Read comment_char and escape_char from source file

Fixes #4

Author: tersmitten

bin/locale-extract-category

@@ -22,13 +22,15 @@ lines = ''
 for line in sys.stdin:
   lines += line
 
-match = re.search(lc_section_pattern, lines, re.DOTALL)
-
-# Lines below should be matched from lines too
-sys.stdout.write('comment_char %\n')
-sys.stdout.write('escape_char  /\n')
-
-if match:
-  sys.stdout.write(match.group(1).lstrip('\n'))
+match_comment_char = re.search(COMMENT_CHAR_PATTERN, lines, re.MULTILINE)
+match_escape_char = re.search(ESCAPE_CHAR_PATTERN, lines, re.MULTILINE)
+match_category = re.search(lc_section_pattern, lines, re.DOTALL)
+
+if match_comment_char:
+  sys.stdout.write(match_comment_char.group(1) + '\n')
+if match_escape_char:
+  sys.stdout.write(match_escape_char.group(1) + '\n')
+if match_category:
+  sys.stdout.write(match_category.group(1).lstrip('\n'))
 
 sys.exit(0)

glibc_locale_tools/glibc_locale_tools.py

@@ -26,6 +26,16 @@
 A re pattern to match a LC_* section (category, e.g. LC_TIME) in a locale file (e.g. nl_NL).
 """
 
+COMMENT_CHAR_PATTERN = r'^(comment_char\s+(.*))$'
+"""
+A re pattern to match a comment_char line.
+"""
+
+ESCAPE_CHAR_PATTERN = r'^(escape_char\s+(.*))$'
+"""
+A re pattern to match a escape_char line.
+"""
+
 
 def unicode_decode(unicode_char):
   """

glibc_locale_tools/test/data/da_DK.LC_TIME.expected

@@ -1,5 +1,5 @@
 comment_char %
-escape_char  /
+escape_char /
 abday       "<U0073><U00F8><U006E>";"<U006D><U0061><U006E>";/
             "<U0074><U0069><U0072>";"<U006F><U006E><U0073>";/
             "<U0074><U006F><U0072>";"<U0066><U0072><U0065>";/

glibc_locale_tools/test/data/en_US.LC_TIME.expected

@@ -1,5 +1,5 @@
 comment_char %
-escape_char  /
+escape_char /
 abday	"<U0053><U0075><U006E>";"<U004D><U006F><U006E>";/
 	"<U0054><U0075><U0065>";"<U0057><U0065><U0064>";/
 	"<U0054><U0068><U0075>";"<U0046><U0072><U0069>";/