feat(#51): add caption support

mdformat_mkdocs/mdit_plugins/__init__.py

@@ -16,6 +16,7 @@
 )
 from ._pymd_abbreviations import PYMD_ABBREVIATIONS_PREFIX, pymd_abbreviations_plugin
 from ._pymd_admon import pymd_admon_plugin
+from ._pymd_captions import PYMD_CAPTIONS_PREFIX, pymd_captions_plugin
 from ._pymd_snippet import PYMD_SNIPPET_PREFIX, pymd_snippet_plugin
 from ._python_markdown_attr_list import (
     PYTHON_MARKDOWN_ATTR_LIST_PREFIX,
@@ -29,6 +30,7 @@
     "MKDOCSTRINGS_CROSSREFERENCE_PREFIX",
     "MKDOCSTRINGS_HEADING_AUTOREFS_PREFIX",
     "PYMD_ABBREVIATIONS_PREFIX",
+    "PYMD_CAPTIONS_PREFIX",
     "PYMD_SNIPPET_PREFIX",
     "PYTHON_MARKDOWN_ATTR_LIST_PREFIX",
     "material_admon_plugin",
@@ -37,6 +39,7 @@
     "mkdocstrings_crossreference_plugin",
     "pymd_abbreviations_plugin",
     "pymd_admon_plugin",
+    "pymd_captions_plugin",
     "pymd_snippet_plugin",
     "python_markdown_attr_list_plugin",
 )

mdformat_mkdocs/mdit_plugins/_pymd_captions.py

@@ -0,0 +1,122 @@
+"""Python-Markdown Extensions Captions.
+
+Matches:
+
+```md
+/// caption
+Default values for config variables.
+///
+```
+
+Docs:
+https://github.com/facelessuser/pymdown-extensions/blob/main/pymdownx/blocks/caption.py
+
+
+"""
+
+from __future__ import annotations
+
+import re
+from typing import TYPE_CHECKING
+
+from mdit_py_plugins.utils import is_code_block
+
+from mdformat_mkdocs._synced.admon_factories._whitespace_admon_factories import (
+    new_token,
+)
+
+if TYPE_CHECKING:
+    from markdown_it import MarkdownIt
+    from markdown_it.rules_block import StateBlock
+
+_CAPTION_START_PATTERN = re.compile(
+    r"\s*///\s*(?P<type>figure-|table-|)caption\s*(\|\s*(?P<number>[\d\.]+))?",
+)
+_CAPTION_END_PATTERN = re.compile(r"^\s*///\s*$")
+_CAPTION_ATTRS_PATTERN = re.compile(r"^s*(?P<attrs>attrs:\s*\{[^}]*\})\s*$")
+PYMD_CAPTIONS_PREFIX = "mkdocs_caption"
+
+
+def _src_in_line(state: StateBlock, line: int) -> tuple[str, int, int]:
+    """Get the source in a given line number."""
+    start_pos = state.bMarks[line] + state.tShift[line]
+    end_pos = state.eMarks[line]
+    return state.src[start_pos:end_pos], start_pos, end_pos
+
+
+def _parse(
+    state: StateBlock,
+    first_line_max_pos: int,
+    start_line: int,
+    end_line: int,
+) -> tuple[int, str, str | None]:
+    """Parse a caption block: optionally read attrs and extract content."""
+    end_match = None
+    max_line = start_line + 1
+    end_pos = -1
+    attrs_text, _, attrs_max_pos = _src_in_line(state, max_line)
+    caption_attrs_match = _CAPTION_ATTRS_PATTERN.match(attrs_text)
+    content_start_pos = (
+        first_line_max_pos + 1 if caption_attrs_match is None else attrs_max_pos + 1
+    )
+    attrs = (
+        caption_attrs_match.group("attrs") if caption_attrs_match is not None else None
+    )
+    if not isinstance(attrs, str):
+        attrs = None
+
+    while end_match is None and max_line <= end_line:
+        line_text, end_pos, _ = _src_in_line(state, max_line)
+        if _CAPTION_END_PATTERN.match(line_text) is None:
+            max_line += 1
+        else:
+            end_match = max_line
+
+    return max_line, state.src[content_start_pos:end_pos], attrs
+
+
+def _material_captions(
+    state: StateBlock,
+    start_line: int,
+    end_line: int,
+    silent: bool,
+) -> bool:
+    """Detect caption blocks and wrap them in a token."""
+    if is_code_block(state, start_line):
+        return False
+
+    first_line_text, _, first_line_max_pos = _src_in_line(state, start_line)
+    start_match = _CAPTION_START_PATTERN.match(first_line_text)
+    if start_match is None:
+        return False
+
+    if silent:
+        return True
+
+    max_line, content, attrs = _parse(state, first_line_max_pos, start_line, end_line)
+
+    with (
+        new_token(state, PYMD_CAPTIONS_PREFIX, "figcaption") as token,
+        new_token(state, "", "p"),
+    ):
+        token.info = start_match.group("type") + "caption"
+        token.meta = {"number": start_match.group("number")}
+        if attrs is not None:
+            token.meta["attrs"] = attrs
+        tkn_inline = state.push("inline", "", 0)
+        tkn_inline.content = content.strip()
+        tkn_inline.map = [start_line, max_line]
+        tkn_inline.children = []
+
+    state.line = max_line + 1
+
+    return True
+
+
+def pymd_captions_plugin(md: MarkdownIt) -> None:
+    md.block.ruler.before(
+        "fence",
+        PYMD_CAPTIONS_PREFIX,
+        _material_captions,
+        {"alt": ["paragraph"]},
+    )

mdformat_mkdocs/plugin.py

@@ -16,6 +16,7 @@
     MKDOCSTRINGS_CROSSREFERENCE_PREFIX,
     MKDOCSTRINGS_HEADING_AUTOREFS_PREFIX,
     PYMD_ABBREVIATIONS_PREFIX,
+    PYMD_CAPTIONS_PREFIX,
     PYMD_SNIPPET_PREFIX,
     PYTHON_MARKDOWN_ATTR_LIST_PREFIX,
     material_admon_plugin,
@@ -24,6 +25,7 @@
     mkdocstrings_crossreference_plugin,
     pymd_abbreviations_plugin,
     pymd_admon_plugin,
+    pymd_captions_plugin,
     pymd_snippet_plugin,
     python_markdown_attr_list_plugin,
 )
@@ -78,6 +80,7 @@ def add_cli_argument_group(group: argparse._ArgumentGroup) -> None:
 def update_mdit(mdit: MarkdownIt) -> None:
     """Update the parser."""
     mdit.use(material_admon_plugin)
+    mdit.use(pymd_captions_plugin)
     mdit.use(material_content_tabs_plugin)
     mdit.use(mkdocstrings_autorefs_plugin)
     mdit.use(pymd_abbreviations_plugin)
@@ -179,6 +182,19 @@ def add_extra_admon_newline(node: RenderTreeNode, context: RenderContext) -> str
     return f"{title}\n\n{''.join(content)}"
 
 
+def render_pymd_caption(node: RenderTreeNode, context: RenderContext) -> str:
+    """Render caption with normalized format."""
+    caption_type = node.info or "caption"
+    attrs = node.meta.get("attrs")
+    number = node.meta.get("number")
+    rendered_content = "".join(
+        child.render(context) for child in node.children[0].children
+    )
+    caption_number = f" | {number}" if number else ""
+    caption_attrs = f"\n    {attrs}" if attrs else ""
+    return f"/// {caption_type}{caption_number}{caption_attrs}\n{rendered_content}\n///"
+
+
 # A mapping from syntax tree node type to a function that renders it.
 # This can be used to overwrite renderer functions of existing syntax
 # or add support for new syntax.
@@ -189,6 +205,7 @@ def add_extra_admon_newline(node: RenderTreeNode, context: RenderContext) -> str
     "admonition_mkdocs_title": render_admon_title,
     "content_tab_mkdocs": add_extra_admon_newline,
     "content_tab_mkdocs_title": render_admon_title,
+    PYMD_CAPTIONS_PREFIX: render_pymd_caption,
     MKDOCSTRINGS_AUTOREFS_PREFIX: _render_meta_content,
     MKDOCSTRINGS_CROSSREFERENCE_PREFIX: _render_cross_reference,
     MKDOCSTRINGS_HEADING_AUTOREFS_PREFIX: _render_heading_autoref,

tests/format/fixtures/text.md

@@ -1735,3 +1735,22 @@ Don't wrap long URLs (fixes: https://github.com/KyleKing/mdformat-mkdocs/issues/
 - a [with space](https://github.com/python/mypy/blob/a3ce6d5307e99a1b6c181eaa7c5cf134c53b7d/test-data/check-protocols)
 - b
 .
+
+Format captions correctly
+.
+|a|b|
+|-|-|
+|c|d|
+
+///  table-caption    |  1.5.2
+A table with letters.
+///
+.
+|a|b|
+|-|-|
+|c|d|
+
+/// table-caption | 1.5.2
+A table with letters.
+///
+.

tests/format/test_wrap.py

@@ -188,6 +188,39 @@
 {: .class1 .class2 .class3 .class4 .class5 .class6 .class7 .class8 .class9 .class10 .class11 .class12 .class13 .class14 .class15 .class16 .class17 .class18 .class19 .class20 }
 """
 
+CASE_CAPTION_WRAP = """
+This line is longer than 40 characters and should be wrapped.
+
+```
+def gcd(a, b):
+  if a == 0: return b
+  elif b == 0: return a
+  if a > b: return gcd(a % b, b)
+  else: return gcd(a, b % a)
+```
+
+///   caption
+Greatest common divisor algorithm.
+///
+"""
+
+CASE_CAPTION_WRAP_TRUE_40 = """
+This line is longer than 40 characters
+and should be wrapped.
+
+```
+def gcd(a, b):
+  if a == 0: return b
+  elif b == 0: return a
+  if a > b: return gcd(a % b, b)
+  else: return gcd(a, b % a)
+```
+
+/// caption
+Greatest common divisor algorithm.
+///
+"""
+
 
 @pytest.mark.parametrize(
     ("text", "expected", "align_lists", "wrap"),
@@ -200,6 +233,7 @@
         (WITH_CODE, WITH_CODE_TRUE_80, True, 80),
         (WITH_ATTR_LIST, WITH_ATTR_LIST_TRUE_80, True, 80),
         (CASE_ATTR_LIST_WRAP, CASE_ATTR_LIST_WRAP_TRUE_80, True, 80),
+        (CASE_CAPTION_WRAP, CASE_CAPTION_WRAP_TRUE_40, True, 40),
     ],
     ids=[
         "CASE_1_FALSE_40",
@@ -210,6 +244,7 @@
         "WITH_CODE_TRUE_80",
         "WITH_ATTR_LIST_TRUE_80",
         "CASE_ATTR_LIST_WRAP_TRUE_80",
+        "CASE_CAPTION_WRAP_TRUE_40",
     ],
 )
 def test_wrap(text: str, expected: str, align_lists: bool, wrap: int):

tests/render/fixtures/pymd_captions.md

@@ -0,0 +1,23 @@
+pymdown captions (https://github.com/KyleKing/mdformat-mkdocs/issues/51)
+.
+# Captions
+
+Captioned content.
+
+/// caption
+First caption.
+///
+
+/// caption
+Second caption.
+///
+.
+<h1>Captions</h1>
+<p>Captioned content.</p>
+<figcaption>
+<p>First caption.</p>
+</figcaption>
+<figcaption>
+<p>Second caption.</p>
+</figcaption>
+.

tests/render/test_render.py

@@ -10,6 +10,7 @@
     mkdocstrings_autorefs_plugin,
     mkdocstrings_crossreference_plugin,
     pymd_abbreviations_plugin,
+    pymd_captions_plugin,
     pymd_snippet_plugin,
     python_markdown_attr_list_plugin,
 )
@@ -32,6 +33,7 @@ def with_plugin(filename, plugins):
         ),
         *with_plugin("mkdocstrings_autorefs.md", [mkdocstrings_autorefs_plugin]),
         *with_plugin("pymd_abbreviations.md", [pymd_abbreviations_plugin]),
+        *with_plugin("pymd_captions.md", [pymd_captions_plugin]),
         *with_plugin(
             "mkdocstrings_crossreference.md",
             [mkdocstrings_crossreference_plugin],