3 # Copyright (C) 2020 Free Software Foundation, Inc.
5 # This file is part of GCC.
7 # GCC is free software; you can redistribute it and/or modify
8 # it under the terms of the GNU General Public License as published by
9 # the Free Software Foundation; either version 3, or (at your option)
12 # GCC is distributed in the hope that it will be useful,
13 # but WITHOUT ANY WARRANTY; without even the implied warranty of
14 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 # GNU General Public License for more details.
17 # You should have received a copy of the GNU General Public License
18 # along with GCC; see the file COPYING. If not, write to
19 # the Free Software Foundation, 51 Franklin Street, Fifth Floor,
20 # Boston, MA 02110-1301, USA.
22 # This script parses a .diff file generated with 'diff -up' or 'diff -cp'
23 # and adds a skeleton ChangeLog file to the file. It does not try to be
24 # too smart when parsing function names, but it produces a reasonable
27 # Author: Martin Liska <mliska@suse.cz>
36 from itertools
import takewhile
40 from unidiff
import PatchSet
44 CO_AUTHORED_BY_PREFIX
= 'co-authored-by: '
46 pr_regex
= re
.compile(r
'(\/(\/|\*)|[Cc*!])\s+(?P<pr>PR [a-z+-]+\/[0-9]+)')
47 prnum_regex
= re
.compile(r
'PR (?P<comp>[a-z+-]+)/(?P<num>[0-9]+)')
48 dr_regex
= re
.compile(r
'(\/(\/|\*)|[Cc*!])\s+(?P<dr>DR [0-9]+)')
49 dg_regex
= re
.compile(r
'{\s+dg-(error|warning)')
50 pr_filename_regex
= re
.compile(r
'(^|[\W_])[Pp][Rr](?P<pr>\d{4,})')
51 identifier_regex
= re
.compile(r
'^([a-zA-Z0-9_#].*)')
52 comment_regex
= re
.compile(r
'^\/\*')
53 struct_regex
= re
.compile(r
'^(class|struct|union|enum)\s+'
54 r
'(GTY\(.*\)\s+)?([a-zA-Z0-9_]+)')
55 macro_regex
= re
.compile(r
'#\s*(define|undef)\s+([a-zA-Z0-9_]+)')
56 super_macro_regex
= re
.compile(r
'^DEF[A-Z0-9_]+\s*\(([a-zA-Z0-9_]+)')
57 fn_regex
= re
.compile(r
'([a-zA-Z_][^()\s]*)\s*\([^*]')
58 template_and_param_regex
= re
.compile(r
'<[^<>]*>')
59 md_def_regex
= re
.compile(r
'\(define.*\s+"(.*)"')
60 bugzilla_url
= 'https://gcc.gnu.org/bugzilla/rest.cgi/bug?id=%s&' \
61 'include_fields=summary,component'
63 function_extensions
= {'.c', '.cpp', '.C', '.cc', '.h', '.inc', '.def', '.md'}
65 # NB: Makefile.in isn't listed as it's not always generated.
66 generated_files
= {'aclocal.m4', 'config.h.in', 'configure'}
69 Generate ChangeLog template for PATCH.
70 PATCH must be generated using diff(1)'s -up or -cp options
71 (or their equivalent in git).
74 script_folder
= os
.path
.realpath(__file__
)
75 root
= os
.path
.dirname(os
.path
.dirname(script_folder
))
78 def find_changelog(path
):
79 folder
= os
.path
.split(path
)[0]
81 if os
.path
.exists(os
.path
.join(root
, folder
, 'ChangeLog')):
83 folder
= os
.path
.dirname(folder
)
86 raise AssertionError()
89 def extract_function_name(line
):
90 if comment_regex
.match(line
):
92 m
= struct_regex
.search(line
)
95 return m
.group(1) + ' ' + m
.group(3)
96 m
= macro_regex
.search(line
)
100 m
= super_macro_regex
.search(line
)
104 m
= fn_regex
.search(line
)
106 # Discard template and function parameters.
108 fn
= re
.sub(template_and_param_regex
, '', fn
)
113 def try_add_function(functions
, line
):
114 fn
= extract_function_name(line
)
115 if fn
and fn
not in functions
:
120 def sort_changelog_files(changed_file
):
121 return (changed_file
.is_added_file
, changed_file
.is_removed_file
)
124 def get_pr_titles(prs
):
126 for idx
, pr
in enumerate(prs
):
127 pr_id
= pr
.split('/')[-1]
128 r
= requests
.get(bugzilla_url
% pr_id
)
129 bugs
= r
.json()['bugs']
131 prs
[idx
] = 'PR %s/%s' % (bugs
[0]['component'], pr_id
)
132 out
= '%s - %s\n' % (prs
[idx
], bugs
[0]['summary'])
133 if out
not in output
:
137 return '\n'.join(output
)
140 def append_changelog_line(out
, relative_path
, text
):
141 line
= f
'\t* {relative_path}:'
142 if len(line
.replace('\t', ' ' * TAB_WIDTH
) + ' ' + text
) <= LINE_LIMIT
:
143 out
+= f
'{line} {text}\n'
150 def get_rel_path_if_prefixed(path
, folder
):
151 if path
.startswith(folder
):
152 return path
[len(folder
):].lstrip('/')
157 def generate_changelog(data
, no_functions
=False, fill_pr_titles
=False,
158 additional_prs
=None):
165 diff
= PatchSet(data
)
168 for apr
in additional_prs
:
169 if not apr
.startswith('PR ') and '/' in apr
:
174 # skip files that can't be parsed
175 if file.path
== '/dev/null':
177 changelog
= find_changelog(file.path
)
178 if changelog
not in changelogs
:
179 changelogs
[changelog
] = []
180 changelog_list
.append(changelog
)
181 changelogs
[changelog
].append(file)
183 # Extract PR entries from newly added tests
184 if 'testsuite' in file.path
and file.is_added_file
:
185 # Only search first ten lines as later lines may
186 # contains commented code which a note that it
187 # has not been tested due to a certain PR or DR.
189 for line
in list(file)[0][0:10]:
190 m
= pr_regex
.search(line
.value
)
195 this_file_prs
.append(pr
.split('/')[-1])
197 m
= dr_regex
.search(line
.value
)
202 this_file_prs
.append(dr
.split('/')[-1])
203 elif dg_regex
.search(line
.value
):
204 # Found dg-warning/dg-error line
206 # PR number in the file name
207 fname
= os
.path
.basename(file.path
)
208 m
= pr_filename_regex
.search(fname
)
212 if pr
not in this_file_prs
and pr2
not in prs
:
216 out
+= get_pr_titles(prs
)
218 # print list of PR entries before ChangeLog entries
226 # sort ChangeLog so that 'testsuite' is at the end
227 for changelog
in sorted(changelog_list
, key
=lambda x
: 'testsuite' in x
):
228 files
= changelogs
[changelog
]
229 out
+= '%s:\n' % os
.path
.join(changelog
, 'ChangeLog')
231 # new and deleted files should be at the end
232 for file in sorted(files
, key
=sort_changelog_files
):
233 assert file.path
.startswith(changelog
)
234 in_tests
= 'testsuite' in changelog
or 'testsuite' in file.path
235 relative_path
= get_rel_path_if_prefixed(file.path
, changelog
)
237 if file.is_added_file
:
238 msg
= 'New test.' if in_tests
else 'New file.'
239 out
= append_changelog_line(out
, relative_path
, msg
)
240 elif file.is_removed_file
:
241 out
= append_changelog_line(out
, relative_path
, 'Removed.')
242 elif hasattr(file, 'is_rename') and file.is_rename
:
243 # A file can be theoretically moved to a location that
244 # belongs to a different ChangeLog. Let user fix it.
246 # Since unidiff 0.7.0, path.file == path.target_file[2:],
247 # it used to be path.source_file[2:]
248 relative_path
= get_rel_path_if_prefixed(file.source_file
[2:],
250 out
= append_changelog_line(out
, relative_path
, 'Moved to...')
251 new_path
= get_rel_path_if_prefixed(file.target_file
[2:],
253 out
+= f
'\t* {new_path}: ...here.\n'
254 elif os
.path
.basename(file.path
) in generated_files
:
255 out
+= '\t* %s: Regenerate.\n' % (relative_path
)
256 append_changelog_line(out
, relative_path
, 'Regenerate.')
260 # Do not add function names for testsuite files
261 extension
= os
.path
.splitext(relative_path
)[1]
262 if not in_tests
and extension
in function_extensions
:
264 modified_visited
= False
267 m
= identifier_regex
.match(line
.value
)
268 if line
.is_added
or line
.is_removed
:
269 # special-case definition in .md files
270 m2
= md_def_regex
.match(line
.value
)
271 if extension
== '.md' and m2
:
273 if fn
not in functions
:
278 if not line
.value
.strip():
280 modified_visited
= True
281 if m
and try_add_function(functions
,
285 elif line
.is_context
:
286 if last_fn
and modified_visited
:
287 try_add_function(functions
, last_fn
)
289 modified_visited
= False
293 modified_visited
= False
295 try_add_function(functions
,
298 out
+= '\t* %s (%s):\n' % (relative_path
, functions
[0])
299 for fn
in functions
[1:]:
300 out
+= '\t(%s):\n' % fn
302 out
+= '\t* %s:\n' % relative_path
307 def update_copyright(data
):
308 current_timestamp
= datetime
.datetime
.now().strftime('%Y-%m-%d')
309 username
= subprocess
.check_output('git config user.name', shell
=True,
310 encoding
='utf8').strip()
311 email
= subprocess
.check_output('git config user.email', shell
=True,
312 encoding
='utf8').strip()
315 diff
= PatchSet(data
)
318 changelog
= os
.path
.join(find_changelog(file.path
), 'ChangeLog')
319 if changelog
not in changelogs
:
320 changelogs
.add(changelog
)
321 with
open(changelog
) as f
:
323 with
open(changelog
, 'w+') as f
:
324 f
.write(f
'{current_timestamp} {username} <{email}>\n\n')
325 f
.write('\tUpdate copyright years.\n\n')
329 def skip_line_in_changelog(line
):
330 if line
.lower().startswith(CO_AUTHORED_BY_PREFIX
) or line
.startswith('#'):
335 if __name__
== '__main__':
336 extra_args
= os
.getenv('GCC_MKLOG_ARGS')
338 sys
.argv
+= json
.loads(extra_args
)
340 parser
= argparse
.ArgumentParser(description
=help_message
)
341 parser
.add_argument('input', nargs
='?',
342 help='Patch file (or missing, read standard input)')
343 parser
.add_argument('-b', '--pr-numbers', action
='store',
344 type=lambda arg
: arg
.split(','), nargs
='?',
345 help='Add the specified PRs (comma separated)')
346 parser
.add_argument('-s', '--no-functions', action
='store_true',
347 help='Do not generate function names in ChangeLogs')
348 parser
.add_argument('-p', '--fill-up-bug-titles', action
='store_true',
349 help='Download title of mentioned PRs')
350 parser
.add_argument('-d', '--directory',
351 help='Root directory where to search for ChangeLog '
353 parser
.add_argument('-c', '--changelog',
354 help='Append the ChangeLog to a git commit message '
356 parser
.add_argument('--update-copyright', action
='store_true',
357 help='Update copyright in ChangeLog files')
358 args
= parser
.parse_args()
359 if args
.input == '-':
362 root
= args
.directory
364 data
= open(args
.input) if args
.input else sys
.stdin
365 if args
.update_copyright
:
366 update_copyright(data
)
368 output
= generate_changelog(data
, args
.no_functions
,
369 args
.fill_up_bug_titles
, args
.pr_numbers
)
371 lines
= open(args
.changelog
).read().split('\n')
372 start
= list(takewhile(skip_line_in_changelog
, lines
))
373 end
= lines
[len(start
):]
374 with
open(args
.changelog
, 'w') as f
:
375 if not start
or not start
[0]:
377 # initial commit subject line 'component: [PRnnnnn]'
378 m
= prnum_regex
.match(prs
[0])
380 title
= f
'{m.group("comp")}: [PR{m.group("num")}]'
381 start
.insert(0, title
)
387 # append 2 empty lines
389 f
.write('\n'.join(start
))
392 f
.write('\n'.join(end
))
394 print(output
, end
='')