##// END OF EJS Templates
only sort small dicts and sets in lib.pretty...
only sort small dicts and sets in lib.pretty The sorting caused performance problems when displaying large collections, even when the result would be truncated.

File last commit:

r12901:a176d05e
r18457:7aba3f1e
Show More
update_whatsnew.py
73 lines | 2.4 KiB | text/x-python | PythonLexer
"""Update the What's New doc (development version)
This collects the snippets from whatsnew/pr/, moves their content into
whatsnew/development.rst (chronologically ordered), and deletes the snippets.
"""
import io
import os
from os.path import dirname, basename, abspath, join as pjoin
from subprocess import check_call, check_output
repo_root = dirname(dirname(abspath(__file__)))
whatsnew_dir = pjoin(repo_root, 'docs', 'source', 'whatsnew')
pr_dir = pjoin(whatsnew_dir, 'pr')
target = pjoin(whatsnew_dir, 'development.rst')
FEATURE_MARK = ".. DO NOT EDIT THIS LINE BEFORE RELEASE. FEATURE INSERTION POINT."
INCOMPAT_MARK = ".. DO NOT EDIT THIS LINE BEFORE RELEASE. INCOMPAT INSERTION POINT."
# 1. Collect the whatsnew snippet files ---------------------------------------
files = set(os.listdir(pr_dir))
# Ignore explanatory and example files
files.difference_update({'README.md',
'incompat-switching-to-perl.rst',
'antigravity-feature.rst'}
)
# Absolute paths
files = {pjoin(pr_dir, f) for f in files}
def getmtime(f):
return check_output(['git', 'log', '-1', '--format="%ai"', '--', f])
files = sorted(files, key=getmtime)
features, incompats = [], []
for path in files:
with io.open(path, encoding='utf-8') as f:
content = f.read().rstrip()
if basename(path).startswith('incompat-'):
incompats.append(content)
else:
features.append(content)
# Put the insertion markers back on the end, so they're ready for next time.
feature_block = '\n\n'.join(features + [FEATURE_MARK])
incompat_block = '\n\n'.join(incompats + [INCOMPAT_MARK])
# 2. Update the target file ---------------------------------------------------
with io.open(target, encoding='utf-8') as f:
content = f.read()
assert content.count(FEATURE_MARK) == 1
assert content.count(INCOMPAT_MARK) == 1
content = content.replace(FEATURE_MARK, feature_block)
content = content.replace(INCOMPAT_MARK, incompat_block)
# Clean trailing whitespace
content = '\n'.join(l.rstrip() for l in content.splitlines())
with io.open(target, 'w', encoding='utf-8') as f:
f.write(content)
# 3. Stage the changes in git -------------------------------------------------
for file in files:
check_call(['git', 'rm', file])
check_call(['git', 'add', target])
print("Merged what's new changes. Check the diff and commit the change.")