switch to fsm implementation of gem2html

1 files changed, 14 insertions, 72 deletions

diff --git a/fsm.py b/fsm.py
index 40f9c2d..901e114 100644
--- a/fsm.py
+++ b/fsm.py
@@ -1,4 +1,6 @@
 import sys
+import urllib.parse
+
 
 class StackFSM(object):
     """
@@ -25,6 +27,7 @@ class StackFSM(object):
     def pop_state(self):
         return self._state_stack.pop()
 
+
 class Parser(object):
 
     def __init__(self, document, output=None):
@@ -38,29 +41,23 @@ class Parser(object):
 
     def parse(self):
         self._fsm.push_state(self.text_state)
-        while self._fsm._current_state() is not None:
+        while self._fsm._current_state() is not None and len(self._document) > self._offset:
             self._fsm.update()
 
     def text_state(self):
-        if len(self._document) <= self._offset:
-            self._fsm.pop_state()
-            return
         line = self._document[self._offset]
         if line.strip() == '':
             self._blanks += 1
         else:
             self._blanks = 0
         if line.strip() == '```':
-            self._fsm.pop_state()
             self._fsm.push_state(self.pre_state)
             self._output.write('<pre>\n')
             self._offset += 1
         elif line.startswith('* '):
-            self._fsm.pop_state()
             self._fsm.push_state(self.list_state)
             self._output.write('<ul>\n')
         elif line.startswith('=>'):
-            self._fsm.pop_state()
             self._fsm.push_state(self.link_state)
             self._output.write('<ul>\n')
         else:
@@ -80,95 +77,40 @@ class Parser(object):
             self._offset += 1
 
     def pre_state(self):
-        if len(self._document) < self._offset:
-            self.pop_state()
-            return
         line = self._document[self._offset]
         if line.strip() == '```':
             self._fsm.pop_state()
-            self._fsm.push_state(self.text_state)
             self._output.write('</pre>\n')
             self._offset += 1
-        elif line.startswith('* '):
-            self._fsm.pop_state()
-            self._fsm.push_state(self.list_state)
-            self._output.write('<ul>\n')
-        elif line.startswith('=>'):
-            self._fsm.pop_state()
-            self._fsm.push_state(self.link_state)
-            self._output.write('<ul>\n')
         else:
             self._output.write(line + '\n')
             self._offset += 1
 
     def list_state(self):
-        if len(self._document) < self._offset:
-            self.pop_state()
-            return
         line = self._document[self._offset]
         if line.startswith('* '):
             self._output.write('<li>{}</li>\n'.format(line[2:]))
             self._offset += 1
         else:
             self._fsm.pop_state()
-            self._fsm.push_state(self.text_state)
             self._output.write('</ul>\n')
 
     def link_state(self):
-        if len(self._document) < self._offset:
-            self.pop_state()
-            return
         line = self._document[self._offset]
         if line.startswith('=>'):
-            parts = line[2:].split(None, 2)
+            parts = line[2:].split(None, 1)
+            url = parts[0]
+            url_parts = urllib.parse.urlsplit(url)
+            if url_parts.scheme in ('gemini', ''):
+                external = ''
+            else:
+                external = ' class="external"'
             if len(parts) == 1:
-                self._output.write('<li><a href="{}">{}</a></li>\n'.format(parts[0], parts[0]))
+                text = url
             else:
-                self._output.write('<li><a href="{}">{}</a></li>\n'.format(parts[0], parts[1]))
+                text = parts[1]
+            self._output.write('<li{}><a href="{}">{}</a></li>\n'.format(external, url, text))
             self._offset += 1
         else:
             self._fsm.pop_state()
-            self._fsm.push_state(self.text_state)
             self._output.write('</ul>\n')
-
-document = """
-# h1
-hello
-hello
-
-## h2
-```
-code
-code
-```
-### h3
-hello
-
-
-hello
-
-### lists
-* hello
-* two
-* three
-
-text
-
-* one
-* two
-* three
-
-### links
-=>https://example.com hello
-=> https://example.com two
-=>  https://example.com three
-
-text
-
-=>https://example.com
-=> https://example.com
-=>  https://example.com
-"""
-
-p = Parser(document.split('\n'))
-p.parse()