org_rw/org_rw.py

@@ -67,7 +67,9 @@ 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 +317,8 @@ class Headline:
         state,
         tags_start,
         tags,
+        space_before_tags,
+        space_after_tags,
         contents,
         children,
         structural,
@@ -340,6 +344,8 @@ 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
@@ -1072,6 +1078,8 @@ class Headline:
             state="",
             tags_start=None,
             tags=[],
+            space_before_tags="",
+            space_after_tags="",
             contents=[],
             children=[],
             structural=[],
@@ -1816,7 +1824,7 @@ def token_list_to_plaintext(tok_list) -> str:
         else:
             assert isinstance(chunk, MarkerToken)
 
-    return "".join(contents).strip()
+    return "".join(contents)
 
 
 def token_list_to_raw(tok_list):
@@ -2182,8 +2190,11 @@ 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 +2214,13 @@ 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 +2247,8 @@ 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,
@@ -2430,27 +2450,19 @@ 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 (
+        yield ("*" * headline.depth + headline.spacing + state + raw_title + tags)
-            "*" * headline.depth
-            + headline.spacing
-            + state
-            + raw_title
-            + tags_padding
-            + tags
-        )
 
         planning = headline.get_planning_line()
         if planning is not None:

tests/14-titles.org

@@ -0,0 +1,12 @@
+#+TITLE: 14-Simple
+#+DESCRIPTION: Org file to evaluate titles
+#+TODO: TODO(t) PAUSED(p) |  DONE(d)
+
+
+* Simple title
+
+* Simple title with tags :tag:
+
+* Simple title with trailing space 
+
+*  Simple title with leading space

tests/test_org.py

@@ -955,6 +955,24 @@ class TestSerde(unittest.TestCase):
         h1_2_h2 = h1_2.children[0]
         self.assertEqual(sorted(h1_2_h2.tags), ["otherh2tag"])
 
+    def test_titles_file(self):
+        with open(os.path.join(DIR, "14-titles.org")) as f:
+            doc = load(f)
+
+        h1, h2, h3, h4 = doc.getTopHeadlines()
+        self.assertEqual(h1.title.get_text(), "Simple title")
+        self.assertEqual(h2.title.get_text(), "Simple title with tags")
+        self.assertEqual(h3.title.get_text(), "Simple title with trailing space")
+        self.assertEqual(h4.title.get_text(), "Simple title with leading space")
+
+    def test_mimic_write_file_14(self):
+        """A goal of this library is to be able to update a file without changing parts not directly modified."""
+        with open(os.path.join(DIR, "14-titles.org")) as f:
+            orig = f.read()
+            doc = loads(orig)
+
+        self.assertEqual(dumps(doc), orig)
+
     def test_update_headline_from_none_to_todo(self):
         orig = "* First entry"
         doc = loads(orig)