aurweb/web/lang/genpopo

#! /usr/bin/python2 -O
# -*- coding: iso-8859-1 -*-

# This script iterates through the script directories
# looking for php scripts that contain  __() functions.
# It creates/appends to the corresponding
# "xxx.po" file in the 'lang' subdirectory and places the
# i18n strings into the file in the proper format.
#
# usage: genpopo [-v] [-f]
#        -v: verbose, print duplicate terms that could be moved to common_po
#        -f: force, overwrite existing translated files, otherwise append
#

import re
import os
import sys

INC_HEADER = """\
<?php
# INSTRUCTIONS TO TRANSLATORS
#
# This file contains the i18n translations for a subset of the
# Arch Linux User Community Repository (AUR).  This is a PHP
# script, and as such, you MUST pay great attention to the syntax.
# If your text contains any double-quotes ("), you MUST escape
# them with a backslash (\).
#

global $_t;
"""

language = 'en'
lang = {}

print_dupes = '-v' in sys.argv
force = '-f' in sys.argv

up = re.compile('_\(\s*"(([^"]|(?<=\\\\)["])+)"')

scriptdirs = ['html', 'lib', 'template']
pofile = '%s.po' % language

current_dir = os.getcwd()

# Iterate through various places where the php files might be.
#
for dir in scriptdirs:
	dir = "../%s" % dir

	if os.path.exists(dir):
		# Find all the PHP files in the current directory.
		#
		files = [x for x in os.listdir(dir)
				if (x[-4:] == '.inc' and x[-7:] != '.po')
				or x[-6:] == '.class'
				or x[-4:] == '.php'
				or x[-6:] == '.phtml'
			]
		os.chdir(dir)

		for file in files:
			f = open(file,'r')
			lines = f.readlines()
			f.close()

			# Parse the file
			print "Parsing %s..." % file
			for line in lines:
				match = up.search(line)
				while match:
					term = match.group(1).replace('\\"','"')
					if print_dupes:
						if term in lang.keys():
							print 'Multiple use of "%s"' % term

					lang[term] = 1
					line = line[match.end(1):]
					match = up.search(line)

		os.chdir(current_dir)

# Generate the .po file if it doesn't already exist.
# If it does exist, only append new stuff to the end.
# If the 'force' option is passed, just overwrite.

if force:
	# Just overwrite any existing files
	print "Generating %s..." % pofile
	
	f = open(pofile,'w')
	f.write(INC_HEADER)

	for term in lang.keys():
		f.write("\n")
		f.write('$_t["%s"]\n  = "%s";\n' % (term, term))

	f.write("\n");
	f.close()
else:
	# Need to leave existing file intact. Only append on new terms.
	mapre = re.compile('^\$_t\["(.*)"\].*$')
	got_match = False

	print "Updating %s..." % pofile

	try:
		f = open(pofile, 'r')
		new_file = 0
	except:
		new_file = 1

	if not new_file:
		contents = f.readlines()
		f.close()

		# Strip beginning/ending empty lines
		while contents[0] == '':
			del contents[0]
		while contents[-1] in ['', "\n", "?>", "?>\n", "\n?>"]:
			del contents[-1]
			
		f = open(pofile,'w')
		f.write("".join(contents))
		f.write("\n");
		f.close()
	else:
		f = open(pofile,'w')
		f.write(INC_HEADER)
		f.write('\n')
		f.close()

	# Read file contents so we can hash what already exists
	try:
		f = open(pofile, 'r')
		new_file = 0
	except:
		new_file = 1

	existing_terms = []
	if not new_file:
		contents = f.readlines()
		f.close()

		# Strip beginning/ending empty lines
		while contents[0] == '':
			del contents[0]
		while contents[-1] in ['', "\n", "?>", "?>\n", "\n?>"]:
			del contents[-1]

		# Collect existing terms
		for line in contents:
			match = mapre.search(line)
			if match:
				existing_terms.append(match.group(1))

	# Append any new terms to EOF
	f = open(pofile, 'w')
	if not new_file:
		f.write("".join(contents))
	else:
		f.write(INC_HEADER)

	for term in lang.keys():
		if term not in existing_terms:
			f.write("\n");
			f.write('$_t["%s"]\n  = "%s";\n' % (term, term))
	f.write("\n");
	f.close()