Initial fix for 'find_by_marker'

Author: elifarley

src/cedarscript_editor/cedarscript_editor.py

@@ -149,6 +149,8 @@ def _update_command(self, cmd: UpdateCommand):
 
         identifier_finder = find_identifier(source_info)
 
+        # TODO test...
+
         match action:
             case MoveClause():
                 # (Check parse_update_command)
@@ -164,11 +166,19 @@ def _update_command(self, cmd: UpdateCommand):
                 # Set range_spec to cover the identifier
                 search_range = restrict_search_range(action, target, identifier_finder)
 
-        marker, search_range = find_marker_or_segment(action, lines, search_range)
-
-        search_range = restrict_search_range_for_marker(
-            marker, action, lines, search_range, identifier_finder
-        )
+        # UPDATE FUNCTION "_check_raw_id_fields_item"
+        # FROM FILE "refactor-benchmark/checks_BaseModelAdminChecks__check_raw_id_fields_item/checks.py"
+        # REPLACE LINE "def _check_raw_id_fields_item(self, obj, field_name, label):"
+        # WITH CONTENT '''
+        # @0:def _check_raw_id_fields_item(obj, field_name, label):
+        # ''';
+        # target = IdentifierFromFile(file_path='refactor-benchmark/checks_BaseModelAdminChecks__check_raw_id_fields_item/checks.py', identifier_type=<MarkerType.FUNCTION: 'function'>, name='_check_raw_id_fields_item', where_clause=None, offset=None)
+        # action = ReplaceClause(region=Marker(type=<MarkerType.LINE: line>, value=def _check_raw_id_fields_item(self, obj, field_name, label):, offset=None))
+        if search_range.line_count:
+            marker, search_range = find_marker_or_segment(action, lines, search_range)
+            search_range = restrict_search_range_for_marker(
+                marker, action, lines, search_range, identifier_finder
+            )
 
         match content:
             case str() | [str(), *_] | (str(), *_):
@@ -309,7 +319,6 @@ def find_marker_or_segment(
 
 
 def restrict_search_range(action, target, identifier_finder: IdentifierFinder) -> RangeSpec:
-    search_range = RangeSpec.EMPTY
     match target:
         case IdentifierFromFile() as identifier_from_file:
             identifier_marker = identifier_from_file.as_marker
@@ -318,12 +327,19 @@ def restrict_search_range(action, target, identifier_finder: IdentifierFinder) -
                 raise ValueError(f"'{identifier_marker}' not found")
             match action:
                 case RegionClause(region=region):
-                    match region:  # BodyOrWhole | Marker | Segment
-                        case BodyOrWhole():
-                            search_range = identifier_boundaries.location_to_search_range(region)
-                        case _:
-                            search_range = identifier_boundaries.location_to_search_range(BodyOrWhole.WHOLE)
-    return search_range
+                    match region:
+                        case BodyOrWhole() | RelativePositionType():
+                            return identifier_boundaries.location_to_search_range(region)
+                        case Marker() as inner_marker:
+                            match identifier_finder(inner_marker):
+                                case IdentifierBoundaries() as inner_boundaries:
+                                    return inner_boundaries.whole
+                                case RangeSpec() as inner_range_spec:
+                                    return inner_range_spec
+                        case _ as invalid: # Marker (LINE) or Segment
+                            # TODO
+                            raise ValueError(f'Not implemented: {invalid}')
+    return RangeSpec.EMPTY
 
 
 def restrict_search_range_for_marker(

src/cedarscript_editor/tree_sitter_identifier_finder.py

@@ -1,7 +1,7 @@
 import logging
 from typing import Callable, TypeAlias, Sequence, NamedTuple, Iterable
 
-from cedarscript_ast_parser import Marker, MarkerType
+from cedarscript_ast_parser import Marker, MarkerType, Segment
 from grep_ast import filename_to_lang
 from text_manipulation.indentation_kit import get_line_indent_count
 from text_manipulation.range_spec import IdentifierBoundaries, RangeSpec
@@ -11,28 +11,37 @@
 
 _log = logging.getLogger(__name__)
 
-IdentifierFinder: TypeAlias = Callable[[Marker], IdentifierBoundaries | None]
+IdentifierFinder: TypeAlias = Callable[[Marker], IdentifierBoundaries | RangeSpec | None]
 
 
-def find_identifier(source_info: tuple[str, str | Sequence[str]]) -> IdentifierFinder:
+def find_identifier(source_info: tuple[str, str | Sequence[str]], search_rage: RangeSpec = RangeSpec.EMPTY) -> IdentifierFinder:
     file_path = source_info[0]
     source = source_info[1]
     if not isinstance(source, str):
         source = '\n'.join(source)
-    return _select_finder(file_path, source)
+    return _select_finder(file_path, source, search_rage)
 
 
-def _select_finder(file_path: str, source: str) -> IdentifierFinder:
+def _select_finder(file_path: str, source: str, search_range: RangeSpec = RangeSpec.EMPTY) -> IdentifierFinder:
     langstr = filename_to_lang(file_path)
     language = get_language(langstr)
     parser = get_parser(langstr)
     _log.info(f"[select_finder] Selected {language}")
 
     tree = parser.parse(bytes(source, "utf-8"))
+    source = source.splitlines()
     query_info = LANG_TO_TREE_SITTER_QUERY[langstr]
 
-    def find_by_marker(marker: Marker) -> IdentifierBoundaries | None:
-        return _find_identifier(language, source, tree, query_info, marker)
+    def find_by_marker(mos: Marker | Segment) -> IdentifierBoundaries | RangeSpec | None:
+        match mos:
+
+            case Marker(MarkerType.LINE) | Segment():
+                # TODO pass IdentifierFinder to enable identifiers as start and/or end of a segment
+                return mos.to_search_range(source, search_range).set_line_count(1)  # returns RangeSpec
+
+            case Marker() as marker:
+                # Returns IdentifierBoundaries
+                return _find_identifier(language, source, tree, query_info, marker)
 
     return find_by_marker
 
@@ -66,7 +75,7 @@ def identifier(self):
         return self.node.text.decode("utf-8")
 
 
-def associate_identifier_parts(captures: Iterable[CaptureInfo], lines: list[str]) -> list[IdentifierBoundaries]:
+def associate_identifier_parts(captures: Iterable[CaptureInfo], lines: Sequence[str]) -> list[IdentifierBoundaries]:
     identifier_map: dict[int, IdentifierBoundaries] = {}
 
     for capture in captures:
@@ -105,7 +114,7 @@ def find_parent_definition(node):
     return None
 
 
-def _find_identifier(language, source: str, tree, query_scm: dict[str, dict[str, str]], marker: Marker) \
+def _find_identifier(language, source: Sequence[str], tree, query_scm: dict[str, dict[str, str]], marker: Marker) \
         -> IdentifierBoundaries | None:
     """
     Find the starting line index of a specified function in the given lines.
@@ -121,7 +130,7 @@ def _find_identifier(language, source: str, tree, query_scm: dict[str, dict[str,
         candidates = language.query(query_scm[marker.type].format(name=marker.value)).captures(tree.root_node)
         candidates: list[IdentifierBoundaries] = capture2identifier_boundaries(
             candidates,
-            source.splitlines()
+            source
         )
     except Exception as e:
         raise ValueError(f"Unable to capture nodes for {marker}: {e}") from e
@@ -145,7 +154,7 @@ def _find_identifier(language, source: str, tree, query_scm: dict[str, dict[str,
     return result
 
 
-def capture2identifier_boundaries(captures, lines: list[str]) -> list[IdentifierBoundaries]:
+def capture2identifier_boundaries(captures, lines: Sequence[str]) -> list[IdentifierBoundaries]:
     captures = [CaptureInfo(c[1], c[0]) for c in captures if not c[1].startswith('_')]
     unique_captures = {}
     for capture in captures:

src/text_manipulation/range_spec.py

@@ -40,7 +40,6 @@ class RangeSpec(NamedTuple):
     indent: int = 0
 
     def __str__(self):
-        """Return a string representation of the RangeSpec."""
         return (f'{self.start}:{self.end}' if self.as_index is None else f'%{self.as_index}') + f'@{self.indent}'
 
     def __lt__(self, other):