Add a post-processing step for <tt> and <code> tags

These tags normally get converted to ` which doesn't work when
there are HTML tags inside. In that case, we need to avoid using
the markdown version and output HTML tags directly.

This solution is modelled after the <pre> hack from
9d4c019.

blogger2ikiwiki.py

@@ -123,6 +123,29 @@ def post_process_pre(text):
     return "\n".join(out)
 
 
+def post_process_tt_code(text):
+    out = []
+
+    lines = text.split("\n")
+    in_pre = False
+    pre_lines = []
+    for line in lines:
+        if "<!-- START TT -->" in line:
+            if "<!-- END TT WITHOUT TAGS -->" in line:
+                out.append(line.replace('<!-- START TT -->', '`').replace('<!-- END TT WITHOUT TAGS -->', '`'))
+            elif "<!-- END TT WITH TAGS -->" in line:
+                out.append(line.replace('<!-- START TT -->', '<tt>').replace('<!-- END TT WITH TAGS -->', '</tt>'))
+        elif "<!-- START CODE -->" in line:
+            if "<!-- END CODE WITHOUT TAGS -->" in line:
+                out.append(line.replace('<!-- START CODE -->', '`').replace('<!-- END CODE WITHOUT TAGS -->', '`'))
+            elif "<!-- END CODE WITH TAGS -->" in line:
+                out.append(line.replace('<!-- START CODE -->', '<code>').replace('<!-- END CODE WITH TAGS -->', '</code>'))
+        else:
+            out.append(line)
+
+    return "\n".join(out)
+
+
 image_regexp = re.compile('\[!\[\]\([^)]+\)\]\(([^)]+)\)')
 htmlimage_regexp = re.compile('(\(http://([^.]+.){2}blogspot.com/[^()]+/)s1600-h/')
 filename_regexp = re.compile('.*/([^/]+\.(jpg|png))')
@@ -162,6 +185,7 @@ def post_process_images(text, image_directory):
 
 def post_process(text, post_filename, is_comment):
     text = post_process_pre(text)
+    text = post_process_tt_code(text)
 
     if not is_comment:
         image_directory = post_filename.split('.mdwn')[0]

html2text.py

@@ -212,6 +212,10 @@ def __init__(self, out=None, baseurl=''):
         self.table = 0
         self.startpre = 0
         self.tags_in_pre = 0
+        self.in_tt = False
+        self.tags_in_tt = False
+        self.in_code = False
+        self.tags_in_code = False
         self.code = False
         self.br_toggle = ''
         self.lastWasNL = 0
@@ -428,23 +432,37 @@ def handle_tag(self, tag, attrs, start):
                 self.p()
 
         if tag in ['em', 'i', 'u'] and not self.ignore_emphasis:
-            if not self.pre:
-                self.o("_")
-            else:
-                self.tags_in_pre = 1
+            if self.pre or self.in_code or self.in_tt:
+                if self.pre:
+                    self.tags_in_pre = 1
+                elif self.in_code:
+                    self.tags_in_code = 1
+                elif self.in_tt:
+                    self.tags_in_tt = 1
+
                 if start:
                     self.o("<i>")
                 else:
                     self.o("</i>")
-        if tag in ['strong', 'b'] and not self.ignore_emphasis:
-            if not self.pre:
-                self.o("**")
             else:
-                self.tags_in_pre = 1
+                self.o("_")
+
+        if tag in ['strong', 'b'] and not self.ignore_emphasis:
+            if self.pre or self.in_code or self.in_tt:
+                if self.pre:
+                    self.tags_in_pre = 1
+                elif self.in_code:
+                    self.tags_in_code = 1
+                elif self.in_tt:
+                    self.tags_in_tt = 1
+
                 if start:
                     self.o("<b>")
                 else:
                     self.o("</b>")
+            else:
+                self.o("**")
+
         if tag in ['del', 'strike', 's']:
             if start:
                 self.o("<"+tag+">")
@@ -456,7 +474,30 @@ def handle_tag(self, tag, attrs, start):
                 # handle some font attributes, but leave headers clean
                 self.handle_emphasis(start, tag_style, parent_style)
 
-        if tag in ["code", "tt"] and not self.pre: self.o('`') #TODO: `` `this` ``
+        #TODO: `` `this` ``
+        if tag == 'tt' and not self.pre:
+            if start:
+                self.in_tt = True
+                self.o('<!-- START TT -->')
+            else:
+                if self.tags_in_tt:
+                    self.o('<!-- END TT WITH TAGS -->')
+                else:
+                    self.o('<!-- END TT WITHOUT TAGS -->')
+                self.in_tt = False
+                self.tags_in_tt = False
+        if tag == 'code' and not self.pre:
+            if start:
+                self.in_code = True
+                self.o('<!-- START CODE -->')
+            else:
+                if self.tags_in_code:
+                    self.o('<!-- END CODE WITH TAGS -->')
+                else:
+                    self.o('<!-- END CODE WITHOUT TAGS -->')
+                self.in_code = False
+                self.tags_in_code = False
+
         if tag == "abbr":
             if start:
                 self.abbr_title = None