any-file to PDF (Phase 1): converters/ dispatcher, images/text/md/csv/json/xml/svg/pdf-passthrough, 112 tests green

converters/md_converter.py

@@ -0,0 +1,54 @@
+"""Markdown (.md / .markdown) -> PDF.
+
+Rendered with the `markdown` library (tables, fenced code, nl2br) and
+typeset by WeasyPrint with app.BASE_CSS injected at the top of <head> so
+the fallback layer sits under any user CSS. The first ATX H1 becomes the
+export name and PDF /Title (it stays in the rendered document for visual
+fidelity); without an H1 the file stem is used.
+
+Empty input renders 0 pages and fails in finalize with
+"no renderable content: ..." — unpinned, acceptable.
+"""
+from __future__ import annotations
+
+import re
+from pathlib import Path
+
+from . import common
+
+_FIRST_H1_RE = re.compile(r"^\s*#\s+(.+?)\s*#*\s*$", re.MULTILINE)
+
+
+def _extract_h1(md_text: str) -> str | None:
+    m = _FIRST_H1_RE.search(md_text)
+    if not m:
+        return None
+    title = re.sub(r"[#*`_]", "", m.group(1))
+    title = " ".join(title.split())
+    return title or None
+
+
+def convert(path: Path, job_id: str, progress_cb=None, output_dir=None,
+            source_name=None) -> str:
+    """Render a Markdown file to a typeset PDF."""
+    import app  # lazy: BASE_CSS lives with the EPUB pipeline
+    import markdown
+
+    text = common.decode_text(path.read_bytes())
+    title = _extract_h1(text)
+    body_html = markdown.markdown(
+        text, extensions=["tables", "fenced_code", "nl2br"]
+    )
+    html_doc = (
+        f"<!DOCTYPE html><html><head><style>\n{app.BASE_CSS}\n</style>"
+        f"</head><body>{body_html}</body></html>"
+    )
+
+    def _build():
+        return html_doc, path.name
+
+    return common.convert_single(
+        path, job_id, build_html=_build, title=title, css="",
+        progress_cb=progress_cb, output_dir=output_dir,
+        name_src=source_name or path.name,
+    )