# to output in gettext .pot format.
#
import sys, os, re, getopt
+if sys.version_info < (2, 4, 0):
+ from sets import Set as set
def relativePath(path, base):
'''return relative path from top source dir'''
def ui_l10n(input_files, output, base):
'''Generate pot file from lib/ui/*'''
output = open(output, 'w')
- Submenu = re.compile(r'^[^#]*Submenu\s+"([^"]*)"')
- Popupmenu = re.compile(r'^[^#]*PopupMenu\s+"[^"]+"\s+"([^"]*)"')
- Toolbar = re.compile(r'^[^#]*Toolbar\s+"[^"]+"\s+"([^"]*)"')
- Item = re.compile(r'[^#]*Item\s+"([^"]*)"')
- TableInsert = re.compile(r'[^#]*TableInsert\s+"([^"]*)"')
+ Submenu = re.compile(r'^[^#]*Submenu\s+"([^"]*)"', re.IGNORECASE)
+ Popupmenu = re.compile(r'^[^#]*PopupMenu\s+"[^"]+"\s+"([^"]*)"', re.IGNORECASE)
+ IconPalette = re.compile(r'^[^#]*IconPalette\s+"[^"]+"\s+"([^"]*)"', re.IGNORECASE)
+ Toolbar = re.compile(r'^[^#]*Toolbar\s+"[^"]+"\s+"([^"]*)"', re.IGNORECASE)
+ Item = re.compile(r'[^#]*Item\s+"([^"]*)"', re.IGNORECASE)
+ TableInsert = re.compile(r'[^#]*TableInsert\s+"([^"]*)"', re.IGNORECASE)
for src in input_files:
input = open(src)
for lineno, line in enumerate(input.readlines()):
string = string.replace('_', ' ')
elif Popupmenu.match(line):
(string,) = Popupmenu.match(line).groups()
+ elif IconPalette.match(line):
+ (string,) = IconPalette.match(line).groups()
elif Toolbar.match(line):
(string,) = Toolbar.match(line).groups()
elif Item.match(line):
output.close()
-def layouts_l10n(input_files, output, base):
+def layouts_l10n(input_files, output, base, layouttranslations):
'''Generate pot file from lib/layouts/*.{layout,inc,module}'''
- out = open(output, 'w')
- Style = re.compile(r'^Style\s+(.*)')
- # include ???LabelString???, but exclude comment lines
- LabelString = re.compile(r'^[^#]*LabelString\S*\s+(.*)')
- GuiName = re.compile(r'\s*GuiName\s+(.*)')
- ListName = re.compile(r'\s*ListName\s+(.*)')
- CategoryName = re.compile(r'\s*Category\s+(.*)')
- NameRE = re.compile(r'DeclareLyXModule.*{(.*)}')
- InsetLayout = re.compile(r'^InsetLayout\s+(.*)')
- DescBegin = re.compile(r'#+\s*DescriptionBegin\s*$')
- DescEnd = re.compile(r'#+\s*DescriptionEnd\s*$')
+ Style = re.compile(r'^\s*Style\s+(.*)\s*$', re.IGNORECASE)
+ # match LabelString, EndLabelString, LabelStringAppendix and maybe others but no comments
+ LabelString = re.compile(r'^[^#]*LabelString\S*\s+(.*)\s*$', re.IGNORECASE)
+ GuiName = re.compile(r'^\s*GuiName\s+(.*)\s*$', re.IGNORECASE)
+ ListName = re.compile(r'^\s*ListName\s+(.*)\s*$', re.IGNORECASE)
+ CategoryName = re.compile(r'^\s*Category\s+(.*)\s*$', re.IGNORECASE)
+ NameRE = re.compile(r'^\s*#\s*\\DeclareLyXModule.*{(.*)}$', re.IGNORECASE)
+ InsetLayout = re.compile(r'^InsetLayout\s+\"?(.*)\"?\s*$', re.IGNORECASE)
+ FlexCheck = re.compile(r'^Flex:(.*)', re.IGNORECASE)
+ DescBegin = re.compile(r'^\s*#DescriptionBegin\s*$', re.IGNORECASE)
+ DescEnd = re.compile(r'^\s*#\s*DescriptionEnd\s*$', re.IGNORECASE)
+ Category = re.compile(r'^\s*#\s*Category:\s+(.*)\s*$', re.IGNORECASE)
+ I18nPreamble = re.compile(r'^\s*((Lang)|(Babel))Preamble\s*$', re.IGNORECASE)
+ EndI18nPreamble = re.compile(r'^\s*End((Lang)|(Babel))Preamble\s*$', re.IGNORECASE)
+ I18nString = re.compile(r'_\(([^\)]+)\)')
+ CounterFormat = re.compile(r'^\s*PrettyFormat\s+"?(.*)"?\s*$', re.IGNORECASE)
+ CiteFormat = re.compile(r'^\s*CiteFormat', re.IGNORECASE)
+ KeyVal = re.compile(r'^\s*_\w+\s+(.*)\s*$')
+ Float = re.compile(r'^\s*Float\s*$', re.IGNORECASE)
+ UsesFloatPkg = re.compile(r'^\s*UsesFloatPkg\s+(.*)\s*$', re.IGNORECASE)
+ IsPredefined = re.compile(r'^\s*IsPredefined\s+(.*)\s*$', re.IGNORECASE)
+ End = re.compile(r'^\s*End', re.IGNORECASE)
+ Comment = re.compile(r'^(.*)#')
+ Translation = re.compile(r'^\s*Translation\s+(.*)\s*$', re.IGNORECASE)
+ KeyValPair = re.compile(r'\s*"(.*)"\s+"(.*)"')
+
+ oldlanguages = []
+ languages = []
+ keyset = set()
+ oldtrans = dict()
+ if layouttranslations:
+ linguas_file = os.path.join(base, 'po/LINGUAS')
+ for line in open(linguas_file).readlines():
+ res = Comment.search(line)
+ if res:
+ line = res.group(1)
+ if line.strip() != '':
+ languages.extend(line.split())
+
+ # read old translations if available
+ try:
+ input = open(output)
+ lang = ''
+ for line in input.readlines():
+ res = Comment.search(line)
+ if res:
+ line = res.group(1)
+ if line.strip() == '':
+ continue
+ res = Translation.search(line)
+ if res:
+ lang = res.group(1)
+ if lang not in languages:
+ oldlanguages.append(lang)
+ languages.append(lang)
+ oldtrans[lang] = dict()
+ continue
+ res = End.search(line)
+ if res:
+ lang = ''
+ continue
+ res = KeyValPair.search(line)
+ if res and lang != '':
+ key = res.group(1).decode('utf-8')
+ val = res.group(2).decode('utf-8')
+ key = key.replace('\\"', '"').replace('\\\\', '\\')
+ val = val.replace('\\"', '"').replace('\\\\', '\\')
+ oldtrans[lang][key] = val
+ keyset.add(key)
+ continue
+ print "Error: Unable to handle line:"
+ print line
+ except IOError:
+ print "Warning: Unable to open %s for reading." % output
+ print " Old translations will be lost."
+
+ # walon is not a known document language
+ # FIXME: Do not hardcode, read from lib/languages!
+ if 'wa' in languages:
+ languages.remove('wa')
+ out = open(output, 'w')
for src in input_files:
readingDescription = False
+ readingI18nPreamble = False
+ readingFloat = False
+ readingCiteFormats = False
+ isPredefined = False
+ usesFloatPkg = True
+ listname = ''
+ floatname = ''
descStartLine = -1
descLines = []
lineno = 0
if res != None:
readingDescription = False
desc = " ".join(descLines)
- writeString(out, src, base, lineno + 1, desc)
+ if not layouttranslations:
+ writeString(out, src, base, lineno + 1, desc)
continue
descLines.append(line[1:].strip())
continue
readingDescription = True
descStartLine = lineno
continue
+ if readingI18nPreamble:
+ res = EndI18nPreamble.search(line)
+ if res != None:
+ readingI18nPreamble = False
+ continue
+ res = I18nString.search(line)
+ if res != None:
+ string = res.group(1)
+ if layouttranslations:
+ keyset.add(string)
+ else:
+ writeString(out, src, base, lineno, string)
+ continue
+ res = I18nPreamble.search(line)
+ if res != None:
+ readingI18nPreamble = True
+ continue
res = NameRE.search(line)
if res != None:
string = res.group(1)
- string = string.replace('\\', '\\\\').replace('"', '')
- if string != "":
- print >> out, '#: %s:%d\nmsgid "%s"\nmsgstr ""\n' % \
- (relativePath(src, base), lineno + 1, string)
+ if not layouttranslations:
+ writeString(out, src, base, lineno + 1, string)
continue
res = Style.search(line)
if res != None:
string = res.group(1)
string = string.replace('_', ' ')
- writeString(out, src, base, lineno, string)
+ # Style means something else inside a float definition
+ if not readingFloat:
+ if not layouttranslations:
+ writeString(out, src, base, lineno, string)
continue
res = LabelString.search(line)
if res != None:
string = res.group(1)
- writeString(out, src, base, lineno, string)
+ if not layouttranslations:
+ writeString(out, src, base, lineno, string)
continue
res = GuiName.search(line)
if res != None:
string = res.group(1)
- writeString(out, src, base, lineno, string)
+ if layouttranslations:
+ # gui name must only be added for floats
+ if readingFloat:
+ floatname = string
+ else:
+ writeString(out, src, base, lineno, string)
continue
res = CategoryName.search(line)
if res != None:
string = res.group(1)
- writeString(out, src, base, lineno, string)
+ if not layouttranslations:
+ writeString(out, src, base, lineno, string)
continue
res = ListName.search(line)
if res != None:
string = res.group(1)
- writeString(out, src, base, lineno, string)
+ if layouttranslations:
+ listname = string.strip('"')
+ else:
+ writeString(out, src, base, lineno, string)
continue
res = InsetLayout.search(line)
if res != None:
string = res.group(1)
string = string.replace('_', ' ')
- writeString(out, src, base, lineno, string)
+ #Flex:xxx is not used in translation
+ #if not layouttranslations:
+ # writeString(out, src, base, lineno, string)
+ m = FlexCheck.search(string)
+ if m:
+ if not layouttranslations:
+ writeString(out, src, base, lineno, m.group(1))
+ continue
+ res = Category.search(line)
+ if res != None:
+ string = res.group(1)
+ if not layouttranslations:
+ writeString(out, src, base, lineno, string)
+ continue
+ res = CounterFormat.search(line)
+ if res != None:
+ string = res.group(1)
+ if not layouttranslations:
+ writeString(out, src, base, lineno, string)
+ continue
+ res = Float.search(line)
+ if res != None:
+ readingFloat = True
+ continue
+ res = IsPredefined.search(line)
+ if res != None:
+ string = res.group(1).lower()
+ if string == 'true':
+ isPredefined = True
+ else:
+ isPredefined = False
+ continue
+ res = UsesFloatPkg.search(line)
+ if res != None:
+ string = res.group(1).lower()
+ if string == 'true':
+ usesFloatPkg = True
+ else:
+ usesFloatPkg = False
+ continue
+ res = CiteFormat.search(line)
+ if res != None:
+ readingCiteFormats = True
continue
+ res = End.search(line)
+ if res != None:
+ # If a float is predefined by the package and it does not need
+ # the float package then it uses the standard babel translations.
+ # This is even true for MarginFigure, MarginTable (both from
+ # tufte-book.layout) and Planotable, Plate (both from aguplus.inc).
+ if layouttranslations and readingFloat and usesFloatPkg and not isPredefined:
+ if floatname != '':
+ keyset.add(floatname)
+ if listname != '':
+ keyset.add(listname)
+ isPredefined = False
+ usesFloatPkg = True
+ listname = ''
+ floatname = ''
+ readingCiteFormats = False
+ readingFloat = False
+ continue
+ if readingCiteFormats:
+ res = KeyVal.search(line)
+ if res != None:
+ val = res.group(1)
+ if not layouttranslations:
+ writeString(out, src, base, lineno, val)
+
+ if layouttranslations:
+ # Extract translations of layout files
+ import polib
+
+ # Sort languages and key to minimize the diff between different runs
+ # with changed translations
+ languages.sort()
+ keys = []
+ for key in keyset:
+ keys.append(key)
+ keys.sort()
+
+ print >> out, '''# This file has been automatically generated by po/lyx_pot.py.
+# PLEASE MODIFY ONLY THE LAGUAGES HAVING NO .po FILE! If you want to regenerate
+# this file from the translations, run `make ../lib/layouttranslations' in po.
+# Python polib library is needed for building the output file.
+#
+# This file should remain fixed during minor LyX releases.
+# For more comments see README.localization file.'''
+ for lang in languages:
+ print >> out, '\nTranslation %s' % lang
+ if lang in oldtrans.keys():
+ trans = oldtrans[lang]
+ else:
+ trans = dict()
+ if not lang in oldlanguages:
+ poname = os.path.join(base, 'po/' + lang + '.po')
+ po = polib.pofile(poname)
+ # Iterate through po entries and not keys for speed reasons.
+ # FIXME: The code is still too slow
+ for entry in po:
+ if not entry.translated():
+ continue
+ if entry.msgid in keys:
+ key = entry.msgid
+ val = entry.msgstr
+ # some translators keep untranslated entries
+ if val != key:
+ trans[key] = val
+ for key in keys:
+ if key in trans.keys():
+ val = trans[key].replace('\\', '\\\\').replace('"', '\\"')
+ key = key.replace('\\', '\\\\').replace('"', '\\"')
+ print >> out, '\t"%s" "%s"' % \
+ (key.encode('utf-8'), val.encode('utf-8'))
+ # also print untranslated entries to help translators
+ elif not lang in oldlanguages:
+ key = key.replace('\\', '\\\\').replace('"', '\\"')
+ print >> out, '\t"%s" "%s"' % \
+ (key.encode('utf-8'), key.encode('utf-8'))
+ print >> out, 'End'
+
out.close()
# get lines that match <string>...</string>
if pat.match(line):
(string,) = pat.match(line).groups()
- string = string.replace('&', '&').replace('<', '<').replace('>', '>')
+ string = string.replace('&', '&').replace('"', '"')
+ string = string.replace('<', '<').replace('>', '>')
string = string.replace('\\', '\\\\').replace('"', r'\"')
+ string = string.replace('
', r'\n')
print >> output, '#: %s:%d\nmsgid "%s"\nmsgstr ""\n' % \
- (relativePath(src, base), lineno+1, string)
+ (relativePath(src, base), lineno+1, string)
input.close()
output.close()
def languages_l10n(input_files, output, base):
- '''Generate pot file from lib/language'''
- output = open(output, 'w')
- # assuming only one language file
- reg = re.compile('[\w-]+\s+[\w"]+\s+"([\w \-\(\)]+)"\s+(true|false)\s+[\w-]+\s+\w+\s+"[^"]*"')
- input = open(input_files[0])
- for lineno, line in enumerate(input.readlines()):
- if line[0] == '#':
- continue
- # From:
- # afrikaans afrikaans "Afrikaans" false iso8859-15 af_ZA ""
- # To:
- # #: lib/languages:2
- # msgid "Afrikaans"
- # msgstr ""
- if reg.match(line):
- print >> output, '#: %s:%d\nmsgid "%s"\nmsgstr ""\n' % \
- (relativePath(input_files[0], base), lineno+1, reg.match(line).groups()[0])
- else:
- print "Error: Unable to handle line:"
- print line
- # No need to abort if the parsing fails (e.g. "ignore" language has no encoding)
- # sys.exit(1)
- input.close()
- output.close()
+ '''Generate pot file from lib/languages'''
+ out = open(output, 'w')
+ GuiName = re.compile(r'^[^#]*GuiName\s+(.*)', re.IGNORECASE)
+
+ for src in input_files:
+ descStartLine = -1
+ descLines = []
+ lineno = 0
+ for line in open(src).readlines():
+ lineno += 1
+ res = GuiName.search(line)
+ if res != None:
+ string = res.group(1)
+ writeString(out, src, base, lineno, string)
+ continue
+
+ out.close()
def external_l10n(input_files, output, base):
'''Generate pot file from lib/external_templates'''
output = open(output, 'w')
- Template = re.compile(r'^Template\s+(.*)')
- GuiName = re.compile(r'\s*GuiName\s+(.*)')
- HelpTextStart = re.compile(r'\s*HelpText\s')
+ Template = re.compile(r'^Template\s+(.*)', re.IGNORECASE)
+ GuiName = re.compile(r'\s*GuiName\s+(.*)', re.IGNORECASE)
+ HelpTextStart = re.compile(r'\s*HelpText\s', re.IGNORECASE)
HelpTextSection = re.compile(r'\s*(\S.*)\s*$')
- HelpTextEnd = re.compile(r'\s*HelpTextEnd\s')
+ HelpTextEnd = re.compile(r'\s*HelpTextEnd\s', re.IGNORECASE)
i = -1
for src in input_files:
input = open(src)
def formats_l10n(input_files, output, base):
'''Generate pot file from configure.py'''
output = open(output, 'w')
- GuiName = re.compile(r'.*\Format\s+\S+\s+\S+\s+"([^"]*)"\s+(\S*)\s+.*')
- GuiName2 = re.compile(r'.*\Format\s+\S+\s+\S+\s+([^"]\S+)\s+(\S*)\s+.*')
+ GuiName = re.compile(r'.*\\Format\s+\S+\s+\S+\s+"([^"]*)"\s+(\S*)\s+.*', re.IGNORECASE)
+ GuiName2 = re.compile(r'.*\\Format\s+\S+\s+\S+\s+([^"]\S+)\s+(\S*)\s+.*', re.IGNORECASE)
input = open(input_files[0])
for lineno, line in enumerate(input.readlines()):
label = ""
output.close()
+def encodings_l10n(input_files, output, base):
+ '''Generate pot file from lib/encodings'''
+ output = open(output, 'w')
+ # assuming only one encodings file
+ # Encoding utf8 utf8 "Unicode (utf8)" UTF-8 variable inputenc
+ reg = re.compile('Encoding [\w-]+\s+[\w-]+\s+"([\w \-\(\)]+)"\s+[\w-]+\s+(fixed|variable)\s+\w+.*')
+ input = open(input_files[0])
+ for lineno, line in enumerate(input.readlines()):
+ if not line.startswith('Encoding'):
+ continue
+ if reg.match(line):
+ print >> output, '#: %s:%d\nmsgid "%s"\nmsgstr ""\n' % \
+ (relativePath(input_files[0], base), lineno+1, reg.match(line).groups()[0])
+ else:
+ print "Error: Unable to handle line:"
+ print line
+ # No need to abort if the parsing fails
+ # sys.exit(1)
+ input.close()
+ output.close()
+
+
+
Usage = '''
-lyx_pot.py [-b|--base top_src_dir] [-o|--output output_file] [-h|--help] -t|--type input_type input_files
+lyx_pot.py [-b|--base top_src_dir] [-o|--output output_file] [-h|--help] [-s|src_file filename] -t|--type input_type input_files
-where
+where
--base:
path to the top source directory. default to '.'
--output:
output pot file, default to './lyx.pot'
+ --src_file
+ filename that contains a list of input files in each line
--input_type can be
ui: lib/ui/*
layouts: lib/layouts/*
+ layouttranslations: create lib/layouttranslations from po/*.po and lib/layouts/*
qt4: qt4 ui files
languages: file lib/languages
+ encodings: file lib/encodings
external: external templates file
formats: formats predefined in lib/configure.py
'''
input_type = None
output = 'lyx.pot'
base = '.'
+ input_files = []
#
- optlist, args = getopt.getopt(sys.argv[1:], 'ht:o:b:',
- ['help', 'type=', 'output=', 'base='])
+ optlist, args = getopt.getopt(sys.argv[1:], 'ht:o:b:s:',
+ ['help', 'type=', 'output=', 'base=', 'src_file='])
for (opt, value) in optlist:
if opt in ['-h', '--help']:
print Usage
base = value
elif opt in ['-t', '--type']:
input_type = value
- if input_type not in ['ui', 'layouts', 'modules', 'qt4', 'languages', 'external', 'formats'] or output is None:
+ elif opt in ['-s', '--src_file']:
+ input_files = [f.strip() for f in open(value)]
+
+ if input_type not in ['ui', 'layouts', 'layouttranslations', 'qt4', 'languages', 'encodings', 'external', 'formats'] or output is None:
print 'Wrong input type or output filename.'
sys.exit(1)
+
+ input_files += args
+
if input_type == 'ui':
- ui_l10n(args, output, base)
+ ui_l10n(input_files, output, base)
elif input_type == 'layouts':
- layouts_l10n(args, output, base)
+ layouts_l10n(input_files, output, base, False)
+ elif input_type == 'layouttranslations':
+ layouts_l10n(input_files, output, base, True)
elif input_type == 'qt4':
- qt4_l10n(args, output, base)
+ qt4_l10n(input_files, output, base)
elif input_type == 'external':
- external_l10n(args, output, base)
+ external_l10n(input_files, output, base)
elif input_type == 'formats':
- formats_l10n(args, output, base)
+ formats_l10n(input_files, output, base)
+ elif input_type == 'encodings':
+ encodings_l10n(input_files, output, base)
else:
- languages_l10n(args, output, base)
+ languages_l10n(input_files, output, base)
+