You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
306 lines
11 KiB
306 lines
11 KiB
# -*- coding: utf-8; mode: python -*-
|
|
##
|
|
## Format
|
|
##
|
|
## ACTION: [AUDIENCE:] COMMIT_MSG [!TAG ...]
|
|
##
|
|
## Description
|
|
##
|
|
## ACTION is one of 'chg', 'fix', 'new'
|
|
##
|
|
## Is WHAT the change is about.
|
|
##
|
|
## 'chg' is for refactor, small improvement, cosmetic changes...
|
|
## 'fix' is for bug fixes
|
|
## 'new' is for new features, big improvement
|
|
##
|
|
## AUDIENCE is optional and one of 'dev', 'usr', 'pkg', 'test', 'doc'
|
|
##
|
|
## Is WHO is concerned by the change.
|
|
##
|
|
## 'dev' is for developpers (API changes, refactors...)
|
|
## 'usr' is for final users (UI changes)
|
|
## 'pkg' is for packagers (packaging changes)
|
|
## 'test' is for testers (test only related changes)
|
|
## 'doc' is for doc guys (doc only changes)
|
|
##
|
|
## COMMIT_MSG is ... well ... the commit message itself.
|
|
##
|
|
## TAGs are additionnal adjective as 'refactor' 'minor' 'cosmetic'
|
|
##
|
|
## They are preceded with a '!' or a '@' (prefer the former, as the
|
|
## latter is wrongly interpreted in github.) Commonly used tags are:
|
|
##
|
|
## 'refactor' is obviously for refactoring code only
|
|
## 'minor' is for a very meaningless change (a typo, adding a comment)
|
|
## 'cosmetic' is for cosmetic driven change (re-indentation, 80-col...)
|
|
## 'wip' is for partial functionality but complete subfunctionality.
|
|
##
|
|
## Example:
|
|
##
|
|
## new: usr: support of bazaar implemented
|
|
## chg: re-indentend some lines !cosmetic
|
|
## new: dev: updated code to be compatible with last version of killer lib.
|
|
## fix: pkg: updated year of licence coverage.
|
|
## new: test: added a bunch of test around user usability of feature X.
|
|
## fix: typo in spelling my name in comment. !minor
|
|
##
|
|
## Please note that multi-line commit message are supported, and only the
|
|
## first line will be considered as the "summary" of the commit message. So
|
|
## tags, and other rules only applies to the summary. The body of the commit
|
|
## message will be displayed in the changelog without reformatting.
|
|
|
|
|
|
##
|
|
## ``ignore_regexps`` is a line of regexps
|
|
##
|
|
## Any commit having its full commit message matching any regexp listed here
|
|
## will be ignored and won't be reported in the changelog.
|
|
##
|
|
ignore_regexps = [
|
|
r'@minor', r'!minor',
|
|
r'@cosmetic', r'!cosmetic',
|
|
r'@refactor', r'!refactor',
|
|
r'@wip', r'!wip',
|
|
r'^([cC]hg|[fF]ix|[nN]ew)\s*:\s*[p|P]kg:',
|
|
r'^([cC]hg|[fF]ix|[nN]ew)\s*:\s*[d|D]ev:',
|
|
r'^(.{3,3}\s*:)?\s*[fF]irst commit.?\s*$',
|
|
r'^$', ## ignore commits with empty messages
|
|
]
|
|
|
|
|
|
## ``section_regexps`` is a list of 2-tuples associating a string label and a
|
|
## list of regexp
|
|
##
|
|
## Commit messages will be classified in sections thanks to this. Section
|
|
## titles are the label, and a commit is classified under this section if any
|
|
## of the regexps associated is matching.
|
|
##
|
|
## Please note that ``section_regexps`` will only classify commits and won't
|
|
## make any changes to the contents. So you'll probably want to go check
|
|
## ``subject_process`` (or ``body_process``) to do some changes to the subject,
|
|
## whenever you are tweaking this variable.
|
|
##
|
|
section_regexps = [
|
|
('**New features**', [
|
|
r'^[aA]dded?\s*:?\s*((dev|use?r|pkg|test|doc)\s*:\s*)?([^\n]*)$',
|
|
r'^[uU]pdated?\s*:?\s*((dev|use?r|pkg|test|doc)\s*:\s*)?([^\n]*)$',
|
|
r'^[cC]hanged?\s*:?\s*((dev|use?r|pkg|test|doc)\s*:\s*)?([^\n]*)$',
|
|
r'^[nN]ew?\s*:?\s*((dev|use?r|pkg|test|doc)\s*:\s*)?([^\n]*)$',
|
|
]),
|
|
('**Fixes**', [
|
|
r'^(?![mM]erge\s*)'
|
|
]
|
|
),
|
|
|
|
]
|
|
|
|
|
|
## ``body_process`` is a callable
|
|
##
|
|
## This callable will be given the original body and result will
|
|
## be used in the changelog.
|
|
##
|
|
## Available constructs are:
|
|
##
|
|
## - any python callable that take one txt argument and return txt argument.
|
|
##
|
|
## - ReSub(pattern, replacement): will apply regexp substitution.
|
|
##
|
|
## - Indent(chars=" "): will indent the text with the prefix
|
|
## Please remember that template engines gets also to modify the text and
|
|
## will usually indent themselves the text if needed.
|
|
##
|
|
## - Wrap(regexp=r"\n\n"): re-wrap text in separate paragraph to fill 80-Columns
|
|
##
|
|
## - noop: do nothing
|
|
##
|
|
## - ucfirst: ensure the first letter is uppercase.
|
|
## (usually used in the ``subject_process`` pipeline)
|
|
##
|
|
## - final_dot: ensure text finishes with a dot
|
|
## (usually used in the ``subject_process`` pipeline)
|
|
##
|
|
## - strip: remove any spaces before or after the content of the string
|
|
##
|
|
## - SetIfEmpty(msg="No commit message."): will set the text to
|
|
## whatever given ``msg`` if the current text is empty.
|
|
##
|
|
## Additionally, you can `pipe` the provided filters, for instance:
|
|
#body_process = Wrap(regexp=r'\n(?=\w+\s*:)') | Indent(chars=" ")
|
|
#body_process = Wrap(regexp=r'\n(?=\w+\s*:)')
|
|
#body_process = noop
|
|
body_process = ReSub(r'((^|\n)[A-Z]\w+(-\w+)*: .*(\n\s+.*)*)+$', r'') | strip
|
|
|
|
|
|
## ``subject_process`` is a callable
|
|
##
|
|
## This callable will be given the original subject and result will
|
|
## be used in the changelog.
|
|
## subject_process = (strip |
|
|
## ReSub(r'^([aA]dd(ed?)?|[nN]ew)(\s?:?\s)(.*)$', r' \4') |
|
|
## ReSub(r'^([cC]hang(ed?)?)(\s?:?\s)(.*)$', r' \4') |
|
|
## ReSub(r'^([fF]ix(ed?)?)(\s?:?\s)(.*)$', r' \4') |
|
|
## ReSub(r'^([uU]pdat(ed?)?)(\s?:?\s)(.*)$', r' \4') |
|
|
## ReSub(r'#(\d{3,4})', r'[#\1](https://github.com/Radarr/Radarr/issues/\1)') |
|
|
## SetIfEmpty("No commit message.") | ucfirst | final_dot)
|
|
|
|
## Available constructs are those listed in ``body_process`` doc.
|
|
subject_process = (strip |
|
|
ReSub(r'^([aA]dd(ed?)?|[nN]ew)(\s?:?\s)(.*)$', r' \4') |
|
|
ReSub(r'^([cC]hang(ed?)?)(\s?:?\s)(.*)$', r' \4') |
|
|
ReSub(r'^([fF]ix(ed?)?)(\s?:?\s)(.*)$', r' \4') |
|
|
ReSub(r'^([uU]pdat(ed?)?)(\s?:?\s)(.*)$', r' \4') |
|
|
ReSub(r'#(\d{3,4})', r'[#\1](https://github.com/Radarr/Radarr/issues/\1)') |
|
|
SetIfEmpty("No commit message.") | ucfirst | final_dot)
|
|
|
|
|
|
## ``tag_filter_regexp`` is a regexp
|
|
##
|
|
## Tags that will be used for the changelog must match this regexp.
|
|
##
|
|
tag_filter_regexp = r'^v[0]+\.[2-9]+\.[0-9]+\.[0-9]{3,4}$'
|
|
|
|
|
|
## ``unreleased_version_label`` is a string or a callable that outputs a string
|
|
##
|
|
## This label will be used as the changelog Title of the last set of changes
|
|
## between last valid tag and HEAD if any.
|
|
unreleased_version_label = "(unreleased)"
|
|
|
|
|
|
## ``output_engine`` is a callable
|
|
##
|
|
## This will change the output format of the generated changelog file
|
|
##
|
|
## Available choices are:
|
|
##
|
|
## - rest_py
|
|
##
|
|
## Legacy pure python engine, outputs ReSTructured text.
|
|
## This is the default.
|
|
##
|
|
## - mustache(<template_name>)
|
|
##
|
|
## Template name could be any of the available templates in
|
|
## ``templates/mustache/*.tpl``.
|
|
## Requires python package ``pystache``.
|
|
## Examples:
|
|
## - mustache("markdown")
|
|
## - mustache("restructuredtext")
|
|
##
|
|
## - makotemplate(<template_name>)
|
|
##
|
|
## Template name could be any of the available templates in
|
|
## ``templates/mako/*.tpl``.
|
|
## Requires python package ``mako``.
|
|
## Examples:
|
|
## - makotemplate("restructuredtext")
|
|
##
|
|
#output_engine = rest_py
|
|
#output_engine = mustache("restructuredtext")
|
|
output_engine = mustache("changelog.tpl")
|
|
#output_engine = makotemplate("restructuredtext")
|
|
|
|
|
|
## ``include_merge`` is a boolean
|
|
##
|
|
## This option tells git-log whether to include merge commits in the log.
|
|
## The default is to include them.
|
|
include_merge = False
|
|
|
|
|
|
## ``log_encoding`` is a string identifier
|
|
##
|
|
## This option tells gitchangelog what encoding is outputed by ``git log``.
|
|
## The default is to be clever about it: it checks ``git config`` for
|
|
## ``i18n.logOutputEncoding``, and if not found will default to git's own
|
|
## default: ``utf-8``.
|
|
#log_encoding = 'utf-8'
|
|
|
|
|
|
## ``publish`` is a callable
|
|
##
|
|
## Sets what ``gitchangelog`` should do with the output generated by
|
|
## the output engine. ``publish`` is a callable taking one argument
|
|
## that is an interator on lines from the output engine.
|
|
##
|
|
## Some helper callable are provided:
|
|
##
|
|
## Available choices are:
|
|
##
|
|
## - stdout
|
|
##
|
|
## Outputs directly to standard output
|
|
## (This is the default)
|
|
##
|
|
## - FileInsertAtFirstRegexMatch(file, pattern, idx=lamda m: m.start())
|
|
##
|
|
## Creates a callable that will parse given file for the given
|
|
## regex pattern and will insert the output in the file.
|
|
## ``idx`` is a callable that receive the matching object and
|
|
## must return a integer index point where to insert the
|
|
## the output in the file. Default is to return the position of
|
|
## the start of the matched string.
|
|
##
|
|
## - FileRegexSubst(file, pattern, replace, flags)
|
|
##
|
|
## Apply a replace inplace in the given file. Your regex pattern must
|
|
## take care of everything and might be more complex. Check the README
|
|
## for a complete copy-pastable example.
|
|
##
|
|
# publish = FileInsertIntoFirstRegexMatch(
|
|
# "CHANGELOG.rst",
|
|
# r'/(?P<rev>[0-9]+\.[0-9]+(\.[0-9]+)?)\s+\([0-9]+-[0-9]{2}-[0-9]{2}\)\n--+\n/',
|
|
# idx=lambda m: m.start(1)
|
|
# )
|
|
#publish = stdout
|
|
|
|
def write_to_file(content):
|
|
with open("CHANGELOG.md", "w+") as f:
|
|
for chunk in content:
|
|
f.write(chunk)
|
|
|
|
publish = write_to_file
|
|
|
|
|
|
## ``revs`` is a list of callable or a list of string
|
|
##
|
|
## callable will be called to resolve as strings and allow dynamical
|
|
## computation of these. The result will be used as revisions for
|
|
## gitchangelog (as if directly stated on the command line). This allows
|
|
## to filter exaclty which commits will be read by gitchangelog.
|
|
##
|
|
## To get a full documentation on the format of these strings, please
|
|
## refer to the ``git rev-list`` arguments. There are many examples.
|
|
##
|
|
## Using callables is especially useful, for instance, if you
|
|
## are using gitchangelog to generate incrementally your changelog.
|
|
##
|
|
## Some helpers are provided, you can use them::
|
|
##
|
|
## - FileFirstRegexMatch(file, pattern): will return a callable that will
|
|
## return the first string match for the given pattern in the given file.
|
|
## If you use named sub-patterns in your regex pattern, it'll output only
|
|
## the string matching the regex pattern named "rev".
|
|
##
|
|
## - Caret(rev): will return the rev prefixed by a "^", which is a
|
|
## way to remove the given revision and all its ancestor.
|
|
##
|
|
## Please note that if you provide a rev-list on the command line, it'll
|
|
## replace this value (which will then be ignored).
|
|
##
|
|
## If empty, then ``gitchangelog`` will act as it had to generate a full
|
|
## changelog.
|
|
##
|
|
## The default is to use all commits to make the changelog.
|
|
#revs = ["^1.0.3", ]
|
|
#revs = [
|
|
# Caret(
|
|
# FileFirstRegexMatch(
|
|
# "CHANGELOG.rst",
|
|
# r"(?P<rev>[0-9]+\.[0-9]+(\.[0-9]+)?)\s+\([0-9]+-[0-9]{2}-[0-9]{2}\)\n--+\n")),
|
|
# "HEAD"
|
|
#]
|
|
revs = ["v0.2.0.134..."]
|