bpo-2504: Add pgettext() and variants to gettext. (GH-7253)
diff --git a/Tools/i18n/msgfmt.py b/Tools/i18n/msgfmt.py
index 63d52d1..3f731e9 100755
--- a/Tools/i18n/msgfmt.py
+++ b/Tools/i18n/msgfmt.py
@@ -5,7 +5,8 @@
This program converts a textual Uniforum-style message catalog (.po file) into
a binary GNU catalog (.mo file). This is essentially the same function as the
-GNU msgfmt program, however, it is a simpler implementation.
+GNU msgfmt program, however, it is a simpler implementation. Currently it
+does not handle plural forms but it does handle message contexts.
Usage: msgfmt.py [OPTIONS] filename.po
@@ -32,12 +33,11 @@
import array
from email.parser import HeaderParser
-__version__ = "1.1"
+__version__ = "1.2"
MESSAGES = {}
-
def usage(code, msg=''):
print(__doc__, file=sys.stderr)
if msg:
@@ -45,15 +45,16 @@
sys.exit(code)
-
-def add(id, str, fuzzy):
+def add(ctxt, id, str, fuzzy):
"Add a non-fuzzy translation to the dictionary."
global MESSAGES
if not fuzzy and str:
- MESSAGES[id] = str
+ if ctxt is None:
+ MESSAGES[id] = str
+ else:
+ MESSAGES[b"%b\x04%b" % (ctxt, id)] = str
-
def generate():
"Return the generated output."
global MESSAGES
@@ -95,10 +96,10 @@
return output
-
def make(filename, outfile):
ID = 1
STR = 2
+ CTXT = 3
# Compute .mo name from .po name and arguments
if filename.endswith('.po'):
@@ -115,7 +116,7 @@
print(msg, file=sys.stderr)
sys.exit(1)
- section = None
+ section = msgctxt = None
fuzzy = 0
# Start off assuming Latin-1, so everything decodes without failure,
@@ -129,8 +130,8 @@
lno += 1
# If we get a comment line after a msgstr, this is a new entry
if l[0] == '#' and section == STR:
- add(msgid, msgstr, fuzzy)
- section = None
+ add(msgctxt, msgid, msgstr, fuzzy)
+ section = msgctxt = None
fuzzy = 0
# Record a fuzzy mark
if l[:2] == '#,' and 'fuzzy' in l:
@@ -138,10 +139,16 @@
# Skip comments
if l[0] == '#':
continue
- # Now we are in a msgid section, output previous section
- if l.startswith('msgid') and not l.startswith('msgid_plural'):
+ # Now we are in a msgid or msgctxt section, output previous section
+ if l.startswith('msgctxt'):
if section == STR:
- add(msgid, msgstr, fuzzy)
+ add(msgctxt, msgid, msgstr, fuzzy)
+ section = CTXT
+ l = l[7:]
+ msgctxt = b''
+ elif l.startswith('msgid') and not l.startswith('msgid_plural'):
+ if section == STR:
+ add(msgctxt, msgid, msgstr, fuzzy)
if not msgid:
# See whether there is an encoding declaration
p = HeaderParser()
@@ -183,7 +190,9 @@
if not l:
continue
l = ast.literal_eval(l)
- if section == ID:
+ if section == CTXT:
+ msgctxt += l.encode(encoding)
+ elif section == ID:
msgid += l.encode(encoding)
elif section == STR:
msgstr += l.encode(encoding)
@@ -194,7 +203,7 @@
sys.exit(1)
# Add last entry
if section == STR:
- add(msgid, msgstr, fuzzy)
+ add(msgctxt, msgid, msgstr, fuzzy)
# Compute output
output = generate()
@@ -206,7 +215,6 @@
print(msg, file=sys.stderr)
-
def main():
try:
opts, args = getopt.getopt(sys.argv[1:], 'hVo:',