org_rw/org_rw.py

@@ -67,7 +67,7 @@ BASE_ENVIRONMENT = {
     ),
 }
 
-HEADLINE_TAGS_RE = re.compile(r"((:(\w|[0-9_@#%])+)+:)\s*$")
+HEADLINE_TAGS_RE = re.compile(r"((?P<space_before_tags>\s+)(:(\w|[0-9_@#%])+)+:)(?P<space_after_tags>\s*)$")
 HEADLINE_RE = re.compile(r"^(?P<stars>\*+)(?P<spacing>\s+)(?P<line>.*?)$")
 KEYWORDS_RE = re.compile(
     r"^(?P<indentation>\s*)#\+(?P<key>[^:\[]+)(\[(?P<options>[^\]]*)\])?:(?P<spacing>\s*)(?P<value>.*)$"
@@ -315,6 +315,8 @@ class Headline:
         state,
         tags_start,
         tags,
+        space_before_tags,
+        space_after_tags,
         contents,
         children,
         structural,
@@ -340,6 +342,8 @@ class Headline:
         self.title = parse_content_block([RawLine(linenum=start_line, line=title)])
         self._state = state
         self.tags_start = tags_start
+        self.space_before_tags = space_before_tags
+        self.space_after_tags = space_after_tags
         self.shallow_tags = tags
         self.contents = contents
         self.children = children
@@ -2182,8 +2186,11 @@ def parse_headline(hl, doc, parent) -> Headline:
 
     if hl_tags is None:
         tags = []
+        space_before_tags = space_after_tags = ''
     else:
-        tags = hl_tags.group(0)[1:-1].split(":")
+        tags = hl_tags.group(0).strip()[1:-1].split(":")
+        space_before_tags = hl_tags.group('space_before_tags') or ''
+        space_after_tags = hl_tags.group('space_after_tags') or ''
         line = HEADLINE_TAGS_RE.sub("", line)
 
     hl_state = None
@@ -2203,6 +2210,13 @@ def parse_headline(hl, doc, parent) -> Headline:
                 is_done = True
                 break
 
+    if len(tags) == 0:
+        # No tags, so title might contain trailing whitespaces, handle it
+        title_ends_with_whitespace_match = re.search(r'\s+$', title)
+        if title_ends_with_whitespace_match is not None:
+            space_before_tags = title_ends_with_whitespace_match.group(0)
+            title = title[:-len(space_before_tags)]
+
     contents = parse_contents(hl["contents"])
 
     if not (isinstance(parent, OrgDoc) or depth > parent.depth):
@@ -2229,6 +2243,8 @@ def parse_headline(hl, doc, parent) -> Headline:
         priority_start=None,
         tags_start=None,
         tags=tags,
+        space_before_tags=space_before_tags,
+        space_after_tags=space_after_tags,
         parent=parent,
         is_todo=is_todo,
         is_done=is_done,
@@ -2430,25 +2446,21 @@ class OrgDoc:
 
     # Writing
     def dump_headline(self, headline, recursive=True):
-        tags = ""
+        tags = headline.space_before_tags
         if len(headline.shallow_tags) > 0:
-            tags = ":" + ":".join(headline.shallow_tags) + ":"
+            tags += ":" + ":".join(headline.shallow_tags) + ":" + headline.space_after_tags
 
         state = ""
         if headline._state:
             state = headline._state["name"] + " "
 
         raw_title = token_list_to_raw(headline.title.contents)
-        tags_padding = ""
-        if not (raw_title.endswith(" ") or raw_title.endswith("\t")) and tags:
-            tags_padding = " "
 
         yield (
             "*" * headline.depth
             + headline.spacing
             + state
             + raw_title
-            + tags_padding
             + tags
         )