org_rw/org_rw.py

@@ -877,6 +877,46 @@ class Headline:
         else:
             raise NotImplementedError()
 
+    def update_raw_contents(self, new_contents):
+        # Clear elements
+        self.keywords = []
+        self.contents = []
+        self.list_items = []
+        self.table_rows = []
+        self.properties = []
+        self.structural = []
+        self.delimiters = []
+        self.scheduled = None
+        self.deadline = None
+        self.closed = None
+
+        reader = OrgDocReader(environment=self.doc.environment)
+        reader.read(new_contents)
+
+        # No need to finalize as we can take the data from the reader instead of from a doc
+        if len(reader.headlines) > 0:
+            # Probably can be done by just adding the headlines to this one's children
+            raise NotImplementedError(
+                "new headlines on raw contents not supported yet. This probably should be simple, see comment on code."
+            )
+
+        for kw in reader.keywords:
+            self.keywords.append(offset_linenum(self.start_line + 1, kw))
+
+        for content in reader.contents:
+            self.contents.append(offset_linenum(self.start_line + 1, content))
+
+        for list_item in reader.list_items:
+            self.list_items.append(offset_linenum(self.start_line + 1, list_item))
+
+        for struct_item in reader.structural:
+            self.structural.append(offset_linenum(self.start_line + 1, struct_item))
+
+        for prop in reader.properties:
+            self.properties.append(offset_linenum(self.start_line + 1, prop))
+
+        # Environment is not used, as it's known
+
     def get_element_in_line(self, linenum):
         for line in self.contents:
             if linenum == line.linenum:
@@ -1074,6 +1114,7 @@ Keyword = collections.namedtuple(
 Property = collections.namedtuple(
     "Property", ("linenum", "match", "key", "value", "options")
 )
+Structural = collections.namedtuple("Structural", ("linenum", "line"))
 
 
 class ListItem:
@@ -1122,6 +1163,19 @@ TableRow = collections.namedtuple(
     ),
 )
 
+ItemWithLineNum = Union[Keyword, RawLine, Property, ListItem, Structural]
+
+
+def offset_linenum(offset: int, item: ItemWithLineNum) -> ItemWithLineNum:
+    if isinstance(item, ListItem):
+        item.linenum += offset
+        return item
+
+    assert isinstance(
+        item, (Keyword, RawLine, Property, Structural)
+    ), "Expected (Keyword|RawLine|Property|Structural), found {}".format(item)
+    return item._replace(linenum=item.linenum + offset)
+
 
 # @TODO How are [YYYY-MM-DD HH:mm--HH:mm] and ([... HH:mm]--[... HH:mm]) differentiated ?
 # @TODO Consider recurrence annotations
@@ -2340,6 +2394,7 @@ class OrgDoc:
         self.headlines: List[Headline] = list(
             map(lambda hl: parse_headline(hl, self, self), headlines)
         )
+        self.environment = environment
 
     @property
     def id(self):
@@ -2524,8 +2579,8 @@ class OrgDocReader:
         self.delimiters: List[DelimiterLine] = []
         self.list_items: List[ListItem] = []
         self.table_rows: List[TableRow] = []
-        self.structural: List = []
-        self.properties: List = []
+        self.structural: List[Structural] = []
+        self.properties: List[Property] = []
         self.current_drawer: Optional[List] = None
         self.environment = environment
 
@@ -2707,7 +2762,7 @@ class OrgDocReader:
     def add_property_drawer_line(self, linenum: int, line: str, match: re.Match):
         if len(self.headline_hierarchy) == 0:
             self.current_drawer = self.properties
-            self.structural.append((linenum, line))
+            self.structural.append(Structural(linenum, line))
         else:
             assert self.headline_hierarchy[-1] is not None
             self.current_drawer = self.headline_hierarchy[-1]["properties"]
@@ -2726,7 +2781,7 @@ class OrgDocReader:
     def add_drawer_end_line(self, linenum: int, line: str, match: re.Match):
         self.current_drawer = None
         if len(self.headline_hierarchy) == 0:
-            self.structural.append((linenum, line))
+            self.structural.append(Structural(linenum, line))
         else:
             assert self.headline_hierarchy[-1] is not None
             self.headline_hierarchy[-1]["structural"].append((linenum, line))

tests/13-update-reparse-test.org

@@ -0,0 +1,22 @@
+#+TITLE: 13-Update reparse
+#+DESCRIPTION: Update-Reparse org file
+#+TODO: TODO(t) PAUSED(p) |  DONE(d)
+
+
+* First level
+  :PROPERTIES:
+  :ID:       13-update-reparse-first-level-id
+  :CREATED:  [2020-01-01 Wed 01:01]
+  :END:
+  First level content
+
+  - A list of items ::
+    - With a sublist
+
+  Something after the list.
+
+** Second level
+   :PROPERTIES:
+   :ID:       13-update-reparse-second-level-id
+   :END:
+   Second level content

tests/test_org.py

@@ -1,4 +1,5 @@
 import os
+import tempfile
 import unittest
 from datetime import datetime as DT
 
@@ -869,6 +870,86 @@ class TestSerde(unittest.TestCase):
 
         self.assertEqual(dumps(doc), orig)
 
+    def test_update_reparse_same_structure(self):
+        with open(os.path.join(DIR, "01-simple.org")) as f:
+            doc = load(f)
+
+        hl = doc.getTopHeadlines()[0]
+        ex = HL(
+            "First level",
+            props=[
+                ("ID", "01-simple-first-level-id"),
+                ("CREATED", DT(2020, 1, 1, 1, 1)),
+            ],
+            content="  First level content\n",
+            children=[
+                HL(
+                    "Second level",
+                    props=[("ID", "01-simple-second-level-id")],
+                    content="\n   Second level content\n",
+                    children=[
+                        HL(
+                            "Third level",
+                            props=[("ID", "01-simple-third-level-id")],
+                            content="\n    Third level content\n",
+                        )
+                    ],
+                )
+            ],
+        )
+
+        # Ground check
+        ex.assert_matches(self, hl)
+
+        # Update
+        lines = list(doc.dump_headline(hl, recursive=False))
+        assert lines[0].startswith("* ")  # Title, skip it
+        content = "\n".join(lines[1:])
+        hl.update_raw_contents(content)
+
+        # Check after update
+        ex.assert_matches(self, hl, accept_trailing_whitespace_changes=True)
+
+    def test_update_reparse_same_values(self):
+        with open(os.path.join(DIR, "13-update-reparse-test.org")) as f:
+            doc = load(f)
+
+        expected_hl_contents = """  :PROPERTIES:
+  :ID:       13-update-reparse-first-level-id
+  :CREATED:  [2020-01-01 Wed 01:01]
+  :END:
+  First level content
+
+  - A list of items ::
+    - With a sublist
+
+  Something after the list.
+"""
+
+        hl = doc.getTopHeadlines()[0]
+        lines = list(doc.dump_headline(hl, recursive=False))
+        assert lines[0].startswith("* ")  # Title, skip it
+        content = "\n".join(lines[1:])
+        self.assertEqual(content, expected_hl_contents)
+
+        # Check after update
+        hl.update_raw_contents(content)
+        self.assertEqual(content, expected_hl_contents)
+
+        # Check after dump and reload
+        with tempfile.NamedTemporaryFile("wt") as f:
+            save = org_rw.dumps(doc)
+            f.write(save)
+            f.flush()
+
+            with open(f.name, "rt") as reader:
+                reloaded = org_rw.load(reader)
+                re_hl = reloaded.getTopHeadlines()[0]
+                lines = list(doc.dump_headline(hl, recursive=False))
+                assert lines[0].startswith("* ")  # Title, skip it
+                content = "\n".join(lines[1:])
+                self.assertEqual(content, expected_hl_contents)
+
     def test_mimic_write_file_13(self):
         with open(os.path.join(DIR, "13-tags.org")) as f:
             orig = f.read()

tests/utils/assertions.py

@@ -67,7 +67,12 @@ class HL:
         self.content = content
         self.children = children
 
-    def assert_matches(self, test_case: unittest.TestCase, doc):
+    def assert_matches(
+        self,
+        test_case: unittest.TestCase,
+        doc,
+        accept_trailing_whitespace_changes=False,
+    ):
         test_case.assertEqual(self.title, get_raw(doc.title))
 
         # Check properties
@@ -84,6 +89,11 @@ class HL:
                         timestamp_to_datetime(doc_props[i].value), prop[1]
                     )
 
+        if accept_trailing_whitespace_changes:
+            test_case.assertEqual(
+                get_raw_contents(doc).rstrip(), self.get_raw().rstrip()
+            )
+        else:
             test_case.assertEqual(get_raw_contents(doc), self.get_raw())
 
         # Check children