Fix for gedit.

rtv/helpers.py

@@ -3,25 +3,36 @@ import os
 import textwrap
 import subprocess
 from datetime import datetime
+from tempfile import NamedTemporaryFile
 
 from . import config
 
 __all__ = ['open_browser', 'clean', 'wrap_text', 'strip_textpad',
-           'strip_subreddit_url', 'humanize_timestamp']
+           'strip_subreddit_url', 'humanize_timestamp', 'open_editor']
 
-def open_editor(filename):
+def open_editor(data=''):
     """
-    Open the given file using the system's default editor.
+    Open a temporary file using the system's default editor.
+
+    The data string will be written to the file before opening. This function
+    will block until the editor has closed. At that point the file will be
+    read and and lines starting with '#' will be stripped.
     """
 
-    editors = [os.getenv('RTV_EDITOR'), os.getenv('EDITOR'), 'nano', 'vi']
-    for program in editors:
-        if program:
-            try:
-                subprocess.Popen([program, filename]).wait()
-                break
-            except OSError:
-                pass
+    with NamedTemporaryFile(prefix='rtv-', suffix='.txt', mode='w') as fp:
+        fp.write(data)
+        fp.flush()
+        editor = os.getenv('RTV_EDITOR') or os.getenv('EDITOR') or 'nano'
+        subprocess.Popen([editor, fp.name]).wait()
+
+        # Open a second file object to read. This appears to be necessary in
+        # order to read the changes made by some editors (gedit). w+ mode does
+        # not work!
+        with open(fp.name) as fp2:
+            text = ''.join(line for line in fp2 if not line.startswith('#'))
+            text = text.rstrip()
+
+    return text
 
 def open_browser(url):
     """

rtv/submission.py

@@ -1,7 +1,6 @@
 import curses
 import sys
 import time
-from tempfile import NamedTemporaryFile
 
 import praw.errors
 
@@ -126,26 +125,18 @@ class SubmissionPage(BasePage):
 
         # Comment out every line of the content
         content = '\n'.join(['# |' + line for line in content.split('\n')])
-        info = {'author': data['author'],
-                'type': data['type'],
-                'content': content}
-        comment_info = COMMENT_FILE.format(**info)
+        comment_info = COMMENT_FILE.format(
+            author=data['author'],
+            type=data['type'].lower(),
+            content=content)
 
         curses.endwin()
-        with NamedTemporaryFile(prefix='rtv-comment-', mode='w+') as fp:
-            fp.write(comment_info)
-            fp.flush()
-            open_editor(fp.name)  # Open the default text editor and block
-            fp.seek(0)
-            comment_text = '\n'.join(line for line in fp.read().split('\n')
-                                     if not line.startswith("#"))
-            comment_text = comment_text.rstrip()
+        comment_text = open_editor(comment_info)
         curses.doupdate()
 
         if not comment_text:
             curses.flash()
             return
-
         try:
             if data['type'] == 'Submission':
                 data['object'].add_comment(comment_text)