added support for parsing standard Markdown links that happen to be internal

src/dragonglass/mparse.py

@@ -2,10 +2,44 @@
 
 import markdown
 import xml.etree.ElementTree as etree
+from urllib.parse import urlparse
+from urllib.parse import unquote as urlunquote
 from markdown.extensions import Extension
 from markdown.inlinepatterns import InlineProcessor, SimpleTagInlineProcessor
 from markdown.preprocessors import Preprocessor
 
+PRIO_BASE = 10000  # priority base for our extensions
+
+
+def is_proper_url(s):
+    """
+    Checks to see if a string is a "proper" URL.
+
+    Args:
+        s (str): The string to be parsed.
+
+    Returns:
+        bool: ``True`` if the string is a proper URL, ``False`` if not.
+    """
+    parseout = urlparse(s)
+    return True if parseout.scheme else False
+
+
+def sanitize_reference(s):
+    """
+    Sanitizes an internal reference to a file by removing URL-quoted characters and any Markdown suffix.
+
+    Args:
+        s (str): Th reference to be sanitized.
+
+    Returns:
+        str: The sanitized reference.
+    """
+    rc = urlunquote(s)
+    if rc.endswith('.md'):
+        rc = rc[:-3]
+    return rc
+
 
 class MetaStripper(Extension):
     """
@@ -23,57 +57,97 @@ class MetaStripper(Extension):
             return lines
 
     def extendMarkdown(self, md):
-        md.preprocessors.register(MetaStripper.MetaStripperProc(md), 'metastripper', 0)
+        md.preprocessors.register(MetaStripper.MetaStripperProc(md), 'metastripper', PRIO_BASE)
 
 
 class ObsidianLinks(Extension):
-    """An extension that processes Obsidian internal links in the [[page name]] format."""
+    """
+    An extension that processes Obsidian internal links in the [[page name]] format, as well as overrides the standard
+    Markdown link processing to handle Obsidian internal links as well as external links.
+    """
     def __init__(self, context, **kwargs):
         super(ObsidianLinks, self).__init__(**kwargs)
         self._context = context
 
+    @property
+    def obsidian_link_classname(self):
+        return 'obsidian-link'
+
+    @property
+    def invalid_reference_classname(self):
+        return 'invalid-reference'
+
+    def _parse_reference(self, contents):
+        text = None
+        t = contents.split('|')
+        if len(t) > 1:
+            text = t[1]
+            contents = t[0]
+
+        node, linktype = self._context.src_index.lookup(contents)
+        if not text:
+            text = contents
+        if node:
+            return node.link_target(self._context.url_prefix), text
+        return None, text
+
     class ObsidianLinksProc(InlineProcessor):
-        def __init__(self, pattern, md, context):
+        def __init__(self, pattern, md, extref):
             super(ObsidianLinks.ObsidianLinksProc, self).__init__(pattern, md)
-            self._context = context
-
-        def parse_reference(self, contents):
-            text = None
-            t = contents.split('|')
-            if len(t) > 1:
-                text = t[1]
-                contents = t[0]
-
-            node, linktype = self._context.src_index.lookup(contents)
-            if not text:
-                text = contents
-            if node:
-                return node.link_target(self._context.url_prefix), text
-            return None, text
+            self._extref = extref
 
         def handleMatch(self, m, data):
-            link, text = self.parse_reference(m.group(0)[2:-2])
+            link, text = self._extref._parse_reference(m.group(1))
             if link is None:
                 el = etree.Element('span')
-                el.set('class', 'invalid-reference')
+                el.set('class', self._extref.invalid_reference_classname)
                 el.text = text
             else:
                 el = etree.Element('a')
                 el.set('href', link)
+                el.set('class', self._extref.obsidian_link_classname)
                 el.text = text
             return el, m.start(0), m.end(0)
 
+    class GenericLinksProc(InlineProcessor):
+        def __init__(self, pattern, md, extref):
+            super(ObsidianLinks.GenericLinksProc, self).__init__(pattern, md)
+            self._extref = extref
+
+        def handleMatch(self, m, data):
+            text = m.group(1)
+            link = m.group(2)
+            if is_proper_url(link):
+                el = etree.Element('a')
+                el.set('href', link)
+                el.text = text
+            else:
+                newlink, _ = self._extref._parse_reference(sanitize_reference(link))
+                if newlink is None:
+                    el = etree.Element('span')
+                    el.set('class', self._extref.invalid_reference_classname)
+                    el.text = text
+                else:
+                    el = etree.Element('a')
+                    el.set('href', newlink)
+                    el.set('class', self._extref.obsidian_link_classname)
+                    el.text = text
+            return el, m.start(0), m.end(0)
+
     def extendMarkdown(self, md):
         OBSLINK_PATTERN = r'\[\[(.*?)\]\]'
-        md.inlinePatterns.register(ObsidianLinks.ObsidianLinksProc(OBSLINK_PATTERN, md, self._context),
-                                   'obsidian_links', 0)
+        GENERICLINK_PATTERN = r'\[(.*?)\]\((.*?)\)'
+        md.inlinePatterns.register(ObsidianLinks.ObsidianLinksProc(OBSLINK_PATTERN, md, self),
+                                   'obsidian_links', PRIO_BASE + 110)
+        md.inlinePatterns.register(ObsidianLinks.GenericLinksProc(GENERICLINK_PATTERN, md, self),
+                                   'obsidian_generic_links', PRIO_BASE + 100)
 
 
 class ObsidianInlines(Extension):
     """An extension that handles the special Obsidian markdown format sequences."""
     def extendMarkdown(self, md):
-        md.inlinePatterns.register(SimpleTagInlineProcessor(r'()~~(.*?)~~', 'del'), 'strikeout', 0)
-        md.inlinePatterns.register(SimpleTagInlineProcessor(r'()\=\=(.*?)\=\=', 'ins'), 'highlight', 0)
+        md.inlinePatterns.register(SimpleTagInlineProcessor(r'()~~(.*?)~~', 'del'), 'strikeout', PRIO_BASE)
+        md.inlinePatterns.register(SimpleTagInlineProcessor(r'()\=\=(.*?)\=\=', 'ins'), 'highlight', PRIO_BASE + 1)
 
 
 def create_markdown_parser(context):