docgen: Don't use stdin and stdout

The encoding on stdin and stdout are usually ascii and may cause problems
with writing unicode to it. Instead of this we open the file directly
and avoid dealing with annoying unicode encode/decode errors.

2 files changed, 21 insertions, 12 deletions

diff --git a/Makefile.am b/Makefile.am
index 8985128..e420332 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -10,7 +10,7 @@ caja-dropbox: caja-dropbox.in serializeimages.py
 	chmod +x caja-dropbox
 
 caja-dropbox.1: caja-dropbox.txt.in caja-dropbox docgen.py
-	python docgen.py $(PACKAGE_VERSION) < caja-dropbox.txt.in > caja-dropbox.txt
+	python docgen.py $(PACKAGE_VERSION) caja-dropbox.txt.in caja-dropbox.txt
 	$(RST2MAN) caja-dropbox.txt > caja-dropbox.1
 
 SUBDIRS = data src
diff --git a/docgen.py b/docgen.py
index 2ed8ad9..e1afa17 100644
--- a/docgen.py
+++ b/docgen.py
@@ -1,25 +1,34 @@
+from __future__ import unicode_literals
 import sys
 import datetime
+import codecs
 
 # heeeheee
 env = {"__name__":"__notmain__"}
 execfile("caja-dropbox", env)
 commands = env["commands"]
 
-f = open("AUTHORS", "r")
-authors = '| ' + f.read().replace('\n', '\n| ')
-f.close()
+with codecs.open("AUTHORS", "r", "utf-8") as afile:
+    authors = '| ' + afile.read().replace('\n', '\n| ')
+
+with codecs.open(sys.argv[2], "r", encoding="utf-8") as infile:
+    instr = infile.read()
 
 formatted_commands = ""
 for cmd in commands:
     split = commands[cmd].__doc__.split('\n', 2)
-    formatted_commands += split[1].decode('ascii').replace(cmd, "`%s`" % cmd).replace("dropbox", "``dropbox``")
-    formatted_commands += split[2].decode('ascii').replace('\n', '\n  | ')
+    formatted_commands += split[1].replace(cmd, "`%s`" % cmd).replace("dropbox", "``dropbox``")
+    formatted_commands += split[2].replace('\n', '\n  | ')
     formatted_commands += '\n\n'
 
-sys.stdout.write(sys.stdin.read().replace\
-                     ('@AUTHORS@', authors).replace\
-                     ('@DATE@', datetime.date.today().isoformat()).replace\
-                     ('@PACKAGE_VERSION@', sys.argv[1]).replace\
-                     ('@SYNOPSIS@', '| '+'\n| '.join(commands[cmd].__doc__.split('\n', 2)[1].decode('ascii').replace(cmd, "`%s`" % cmd).replace("dropbox", "``dropbox``") for cmd in commands)).replace\
-                     ('@COMMANDS@', formatted_commands))
+replace = {'@AUTHORS@': authors,
+           '@DATE@': datetime.date.today().isoformat(),
+           '@PACKAGE_VERSION@': sys.argv[1],
+           '@SYNOPSIS@':  '| '+'\n| '.join(commands[cmd].__doc__.split('\n', 2)[1].replace(cmd, "`%s`" % cmd).replace("dropbox", "``dropbox``") for cmd in commands),
+           '@COMMANDS@': formatted_commands}
+
+for r in replace.keys():
+    instr = instr.replace(r, replace[r])
+
+with codecs.open(sys.argv[3], "w", encoding="utf-8") as outfile:
+    outfile.write(instr)