shithub: choc

Info  •  Files  •  Log  •  Branches

man/{docgen,simplecpp}: Force UTF-8 reading and writing regardless of locale

When encountering non-ASCII UTF-8 characters in source files (as
exists in Crispy Doom), the Python scripts would fail to work with
a UnicodeDecodeError if the locale was not a UTF-8 one (eg, the C
locale as encountered in chroot build environments).

Let's just use UTF-8 on both open() and replace print() statements with
sys.stdout.buffer.write().  This guarantees that all input and output is
UTF-8 regardless.

--- a/man/docgen

+++ b/man/docgen

@@ -300,7 +300,7 @@

 # Read list of wiki pages

 def read_wikipages():

-    f = open("wikipages")

+    f = open("wikipages", encoding='UTF-8')

     try:

         for line in f:

@@ -356,7 +356,7 @@

     current_config_file = None

-    f = open(file)

+    f = open(file, encoding='UTF-8')

     try:

         param = None

@@ -422,7 +422,7 @@

         process_file(path)

 def print_template(template_file, content):

-    f = open(template_file)

+    f = open(template_file, encoding='UTF-8')

     try:

         for line in f:

@@ -432,7 +432,7 @@

                 print_template(filename, content)

             else:

                 line = line.replace("@content", content)

-                print(line.rstrip())

+                sys.stdout.buffer.write(line.rstrip().encode('UTF-8'))

     finally:

         f.close()

@@ -452,7 +452,7 @@

     read_wikipages()

     for t in targets:

-        print(t.wiki_output())

+        sys.stdout.buffer.write(t.wiki_output().encode('UTF-8'))

 def plaintext_output(targets, template_file):

--- a/man/simplecpp

+++ b/man/simplecpp

@@ -76,7 +76,7 @@

 		raise Exception("Mismatched #if in '%s'" % stream.name)

 def parse_file(filename):

-	f = open(filename)

+	f = open(filename, encoding='UTF-8')

 	try:

 		parse_stream(f)