curl/scripts/release-notes.pl

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

244 lines
6.2 KiB
Perl
Raw Normal View History

#!/usr/bin/env perl
#***************************************************************************
# _ _ ____ _
# Project ___| | | | _ \| |
# / __| | | | |_) | |
# | (__| |_| | _ <| |___
# \___|\___/|_| \_\_____|
#
# Copyright (C) Daniel Stenberg, <daniel@haxx.se>, et al.
#
# This software is licensed as described in the file COPYING, which
# you should have received as part of this distribution. The terms
2020-11-04 21:02:01 +08:00
# are also available at https://curl.se/docs/copyright.html.
#
# You may opt to use, copy, modify, merge, publish, distribute and/or sell
# copies of the Software, and permit persons to whom the Software is
# furnished to do so, under the terms of the COPYING file.
#
# This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
# KIND, either express or implied.
#
# SPDX-License-Identifier: curl
#
###########################################################################
###############################################
#
# ==== How to use this script ====
#
# 1. Get recent commits added to RELEASE-NOTES:
#
# $ ./scripts/release-notes.pl
#
# 2. Edit RELEASE-NOTES and remove all entries that don't belong. Unused
# references below will be cleaned up in the next step. Make sure to move
# "changes" up to the changes section. All entries will by default be listed
# under bug-fixes as this script can't know where to put them.
#
# 3. Run the cleanup script and let it sort the entries and remove unused
# references from lines you removed in step (2):
#
# $ ./scripts/release-notes.pl cleanup
#
# 4. Reload RELEASE-NOTES and verify that things look okay. The cleanup
# procedure can and should be re-run when lines are removed or rephrased.
#
# 5. Run ./scripts/contributors.sh and update the contributor list of names
# The list can also be extended or edited manually.
#
# 6. Run ./scripts/delta and update the contributor count at the top, and
# double-check/update the other counters.
#
# 7. Commit the file using "RELEASE-NOTES: synced" as commit message.
#
################################################
my $cleanup = ($ARGV[0] eq "cleanup");
my @gitlog=`git log @^{/RELEASE-NOTES:.synced}..` if(!$cleanup);
my @releasenotes=`cat RELEASE-NOTES`;
my @o; # the entire new RELEASE-NOTES
my @refused; # [num] = [2 bits of use info]
my @refs; # [number] = [URL]
for my $l (@releasenotes) {
if($l =~ /^ o .*\[(\d+)\]/) {
# referenced, set bit 0
$refused[$1]=1;
}
elsif($l =~ /^ \[(\d+)\] = (.*)/) {
# listed in a reference, set bit 1
$refused[$1] |= 2;
$refs[$1] = $2;
}
}
# Return a new fresh reference number
sub getref {
for my $r (1 .. $#refs) {
if(!$refused[$r] & 1) {
return $r;
}
}
# add at the end
return $#refs + 1;
}
# '#num'
# 'num'
# 'https://github.com/curl/curl/issues/6939'
# 'https://github.com/curl/curl-www/issues/69'
# 'https://elsewhere.example.com/discussion'
sub extract {
my ($ref)=@_;
if($ref =~ /^(\#|)(\d+)/) {
# return the plain number
return $2;
}
elsif($ref =~ /^https:\/\/github.com\/curl\/curl\/.*\/(\d+)/) {
# return the plain number
return $1;
}
elsif($ref =~ /:\/\//) {
# contains a '://', return the URL
return $ref;
}
# false alarm, not a valid line
}
my $short;
my $first;
for my $l (@gitlog) {
chomp $l;
if($l =~ /^commit/) {
if($first) {
onecommit($short);
}
# starts a new commit
undef @fixes;
undef @closes;
undef @bug;
$short = "";
$first = 0;
}
elsif(($l =~ /^ (.*)/) && !$first) {
# first line
$short = $1;
$short =~ s/ ?\[(ci skip|skip ci)\]//g;
$first = 1;
push @line, $short;
}
elsif(($l =~ /^ (.*)/) && $first) {
# not the first
my $line = $1;
if($line =~ /^Fixes(:|) *(.*)/i) {
my $ref = extract($2);
push @fixes, $ref if($ref);
}
elsif($line =~ /^Clo(s|)es(:|) *(.*)/i) {
my $ref = extract($3);
push @closes, $ref if($ref);
}
elsif($line =~ /^Bug: (.*)/i) {
my $ref = extract($1);
push @bug, $ref if($ref);
}
}
}
if($first) {
onecommit($short);
}
# call at the end of a parsed commit
sub onecommit {
my ($short)=@_;
my $ref;
if($bug[0]) {
$ref = $bug[0];
}
elsif($fixes[0]) {
$ref = $fixes[0];
}
elsif($closes[0]) {
$ref = $closes[0];
}
if($ref =~ /^#?(\d+)/) {
2020-11-04 21:02:01 +08:00
$ref = "https://curl.se/bug/?i=$1"
}
if($ref) {
my $r = getref();
$refs[$r] = $ref;
$moreinfo{$short}=$r;
$refused[$r] |= 1;
}
}
#### Output the new RELEASE-NOTES
my @bullets;
for my $l (@releasenotes) {
if(($l =~ /^This release includes the following bugfixes:/) && !$cleanup) {
push @o, $l;
push @o, "\n";
for my $f (@line) {
push @o, sprintf " o %s%s\n", $f,
$moreinfo{$f}? sprintf(" [%d]", $moreinfo{$f}): "";
$refused[$moreinfo{$f}]=3;
}
push @o, " --- new entries are listed above this ---";
next;
}
elsif($cleanup) {
if($l =~ /^ --- new entries are listed/) {
# ignore this if still around
next;
}
elsif($l =~ /^ o .*/) {
push @bullets, $l;
next;
}
elsif($bullets[0]) {
# output them case insensitively
for my $b (sort { "\L$a" cmp "\L$b" } @bullets) {
push @o, $b;
}
undef @bullets;
}
}
if($l =~ /^ \[(\d+)\] = /) {
# stop now
last;
}
else {
push @o, $l;
}
}
my @srefs;
my $ln;
for my $n (1 .. $#refs) {
my $r = $refs[$n];
if($r && ($refused[$n] & 1)) {
push @o, sprintf " [%d] = %s\n", $n, $r;
}
}
open(O, ">RELEASE-NOTES");
for my $l (@o) {
print O $l;
}
close(O);
exit;
# Debug: show unused references
for my $r (1 .. $#refs) {
if($refused[$r] != 3) {
printf "%s is %d!\n", $r, $refused[$r];
}
}