Support reading labelled code cells

mkquartodocs/extension.py

@@ -24,14 +24,14 @@
 
 # `:::: {.cell execution_count="1"}`
 # `:::::: {.cell layout-align="default"}` <-  happens in mermaid diagrams
-CELL_REGEX: Final = re.compile(r"^(:{3,}) \{\.cell .*}\s*$")
+CELL_REGEX: Final = re.compile(r"^(:{3,}) \{(#[-\w_]+)?\.cell .*}\s*$")
 CELL_END: Final = re.compile(r"^(:{3,})$")
 
 # In theory it would be a 'cell-output' but there are other kinds of out
 # Cells contain multiple cell elements.
 # `::: {.cell-output .cell-output-stdout}`
 CELL_ELEM_REGEX: Final = re.compile(
-    r"^(:{3,}) \{(.cell-\w+)\s?(\.cell-[\w-]+)?( execution_count=\"\d+\")?\}$"
+    r"^(:{3,}) \{(#[-\w_]+)?\s*(.cell-\w+)\s?(\.cell-[\w-]+)?( execution_count=\"\d+\")?\}$"
 )
 # `::::: cell-output-display`
 CELL_ELEM_ALT_REGEX: Final = re.compile(r"^(:{3,})\s*(cell-output-display\s*)$")
@@ -136,6 +136,7 @@ class BlockContext:
     attributes: list[str]
     start: Cursor
     end: Cursor | UnknownEnd
+    label: str | None = None
 
     def __post_init__(self):
         log.info(f"BlockContext: {self}")
@@ -149,25 +150,37 @@ def _from_cell_start_line(cls, line: str, line_number: int) -> BlockContext:
         sr = CELL_REGEX.search(line)
         grp = sr.groups()
         assert all(w == ":" for w in grp[0]), f"{grp[0]} should be :"
+        first_cell_grp = 1
+        label = None
+        if len(grp) > 1 and grp [1] and grp[1].startswith("#"):
+            label = grp[1][1:]
+            first_cell_grp = 2
         return BlockContext(
             block_type=BlockType.CELL,
             delimiter=grp[0],
-            attributes=grp[1:],
+            attributes=grp[first_cell_grp:],
             start=Cursor(line=line_number, col=0),
             end=UnknownEnd(),
+            label=label,
         )
 
     @classmethod
     def _from_cell_element_line(cls, line: str, line_number: int) -> BlockContext:
         """This function assumes that you already checked that the line matches CELL_ELEM_REGEX"""
         sr = CELL_ELEM_REGEX.search(line)
         grp = sr.groups()
+        first_cell_grp = 1
+        label = None
+        if len(grp) > 1 and grp [1] and grp[1].startswith("#"):
+            label = grp[1][1:]
+            first_cell_grp = 2
         return BlockContext(
             block_type=BlockType.CELL_ELEM,
             delimiter=grp[0],
-            attributes=grp[1:],
+            attributes=grp[first_cell_grp:],
             start=Cursor(line=line_number, col=0),
             end=UnknownEnd(),
+            label=label,
         )
 
     @classmethod
@@ -240,6 +253,7 @@ def with_end(self, end: Cursor | UnknownEnd) -> BlockContext:
             attributes=self.attributes,
             start=self.start,
             end=end,
+            label=self.label,
         )
 
     def find_end(
@@ -358,8 +372,16 @@ def _into_output_lines_cell_elem(self, file_content: FileContent) -> list[str]:
             # markdown in html attribute and leave as is.
             non_empty = [line for line in internal if line.strip()]
             if internal[0].startswith("<div>") and non_empty[-1].endswith("</div>"):
+                print("FOO")
+                print(self.label)
                 if "style" in internal[1]:
-                    internal[0] = '<div markdown="block">'
+                    if self.label:
+                        internal[0] = f'<div markdown="block" id="{self.label}">'
+                    else:
+                        internal[0] = '<div markdown="block">'
+                else:
+                    if self.label:
+                        internal[0] = f'<div id="{self.label}">'
 
                 last_end = self.end