feat: markdown parser task lists and blockquotes (#28)

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "dacli"
-version = "0.4.26"
+version = "0.4.27"
 description = "Documentation Access CLI - Navigate and query large documentation projects"
 readme = "README.md"
 license = { text = "MIT" }

src/dacli/__init__.py

@@ -5,4 +5,4 @@
 """
 
 
-__version__ = "0.4.26"
+__version__ = "0.4.27"

src/dacli/api/models.py

@@ -9,7 +9,9 @@
 from pydantic import BaseModel, Field
 
 # Valid element types for GET /elements endpoint
-VALID_ELEMENT_TYPES = frozenset(["admonition", "code", "image", "list", "plantuml", "table"])
+VALID_ELEMENT_TYPES = frozenset(
+    ["admonition", "blockquote", "code", "image", "list", "plantuml", "table"]
+)
 
 
 class LocationResponse(BaseModel):

src/dacli/cli.py

@@ -609,7 +609,7 @@ def search(ctx: CliContext, query: str, scope: str | None, max_results: int):
 """)
 @click.argument("section_path", required=False, default=None)
 @click.option("--type", "element_type", default=None,
-              help="Element type: admonition, code, image, list, plantuml, table")
+              help="Element type: admonition, blockquote, code, image, list, plantuml, table")
 @click.option("--recursive", is_flag=True, default=False,
               help="Include elements from child sections")
 @click.option("--include-content", is_flag=True, default=False,

src/dacli/markdown_parser.py

@@ -41,9 +41,13 @@
 IMAGE_PATTERN = re.compile(r"!\[([^\]]*)\]\(([^)\s]+)(?:\s+\"([^\"]*)\")?\)")
 
 # List patterns
+TASK_LIST_PATTERN = re.compile(r"^[\s]*-\s+\[([ xX])\]\s+.+$")
 UNORDERED_LIST_PATTERN = re.compile(r"^[\s]*[-*+]\s+.+$")
 ORDERED_LIST_PATTERN = re.compile(r"^[\s]*\d+\.\s+.+$")
 
+# Blockquote pattern
+BLOCKQUOTE_PATTERN = re.compile(r"^>\s?(.*)$")
+
 # Setext heading patterns (for warning detection - not supported)
 # H1: line of text followed by line of ='s (at least 3)
 SETEXT_H1_UNDERLINE = re.compile(r"^={3,}\s*$")
@@ -535,6 +539,11 @@ def _parse_elements(
         current_list_type: str | None = None
         current_list_element: Element | None = None
 
+        # Track blockquote state
+        in_blockquote = False
+        blockquote_start_line = 0
+        blockquote_content: list[str] = []
+
         # Content tracking for Issue #159
         table_content: list[str] = []
         list_content: list[str] = []
@@ -681,8 +690,67 @@ def _parse_elements(
                 current_list_element = None
                 continue
 
-            # Handle lists (unordered and ordered)
+            # Handle blockquotes (Issue #28)
+            blockquote_match = BLOCKQUOTE_PATTERN.match(line)
+            if blockquote_match and not in_code_block and not in_table:
+                # Finalize any open list before starting blockquote
+                if current_list_element is not None and list_content:
+                    current_list_element.attributes["content"] = "\n".join(list_content)
+                current_list_type = None
+                current_list_element = None
+
+                if not in_blockquote:
+                    in_blockquote = True
+                    blockquote_start_line = line_num
+                    blockquote_content = []
+                blockquote_content.append(line)
+                continue
+
+            if in_blockquote:
+                # Non-blockquote line ends the blockquote
+                elements.append(
+                    Element(
+                        type="blockquote",
+                        source_location=SourceLocation(
+                            file=file_path,
+                            line=blockquote_start_line,
+                            end_line=blockquote_start_line + len(blockquote_content) - 1,
+                        ),
+                        attributes={
+                            "content": "\n".join(blockquote_content),
+                        },
+                        parent_section=current_section_path,
+                    )
+                )
+                in_blockquote = False
+                blockquote_content = []
+
+            # Handle lists (unordered, ordered, and task lists)
             if not in_code_block and not in_table:
+                # Check for task list (- [ ] or - [x]) before unordered list
+                # since task list items also match unordered list pattern
+                if TASK_LIST_PATTERN.match(line):
+                    if current_list_type != "task":
+                        # Save previous list content if any (Issue #159)
+                        if current_list_element is not None and list_content:
+                            current_list_element.attributes["content"] = "\n".join(list_content)
+                        current_list_type = "task"
+                        list_content = []
+                        element = Element(
+                            type="list",
+                            source_location=SourceLocation(
+                                file=file_path, line=line_num, end_line=line_num
+                            ),
+                            attributes={"list_type": "task"},
+                            parent_section=current_section_path,
+                        )
+                        elements.append(element)
+                        current_list_element = element
+                    list_content.append(line)
+                    if current_list_element is not None:
+                        current_list_element.source_location.end_line = line_num
+                    continue
+
                 # Check for unordered list (*, -, +)
                 if UNORDERED_LIST_PATTERN.match(line):
                     if current_list_type != "unordered":
@@ -741,6 +809,23 @@ def _parse_elements(
                     current_list_type = None
                     current_list_element = None
 
+        # Handle blockquote at end of file (Issue #28)
+        if in_blockquote and blockquote_content:
+            elements.append(
+                Element(
+                    type="blockquote",
+                    source_location=SourceLocation(
+                        file=file_path,
+                        line=blockquote_start_line,
+                        end_line=blockquote_start_line + len(blockquote_content) - 1,
+                    ),
+                    attributes={
+                        "content": "\n".join(blockquote_content),
+                    },
+                    parent_section=current_section_path,
+                )
+            )
+
         # Handle unclosed code block at end of file
         if in_code_block:
             logger.warning(

src/dacli/mcp_app.py

@@ -351,7 +351,7 @@ def get_elements(
         the documentation, such as code examples, tables, or diagrams.
 
         Args:
-            element_type: Filter by type - 'admonition', 'code', 'image',
+            element_type: Filter by type - 'admonition', 'blockquote', 'code', 'image',
                           'list', 'plantuml', 'table'. None returns all elements.
             section_path: Filter by section path (e.g., '/architecture').
             recursive: If True, include elements from child sections.

src/dacli/models.py

@@ -85,7 +85,7 @@ class Element:
         index: 0-based index within the parent section (set during indexing)
     """
 
-    type: Literal["code", "table", "image", "plantuml", "admonition", "list"]
+    type: Literal["code", "table", "image", "plantuml", "admonition", "list", "blockquote"]
     source_location: SourceLocation
     attributes: dict[str, Any]
     parent_section: str