and fix outdated commands
      release.sh: add --kup to upload release tarball to kernel.org
      release.sh: update version files make commit optional
      Add git-contributors script to notify about merges
      git-contributors: better handling of hash mark/multiple emails
      git-contributors: make revspec required and shebang fix
      release.sh: generate ANNOUNCE email
      release.sh: add -f to generate for-next update email
      libxfs-apply: drop Cc: to stable release list
      gitignore: ignore a few newly generated files

 .gitignore                |   2 +
 release.sh                | 185 +++++++++++++++++++++++++++++++++++++++++++---
 tools/git-contributors.py | 168 +++++++++++++++++++++++++++++++++++++++++
 tools/libxfs-apply        |   1 +
 4 files changed, 344 insertions(+), 12 deletions(-)
---
base-commit: eff7226942a59fc78e8ecd7577657c30ed0cf9a8
change-id: 20241111-update-release-13ef17c008a5

Best regards,
-- 
Andrey Albershteyn <[email protected]>
#1 lore raw

PATCH v4 1/10 release.sh: add signing and fix outdated commands

---
 release.sh | 29 ++++++++++++++++++++++++-----
 1 file changed, 24 insertions(+), 5 deletions(-)
release.sh +24 -5
index 577257a354d442e1cc0a2b9381b11ffbe2f64a71..b15ed610082f34928827ab0547db944cf559cef4 100755
@@ -9,6 +9,8 @@
 # configure.ac (with new version string)
 # debian/changelog (with new release entry, only for release version)
 
+set -e
+
 . ./VERSION
 
 version=${PKG_MAJOR}.${PKG_MINOR}.${PKG_REVISION}
@@ -16,21 +18,38 @@ date=`date +"%-d %B %Y"`
 
 echo "Cleaning up"
 make realclean
+rm -rf "xfsprogs-${version}.tar" \
+	"xfsprogs-${version}.tar.gz" \
+	"xfsprogs-${version}.tar.asc" \
+	"xfsprogs-${version}.tar.sign"
 
 echo "Updating CHANGES"
 sed -e "s/${version}.*/${version} (${date})/" doc/CHANGES > doc/CHANGES.tmp && \
 	mv doc/CHANGES.tmp doc/CHANGES
 
 echo "Commiting CHANGES update to git"
-git commit -a -m "${version} release"
+git commit --all --signoff --message="xfsprogs: Release v${version}
+
+Update all the necessary files for a v${version} release."
 
 echo "Tagging git repository"
-git tag -a -m "${version} release" v${version}
+git tag --annotate --sign --message="Release v${version}" v${version}
 
 echo "Making source tarball"
 make dist
+gunzip -k "xfsprogs-${version}.tar.gz"
 
-#echo "Sign the source tarball"
-#gpg --detach-sign xfsprogs-${version}.tar.gz
+echo "Sign the source tarball"
+gpg \
+	--detach-sign \
+	--armor \
+	"xfsprogs-${version}.tar"
 
-echo "Done.  Please remember to push out tags using \"git push --tags\""
+echo "Verify signature"
+gpg \
+	--verify \
+	"xfsprogs-${version}.tar.asc"
+
+mv "xfsprogs-${version}.tar.asc" "xfsprogs-${version}.tar.sign"
+
+echo "Done. Please remember to push out tags using \"git push origin v${version}\""
Signature · 1 line
2.47.2
#2 lore raw

PATCH v4 2/10 release.sh: add --kup to upload release tarball to kernel.org

In reply to Andrey Albershteyn

Add kup support so that the maintainer can push the newly formed release tarballs to kernel.org.

---
 release.sh | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)
release.sh +31 -0
index b15ed610082f34928827ab0547db944cf559cef4..b036c3241b3f67bfb2435398e6a17ea4c6a6eebe 100755
@@ -16,6 +16,30 @@ set -e
 version=${PKG_MAJOR}.${PKG_MINOR}.${PKG_REVISION}
 date=`date +"%-d %B %Y"`
 
+KUP=0
+
+help() {
+	echo "$(basename) - create xfsprogs release"
+	printf "\t[--kup|-k] upload final tarball with KUP\n"
+}
+
+while [ $# -gt 0 ]; do
+	case "$1" in
+		--kup|-k)
+			KUP=1
+			;;
+		--help|-h)
+			help
+			exit 0
+			;;
+		*)
+			>&2 printf "Error: Invalid argument\n"
+			exit 1
+			;;
+		esac
+	shift
+done
+
 echo "Cleaning up"
 make realclean
 rm -rf "xfsprogs-${version}.tar" \
@@ -52,4 +76,11 @@ gpg \
 
 mv "xfsprogs-${version}.tar.asc" "xfsprogs-${version}.tar.sign"
 
+if [ $KUP -eq 1 ]; then
+	kup put \
+		xfsprogs-${version}.tar.gz \
+		xfsprogs-${version}.tar.sign \
+		pub/linux/utils/fs/xfs/xfsprogs/
+fi;
+
 echo "Done. Please remember to push out tags using \"git push origin v${version}\""
Signature · 1 line
2.47.2
#3 lore raw

PATCH v4 3/10 release.sh: update version files make commit optional

In reply to Andrey Albershteyn

Based on ./VERSION script updates all other files. For ./doc/changelog script asks maintainer to fill it manually as not all changes goes into changelog.

--no-commit|-n flag is handy when something got into the version commit and need to be changed manually. Then ./release.sh -c will use fixed history

---
 release.sh | 75 ++++++++++++++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 58 insertions(+), 17 deletions(-)
release.sh +58 -17
index b036c3241b3f67bfb2435398e6a17ea4c6a6eebe..c3f74bc8dc587a40d867dd22e1e4c6e4aabb8997 100755
@@ -11,16 +11,33 @@
 
 set -e
 
-. ./VERSION
-
-version=${PKG_MAJOR}.${PKG_MINOR}.${PKG_REVISION}
-date=`date +"%-d %B %Y"`
-
 KUP=0
+COMMIT=1
 
 help() {
 	echo "$(basename) - create xfsprogs release"
 	printf "\t[--kup|-k] upload final tarball with KUP\n"
+	printf "\t[--no-commit|-n] don't create release commit\n"
+}
+
+update_version() {
+	echo "Updating version files"
+	# doc/CHANGES
+	header="xfsprogs-${version} ($(date +'%d %b %Y'))"
+	sed -i "1s/^/$header\n\t<TODO list user affecting changes>\n\n/" doc/CHANGES
+	$EDITOR doc/CHANGES
+
+	# ./configure.ac
+	CONF_AC="AC_INIT([xfsprogs],[${version}],[[email protected]])"
+	sed -i "s/^AC_INIT.*/$CONF_AC/" ./configure.ac
+
+	# ./debian/changelog
+	sed -i "1s/^/\n/" ./debian/changelog
+	sed -i "1s/^/ -- Nathan Scott <[email protected]>  `date -R`\n/" ./debian/changelog
+	sed -i "1s/^/\n/" ./debian/changelog
+	sed -i "1s/^/  * New upstream release\n/" ./debian/changelog
+	sed -i "1s/^/\n/" ./debian/changelog
+	sed -i "1s/^/xfsprogs (${version}-1) unstable; urgency=low\n/" ./debian/changelog
 }
 
 while [ $# -gt 0 ]; do
@@ -28,6 +45,9 @@ while [ $# -gt 0 ]; do
 		--kup|-k)
 			KUP=1
 			;;
+		--no-commit|-n)
+			COMMIT=0
+			;;
 		--help|-h)
 			help
 			exit 0
@@ -40,6 +60,36 @@ while [ $# -gt 0 ]; do
 	shift
 done
 
+if [ -z "$EDITOR" ]; then
+	EDITOR=$(command -v vi)
+fi
+
+if [ $COMMIT -eq 1 ]; then
+	if git diff --exit-code ./VERSION > /dev/null; then
+		$EDITOR ./VERSION
+	fi
+fi
+
+. ./VERSION
+
+version=${PKG_MAJOR}.${PKG_MINOR}.${PKG_REVISION}
+date=`date +"%-d %B %Y"`
+
+if [ $COMMIT -eq 1 ]; then
+	update_version
+
+	git diff --color=always | less -r
+	[[ "$(read -e -p 'All good? [Y/n]> '; echo $REPLY)" == [Nn]* ]] && exit 0
+
+	echo "Commiting new version update to git"
+	git commit --all --signoff --message="xfsprogs: Release v${version}
+
+Update all the necessary files for a v${version} release."
+
+	echo "Tagging git repository"
+	git tag --annotate --sign --message="Release v${version}" v${version}
+fi
+
 echo "Cleaning up"
 make realclean
 rm -rf "xfsprogs-${version}.tar" \
@@ -47,17 +97,6 @@ rm -rf "xfsprogs-${version}.tar" \
 	"xfsprogs-${version}.tar.asc" \
 	"xfsprogs-${version}.tar.sign"
 
-echo "Updating CHANGES"
-sed -e "s/${version}.*/${version} (${date})/" doc/CHANGES > doc/CHANGES.tmp && \
-	mv doc/CHANGES.tmp doc/CHANGES
-
-echo "Commiting CHANGES update to git"
-git commit --all --signoff --message="xfsprogs: Release v${version}
-
-Update all the necessary files for a v${version} release."
-
-echo "Tagging git repository"
-git tag --annotate --sign --message="Release v${version}" v${version}
 
 echo "Making source tarball"
 make dist
@@ -83,4 +122,6 @@ if [ $KUP -eq 1 ]; then
 		pub/linux/utils/fs/xfs/xfsprogs/
 fi;
 
-echo "Done. Please remember to push out tags using \"git push origin v${version}\""
+echo ""
+echo "Done. Please remember to push out tags and the branch."
+printf "\tgit push origin v${version} master:master master:for-next\n"
Signature · 1 line
2.47.2
#4 lore raw

PATCH v4 4/10 Add git-contributors script to notify about merges

In reply to Andrey Albershteyn

Add python script used to collect emails over all changes merged in the next release.

---
 tools/git-contributors.py | 101 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 101 insertions(+)
tools/git-contributors.py new file +101 -0
new file mode 100755
index 0000000000000000000000000000000000000000..70ac8abb26c8ce65de336c5ae48abcfee39508b2
@@ -0,0 +1,101 @@
+#!/usr/bin/python3
+
+# List all contributors to a series of git commits.
+# Copyright(C) 2025 Oracle, All Rights Reserved.
+# Licensed under GPL 2.0 or later
+
+import re
+import subprocess
+import io
+import sys
+import argparse
+import email.utils
+
+DEBUG = False
+
+def backtick(args):
+    '''Generator function that yields lines of a program's stdout.'''
+    if DEBUG:
+        print(' '.join(args))
+    p = subprocess.Popen(args, stdout = subprocess.PIPE)
+    for line in io.TextIOWrapper(p.stdout, encoding="utf-8"):
+        yield line
+
+class find_developers(object):
+    def __init__(self):
+        tags = '%s|%s|%s|%s|%s|%s|%s|%s' % (
+            'signed-off-by',
+            'acked-by',
+            'cc',
+            'reviewed-by',
+            'reported-by',
+            'tested-by',
+            'suggested-by',
+            'reported-and-tested-by')
+        # some tag, a colon, a space, and everything after that
+        regex1 = r'^(%s):\s+(.+)$' % tags
+
+        self.r1 = re.compile(regex1, re.I)
+
+    def run(self, lines):
+        addr_list = []
+
+        for line in lines:
+            l = line.strip()
+
+            # emailutils can handle abominations like:
+            #
+            # Reviewed-by: Bogus J. Simpson <[email protected]>
+            # Reviewed-by: "Bogus J. Simpson" <[email protected]>
+            # Reviewed-by: [email protected]
+            # Cc: <[email protected]> # v6.9
+            # Tested-by: Moo Cow <[email protected]> # powerpc
+            m = self.r1.match(l)
+            if not m:
+                continue
+            (name, addr) = email.utils.parseaddr(m.expand(r'\g<2>'))
+
+            # This last split removes anything after a hash mark,
+            # because someone could have provided an improperly
+            # formatted email address:
+            #
+            # Cc: [email protected] # v6.19+
+            #
+            # emailutils doesn't seem to catch this, and I can't
+            # fully tell from RFC2822 that this isn't allowed.  I
+            # think it is because dtext doesn't forbid spaces or
+            # hash marks.
+            addr_list.append(addr.split('#')[0])
+
+        return sorted(set(addr_list))
+
+def main():
+    global DEBUG
+
+    parser = argparse.ArgumentParser(description = "List email addresses of contributors to a series of git commits.")
+    parser.add_argument("revspec", nargs = '?', default = None, \
+            help = "git revisions to process.")
+    parser.add_argument("--separator", type = str, default = '\n', \
+            help = "Separate each email address with this string.")
+    parser.add_argument('--debug', action = 'store_true', default = False, \
+            help = argparse.SUPPRESS)
+    args = parser.parse_args()
+
+    if args.debug:
+        DEBUG = True
+
+    fd = find_developers()
+    if args.revspec:
+        # read git commits from repo
+        contributors = fd.run(backtick(['git', 'log', '--pretty=medium',
+                  args.revspec]))
+    else:
+        # read patch from stdin
+        contributors = fd.run(sys.stdin.readlines())
+
+    print(args.separator.join(sorted(contributors)))
+    return 0
+
+if __name__ == '__main__':
+    sys.exit(main())
+
Signature · 1 line
2.47.2
#5 lore raw

PATCH v4 5/10 git-contributors: better handling of hash mark/multiple emails

In reply to Andrey Albershteyn

Better handling of hash mark, tags with multiple emails and not quoted names in emails. See comments in the script.

---
 tools/git-contributors.py | 109 ++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 90 insertions(+), 19 deletions(-)
tools/git-contributors.py +90 -19
index 70ac8abb26c8ce65de336c5ae48abcfee39508b2..1a0f2b80e3dad9124b86b29f8507389ef91fe813 100755
@@ -37,35 +37,106 @@ class find_developers(object):
 
         self.r1 = re.compile(regex1, re.I)
 
+        # regex to guess if this is a list of multiple addresses.
+        # Not sure why the initial "^.*" is needed here.
+        self.r2 = re.compile(r'^.*,[^,]*@[^@]*,[^,]*@', re.I)
+
+        # regex to match on anything inside a pair of angle brackets
+        self.r3 = re.compile(r'^.*<(.+)>', re.I)
+
+    def _handle_addr(self, addr):
+        # The next split removes everything after an octothorpe (hash
+        # mark), because someone could have provided an improperly
+        # formatted email address:
+        #
+        # Cc: [email protected] # v6.19+
+        #
+        # This, according to my reading of RFC5322, is allowed because
+        # octothorpes can be part of atom text.  However, it is
+        # interepreted as if there weren't any whitespace
+        # ("[email protected]#v6.19+").  The grammar allows for
+        # this form, even though this is not a correct Internet domain
+        # name.
+        #
+        # Worse, if you follow the format specified in the kernel's
+        # SubmittingPatches file:
+        #
+        # Cc: <[email protected]> # v6.9
+        #
+        # emailutils will not know how to parse this, and returns empty
+        # strings.  I think this is because the angle-addr
+        # specification allows only whitespace between the closing
+        # angle bracket and the CRLF.
+        #
+        # Hack around both problems by ignoring everything after an
+        # octothorpe, no matter where it occurs in the string.  If
+        # someone has one in their name or the email address, too bad.
+        a = addr.split('#')[0]
+
+        # emailutils can extract email addresses from headers that
+        # roughly follow the destination address field format:
+        #
+        # Reviewed-by: Bogus J. Simpson <[email protected]>
+        # Reviewed-by: "Bogus J. Simpson" <[email protected]>
+        # Reviewed-by: [email protected]
+        #
+        # Use it to extract the email address, because we don't care
+        # about the display name.
+        (name, addr) = email.utils.parseaddr(a)
+        if DEBUG:
+            print(f'A:{a}:NAME:{name}:ADDR:{addr}:')
+        if len(addr) > 0:
+            return addr
+
+        # If emailutils fails to find anything, let's see if there's
+        # a sequence of characters within angle brackets and hope that
+        # is an email address.  This works around things like:
+        #
+        # Reported-by: Xu, Wen <[email protected]>
+        #
+        # Which should have had the name in quotations because there's
+        # a comma.
+        m = self.r3.match(a)
+        if m:
+            addr = m.expand(r'\g<1>')
+            if DEBUG:
+                print(f"M3:{addr}:M:{m}:")
+            return addr
+
+        # No idea, just spit the whole thing out and hope for the best.
+        return a
+
     def run(self, lines):
         addr_list = []
 
         for line in lines:
             l = line.strip()
 
-            # emailutils can handle abominations like:
-            #
-            # Reviewed-by: Bogus J. Simpson <[email protected]>
-            # Reviewed-by: "Bogus J. Simpson" <[email protected]>
-            # Reviewed-by: [email protected]
-            # Cc: <[email protected]> # v6.9
-            # Tested-by: Moo Cow <[email protected]> # powerpc
+            # First, does this line match any of the headers we
+            # know about?
             m = self.r1.match(l)
             if not m:
                 continue
-            (name, addr) = email.utils.parseaddr(m.expand(r'\g<2>'))
+            rightside = m.expand(r'\g<2>')
 
-            # This last split removes anything after a hash mark,
-            # because someone could have provided an improperly
-            # formatted email address:
-            #
-            # Cc: [email protected] # v6.19+
-            #
-            # emailutils doesn't seem to catch this, and I can't
-            # fully tell from RFC2822 that this isn't allowed.  I
-            # think it is because dtext doesn't forbid spaces or
-            # hash marks.
-            addr_list.append(addr.split('#')[0])
+            n = self.r2.match(rightside)
+            if n:
+                # Break the line into an array of addresses,
+                # delimited by commas, then handle each
+                # address.
+                addrs = rightside.split(',')
+                if DEBUG:
+                    print(f"0LINE:{rightside}:ADDRS:{addrs}:M:{n}")
+                for addr in addrs:
+                    a = self._handle_addr(addr)
+                    addr_list.append(a)
+            else:
+                # Otherwise treat the line as a single email
+                # address.
+                if DEBUG:
+                    print(f"1LINE:{rightside}:M:{n}")
+                a = self._handle_addr(rightside)
+                addr_list.append(a)
 
         return sorted(set(addr_list))
 
Signature · 1 line
2.47.2
Reviewed-by on v4 5/10
#6 lore raw

Re: [PATCH v4 05/10] git-contributors: better handling of hash mark/multiple emails

Quoted 5 lines · Andrey Albershteyn

Better handling of hash mark, tags with multiple emails and not quoted names in emails. See comments in the script.

Matches my original git-contributors script, good enough for now

--D

Quoted 139 lines
---
 tools/git-contributors.py | 109 ++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 90 insertions(+), 19 deletions(-)
tools/git-contributors.py +90 -19
index 70ac8abb26c8ce65de336c5ae48abcfee39508b2..1a0f2b80e3dad9124b86b29f8507389ef91fe813 100755
@@ -37,35 +37,106 @@ class find_developers(object):
 
         self.r1 = re.compile(regex1, re.I)
 
+        # regex to guess if this is a list of multiple addresses.
+        # Not sure why the initial "^.*" is needed here.
+        self.r2 = re.compile(r'^.*,[^,]*@[^@]*,[^,]*@', re.I)
+
+        # regex to match on anything inside a pair of angle brackets
+        self.r3 = re.compile(r'^.*<(.+)>', re.I)
+
+    def _handle_addr(self, addr):
+        # The next split removes everything after an octothorpe (hash
+        # mark), because someone could have provided an improperly
+        # formatted email address:
+        #
+        # Cc: [email protected] # v6.19+
+        #
+        # This, according to my reading of RFC5322, is allowed because
+        # octothorpes can be part of atom text.  However, it is
+        # interepreted as if there weren't any whitespace
+        # ("[email protected]#v6.19+").  The grammar allows for
+        # this form, even though this is not a correct Internet domain
+        # name.
+        #
+        # Worse, if you follow the format specified in the kernel's
+        # SubmittingPatches file:
+        #
+        # Cc: <[email protected]> # v6.9
+        #
+        # emailutils will not know how to parse this, and returns empty
+        # strings.  I think this is because the angle-addr
+        # specification allows only whitespace between the closing
+        # angle bracket and the CRLF.
+        #
+        # Hack around both problems by ignoring everything after an
+        # octothorpe, no matter where it occurs in the string.  If
+        # someone has one in their name or the email address, too bad.
+        a = addr.split('#')[0]
+
+        # emailutils can extract email addresses from headers that
+        # roughly follow the destination address field format:
+        #
+        # Reviewed-by: Bogus J. Simpson <[email protected]>
+        # Reviewed-by: "Bogus J. Simpson" <[email protected]>
+        # Reviewed-by: [email protected]
+        #
+        # Use it to extract the email address, because we don't care
+        # about the display name.
+        (name, addr) = email.utils.parseaddr(a)
+        if DEBUG:
+            print(f'A:{a}:NAME:{name}:ADDR:{addr}:')
+        if len(addr) > 0:
+            return addr
+
+        # If emailutils fails to find anything, let's see if there's
+        # a sequence of characters within angle brackets and hope that
+        # is an email address.  This works around things like:
+        #
+        # Reported-by: Xu, Wen <[email protected]>
+        #
+        # Which should have had the name in quotations because there's
+        # a comma.
+        m = self.r3.match(a)
+        if m:
+            addr = m.expand(r'\g<1>')
+            if DEBUG:
+                print(f"M3:{addr}:M:{m}:")
+            return addr
+
+        # No idea, just spit the whole thing out and hope for the best.
+        return a
+
     def run(self, lines):
         addr_list = []
 
         for line in lines:
             l = line.strip()
 
-            # emailutils can handle abominations like:
-            #
-            # Reviewed-by: Bogus J. Simpson <[email protected]>
-            # Reviewed-by: "Bogus J. Simpson" <[email protected]>
-            # Reviewed-by: [email protected]
-            # Cc: <[email protected]> # v6.9
-            # Tested-by: Moo Cow <[email protected]> # powerpc
+            # First, does this line match any of the headers we
+            # know about?
             m = self.r1.match(l)
             if not m:
                 continue
-            (name, addr) = email.utils.parseaddr(m.expand(r'\g<2>'))
+            rightside = m.expand(r'\g<2>')
 
-            # This last split removes anything after a hash mark,
-            # because someone could have provided an improperly
-            # formatted email address:
-            #
-            # Cc: [email protected] # v6.19+
-            #
-            # emailutils doesn't seem to catch this, and I can't
-            # fully tell from RFC2822 that this isn't allowed.  I
-            # think it is because dtext doesn't forbid spaces or
-            # hash marks.
-            addr_list.append(addr.split('#')[0])
+            n = self.r2.match(rightside)
+            if n:
+                # Break the line into an array of addresses,
+                # delimited by commas, then handle each
+                # address.
+                addrs = rightside.split(',')
+                if DEBUG:
+                    print(f"0LINE:{rightside}:ADDRS:{addrs}:M:{n}")
+                for addr in addrs:
+                    a = self._handle_addr(addr)
+                    addr_list.append(a)
+            else:
+                # Otherwise treat the line as a single email
+                # address.
+                if DEBUG:
+                    print(f"1LINE:{rightside}:M:{n}")
+                a = self._handle_addr(rightside)
+                addr_list.append(a)
 
         return sorted(set(addr_list))
 
Signature · 2 lines
-- 
2.47.2
#7 lore raw

PATCH v4 6/10 git-contributors: make revspec required and shebang fix

In reply to Andrey Albershteyn

Without default value script will show help instead of just hanging waiting for input on stdin.

Shebang fix for system with different python location than the /usr/bin one.

Cut leading delimiter from the final CC string.

---
 tools/git-contributors.py | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
tools/git-contributors.py +2 -6
index 1a0f2b80e3dad9124b86b29f8507389ef91fe813..01177a9af749776ce4ac982f29f8f9302904d820 100755
@@ -1,4 +1,4 @@
-#!/usr/bin/python3
+#!/usr/bin/env python3
 
 # List all contributors to a series of git commits.
 # Copyright(C) 2025 Oracle, All Rights Reserved.
@@ -144,8 +144,7 @@ def main():
     global DEBUG
 
     parser = argparse.ArgumentParser(description = "List email addresses of contributors to a series of git commits.")
-    parser.add_argument("revspec", nargs = '?', default = None, \
-            help = "git revisions to process.")
+    parser.add_argument("revspec", help = "git revisions to process.")
     parser.add_argument("--separator", type = str, default = '\n', \
             help = "Separate each email address with this string.")
     parser.add_argument('--debug', action = 'store_true', default = False, \
@@ -160,9 +159,6 @@ def main():
         # read git commits from repo
         contributors = fd.run(backtick(['git', 'log', '--pretty=medium',
                   args.revspec]))
-    else:
-        # read patch from stdin
-        contributors = fd.run(sys.stdin.readlines())
 
     print(args.separator.join(sorted(contributors)))
     return 0
Signature · 1 line
2.47.2
Reviewed-by on v4 6/10
#8 lore raw

Re: [PATCH v4 06/10] git-contributors: make revspec required and shebang fix

Quoted 11 lines · Andrey Albershteyn

Without default value script will show help instead of just hanging waiting for input on stdin.

Shebang fix for system with different python location than the /usr/bin one.

Cut leading delimiter from the final CC string.

---
 tools/git-contributors.py | 8 ++------

Looks good,

--D

Quoted 37 lines
 1 file changed, 2 insertions(+), 6 deletions(-)
tools/git-contributors.py +2 -6
index 1a0f2b80e3dad9124b86b29f8507389ef91fe813..01177a9af749776ce4ac982f29f8f9302904d820 100755
@@ -1,4 +1,4 @@
-#!/usr/bin/python3
+#!/usr/bin/env python3
 
 # List all contributors to a series of git commits.
 # Copyright(C) 2025 Oracle, All Rights Reserved.
@@ -144,8 +144,7 @@ def main():
     global DEBUG
 
     parser = argparse.ArgumentParser(description = "List email addresses of contributors to a series of git commits.")
-    parser.add_argument("revspec", nargs = '?', default = None, \
-            help = "git revisions to process.")
+    parser.add_argument("revspec", help = "git revisions to process.")
     parser.add_argument("--separator", type = str, default = '\n', \
             help = "Separate each email address with this string.")
     parser.add_argument('--debug', action = 'store_true', default = False, \
@@ -160,9 +159,6 @@ def main():
         # read git commits from repo
         contributors = fd.run(backtick(['git', 'log', '--pretty=medium',
                   args.revspec]))
-    else:
-        # read patch from stdin
-        contributors = fd.run(sys.stdin.readlines())
 
     print(args.separator.join(sorted(contributors)))
     return 0
Signature · 2 lines
-- 
2.47.2
#9 lore raw

PATCH v4 7/10 release.sh: generate ANNOUNCE email

In reply to Andrey Albershteyn
---
 release.sh | 54 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 54 insertions(+)
release.sh +54 -0
index c3f74bc8dc587a40d867dd22e1e4c6e4aabb8997..696d3ec107ca0cc11ed565734ca7423acfd6d858 100755
@@ -13,11 +13,13 @@ set -e
 
 KUP=0
 COMMIT=1
+LAST_HEAD=""
 
 help() {
 	echo "$(basename) - create xfsprogs release"
 	printf "\t[--kup|-k] upload final tarball with KUP\n"
 	printf "\t[--no-commit|-n] don't create release commit\n"
+	printf "\t[--last-head|-l] commit of the last release\n"
 }
 
 update_version() {
@@ -40,6 +42,48 @@ update_version() {
 	sed -i "1s/^/xfsprogs (${version}-1) unstable; urgency=low\n/" ./debian/changelog
 }
 
+prepare_mail() {
+	branch="$1"
+	mail_file=$(mktemp)
+	if [ -n "$LAST_HEAD" ]; then
+		if [ $branch == "master" ]; then
+			reason="$(git describe --abbrev=0 $branch) released"
+		else
+			reason="for-next updated to $(git log --oneline --format="%h" -1 $branch)"
+		fi;
+		cat << EOF > $mail_file
+To: [email protected]
+Cc: $(./tools/git-contributors.py $LAST_HEAD..$branch --delimiter ', ')
+Subject: [ANNOUNCE] xfsprogs: $reason
+
+Hi folks,
+
+The xfsprogs $branch branch in repository at:
+
+	git://git.kernel.org/pub/scm/fs/xfs/xfsprogs-dev.git
+
+has just been updated.
+
+Patches often get missed, so if your outstanding patches are properly reviewed
+on the list and not included in this update, please let me know.
+
+The for-next branch has also been updated to match the state of master.
+
+The new head of the $branch branch is commit:
+
+$(git log --oneline --format="%H" -1 $branch)
+
+New commits:
+
+$(git shortlog --format="[%h] %s" $LAST_HEAD..$branch)
+
+Code Diffstat:
+
+$(git diff --stat --summary -C -M $LAST_HEAD..$branch)
+EOF
+	fi
+}
+
 while [ $# -gt 0 ]; do
 	case "$1" in
 		--kup|-k)
@@ -48,6 +92,10 @@ while [ $# -gt 0 ]; do
 		--no-commit|-n)
 			COMMIT=0
 			;;
+		--last-head|-l)
+			LAST_HEAD=$2
+			shift
+			;;
 		--help|-h)
 			help
 			exit 0
@@ -122,6 +170,12 @@ if [ $KUP -eq 1 ]; then
 		pub/linux/utils/fs/xfs/xfsprogs/
 fi;
 
+prepare_mail "master"
+
 echo ""
 echo "Done. Please remember to push out tags and the branch."
 printf "\tgit push origin v${version} master:master master:for-next\n"
+if [ -n "$LAST_HEAD" ]; then
+	echo "Command to send ANNOUNCE email"
+	printf "\tneomutt -H $mail_file\n"
+fi
Signature · 1 line
2.47.2
#10 lore raw

PATCH v4 8/10 release.sh: add -f to generate for-next update email

In reply to Andrey Albershteyn

Add --for-next/-f to generate ANNOUNCE email for for-next branch update. This doesn't require new commit/tarball/tags, so skip it.

---
 release.sh | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
release.sh +16 -0
index 696d3ec107ca0cc11ed565734ca7423acfd6d858..b7461e958e6f55f7b8ccf31b53e2d304db11789e 100755
@@ -14,12 +14,14 @@ set -e
 KUP=0
 COMMIT=1
 LAST_HEAD=""
+FOR_NEXT=0
 
 help() {
 	echo "$(basename) - create xfsprogs release"
 	printf "\t[--kup|-k] upload final tarball with KUP\n"
 	printf "\t[--no-commit|-n] don't create release commit\n"
 	printf "\t[--last-head|-l] commit of the last release\n"
+	printf "\t[--for-next|-f] generate announce email for for-next update\n"
 }
 
 update_version() {
@@ -96,6 +98,9 @@ while [ $# -gt 0 ]; do
 			LAST_HEAD=$2
 			shift
 			;;
+		--for-next|-f)
+			FOR_NEXT=1
+			;;
 		--help|-h)
 			help
 			exit 0
@@ -108,6 +113,17 @@ while [ $# -gt 0 ]; do
 	shift
 done
 
+if [ $FOR_NEXT -eq 1 ]; then
+	echo "Push your for-next branch:"
+	printf "\tgit push origin for-next:for-next\n"
+	prepare_mail "for-next"
+	if [ -n "$LAST_HEAD" ]; then
+		echo "Command to send ANNOUNCE email"
+		printf "\tneomutt -H $mail_file\n"
+	fi
+	exit 0
+fi
+
 if [ -z "$EDITOR" ]; then
 	EDITOR=$(command -v vi)
 fi
Signature · 1 line
2.47.2
#11 lore raw

PATCH v4 9/10 libxfs-apply: drop Cc: to stable release list

In reply to Andrey Albershteyn

These Cc: tags are intended for kernel commits which need to be backported to stable kernels. Maintainers of stable kernel aren't interested in xfsprogs syncs.

---
 tools/libxfs-apply | 1 +
 1 file changed, 1 insertion(+)
tools/libxfs-apply +1 -0
index 097a695f942bb832c2fb1456a0fd8c28c025d1a6..e9672e572d23af296dccfe6499eda9b909f44afd 100755
@@ -254,6 +254,7 @@ fixup_header_format()
 		}
 		/^Date:/ { date_seen=1; next }
 		/^difflib/ { next }
+		/[Cc]{2}: <[email protected]>?.*/ { next }
 
 		// {
 			if (date_seen == 0)
Signature · 1 line
2.47.2
#12 lore raw

Re: [PATCH v4 09/10] libxfs-apply: drop Cc: to stable release list

Quoted 14 lines · Andrey Albershteyn

These Cc: tags are intended for kernel commits which need to be backported to stable kernels. Maintainers of stable kernel aren't interested in xfsprogs syncs.

---
 tools/libxfs-apply | 1 +
 1 file changed, 1 insertion(+)
tools/libxfs-apply +0 -0
index 097a695f942bb832c2fb1456a0fd8c28c025d1a6..e9672e572d23af296dccfe6499eda9b909f44afd 100755

tools/libxfs-apply fixup_header_format() · line 257

Quoted 4 lines
tools/libxfs-apply +1 -0
@@ -254,6 +254,7 @@ fixup_header_format()
 		}
 		/^Date:/ { date_seen=1; next }
 		/^difflib/ { next }
+		/[Cc]{2}: <[email protected]>?.*/ { next }

You might want to ignore the angle brackets, because some people do:

which is valid rfc822 even if SubmittingPatches says not to do that. Annoyingly, other parts of the documentation lay that out as an example.

		/[Cc]{2}:.*[email protected]/ { next }

<shrug>

--D

Quoted 8 lines
tools/libxfs-apply +0 -0
 		// {
 			if (date_seen == 0)
 
Signature · 1 line
2.47.2
#13 lore raw

Re: [PATCH v4 09/10] libxfs-apply: drop Cc: to stable release list

Quoted 10 lines · Darrick J. Wong
Quoted 9 lines · Andrey Albershteyn

These Cc: tags are intended for kernel commits which need to be backported to stable kernels. Maintainers of stable kernel aren't interested in xfsprogs syncs.

---
 tools/libxfs-apply | 1 +
 1 file changed, 1 insertion(+)
Quoted 11 lines

tools/libxfs-apply fixup_header_format() · line 257

Quoted 9 lines
tools/libxfs-apply +1 -0
index 097a695f942bb832c2fb1456a0fd8c28c025d1a6..e9672e572d23af296dccfe6499eda9b909f44afd 100755
@@ -254,6 +254,7 @@ fixup_header_format()
 		}
 		/^Date:/ { date_seen=1; next }
 		/^difflib/ { next }
+		/[Cc]{2}: <[email protected]>?.*/ { next }

You might want to ignore the angle brackets, because some people do:

The ? does that :) One or zero of <>

Quoted 21 lines

which is valid rfc822 even if SubmittingPatches says not to do that. Annoyingly, other parts of the documentation lay that out as an example.

		/[Cc]{2}:.*[email protected]/ { next }

<shrug>

--D

Quoted 8 lines
tools/libxfs-apply +0 -0
 		// {
 			if (date_seen == 0)
 
Signature · 1 line
2.47.2
Signature · 1 line
- Andrey
Reviewed-by
#14 lore raw

Re: [PATCH v4 09/10] libxfs-apply: drop Cc: to stable release list

Quoted 23 lines · Andrey Albershteyn
Quoted 10 lines · Darrick J. Wong
Quoted 9 lines · Andrey Albershteyn

These Cc: tags are intended for kernel commits which need to be backported to stable kernels. Maintainers of stable kernel aren't interested in xfsprogs syncs.

---
 tools/libxfs-apply | 1 +
 1 file changed, 1 insertion(+)
Quoted 11 lines

tools/libxfs-apply fixup_header_format() · line 257

Quoted 9 lines
tools/libxfs-apply +1 -0
index 097a695f942bb832c2fb1456a0fd8c28c025d1a6..e9672e572d23af296dccfe6499eda9b909f44afd 100755
@@ -254,6 +254,7 @@ fixup_header_format()
 		}
 		/^Date:/ { date_seen=1; next }
 		/^difflib/ { next }
+		/[Cc]{2}: <[email protected]>?.*/ { next }

You might want to ignore the angle brackets, because some people do:

Quoted 1 line

The ? does that :) One or zero of <>

Aha, I missed that! Thanks for the correction.

--D

Quoted 26 lines
Quoted 21 lines

which is valid rfc822 even if SubmittingPatches says not to do that. Annoyingly, other parts of the documentation lay that out as an example.

		/[Cc]{2}:.*[email protected]/ { next }

<shrug>

--D

Quoted 8 lines
tools/libxfs-apply +0 -0
 		// {
 			if (date_seen == 0)
 
Signature · 1 line
2.47.2
Signature · 1 line
- Andrey
#15 lore raw

PATCH v4 10/10 gitignore: ignore a few newly generated files

In reply to Andrey Albershteyn

These files are generated from corresponding *.in templates.

---
 .gitignore | 2 ++
 1 file changed, 2 insertions(+)
.gitignore +2 -0
index 756867124a021b195a10fc2a8a598f16aa6514c4..5d971200d5bfb285e680427de193f81d8ab77c06 100644
@@ -65,12 +65,14 @@ cscope.*
 /mdrestore/xfs_mdrestore
 /mkfs/fstyp
 /mkfs/mkfs.xfs
+/mkfs/xfs_protofile
 /quota/xfs_quota
 /repair/xfs_repair
 /rtcp/xfs_rtcp
 /spaceman/xfs_spaceman
 /scrub/xfs_scrub
 /scrub/xfs_scrub_all
+/scrub/xfs_scrub_all.timer
 /scrub/xfs_scrub_fail
 /scrub/*.cron
 /scrub/*.service
Signature · 1 line
2.47.2
Reviewed-by on v4 10/10
#16 lore raw

Re: [PATCH v4 10/10] gitignore: ignore a few newly generated files

Quoted 3 lines · Andrey Albershteyn

These files are generated from corresponding *.in templates.

--D

Quoted 28 lines
---
 .gitignore | 2 ++
 1 file changed, 2 insertions(+)
.gitignore +2 -0
index 756867124a021b195a10fc2a8a598f16aa6514c4..5d971200d5bfb285e680427de193f81d8ab77c06 100644
@@ -65,12 +65,14 @@ cscope.*
 /mdrestore/xfs_mdrestore
 /mkfs/fstyp
 /mkfs/mkfs.xfs
+/mkfs/xfs_protofile
 /quota/xfs_quota
 /repair/xfs_repair
 /rtcp/xfs_rtcp
 /spaceman/xfs_spaceman
 /scrub/xfs_scrub
 /scrub/xfs_scrub_all
+/scrub/xfs_scrub_all.timer
 /scrub/xfs_scrub_fail
 /scrub/*.cron
 /scrub/*.service
Signature · 2 lines
-- 
2.47.2