kallithea Files · scripts/logformat.py

Files @ 210e76d69b62

Branch filter:

Location: kallithea/scripts/logformat.py

210e76d69b62 1.8 KiB text/x-python Show Annotation Show as Raw Download as Raw

mads

mysql: bump charset to to 'utf8mb4' to get full UTF-8 support

We used to use 'utf8', but in MySQL, this is just an alias for 'utf8mb3' which
isn't full unicode. 'utf8mb4' has less surprises (especially if used with the
'utf8mb4_unicode_ci' collation).

MySQL character sets for server, database, tables, and connection are set
independently. Until now, we have specified 'utf8' when creating tables to
overrule the database charset and in the default MySQL connection URL.

For existing databases, it is recommended to change encoding (and collation) by
altering the each of the tables inside it as described on
https://stackoverflow.com/questions/6115612/how-to-convert-an-entire-mysql-database-characterset-and-collation-to-utf-8 .

#!/usr/bin/env python3

import re
import sys


logre = r'''
(log\.(?:error|info|warning|debug)
[(][ \n]*
)
%s
(
[ \n]*[)]
)
'''


res = [
    # handle % () - keeping spaces around the old %
    (re.compile(logre % r'''("[^"]*"|'[^']*')   ([\n ]*) %  ([\n ]*) \( ( (?:[^()]|\n)* (?: \( (?:[^()]|\n)* \) (?:[^()]|\n)* )* ) \) ''', flags=re.MULTILINE | re.VERBOSE), r'\1\2,\3\4\5\6'),
    # handle % without () - keeping spaces around the old %
    (re.compile(logre % r'''("[^"]*"|'[^']*')   ([\n ]*) %  ([\n ]*)    ( (?:[^()]|\n)* (?: \( (?:[^()]|\n)* \) (?:[^()]|\n)* )* )    ''', flags=re.MULTILINE | re.VERBOSE), r'\1\2,\3\4\5\6'),
    # remove extra space if it is on next line
    (re.compile(logre % r'''("[^"]*"|'[^']*') , (\n [ ]) ([ ][\n ]*)    ( (?:[^()]|\n)* (?: \( (?:[^()]|\n)* \) (?:[^()]|\n)* )* )    ''', flags=re.MULTILINE | re.VERBOSE), r'\1\2,\3\4\5\6'),
    # remove extra space if it is on same line
    (re.compile(logre % r'''("[^"]*"|'[^']*') , [ ]+  () (   [\n ]+)    ( (?:[^()]|\n)* (?: \( (?:[^()]|\n)* \) (?:[^()]|\n)* )* )    ''', flags=re.MULTILINE | re.VERBOSE), r'\1\2,\3\4\5\6'),
    # remove trailing , and space
    (re.compile(logre % r'''("[^"]*"|'[^']*') ,       () (   [\n ]*)    ( (?:[^()]|\n)* (?: \( (?:[^()]|\n)* \) (?:[^()]|\n)* )* [^(), \n] ) [ ,]*''', flags=re.MULTILINE | re.VERBOSE), r'\1\2,\3\4\5\6'),
    ]


def rewrite(f):
    s = open(f).read()
    for r, t in res:
        s = r.sub(t, s)
    open(f, 'w').write(s)


if __name__ == '__main__':
    if len(sys.argv) < 2:
        print('Cleanup of superfluous % formatting of log statements.')
        print('Usage:')
        print('''  hg revert `hg loc '*.py'|grep -v logformat.py` && scripts/logformat.py `hg loc '*.py'` && hg diff''')
        raise SystemExit(1)

    for f in sys.argv[1:]:
        rewrite(f)