aurweb/web/utils/genpopo

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

# this script iterates through the 'html' and 'lib' directories
# looking for php scripts that contain a include_once("xxx_po.inc")
# line and __() functions.  It creates/appends to the corresponding
# "xxx_po.inc" 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
#

INC_HEADER = """\
<?
# 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 the backslash character (\).
#

include_once("translator.inc");
global $_t;
"""


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

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

lang = { 'common_po.inc': {} }

current_dir = os.getcwd()

# Find the common_po.inc file.
#
common = {}
for dir in ['../lang', 'lang']:
	if os.path.exists(dir):
		os.chdir(dir)
		if os.path.exists('common_po.list'):
			f = open('common_po.list','r')
			lines = f.readlines()
			f.close()
			for line in lines:
				if line[0] != '#':
					common[line[:-1]] = 0
					lang['common_po.inc'][line[:-1]] = 1
			os.chdir(current_dir)
			break
		os.chdir(current_dir)
else:
	print "Can't find common_po.list file."
	raise SystemExit

# Find the lang directory.
#
for dir in ['../lang', 'lang']:
	if os.path.exists(dir):
		lang_dir = dir
		break
else:
	print "Can't find the lang directory."
	raise SystemExit

# Iterate through various places where the php files might be.
#
for dir in ['../html', '../lib', 'html', 'lib']:

	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.inc')
					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()

			# Is this file one we need to parse for internationalized strings?
			#
			parse_file = 0
			for line in lines:
				match = re.search("include(_once|)\s*\(\s*[\"']([A-Za-z_]+_po.inc)[\"']\s*\);",line)
				if match and match.group(2) != "common_po.inc":
					po = match.group(2)
					if not lang.has_key(po):
						lang[po] = {}
					parse_file = 1
					break

			# If we need to parse the file, do so.
			#
			if parse_file:
				print "Parsing %s..." % file
				for line in lines:
					match = up.search(line)
					while match:
						term = match.group(1).replace('\\"','"')
						if common.has_key(term):
							common[term] += 1
						else:
							if print_dupes:
								for key in lang.keys():
									if key != po and lang[key].has_key(term):
										print "...Duplicate term: \"%s\" is also in %s." % (term,key)
							lang[po][term] = 1
						line = line[match.end(1):]
						match = up.search(line)

		os.chdir(current_dir)

# Now generate all the .inc files if they don't already exist.
# if they do exist, only append new stuff to the end.  If the 'force'
# option is passed, just overwrite the entire thing.
#
os.chdir(lang_dir)
if force:
	# just going to overwrite any existing files
	#
	for po in lang.keys():
		print "Generating %s..." % po

		f = open(po,'w')
		f.write(INC_HEADER)

		for term in lang[po].keys():
			f.write("\n")
			f.write('$_t["en"]["%s"] = "%s";\n' % (term, term))
			f.write('# $_t["es"]["%s"] = "--> Traducci<63>n espa<70>ola aqu<71>. <--";\n' % term)
			f.write('# $_t["fr"]["%s"] = "--> Traduction fran<61>aise ici. <--";\n' % term)
			f.write('# $_t["de"]["%s"] = "--> Deutsche <20>bersetzung hier. <--";\n' % term)

		f.write("\n");
		f.write("?>");
		f.close()
else:
	# need to leave existing file intact, and only append on terms that are new
	#
	mapre = re.compile('^\$_t\["en"\]\["(.*)"\].*$')
	for po in lang.keys():
		print "Updating %s..." % po
		# first read in file contents so we can hash what already exists
		#
		try:
			f = open(po, 'r')
			new_file = 0
		except:
			new_file = 1

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

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

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

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

		for term in lang[po].keys():
			if term not in existing_terms:
				f.write("\n");
				f.write('$_t["en"]["%s"] = "%s";\n' % (term, term))
				f.write('# $_t["es"]["%s"] = "--> Traducci<63>n espa<70>ola aqu<71>. <--";\n' % term)
				f.write('# $_t["fr"]["%s"] = "--> Traduction fran<61>aise ici. <--";\n' % term)
				f.write('# $_t["de"]["%s"] = "--> Deutsche <20>bersetzung hier. <--";\n' % term)
		f.write("\n?>");
		f.close()

# Print out warnings for unused and little-used common entries.
#
for key in common.keys():
	if common[key] == 1:
		print "Warning: common entry '%s' is only used once." % key
for key in common.keys():
	if common[key] == 0:
		print "Warning: unused common entry '%s'." % key

# vim: ts=2 sw=2 noet ft=python