Handle Jekyll-format files in OCaml
Fix JSON feed differences: slugs, PDFs, DOI cache
- Normalize slugs to match Jekyll behavior (dots → dashes)
so geotessera-python-0.7 becomes geotessera-python-0-7
- Add filesystem PDF check to JSON feed attachments
matching arod_papers.ml behavior for static/papers/*.pdf
- Add DOI cache infrastructure for external references:
- New bushel_doi_entry.ml module for parsing data/doi.yml
- Load DOI entries in bushel_loader.ml
- Complete note_references with external DOI URL scanning
- Scan for doi.org URLs and publisher URLs (Elsevier, IEEE, etc.)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
+12
-2
+12
-2
lib/frontmatter.ml
+12
-2
lib/frontmatter.ml
···
36
36
with _ -> None
37
37
else None
38
38
39
+
(** Normalize a slug to match Jekyll's slug_of_string behavior:
40
+
map all non-alphanumeric characters to hyphens and lowercase. *)
41
+
let normalize_slug s =
42
+
let mapped = String.map (fun c ->
43
+
match c with
44
+
| 'a'..'z' | 'A'..'Z' | '0'..'9' -> c
45
+
| _ -> '-'
46
+
) s in
47
+
String.lowercase_ascii mapped
48
+
39
49
let slug_of_fname fname =
40
50
let basename = Filename.basename fname in
41
51
let no_ext = Filename.chop_extension basename in
42
52
match parse_date_prefix no_ext with
43
-
| Some (date, slug) -> Ok (slug, Some date)
44
-
| None -> Ok (no_ext, None)
53
+
| Some (date, slug) -> Ok (normalize_slug slug, Some date)
54
+
| None -> Ok (normalize_slug no_ext, None)
45
55
46
56
(** Parse frontmatter using yamlrw's streaming parser.
47
57
Uses multi-document support to find the document boundary,