sync-localization-compat.py

#
# (c) Morthimer McMare  a.k.a. JSO_x, 2023-2026
#
# Base files by PROPHESSOR and Morthimer McMare:
#   https://github.com/DRRP-Team/DRRP/blob/master/tools/updateOldLocalization.py
#   https://github.com/LLDM-Doom-Modding/ZChecker/blob/master/SyncLocalization.py
#

import csv
import sys
from pathlib import Path


def escape_incompatible_sequences( text: str ) -> str:
	return text.replace( '"', '\\"' ).replace( '\n', '\\n' ).replace( "ё", "е" ).replace( "—", "-" )


def parse_csv( csvrows ) -> tuple[dict[str, list[str]], list[str]]:
	outparsed = {}
	outheader = []
	header = next( csvrows )
	headeridskip = None

	# Most of default manpages CSV language headers are in format:
	#   "Identifier, Remark, Identifier, default, lang1, lang2, ..."
	#
	# GZDoom will skip first "Identifier" column in this case, so I use this
	# behaviour to add a Remark field before the ID.
	if ( header[ 0 ] == 'Identifier' and 'Identifier' in header[ 1 : ] ):
		headeridskip = 0
		print( f"  Skipped dummy column 'Identifier' at index {headeridskip}." )

	columns = [[] for _ in header]

	# Convert CSV to the list of columns:
	for row in csvrows:
		for col, cell in enumerate( row ):
			if col == headeridskip:
				continue

			#print( col, cell )
			columns[ col ].append( cell )

	if headeridskip is not None:
		outheader = header[ : headeridskip ] + header[ headeridskip + 1 : ]
		columns = columns[ : headeridskip ] + columns[ headeridskip + 1 : ]
		#print( f'outheader: {outheader} (len {len(outheader)})' )
	else:
		outheader = header

	#print( "\n=========" )

	# Convert columns to `dict<str('Header'), list>`:
	for idx, i in enumerate( outheader ):
		#print( f"parsed[ {i} ] = columns[ {idx} ]: {columns[idx]} (len {len(columns[idx])})", end='\n\n' )
		outparsed[ i ] = columns[ idx ]

	return outparsed, outheader
pass # of def parse_csv( csvrows )


csvlangfiles = list( Path( "." ).glob( "LANGUAGE.*.csv" ) )
#csvlangfiles = [ "LANGUAGE.manpages_general.csv" ]

parsedcolumns = {}

for csvfileidx, csvfile in enumerate( csvlangfiles ):
	curparsedcolumns = {}
	curheader = []

	print( f'Parsing CSV file "{csvfile}"...' )

	with open( str(csvfile), mode='r', encoding='utf-8' ) as infile:
		(curparsedcolumns, curheader) = parse_csv( csv.reader( infile ) )

	for i in curheader:
		if not parsedcolumns.get( i ):
			if csvfileidx == 0:
				print( f"  Added new column '{i}'." )
				parsedcolumns[ i ] = list()
			else:
				print( f"  Error: extra column with header '{i}' in file '{csvfile}'." )
				exit( 1 )

	# Adds everything to the global columns:
	for k, v in curparsedcolumns.items():
		parsedcolumns[ k ] += v


if not parsedcolumns.get( 'Identifier' ) or not parsedcolumns.get( 'Remark' ):
	print( "No column 'Identifier' or 'Remark' in parsedcolumns." )
	exit( 1 ) # Wrong format?


compat_filename_remapping = {
	"ru": "rus",
	"default": "enu"
}
compat_headers_remapping = {
	"ru": "ru",
	"default": "enu default"
}
compat_encoding_remapping = {
	"ru": "CP1251"
}

print( "Resulting parsed columns: " + str( [i[0] for i in parsedcolumns.items()] ) )

for column in parsedcolumns.items():
	colname = column[ 0 ]

	if colname == 'Identifier' or colname == 'Remark':
		print( "  Skipped non-language column '" + colname + "'." )
		continue

	# Configuring output file:
	compatfilesuffix = compat_filename_remapping.get( colname, colname )
	compatfileheader = compat_headers_remapping.get( colname, colname )
	compatfileencoding = compat_encoding_remapping.get( colname, 'CP1252' )

	compatfilename = "LANGUAGE.gzdoom330." + compatfilesuffix
	print( f'  Creating LANGUAGE lump "{compatfilename}" for \'{colname}\' with encoding "{compatfileencoding}" and header [{compatfileheader}].' )

	# Write to current output language file:
	with open( compatfilename, mode='w', encoding=compatfileencoding ) as outfile:
		#if lfprint is not None:
		#	del lfprint
		def lwrite( *args, sep=' ', end='\n' ):
			outfile.write( sep.join( map( str, args ) ) + end )

		# Write headers:
		lwrite( f"// This file is automatically generated by {sys.argv[0]}." )
		lwrite( "// Update localization lines in \"LANGUAGE.*.csv\" instead.", end = '\n\n' )
		lwrite( "[" + compatfileheader + "]\n\n" )

		#print( "\nparsedcolumns['Identifier'], column, parsedcolumns['Remark'] ==", len(parsedcolumns['Identifier']), len(column), len(parsedcolumns['Remark']), '\n' )

		for (lid, text, remark) in zip( parsedcolumns[ 'Identifier' ], column[ 1 ], parsedcolumns[ 'Remark' ], strict = True ):
			if not lid:
				if remark:
					lwrite( "// %s" % (remark) )
				else:
					lwrite()

			else:
				filteredtext = escape_incompatible_sequences( text )
				lwrite( "%-37s = \"%s\";" % (lid, filteredtext), end = f' // {remark}\n' if remark else '\n' )


print( "Updated successfully." )