scripts/i18n: add command 'normalize-po-files'

The translation files in the Kallithea repository contained references to
the location(s) of each string in the repository. This is useful to
translators, but is not needed for all other users.

The big problem with that information is that it changes very commonly as a
result of normal development in Kallithea, causing a lot of unimportant
delta in the Kallithea repository, thus causing unnecessary repository
growth.

In this commit, a basic version of the script is added, only containing the
code to normalize the translation files by removing generated and outdated
data.

This can be used to check or ensure internal consistency between code and
translations, by extracting and merging and then removing most of it again with
normalize-po-files:

  ./setup.py extract_messages

  for po in kallithea/i18n/*/LC_MESSAGES/kallithea.po; do msgmerge --width=76 --backup=none --previous --update $po kallithea/i18n/kallithea.pot ; done

  scripts/i18n normalize-po-files kallithea/i18n/kallithea.pot kallithea/i18n/*/LC_MESSAGES/kallithea.po

Includes contributions from Mads Kiilerich.

scripts/i18n

@@ -21,6 +21,14 b' import i18n_utils'
 
 """
 Tool for maintenance of .po and .pot files
+
+Normally, the i18n-related files contain for each translatable string a
+reference to all the source code locations where this string is found. This
+meta data is useful for translators to assess how strings are used, but is not
+relevant for normal development nor for running Kallithea. Such meta data, or
+derived data like kallithea.pot, will inherently be outdated, and create
+unnecessary churn and repository growth, making it harder to spot actual and
+important changes.
 """
 
 @click.group()
@@ -30,5 +38,16 b' def cli(debug):'
         i18n_utils.do_debug = True
     pass
 
[e-mail]()
[e-mail]('po_files', nargs=-1)
+def normalize_po_files(po_files):
+    """Normalize the specified .po and .pot files.
+
+    Only actual translations and essential headers will be preserved.
+    """
+    for po_file in po_files:
+        i18n_utils._normalize_po_file(po_file, strip=True)
+
+
 if __name__ == '__main__':
     cli()

scripts/i18n_utils.py

@@ -13,6 +13,8 b''
 
 from __future__ import print_function
 
+import os
+import re
 import subprocess
 
 
@@ -25,3 +27,141 b' def debug(*args, **kwargs):'
 def runcmd(cmd, *args, **kwargs):
     debug('... Executing command: %s' % ' '.join(cmd))
     subprocess.check_call(cmd, *args, **kwargs)
+
+header_comment_strip_re = re.compile(r'''
+    ^
+    [#][ ]Translations[ ]template[ ]for[ ]Kallithea[.] \n
+    |
+    ^
+    [#][ ]FIRST[ ]AUTHOR[ ]<EMAIL@ADDRESS>,[ ]\d+[.] \n
+    [#] \n
+    [#],[ ]fuzzy \n
+    ''',
+    re.MULTILINE|re.VERBOSE)
+
+header_normalize_re = re.compile(r'''
+    ^ "
+    (POT-Creation-Date|PO-Revision-Date|Last-Translator|Language-Team|X-Generator|Generated-By|Project-Id-Version):
+    [ ][^\\]*\\n
+    " \n
+    ''',
+    re.MULTILINE|re.IGNORECASE|re.VERBOSE)
+
+def _normalize_po(raw_content):
+    r"""
+    >>> print(_normalize_po(r'''
+    ... # header comment
+    ...
+    ...
+    ... # comment before header
+    ... msgid ""
+    ... msgstr "yada"
+    ... "POT-Creation-Date: 2019-05-04 21:13+0200\n"
+    ... "MIME-Version: "
+    ... "1.0\n"
+    ... "Last-Translator: Jabba"
+    ... "the Hutt\n"
+    ... "X-Generator: Weblate 1.2.3\n"
+    ...
+    ... # comment, but not in header
+    ... msgid "None"
+    ... msgstr "Ingen"
+    ...
+    ...
+    ... line 2
+    ... # third comment
+    ...
+    ... msgid "Special"
+    ... msgstr ""
+    ...
+    ... msgid "Specialist"
+    ... # odd comment
+    ... msgstr ""
+    ... "Expert"
+    ...
+    ... # crazy fuzzy auto translation by msgmerge, using foo for bar
+    ... #, fuzzy
+    ... #| msgid "some foo string"
+    ... msgid "some bar string."
+    ... msgstr "translation of foo string"
+    ...
+    ... msgid "%d minute"
+    ... msgid_plural "%d minutes"
+    ... msgstr[0] "minut"
+    ... msgstr[1] "minutter"
+    ... msgstr[2] ""
+    ...
+    ... msgid "%d year"
+    ... msgid_plural "%d years"
+    ... msgstr[0] ""
+    ... msgstr[1] ""
+    ...
+    ... # last comment
+    ... ''') + '^^^')
+    # header comment
+    <BLANKLINE>
+    <BLANKLINE>
+    # comment before header
+    <BLANKLINE>
+    msgid ""
+    msgstr "yada"
+    "MIME-Version: "
+    "1.0\n"
+    <BLANKLINE>
+    msgid "None"
+    msgstr "Ingen"
+    <BLANKLINE>
+    line 2
+    <BLANKLINE>
+    msgid "Specialist"
+    msgstr ""
+    "Expert"
+    <BLANKLINE>
+    msgid "%d minute"
+    msgid_plural "%d minutes"
+    msgstr[0] "minut"
+    msgstr[1] "minutter"
+    msgstr[2] ""
+    ^^^
+    """
+    header_start = raw_content.find('\nmsgid ""\n') + 1
+    header_end = raw_content.find('\n\n', header_start) + 1 or len(raw_content)
+    chunks = [
+        header_comment_strip_re.sub('', raw_content[0:header_start])
+            .strip(),
+        '',
+        header_normalize_re.sub('', raw_content[header_start:header_end])
+            .strip(),
+        '']  # preserve normalized header
+    # all chunks are separated by empty line
+    for raw_chunk in raw_content[header_end:].split('\n\n'):
+        if '\n#, fuzzy' in raw_chunk:  # might be like "#, fuzzy, python-format"
+            continue  # drop crazy auto translation that is worse than useless
+        # strip all comment lines from chunk
+        chunk_lines = [
+            line
+            for line in raw_chunk.splitlines()
+            if line
+            and not line.startswith('#')
+        ]
+        if not chunk_lines:
+            continue
+        # check lines starting from first msgstr, skip chunk if no translation lines
+        msgstr_i = [i for i, line in enumerate(chunk_lines) if line.startswith('msgstr')]
+        if (
+            chunk_lines[0].startswith('msgid') and
+            msgstr_i and
+            all(line.endswith(' ""') for line in chunk_lines[msgstr_i[0]:])
+        ):  # skip translation chunks that doesn't have any actual translations
+            continue
+        chunks.append('\n'.join(chunk_lines) + '\n')
+    return '\n'.join(chunks)
+
+def _normalize_po_file(po_file, strip=False):
+    if strip:
+        po_tmp = po_file + '.tmp'
+        with open(po_file, 'r') as src, open(po_tmp, 'w') as dest:
+            raw_content = src.read()
+            normalized_content = _normalize_po(raw_content)
+            dest.write(normalized_content)
+        os.rename(po_tmp, po_file)