appypod-rattail/gen/po.py

338 lines
14 KiB
Python
Raw Normal View History

2009-06-29 07:06:01 -05:00
# ------------------------------------------------------------------------------
import os, re, time, copy
from .utils import produceNiceMessage
2009-06-29 07:06:01 -05:00
# ------------------------------------------------------------------------------
poHeader = '''msgid ""
msgstr ""
"Project-Id-Version: %s\\n"
"POT-Creation-Date: %s\\n"
"MIME-Version: 1.0\\n"
"Content-Type: text/plain; charset=utf-8\\n"
"Content-Transfer-Encoding: 8bit\\n"
"Plural-Forms: nplurals=1; plural=0\\n"
"Language-code: %s\\n"
"Language-name: %s\\n"
"Preferred-encodings: utf-8 latin1\\n"
"Domain: %s\\n"
%s
'''
fallbacks = {'en': 'en-us en-ca',
'fr': 'fr-be fr-ca fr-lu fr-mc fr-ch fr-fr'}
2009-06-29 07:06:01 -05:00
# Default values for i18n labels whose ids are not fixed.
CONFIG = "Configuration panel for product '%s'"
CONFIRM = 'Are you sure ?'
2009-06-29 07:06:01 -05:00
# ------------------------------------------------------------------------------
class PoMessage:
'''Represents a i18n message (po format).'''
def __init__(self, id, msg, default, fuzzy=False, comments=[],
niceDefault=False):
2009-06-29 07:06:01 -05:00
self.id = id
self.msg = msg
self.default = default
if niceDefault: self.produceNiceDefault()
2009-06-29 07:06:01 -05:00
self.fuzzy = fuzzy # True if the default value has changed in the pot
# file: the msg in the po file needs to be translated again.
self.comments = comments
def update(self, newMsg, isPot, language):
'''Updates me with new values from p_newMsg. If p_isPot is True (which
means that the current file is a pot file), I do not care about
filling self.msg.'''
if isPot:
self.msg = ""
if not self.default:
self.default = newMsg.default
# It means that if the default message has changed, we will not
# update it in the pot file. We will always keep the one that
# the user may have changed in the pot file. We will write a
# default message only when no default message is defined.
else:
# newMsg comes from a pot file or from a base po file (like a
# standard Appy po file). We must update the corresponding
2009-06-29 07:06:01 -05:00
# message in the current po file.
oldDefault = self.default
if self.default != newMsg.default:
# The default value has changed in the pot file
oldDefault = self.default
2009-06-29 07:06:01 -05:00
self.default = newMsg.default
self.fuzzy = False
2009-06-29 07:06:01 -05:00
if self.msg.strip():
self.fuzzy = True
# We mark the message as "fuzzy" (=may need to be rewritten
# because the default value has changed) only if the user
# has already entered a message. Else, this has no sense to
# rewrite the empty message.
if not oldDefault.strip():
# This is a strange case: the old default value did not
# exist. Maybe was this PO file generated from some
# tool, but simply without any default value. So in
# this case, we do not consider the label as fuzzy.
self.fuzzy = False
# If p_newMsg contains a message, and no message is defined for
# self, copy it.
if newMsg.msg and not self.msg:
self.msg = newMsg.msg
# For english, the the default value from a pot file can be used as
# value for the po file.
2009-06-29 07:06:01 -05:00
if (language == 'en'):
if not self.msg:
# Put the default message into msg for english
self.msg = self.default
if self.fuzzy and (self.msg == oldDefault):
# The message was equal to the old default value. It means
# that the user did not change it, because for English we
# fill by default the message with the default value (see
# code just above). So in this case, the message was not
# really fuzzy.
self.fuzzy = False
self.msg = self.default
def produceNiceDefault(self):
'''Transforms self.default into a nice msg.'''
self.default = produceNiceMessage(self.default)
def generate(self):
'''Produces myself as I must appear in a po(t) file.'''
res = ''
for comment in self.comments:
res += comment + '\n'
if self.default != None:
res = '#. Default: "%s"\n' % self.default
if self.fuzzy:
res += '#, fuzzy\n'
res += 'msgid "%s"\n' % self.id
res += 'msgstr "%s"\n' % self.msg
return res
def __repr__(self):
return '<i18n msg id="%s", msg="%s", default="%s">' % \
(self.id, self.msg, self.default)
def getMessage(self):
'''Returns self.msg, but with some replacements.'''
return self.msg.replace('<br/>', '\n').replace('\\"', '"')
class PoMessages:
'''A list of po messages under construction.'''
def __init__(self):
# The list of messages
self.messages = []
# A dict of message ids, useful for efficiently checking if an id is
# already in the list or not.
self.ids = {}
def append(self, id, default, nice=True):
'''Creates a new PoMessage and adds it to self.messages. If p_nice is
True, it produces a nice default value for the message.'''
# Avoir creating duplicate ids
if id in self.ids: return
message = PoMessage(id, '', default, niceDefault=nice)
self.messages.append(message)
self.ids[id] = True
def get(self): return self.messages
2009-06-29 07:06:01 -05:00
class PoHeader:
def __init__(self, name, value):
self.name = name
self.value = value
def generate(self):
'''Generates the representation of myself into a po(t) file.'''
return '"%s: %s\\n"\n' % (self.name, self.value)
class PoFile:
'''Represents a i18n file.'''
def __init__(self, fileName):
self.fileName = fileName
self.isPot = fileName.endswith('.pot')
self.messages = [] # Ordered list of messages
self.messagesDict = {} # Dict of the same messages, indexed by msgid
self.headers = []
self.headersDict = {}
# Get application name, domain name and language from fileName
self.applicationName = ''
self.language = ''
self.domain = ''
baseName = os.path.splitext(os.path.basename(fileName))[0]
elems = baseName.split('-')
if self.isPot:
if len(elems) == 1:
self.applicationName = self.domain = elems[0]
2009-06-29 07:06:01 -05:00
else:
self.applicationName, self.domain = elems
2009-06-29 07:06:01 -05:00
else:
if len(elems) == 1:
self.applicationName = self.domain = ''
self.language = elems[0]
elif len(elems) == 2:
self.applicationName = self.domain = elems[0]
2009-06-29 07:06:01 -05:00
self.language = elems[1]
else:
self.applicationName, self.domain, self.language = elems
self.generated = False # Will be True during the generation process,
# once this file will have been generated.
2009-06-29 07:06:01 -05:00
def addMessage(self, newMsg, needsCopy=True):
if needsCopy:
res = copy.copy(newMsg)
else:
res = newMsg
2009-06-29 07:06:01 -05:00
self.messages.append(res)
self.messagesDict[res.id] = res
return res
def addHeader(self, newHeader):
self.headers.append(newHeader)
self.headersDict[newHeader.name] = newHeader
def update(self, newMessages, removeNotNewMessages=False,
keepExistingOrder=True):
'''Updates the existing messages with p_newMessages.
If p_removeNotNewMessages is True, all messages in self.messages
that are not in newMessages will be removed, excepted if they start
with "custom_". If p_keepExistingOrder is False, self.messages will
be sorted according to p_newMessages. Else, newMessages that are not
yet in self.messages will be appended to the end of self.messages.'''
2009-06-29 07:06:01 -05:00
# First, remove not new messages if necessary
newIds = [m.id for m in newMessages]
removedIds = []
if removeNotNewMessages:
i = len(self.messages)-1
while i >= 0:
oldId = self.messages[i].id
if not oldId.startswith('custom_') and (oldId not in newIds):
2009-06-29 07:06:01 -05:00
del self.messages[i]
del self.messagesDict[oldId]
removedIds.append(oldId)
i -= 1
if keepExistingOrder:
# Update existing messages and add inexistent messages to the end.
for newMsg in newMessages:
if newMsg.id in self.messagesDict:
2009-06-29 07:06:01 -05:00
msg = self.messagesDict[newMsg.id]
else:
msg = self.addMessage(newMsg)
msg.update(newMsg, self.isPot, self.language)
else:
# Keep the list of all old messages not being in new messages.
# We will append them at the end of the new messages.
notNewMessages = [m for m in self.messages if m.id not in newIds]
del self.messages[:]
for newMsg in newMessages:
if newMsg.id in self.messagesDict:
2009-06-29 07:06:01 -05:00
msg = self.messagesDict[newMsg.id]
self.messages.append(msg)
else:
msg = self.addMessage(newMsg)
msg.update(newMsg, self.isPot, self.language)
# Append the list of old messages to the end
self.messages += notNewMessages
return removedIds
def generateHeaders(self, f):
if not self.headers:
creationTime = time.strftime("%Y-%m-%d %H:%M-%S", time.localtime())
fb = ''
if not self.isPot:
# I must add fallbacks
if self.language in fallbacks:
2009-06-29 07:06:01 -05:00
fb = '"X-is-fallback-for: %s\\n"' % fallbacks[self.language]
f.write(poHeader % (self.applicationName, creationTime,
self.language, self.language, self.domain, fb))
else:
# Some headers were already found, we dump them as is
f.write('msgid ""\nmsgstr ""\n')
for header in self.headers:
f.write(header.generate())
f.write('\n')
def generate(self):
'''Generates the corresponding po or pot file.'''
folderName = os.path.dirname(self.fileName)
if not os.path.exists(folderName):
os.makedirs(folderName)
f = file(self.fileName, 'w')
self.generateHeaders(f)
for msg in self.messages:
f.write(msg.generate())
f.write('\n')
f.close()
self.generated = True
def getPoFileName(self, language):
'''Gets the name of the po file that corresponds to this pot file and
the given p_language.'''
if self.applicationName == self.domain:
res = '%s-%s.po' % (self.applicationName, language)
else:
res = '%s-%s-%s.po' % (self.applicationName, self.domain, language)
return res
def getCustomMessages(self):
'''Returns, the list of messages from self.messages whose ID starts with
"custom_".'''
return [m for m in self.messages if m.id.startswith('custom_')]
2009-06-29 07:06:01 -05:00
class PoParser:
'''Allows to parse a i18n file. The result is produced in self.res as a
PoFile instance.'''
def __init__(self, fileName):
self.res = PoFile(fileName)
# Regular expressions for msgIds, msgStrs and default values.
re_default = re.compile('#\.\s+Default\s*:\s*"(.*)"')
re_fuzzy = re.compile('#,\s+fuzzy')
re_id = re.compile('msgid\s+"(.*)"')
re_msg = re.compile('msgstr\s+"(.*)"')
def parse(self):
'''Parses all i18n messages in the file, stores it in
self.res.messages and returns self.res.'''
2009-06-29 07:06:01 -05:00
f = file(self.res.fileName)
# Currently parsed values
msgDefault = msgFuzzy = msgId = msgStr = None
comments = []
# Walk every line of the po(t) file
for line in f:
lineContent = line.strip()
if lineContent and (not lineContent.startswith('"')):
r = self.re_default.match(lineContent)
if r:
msgDefault = r.group(1)
else:
r = self.re_fuzzy.match(lineContent)
if r:
msgFuzzy = True
else:
r = self.re_id.match(lineContent)
if r:
msgId = r.group(1)
else:
r = self.re_msg.match(lineContent)
if r:
msgStr = r.group(1)
else:
if lineContent.startswith('#'):
comments.append(lineContent.strip())
if msgStr != None:
if not ((msgId == '') and (msgStr == '')):
poMsg = PoMessage(msgId, msgStr, msgDefault, msgFuzzy,
comments)
self.res.addMessage(poMsg)
msgDefault = msgFuzzy = msgId = msgStr = None
comments = []
if lineContent.startswith('"'):
# It is a header value
name, value = lineContent.strip('"').split(':', 1)
if value.endswith('\\n'):
value = value[:-2]
self.res.addHeader(PoHeader(name.strip(), value.strip()))
f.close()
return self.res
# ------------------------------------------------------------------------------