1

#!/usr/bin/env python

     2

# -*- coding: utf-8 -*-

     3

     4

# This file is part of Beremiz

     5

# Copyright (C) 2021: Edouard TISSERANT

     6

#

     7

# See COPYING file for copyrights details.

     8

     9

from __future__ import absolute_import

    10

from lxml import etree

    11

import os

    12

import sys

    13

import subprocess

    14

import time

    15

import ast

    16

import wx

    17

import re

    18

    19

# to have it for python 2, had to install 

    20

# https://pypi.org/project/pycountry/18.12.8/

    21

# python2 -m pip install pycountry==18.12.8 --user

    22

import pycountry

    23

    24

cmd_parser = re.compile(r'(?:"([^"]+)"\s*|([^\s]+)\s*)?')

    25

    26

def open_pofile(pofile):

    27

    """ Opens PO file with POEdit """

    28

    29

    if sys.platform.startswith('win'):

    30

        from six.moves import winreg

    31

        poedit_cmd = None

    32

        try:

    33

            poedit_cmd = winreg.QueryValue(winreg.HKEY_LOCAL_MACHINE,

    34

                                           'SOFTWARE\\Classes\\poedit\\shell\\open\\command')

    35

            cmd = re.findall(cmd_parser, poedit_cmd)

    36

            dblquote_value,smpl_value = cmd[0]

    37

            poedit_path = dblquote_value+smpl_value

    38

        except OSError:

    39

            poedit_path = None

    40

    41

    else:

    42

        try:

    43

            poedit_path = subprocess.check_output("command -v poedit", shell=True).strip()

    44

        except subprocess.CalledProcessError:

    45

            poedit_path = None

    46

    47

    if poedit_path is None:

    48

        wx.MessageBox("POEdit is not found or installed !")

    49

    else:

    50

        subprocess.Popen([poedit_path,pofile])

    51

    52

def EtreeToMessages(msgs):

    53

    """ Converts XML tree from 'extract_i18n' templates into a list of tuples """

    54

    messages = []

    55

    56

    for msg in msgs:

    57

        messages.append((

    58

            "\n".join([line.text for line in msg]),

    59

            msg.get("label"), msg.get("id")))

    60

    61

    return messages

    62

    63

def SaveCatalog(fname, messages):

    64

    """ Save messages given as list of tupple (msg,label,id) in POT file """

    65

    w = POTWriter()

    66

    w.ImportMessages(messages)

    67

    68

    with open(fname, 'w') as POT_file:

    69

        w.write(POT_file)

    70

    71

def ReadTranslations(dirpath):

    72

    """ Read all PO files from a directory and return a list of (langcode, translation_dict) tuples """

    73

    74

    po_files = [fname for fname in os.listdir(dirpath) if fname.endswith(".po")]

    75

    po_files.sort()

    76

    77

    translations = []

    78

    for po_fname in po_files:

    79

        r = POReader()

    80

        with open(os.path.join(dirpath, po_fname), 'r') as PO_file:

    81

            r.read(PO_file)

    82

            translations.append((po_fname[:-3], r.get_messages()))

    83

    return translations

    84

    85

def MatchTranslations(translations, messages, errcallback):

    86

    """

    87

    Matches translations against original message catalog,

    88

    warn about inconsistancies,

    89

    returns list of langs, and a list of (msgid, [translations]) tuples

    90

    """

    91

    translated_messages = []

    92

    broken_lang = set()

    93

    for msgid,label,svgid in messages:

    94

        translated_message = []

    95

        for langcode,translation in translations:

    96

            msg = translation.pop(msgid, None)

    97

            if msg is None:

    98

                broken_lang.add(langcode)

   100

                translated_message.append(msgid)

   101

            else:

   102

                translated_message.append(msg)

   103

        translated_messages.append((msgid,translated_message))

   104

    langs = []

   105

    for langcode,translation in translations:

   106

        try:

   107

            l,c = langcode.split("_")

   108

            language_name = pycountry.languages.get(alpha_2 = l).name

   109

            country_name = pycountry.countries.get(alpha_2 = c).name

   110

            langname = "{} ({})".format(language_name, country_name)

   111

        except:

   112

            try:

   113

                langname = pycountry.languages.get(alpha_2 = langcode).name

   114

            except:

   115

                langname = langcode

   116

   117

        langs.append((langname,langcode))

   118

   119

        broken = False

   120

        for msgid, msg in translation.iteritems():

   121

            broken = True

   122

            errcallback(_('{}: Unused translation "{}":"{}"\n').format(langcode,msgid,msg))

   123

        if broken or langcode in broken_lang:

   125

   126

   127

    return langs,translated_messages

   128

   129

   130

def TranslationToEtree(langs,translated_messages):

   131

   132

    result = etree.Element("translations")

   133

   134

    langsroot = etree.SubElement(result, "langs")

   135

    for name, code in langs:

   136

        langel = etree.SubElement(langsroot, "lang", {"code":code})

   137

        langel.text = name

   138

   139

    msgsroot = etree.SubElement(result, "messages")

   140

    for msgid, msgs in translated_messages:

   141

        msgidel = etree.SubElement(msgsroot, "msgid")

   142

        for msg in msgs:

   143

            msgel = etree.SubElement(msgidel, "msg")

   144

            for line in msg.split("\n"):

   145

                lineel = etree.SubElement(msgel, "line")

   146

                lineel.text = escape(line.encode("utf-8")).decode("utf-8")

   147

   148

    return result

   149

   150

   151

   152

locpfx = '#:svghmi.svg:'

   153

   154

pot_header = '''\

   155

# SOME DESCRIPTIVE TITLE.

   156

# Copyright (C) YEAR ORGANIZATION

   157

# FIRST AUTHOR <EMAIL@ADDRESS>, YEAR.

   158

#

   159

msgid ""

   160

msgstr ""

   161

"Project-Id-Version: PACKAGE VERSION\\n"

   162

"POT-Creation-Date: %(time)s\\n"

   163

"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\\n"

   164

"Last-Translator: FULL NAME <EMAIL@ADDRESS>\\n"

   165

"Language-Team: LANGUAGE <LL@li.org>\\n"

   166

"MIME-Version: 1.0\\n"

   167

"Content-Type: text/plain; charset=UTF-8\\n"

   168

"Content-Transfer-Encoding: 8bit\\n"

   169

"Generated-By: SVGHMI 1.0\\n"

   170

   171

'''

   172

escapes = []

   173

   174

def make_escapes(pass_iso8859):

   175

    global escapes

   176

    escapes = [chr(i) for i in range(256)]

   177

    if pass_iso8859:

   178

        # Allow iso-8859 characters to pass through so that e.g. 'msgid

   179

        # "Höhe"' would result not result in 'msgid "H\366he"'.  Otherwise we

   180

        # escape any character outside the 32..126 range.

   181

        mod = 128

   182

    else:

   183

        mod = 256

   184

    for i in range(mod):

   185

        if not(32 <= i <= 126):

   186

            escapes[i] = "\\%03o" % i

   187

    escapes[ord('\\')] = '\\\\'

   188

    escapes[ord('\t')] = '\\t'

   189

    escapes[ord('\r')] = '\\r'

   190

    escapes[ord('\n')] = '\\n'

   191

    escapes[ord('\"')] = '\\"'

   192

   193

make_escapes(pass_iso8859 = True)

   194

   195

EMPTYSTRING = ''

   196

   197

def escape(s):

   198

    global escapes

   199

    s = list(s)

   200

    for i in range(len(s)):

   201

        s[i] = escapes[ord(s[i])]

   202

    return EMPTYSTRING.join(s)

   203

   204

def normalize(s):

   205

    # This converts the various Python string types into a format that is

   206

    # appropriate for .po files, namely much closer to C style.

   207

    lines = s.split('\n')

   208

    if len(lines) == 1:

   209

        s = '"' + escape(s) + '"'

   210

    else:

   211

        if not lines[-1]:

   212

            del lines[-1]

   213

            lines[-1] = lines[-1] + '\n'

   214

        for i in range(len(lines)):

   215

            lines[i] = escape(lines[i])

   216

        lineterm = '\\n"\n"'

   217

        s = '""\n"' + lineterm.join(lines) + '"'

   218

    return s

   219

   220

   221

class POTWriter:

   222

    def __init__(self):

   223

        self.__messages = {}

   224

   225

    def ImportMessages(self, msgs):

   226

        for  msg, label, svgid in msgs:

   227

            self.addentry(msg.encode("utf-8"), label, svgid)

   228

   229

    def addentry(self, msg, label, svgid):

   230

        entry = (label, svgid)

   231

        self.__messages.setdefault(msg, set()).add(entry)

   232

   233

    def write(self, fp):

   234

        timestamp = time.strftime('%Y-%m-%d %H:%M+%Z')

   235

        print >> fp, pot_header % {'time': timestamp}

   236

        reverse = {}

   237

        for k, v in self.__messages.items():

   238

            keys = list(v)

   239

            keys.sort()

   240

            reverse.setdefault(tuple(keys), []).append((k, v))

   241

        rkeys = reverse.keys()

   242

        rkeys.sort()

   243

        for rkey in rkeys:

   244

            rentries = reverse[rkey]

   245

            rentries.sort()

   246

            for k, v in rentries:

   247

                v = list(v)

   248

                v.sort()

   249

                locline = locpfx

   250

                for label, svgid in v:

   251

                    d = {'label': label, 'svgid': svgid}

   252

                    s = _(' %(label)s:%(svgid)s') % d

   253

                    if len(locline) + len(s) <= 78:

   254

                        locline = locline + s

   255

                    else:

   256

                        print >> fp, locline

   257

                        locline = locpfx + s

   258

                if len(locline) > len(locpfx):

   259

                    print >> fp, locline

   260

                print >> fp, 'msgid', normalize(k)

   261

                print >> fp, 'msgstr ""\n'

   262

   263

   264

class POReader:

   265

    def __init__(self):

   266

        self.__messages = {}

   267

   268

    def get_messages(self):

   269

        return self.__messages

   270

   271

    def add(self, msgid, msgstr, fuzzy):

   272

        "Add a non-fuzzy translation to the dictionary."

   273

        if not fuzzy and msgstr and msgid:

   274

            self.__messages[msgid.decode('utf-8')] = msgstr.decode('utf-8')

   275

   276

    def read(self, fp):

   277

        ID = 1

   278

        STR = 2

   279

   280

        lines = fp.readlines()

   281

        section = None

   282

        fuzzy = 0

   283

   284

        # Parse the catalog

   285

        lno = 0

   286

        for l in lines:

   287

            lno += 1

   288

            # If we get a comment line after a msgstr, this is a new entry

   289

            if l[0] == '#' and section == STR:

   290

                self.add(msgid, msgstr, fuzzy)

   291

                section = None

   292

                fuzzy = 0

   293

            # Record a fuzzy mark

   294

            if l[:2] == '#,' and 'fuzzy' in l:

   295

                fuzzy = 1

   296

            # Skip comments

   297

            if l[0] == '#':

   298

                continue

   299

            # Now we are in a msgid section, output previous section

   300

            if l.startswith('msgid') and not l.startswith('msgid_plural'):

   301

                if section == STR:

   302

                    self.add(msgid, msgstr, fuzzy)

   303

                section = ID

   304

                l = l[5:]

   305

                msgid = msgstr = ''

   306

                is_plural = False

   307

            # This is a message with plural forms

   308

            elif l.startswith('msgid_plural'):

   309

                if section != ID:

   310

                    print >> sys.stderr, 'msgid_plural not preceded by msgid on %s:%d' %\

   311

                        (infile, lno)

   312

                    sys.exit(1)

   313

                l = l[12:]

   314

                msgid += '\0' # separator of singular and plural

   315

                is_plural = True

   316

            # Now we are in a msgstr section

   317

            elif l.startswith('msgstr'):

   318

                section = STR

   319

                if l.startswith('msgstr['):

   320

                    if not is_plural:

   321

                        print >> sys.stderr, 'plural without msgid_plural on %s:%d' %\

   322

                            (infile, lno)

   323

                        sys.exit(1)

   324

                    l = l.split(']', 1)[1]

   325

                    if msgstr:

   326

                        msgstr += '\0' # Separator of the various plural forms

   327

                else:

   328

                    if is_plural:

   329

                        print >> sys.stderr, 'indexed msgstr required for plural on  %s:%d' %\

   330

                            (infile, lno)

   331

                        sys.exit(1)

   332

                    l = l[6:]

   333

            # Skip empty lines

   334

            l = l.strip()

   335

            if not l:

   336

                continue

   337

            l = ast.literal_eval(l)

   338

            if section == ID:

   339

                msgid += l

   340

            elif section == STR:

   341

                msgstr += l

   342

            else:

   343

                print >> sys.stderr, 'Syntax error on %s:%d' % (infile, lno), \

   344

                      'before:'

   345

                print >> sys.stderr, l

   346

                sys.exit(1)

   347

        # Add last entry

   348

        if section == STR:

   349

            self.add(msgid, msgstr, fuzzy)

   350

   351