GH-72: Standardize markdown support

examples/basic_site.py

@@ -30,7 +30,7 @@
     Rule(
         REMatcher(r'.*\.md'),
         [OutputDirPathCalc('.html'), None],
-        JinjaMarkdownStep()
+        JinjaMarkdownStep(frontmatter_parser='simple')
     ),
     # Copy everything else in static/ directories through.
     Rule(

examples/basic_site/index.md

@@ -1,5 +1,7 @@
+---
 template: base.jinja.html
 footnote: Generated with Anchovy!
+---
 
 # Welcome to Example Site 1!
 

examples/code_index.py

@@ -28,8 +28,8 @@
 
 
 MARKDOWN_TEMPLATE = """---
-template = "base.jinja.html"
-footnote = "Generated from {path} by Anchovy."
+template: base.jinja.html
+footnote: Generated from {path} by Anchovy.
 ---
 # {path}
 

examples/gallery.py

@@ -31,7 +31,7 @@
     Rule(
         REMatcher(r'.*\.md'),
         [OutputDirPathCalc('.html'), None],
-        JinjaMarkdownStep()
+        JinjaMarkdownStep(frontmatter_parser='toml')
     ),
     # Convert JPG files to WebP...
     Rule(

examples/gallery/index.md

@@ -1,11 +1,13 @@
-template: base.jinja.html
-footnote: Generated with Anchovy!
+---
+template = "base.jinja.html"
+footnote = "Generated with Anchovy!"
+---
 
 # Photo Gallery
 
 <div class="gallery">
-    <a href="static/images/home-sweet-home-cdk.webp"><img src="static/images/home-sweet-home-cdk.thumb.png" alt="A Vintage Automobile"></a>
-    <a href="static/images/nice-wheels-cdk.webp"><img src="static/images/nice-wheels-cdk.thumb.png" alt="A Old Wood-Sided House"></a>
-    <a href="static/images/stormy-cdk.webp"><img src="static/images/stormy-cdk.thumb.png" alt="A Boardwalk and Damaged Pines During a Storm"></a>
-    <a href="static/images/quayside-newcastle.webp"><img src="static/images/quayside-newcastle.thumb.png" alt="A Quay in Newcastle, England"></a>
+    <a href="static/images/home-sweet-home-cdk.webp"><img src="static/images/home-sweet-home-cdk.thumb.webp" alt="A Vintage Automobile"></a>
+    <a href="static/images/nice-wheels-cdk.webp"><img src="static/images/nice-wheels-cdk.thumb.webp" alt="A Old Wood-Sided House"></a>
+    <a href="static/images/stormy-cdk.webp"><img src="static/images/stormy-cdk.thumb.webp" alt="A Boardwalk and Damaged Pines During a Storm"></a>
+    <a href="static/images/quayside-newcastle.webp"><img src="static/images/quayside-newcastle.thumb.webp" alt="A Quay in Newcastle, England"></a>
 </div>

pyproject.toml

@@ -71,9 +71,9 @@ all = ["anchovy[base]"]
 # ignore this option.
 cq = [
     "anchovy[all]",
-    "commonmark>=0.9.1", "markdown>=3.4.3", "mistletoe>=1.1.0",
     "tqdm>=4.65.0",
     "minify-html>=0.11.1",
+    "ruamel.yaml>=0.18.5",
     "pylint",
     "pyright",
     "pytest",

src/anchovy/components/md_frontmatter.py

@@ -0,0 +1,51 @@
+import sys
+import typing as t
+
+
+def simple_frontmatter_parser(content: str) -> dict:
+    """
+    Read metadata from the front of a markdown-formatted text in a very simple
+    YAML-like format, without value parsing.
+    """
+    meta = {}
+    lines = content.splitlines()
+
+    for line in lines:
+        if ':' not in line:
+            break
+        key, value = line.split(':', 1)
+        if not key.isidentifier():
+            break
+        meta[key.strip()] = value.strip()
+
+    print(meta, '...')
+    return meta
+
+
+def get_toml_frontmatter_parser():
+    if sys.version_info < (3, 11):
+        import tomli as tomllib
+    else:
+        import tomllib
+    return tomllib.loads
+
+
+def get_yaml_frontmatter_parser():
+    from ruamel.yaml import YAML
+    return YAML(typ='safe').load
+
+
+FrontMatterParser = t.Callable[[str], dict]
+FrontMatterParserName = t.Literal['simple', 'toml', 'yaml']
+
+FRONTMATTER_PARSER_FACTORIES: dict[FrontMatterParserName, t.Callable[[], FrontMatterParser]] = {
+    'simple': lambda: simple_frontmatter_parser,
+    'toml': get_toml_frontmatter_parser,
+    'yaml': get_yaml_frontmatter_parser,
+}
+
+
+def get_frontmatter_parser(parser) -> FrontMatterParser:
+    if callable(parser):
+        return parser
+    return FRONTMATTER_PARSER_FACTORIES[parser]()

src/anchovy/components/md_rendering.py

@@ -3,16 +3,12 @@
 """
 from __future__ import annotations
 
-import sys
 import typing as t
 
 from markdown_it.common.utils import escapeHtml, unescapeAll
 from markdown_it.renderer import RendererHTML
 
-if sys.version_info < (3, 11):
-    import tomli as tomllib
-else:
-    import tomllib
+from .md_frontmatter import simple_frontmatter_parser, FrontMatterParser
 
 if t.TYPE_CHECKING:
     from collections.abc import Sequence
@@ -60,6 +56,10 @@ class AnchovyRendererHTML(RendererHTML):
     A customized markdown-it-py HTML renderer, with hooks for better pygments
     integration and toml frontmatter support.
     """
+    def __init__(self, parser: t.Any = None):
+        super().__init__(parser)
+        self.front_matter_parser: FrontMatterParser = simple_frontmatter_parser
+
     # https://github.com/executablebooks/markdown-it-py/issues/256
     def fence(self, tokens: Sequence[Token], idx: int, options: OptionsDict, env: EnvType):
         """
@@ -76,10 +76,13 @@ def fence(self, tokens: Sequence[Token], idx: int, options: OptionsDict, env: En
             or escapeHtml(token.content)
         )
 
+    def set_front_matter_parser(self, parser: FrontMatterParser):
+        self.front_matter_parser = parser
+
     def front_matter(self, tokens: Sequence[Token], idx: int, _options: OptionsDict, env: EnvType):
         """
         Handles parsing markdown frontmatter using TOML.
         """
-        parsed = tomllib.loads(tokens[idx].content)
+        parsed = self.front_matter_parser(tokens[idx].content)
         env['anchovy_meta'].update(parsed)
         return ''

src/anchovy/custody.py

@@ -63,6 +63,7 @@ class Custodian:
     Class for managing custody info and intelligent rebuilds.
     """
     encoding = 'utf-8'
+    newline = '\n'
     context: 'Context'
 
     def __init__(self,
@@ -193,7 +194,7 @@ def dump_file(self, path: Path):
             'graph': self.graph,
             'meta': self.meta,
         }
-        with path.open('w', encoding=self.encoding) as file:
+        with path.open('w', encoding=self.encoding, newline=self.newline) as file:
             json.dump(data, file, indent=2)
 
     def update_meta(self, entry: CustodyEntry):