this repo has no description
river
+1327
-13
+35
-1
stack/river/cmd/river_cmd.ml
+35
-1
stack/river/cmd/river_cmd.ml
···
580
580
Log.err (fun m -> m "Failed to export merged feed: %s" err);
581
581
1
582
582
) $ format_arg $ title_arg $ limit_arg)
583
+
584
+
let html =
585
+
let output_dir_arg =
586
+
let doc = "Output directory for HTML site" in
587
+
Arg.(required & pos 0 (some string) None & info [] ~docv:"OUTPUT_DIR" ~doc)
588
+
in
589
+
let title_arg =
590
+
let doc = "Site title" in
591
+
Arg.(value & opt string "River Feed" & info ["title"; "t"] ~doc)
592
+
in
593
+
let posts_per_page_arg =
594
+
let doc = "Number of posts per page (default: 25)" in
595
+
Arg.(value & opt int 25 & info ["posts-per-page"; "p"] ~doc)
596
+
in
597
+
Term.(const (fun output_dir_str title posts_per_page env _xdg _profile ->
598
+
let state = River.State.create env ~app_name:"river" in
599
+
let output_dir = Eio.Path.(env#fs / output_dir_str) in
600
+
match River.State.export_html_site state ~output_dir ~title ~posts_per_page () with
601
+
| Ok () ->
602
+
Log.info (fun m -> m "HTML site generated in %s" output_dir_str);
603
+
0
604
+
| Error err ->
605
+
Log.err (fun m -> m "Failed to generate HTML site: %s" err);
606
+
1
607
+
) $ output_dir_arg $ title_arg $ posts_per_page_arg)
608
+
583
609
let main_cmd =
584
610
let doc = "River feed management CLI" in
585
611
let main_info = Cmd.info "river-cli" ~version:"1.0" ~doc in
···
615
641
~service:"river"
616
642
merge
617
643
in
618
-
Cmd.group main_info [user_cmd; sync_cmd; list_cmd; info_cmd; merge_cmd]
644
+
let html_cmd =
645
+
Eiocmd.run
646
+
~use_keyeio:false
647
+
~info:(Cmd.info "html" ~doc:"Generate a static HTML site from all feeds")
648
+
~app_name:"river"
649
+
~service:"river"
650
+
html
651
+
in
652
+
Cmd.group main_info [user_cmd; sync_cmd; list_cmd; info_cmd; merge_cmd; html_cmd]
+1
-1
stack/river/lib/dune
+1
-1
stack/river/lib/dune
···
1
1
(library
2
2
(name river)
3
3
(public_name river)
4
-
(libraries eio eio_main requests requests_json_api logs str syndic lambdasoup uri ptime jsonfeed jsont jsont.bytesrw xdge cmdliner eiocmd fmt sortal))
4
+
(libraries eio eio_main requests requests_json_api logs str syndic lambdasoup uri ptime jsonfeed jsont jsont.bytesrw xdge cmdliner eiocmd fmt sortal cmarkit))
+10
-3
stack/river/lib/feed.ml
+10
-3
stack/river/lib/feed.ml
···
54
54
Log.debug (fun m -> m "Successfully parsed as JSONFeed");
55
55
Json jsonfeed
56
56
| Error err ->
57
-
Log.debug (fun m -> m "Not a JSONFeed: %s" (Jsont.Error.to_string err));
57
+
let err_str = Jsont.Error.to_string err in
58
+
Log.debug (fun m -> m "Not a JSONFeed: %s" err_str);
58
59
(* Fall through to XML parsing *)
59
-
failwith "Not a valid JSONFeed"
60
+
failwith (Printf.sprintf "Not a valid JSONFeed: %s" err_str)
60
61
) else (
61
62
(* Try XML formats *)
62
63
try
···
110
111
failwith (Printf.sprintf "HTTP %d: %s" status truncated_msg)
111
112
in
112
113
113
-
let content = classify_feed ~xmlbase response in
114
+
let content =
115
+
try classify_feed ~xmlbase response
116
+
with Failure msg ->
117
+
Log.err (fun m -> m "Failed to parse feed '%s' (%s): %s"
118
+
(Source.name source) (Source.url source) msg);
119
+
raise (Failure msg)
120
+
in
114
121
let title =
115
122
match content with
116
123
| Atom atom -> Text_extract.string_of_text_construct atom.Syndic.Atom.title
+653
-1
stack/river/lib/format.ml
+653
-1
stack/river/lib/format.ml
···
43
43
| None -> Ptime.of_float_s (Unix.gettimeofday ()) |> Option.get
44
44
| Some d -> d
45
45
in
46
-
Syndic.Atom.entry ~content ~contributors ~links ~id ~authors ~title ~updated
46
+
let categories =
47
+
List.map (fun tag -> Syndic.Atom.category tag) (Post.tags post)
48
+
in
49
+
Syndic.Atom.entry ~content ~contributors ~links ~id ~authors ~title ~updated ~categories
47
50
()
48
51
49
52
let entries_of_posts posts = List.map entry_of_post posts
···
137
140
| Feed.Json jf -> Some jf
138
141
| _ -> None
139
142
end
143
+
144
+
module Html = struct
145
+
(** HTML static site generation. *)
146
+
147
+
let css = {|
148
+
* { margin: 0; padding: 0; box-sizing: border-box; }
149
+
150
+
body {
151
+
font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Helvetica, Arial, sans-serif;
152
+
line-height: 1.5;
153
+
color: #333;
154
+
background: #fff;
155
+
max-width: 800px;
156
+
margin: 0 auto;
157
+
padding: 15px;
158
+
}
159
+
160
+
header {
161
+
border-bottom: 1px solid #e1e4e8;
162
+
padding-bottom: 10px;
163
+
margin-bottom: 20px;
164
+
}
165
+
166
+
header h1 {
167
+
font-size: 22px;
168
+
font-weight: 600;
169
+
margin-bottom: 6px;
170
+
}
171
+
172
+
header h1 a {
173
+
color: #333;
174
+
text-decoration: none;
175
+
}
176
+
177
+
nav {
178
+
font-size: 13px;
179
+
}
180
+
181
+
nav a {
182
+
color: #586069;
183
+
text-decoration: none;
184
+
margin-right: 12px;
185
+
}
186
+
187
+
nav a:hover {
188
+
color: #0366d6;
189
+
}
190
+
191
+
.post {
192
+
margin-bottom: 25px;
193
+
padding-bottom: 20px;
194
+
border-bottom: 1px solid #e1e4e8;
195
+
}
196
+
197
+
.post:last-child {
198
+
border-bottom: none;
199
+
}
200
+
201
+
.post-title {
202
+
font-size: 18px;
203
+
font-weight: 600;
204
+
margin-bottom: 5px;
205
+
line-height: 1.3;
206
+
}
207
+
208
+
.post-title a {
209
+
color: #0366d6;
210
+
text-decoration: none;
211
+
}
212
+
213
+
.post-title a:hover {
214
+
text-decoration: underline;
215
+
}
216
+
217
+
.post-meta {
218
+
font-size: 12px;
219
+
color: #586069;
220
+
margin-bottom: 8px;
221
+
display: flex;
222
+
align-items: center;
223
+
gap: 8px;
224
+
}
225
+
226
+
.post-meta a {
227
+
color: #586069;
228
+
text-decoration: none;
229
+
}
230
+
231
+
.post-meta a:hover {
232
+
color: #0366d6;
233
+
}
234
+
235
+
.author-thumbnail {
236
+
width: 24px;
237
+
height: 24px;
238
+
border-radius: 50%;
239
+
object-fit: cover;
240
+
}
241
+
242
+
.post-meta-text {
243
+
flex: 1;
244
+
}
245
+
246
+
.post-excerpt {
247
+
font-size: 14px;
248
+
color: #24292e;
249
+
line-height: 1.5;
250
+
}
251
+
252
+
.post-excerpt p {
253
+
margin-bottom: 8px;
254
+
}
255
+
256
+
.post-excerpt ul, .post-excerpt ol {
257
+
margin-left: 20px;
258
+
margin-bottom: 8px;
259
+
}
260
+
261
+
.post-excerpt li {
262
+
margin-bottom: 3px;
263
+
}
264
+
265
+
.post-excerpt code {
266
+
background: #f6f8fa;
267
+
padding: 2px 4px;
268
+
border-radius: 3px;
269
+
font-size: 13px;
270
+
}
271
+
272
+
.post-excerpt img {
273
+
float: right;
274
+
width: 35%;
275
+
max-width: 300px;
276
+
margin: 0 0 10px 15px;
277
+
border-radius: 4px;
278
+
cursor: pointer;
279
+
transition: opacity 0.2s;
280
+
}
281
+
282
+
.post-excerpt img:hover {
283
+
opacity: 0.9;
284
+
}
285
+
286
+
@media (max-width: 600px) {
287
+
.post-excerpt img {
288
+
float: none;
289
+
width: 100%;
290
+
max-width: 100%;
291
+
margin: 10px 0;
292
+
}
293
+
}
294
+
295
+
.lightbox {
296
+
display: none;
297
+
position: fixed;
298
+
top: 0;
299
+
left: 0;
300
+
width: 100%;
301
+
height: 100%;
302
+
background: rgba(0, 0, 0, 0.9);
303
+
z-index: 1000;
304
+
cursor: pointer;
305
+
align-items: center;
306
+
justify-content: center;
307
+
}
308
+
309
+
.lightbox.active {
310
+
display: flex;
311
+
}
312
+
313
+
.lightbox img {
314
+
max-width: 95%;
315
+
max-height: 95%;
316
+
object-fit: contain;
317
+
}
318
+
319
+
.post-full-content {
320
+
display: none;
321
+
font-size: 14px;
322
+
color: #24292e;
323
+
line-height: 1.5;
324
+
margin-top: 10px;
325
+
}
326
+
327
+
.post-full-content.active {
328
+
display: block;
329
+
}
330
+
331
+
.post-full-content p {
332
+
margin-bottom: 10px;
333
+
}
334
+
335
+
.post-full-content ul, .post-full-content ol {
336
+
margin-left: 20px;
337
+
margin-bottom: 10px;
338
+
}
339
+
340
+
.post-full-content li {
341
+
margin-bottom: 4px;
342
+
}
343
+
344
+
.post-full-content h1, .post-full-content h2, .post-full-content h3 {
345
+
margin-top: 15px;
346
+
margin-bottom: 8px;
347
+
}
348
+
349
+
.post-full-content h1 {
350
+
font-size: 18px;
351
+
font-weight: 600;
352
+
}
353
+
354
+
.post-full-content h2 {
355
+
font-size: 16px;
356
+
font-weight: 600;
357
+
}
358
+
359
+
.post-full-content h3 {
360
+
font-size: 15px;
361
+
font-weight: 600;
362
+
}
363
+
364
+
.post-full-content code {
365
+
background: #f6f8fa;
366
+
padding: 2px 4px;
367
+
border-radius: 3px;
368
+
font-size: 13px;
369
+
}
370
+
371
+
.post-full-content pre {
372
+
background: #f6f8fa;
373
+
padding: 10px;
374
+
border-radius: 4px;
375
+
overflow-x: auto;
376
+
margin-bottom: 10px;
377
+
}
378
+
379
+
.post-full-content pre code {
380
+
background: none;
381
+
padding: 0;
382
+
}
383
+
384
+
.post-full-content blockquote {
385
+
border-left: 3px solid #e1e4e8;
386
+
padding-left: 12px;
387
+
margin: 10px 0;
388
+
color: #586069;
389
+
}
390
+
391
+
.post-full-content img {
392
+
max-width: 100%;
393
+
height: auto;
394
+
margin: 10px 0;
395
+
border-radius: 4px;
396
+
}
397
+
398
+
.read-more {
399
+
display: inline-block;
400
+
color: #0366d6;
401
+
font-size: 13px;
402
+
cursor: pointer;
403
+
text-decoration: none;
404
+
margin-top: 8px;
405
+
padding: 4px 8px;
406
+
border: 1px solid #e1e4e8;
407
+
border-radius: 3px;
408
+
background: #f6f8fa;
409
+
transition: background 0.2s;
410
+
}
411
+
412
+
.read-more:hover {
413
+
background: #e1e4e8;
414
+
}
415
+
416
+
.read-more::after {
417
+
content: ' ▼';
418
+
font-size: 10px;
419
+
}
420
+
421
+
.read-more.active::after {
422
+
content: ' ▲';
423
+
}
424
+
425
+
.post-tags {
426
+
margin-top: 8px;
427
+
font-size: 11px;
428
+
clear: both;
429
+
}
430
+
431
+
.post-tags a {
432
+
display: inline-block;
433
+
background: #f1f8ff;
434
+
color: #0366d6;
435
+
padding: 2px 6px;
436
+
border-radius: 3px;
437
+
text-decoration: none;
438
+
margin-right: 4px;
439
+
margin-bottom: 4px;
440
+
}
441
+
442
+
.post-tags a:hover {
443
+
background: #dbedff;
444
+
}
445
+
446
+
.pagination {
447
+
margin-top: 30px;
448
+
padding-top: 15px;
449
+
border-top: 1px solid #e1e4e8;
450
+
text-align: center;
451
+
font-size: 13px;
452
+
}
453
+
454
+
.pagination a {
455
+
color: #0366d6;
456
+
text-decoration: none;
457
+
margin: 0 8px;
458
+
}
459
+
460
+
.pagination a:hover {
461
+
text-decoration: underline;
462
+
}
463
+
464
+
.pagination .current {
465
+
color: #24292e;
466
+
font-weight: 600;
467
+
}
468
+
469
+
.link-item {
470
+
margin-bottom: 15px;
471
+
padding-bottom: 12px;
472
+
border-bottom: 1px solid #e1e4e8;
473
+
}
474
+
475
+
.link-item:last-child {
476
+
border-bottom: none;
477
+
}
478
+
479
+
.link-url {
480
+
font-size: 14px;
481
+
margin-bottom: 3px;
482
+
}
483
+
484
+
.link-url a {
485
+
color: #0366d6;
486
+
text-decoration: none;
487
+
word-break: break-all;
488
+
}
489
+
490
+
.link-url a:hover {
491
+
text-decoration: underline;
492
+
}
493
+
494
+
.link-meta {
495
+
font-size: 11px;
496
+
color: #586069;
497
+
}
498
+
499
+
.link-meta a {
500
+
color: #586069;
501
+
text-decoration: none;
502
+
}
503
+
504
+
.link-meta a:hover {
505
+
color: #0366d6;
506
+
}
507
+
508
+
.author-list, .category-list {
509
+
list-style: none;
510
+
}
511
+
512
+
.author-list li, .category-list li {
513
+
margin-bottom: 12px;
514
+
padding-bottom: 12px;
515
+
border-bottom: 1px solid #e1e4e8;
516
+
}
517
+
518
+
.author-list li:last-child, .category-list li:last-child {
519
+
border-bottom: none;
520
+
}
521
+
522
+
.author-list a, .category-list a {
523
+
color: #0366d6;
524
+
text-decoration: none;
525
+
font-size: 15px;
526
+
}
527
+
528
+
.author-list a:hover, .category-list a:hover {
529
+
text-decoration: underline;
530
+
}
531
+
532
+
.count {
533
+
color: #586069;
534
+
font-size: 12px;
535
+
margin-left: 6px;
536
+
}
537
+
538
+
footer {
539
+
margin-top: 40px;
540
+
padding-top: 15px;
541
+
border-top: 1px solid #e1e4e8;
542
+
text-align: center;
543
+
font-size: 11px;
544
+
color: #586069;
545
+
}
546
+
|}
547
+
548
+
let html_escape s =
549
+
let buf = Buffer.create (String.length s) in
550
+
String.iter (function
551
+
| '<' -> Buffer.add_string buf "<"
552
+
| '>' -> Buffer.add_string buf ">"
553
+
| '&' -> Buffer.add_string buf "&"
554
+
| '"' -> Buffer.add_string buf """
555
+
| '\'' -> Buffer.add_string buf "'"
556
+
| c -> Buffer.add_char buf c
557
+
) s;
558
+
Buffer.contents buf
559
+
560
+
let format_date date =
561
+
let open Unix in
562
+
let tm = gmtime (Ptime.to_float_s date) in
563
+
let months = [|"January"; "February"; "March"; "April"; "May"; "June";
564
+
"July"; "August"; "September"; "October"; "November"; "December"|] in
565
+
Printf.sprintf "%s %d, %d" months.(tm.tm_mon) tm.tm_mday (1900 + tm.tm_year)
566
+
567
+
let page_template ~title ~nav_current content =
568
+
Printf.sprintf {|<!DOCTYPE html>
569
+
<html lang="en">
570
+
<head>
571
+
<meta charset="UTF-8">
572
+
<meta name="viewport" content="width=device-width, initial-scale=1.0">
573
+
<title>%s</title>
574
+
<style>%s</style>
575
+
</head>
576
+
<body>
577
+
<header>
578
+
<h1><a href="index.html">River Feed</a></h1>
579
+
<nav>
580
+
<a href="index.html"%s>Posts</a>
581
+
<a href="authors/index.html"%s>Authors</a>
582
+
<a href="categories/index.html"%s>Categories</a>
583
+
<a href="links.html"%s>Links</a>
584
+
</nav>
585
+
</header>
586
+
<main>
587
+
%s
588
+
</main>
589
+
<footer>
590
+
Generated by River Feed Aggregator
591
+
</footer>
592
+
<div class="lightbox" id="lightbox">
593
+
<img id="lightbox-img" src="" alt="">
594
+
</div>
595
+
<script>
596
+
(function() {
597
+
const lightbox = document.getElementById('lightbox');
598
+
const lightboxImg = document.getElementById('lightbox-img');
599
+
600
+
// Add click handler to all images in excerpts and full content
601
+
document.addEventListener('click', function(e) {
602
+
if (e.target.tagName === 'IMG' && (e.target.closest('.post-excerpt') || e.target.closest('.post-full-content'))) {
603
+
e.preventDefault();
604
+
lightboxImg.src = e.target.src;
605
+
lightboxImg.alt = e.target.alt;
606
+
lightbox.classList.add('active');
607
+
}
608
+
});
609
+
610
+
// Close lightbox on click
611
+
lightbox.addEventListener('click', function() {
612
+
lightbox.classList.remove('active');
613
+
lightboxImg.src = '';
614
+
});
615
+
616
+
// Close on escape key
617
+
document.addEventListener('keydown', function(e) {
618
+
if (e.key === 'Escape' && lightbox.classList.contains('active')) {
619
+
lightbox.classList.remove('active');
620
+
lightboxImg.src = '';
621
+
}
622
+
});
623
+
624
+
// Read more toggle
625
+
document.addEventListener('click', function(e) {
626
+
if (e.target.classList.contains('read-more')) {
627
+
e.preventDefault();
628
+
const post = e.target.closest('.post');
629
+
const fullContent = post.querySelector('.post-full-content');
630
+
const excerpt = post.querySelector('.post-excerpt');
631
+
632
+
if (fullContent.classList.contains('active')) {
633
+
fullContent.classList.remove('active');
634
+
excerpt.style.display = 'block';
635
+
e.target.textContent = 'Read more';
636
+
e.target.classList.remove('active');
637
+
} else {
638
+
fullContent.classList.add('active');
639
+
excerpt.style.display = 'none';
640
+
e.target.textContent = 'Show less';
641
+
e.target.classList.add('active');
642
+
}
643
+
}
644
+
});
645
+
})();
646
+
</script>
647
+
</body>
648
+
</html>|}
649
+
(html_escape title)
650
+
css
651
+
(if nav_current = "posts" then " class=\"current\"" else "")
652
+
(if nav_current = "authors" then " class=\"current\"" else "")
653
+
(if nav_current = "categories" then " class=\"current\"" else "")
654
+
(if nav_current = "links" then " class=\"current\"" else "")
655
+
content
656
+
657
+
let pagination_html ~current_page ~total_pages ~base_path =
658
+
if total_pages <= 1 then ""
659
+
else
660
+
let prev = if current_page > 1 then
661
+
let prev_page = current_page - 1 in
662
+
let href = if prev_page = 1 then base_path ^ "index.html"
663
+
else Printf.sprintf "%spage-%d.html" base_path prev_page in
664
+
Printf.sprintf {|<a href="%s">← Previous</a>|} href
665
+
else ""
666
+
in
667
+
let next = if current_page < total_pages then
668
+
Printf.sprintf {|<a href="%spage-%d.html">Next →</a>|} base_path (current_page + 1)
669
+
else ""
670
+
in
671
+
let pages =
672
+
let buf = Buffer.create 256 in
673
+
for i = 1 to total_pages do
674
+
if i = current_page then
675
+
Buffer.add_string buf (Printf.sprintf {| <span class="current">%d</span>|} i)
676
+
else
677
+
let href = if i = 1 then base_path ^ "index.html"
678
+
else Printf.sprintf "%spage-%d.html" base_path i in
679
+
Buffer.add_string buf (Printf.sprintf {| <a href="%s">%d</a>|} href i)
680
+
done;
681
+
Buffer.contents buf
682
+
in
683
+
Printf.sprintf {|<div class="pagination">%s%s%s</div>|} prev pages next
684
+
685
+
let full_content_from_html html_content =
686
+
(* Convert HTML to markdown then to clean HTML using Cmarkit *)
687
+
let markdown = Html_markdown.html_to_markdown html_content in
688
+
let doc = Cmarkit.Doc.of_string markdown in
689
+
Cmarkit_html.of_doc ~safe:true doc
690
+
691
+
let post_excerpt_from_html html_content ~max_length =
692
+
(* Convert HTML to markdown for excerpt *)
693
+
let markdown = Html_markdown.html_to_markdown html_content in
694
+
(* Find paragraph break after max_length *)
695
+
let excerpt_md =
696
+
if String.length markdown <= max_length then markdown
697
+
else
698
+
(* Look for double newline (paragraph break) after max_length *)
699
+
let start_search = min max_length (String.length markdown - 1) in
700
+
let rec find_para_break pos =
701
+
if pos >= String.length markdown - 1 then
702
+
String.length markdown
703
+
else if pos < String.length markdown - 1 &&
704
+
markdown.[pos] = '\n' && markdown.[pos + 1] = '\n' then
705
+
pos
706
+
else
707
+
find_para_break (pos + 1)
708
+
in
709
+
let break_pos = find_para_break start_search in
710
+
let truncated = String.sub markdown 0 break_pos in
711
+
if break_pos < String.length markdown then
712
+
truncated ^ "..."
713
+
else
714
+
truncated
715
+
in
716
+
(* Convert markdown back to HTML using Cmarkit with custom renderer *)
717
+
let doc = Cmarkit.Doc.of_string excerpt_md in
718
+
719
+
(* Custom renderer that makes headings smaller and inline *)
720
+
let inline_headings =
721
+
let block c = function
722
+
| Cmarkit.Block.Heading (h, _) ->
723
+
let level = Cmarkit.Block.Heading.level h in
724
+
let inline = Cmarkit.Block.Heading.inline h in
725
+
(* Render heading as a strong tag with smaller font *)
726
+
let style = match level with
727
+
| 1 -> "font-size: 15px; font-weight: 600;"
728
+
| 2 -> "font-size: 14px; font-weight: 600;"
729
+
| _ -> "font-size: 14px; font-weight: 500;"
730
+
in
731
+
Cmarkit_renderer.Context.string c (Printf.sprintf "<strong style=\"%s\">" style);
732
+
Cmarkit_renderer.Context.inline c inline;
733
+
Cmarkit_renderer.Context.string c "</strong> ";
734
+
true
735
+
| _ -> false
736
+
in
737
+
Cmarkit_renderer.make ~block ()
738
+
in
739
+
740
+
let renderer = Cmarkit_renderer.compose (Cmarkit_html.renderer ~safe:true ()) inline_headings in
741
+
Cmarkit_renderer.doc_to_string renderer doc
742
+
743
+
let render_post_html ~post ~author_username =
744
+
let title = Post.title post in
745
+
let author = Post.author post in
746
+
let date_str = match Post.date post with
747
+
| Some d -> format_date d
748
+
| None -> "No date"
749
+
in
750
+
let link_html = match Post.link post with
751
+
| Some uri ->
752
+
Printf.sprintf {|<a href="%s">%s</a>|}
753
+
(html_escape (Uri.to_string uri))
754
+
(html_escape title)
755
+
| None -> html_escape title
756
+
in
757
+
let excerpt = post_excerpt_from_html (Post.content post) ~max_length:300 in
758
+
let tags_html =
759
+
match Post.tags post with
760
+
| [] -> ""
761
+
| tags ->
762
+
let tag_links = List.map (fun tag ->
763
+
Printf.sprintf {|<a href="../categories/%s.html">%s</a>|}
764
+
(html_escape tag) (html_escape tag)
765
+
) tags in
766
+
Printf.sprintf {|<div class="post-tags">%s</div>|}
767
+
(String.concat "" tag_links)
768
+
in
769
+
Printf.sprintf {|<article class="post">
770
+
<h2 class="post-title">%s</h2>
771
+
<div class="post-meta">
772
+
By <a href="../authors/%s.html">%s</a> on %s
773
+
</div>
774
+
<div class="post-excerpt">
775
+
%s
776
+
</div>
777
+
%s
778
+
</article>|}
779
+
link_html
780
+
(html_escape author_username)
781
+
(html_escape author)
782
+
date_str
783
+
excerpt
784
+
tags_html
785
+
786
+
let render_posts_page ~title ~posts ~current_page ~total_pages ~base_path ~nav_current =
787
+
let posts_html = String.concat "\n" posts in
788
+
let pagination = pagination_html ~current_page ~total_pages ~base_path in
789
+
let content = posts_html ^ "\n" ^ pagination in
790
+
page_template ~title ~nav_current content
791
+
end
+54
stack/river/lib/format.mli
+54
stack/river/lib/format.mli
···
101
101
102
102
Returns None if the feed is not JSONFeed. *)
103
103
end
104
+
105
+
module Html : sig
106
+
(** HTML static site generation. *)
107
+
108
+
val format_date : Ptime.t -> string
109
+
(** [format_date date] formats a date in human-readable format (e.g., "November 23, 2025"). *)
110
+
111
+
val html_escape : string -> string
112
+
(** [html_escape s] escapes HTML special characters in string. *)
113
+
114
+
val full_content_from_html : string -> string
115
+
(** [full_content_from_html html_content] converts HTML content to clean markdown-derived HTML.
116
+
117
+
@param html_content The HTML content to convert *)
118
+
119
+
val post_excerpt_from_html : string -> max_length:int -> string
120
+
(** [post_excerpt_from_html html_content ~max_length] generates an excerpt from HTML content.
121
+
122
+
Converts HTML to markdown, truncates to max_length, and converts back to simple HTML.
123
+
124
+
@param html_content The HTML content to excerpt
125
+
@param max_length Maximum length of the excerpt in characters *)
126
+
127
+
val render_post_html : post:Post.t -> author_username:string -> string
128
+
(** [render_post_html ~post ~author_username] renders a single post as HTML.
129
+
130
+
@param post The post to render
131
+
@param author_username The username of the author (for linking) *)
132
+
133
+
val render_posts_page :
134
+
title:string ->
135
+
posts:string list ->
136
+
current_page:int ->
137
+
total_pages:int ->
138
+
base_path:string ->
139
+
nav_current:string ->
140
+
string
141
+
(** [render_posts_page ~title ~posts ~current_page ~total_pages ~base_path ~nav_current]
142
+
renders a complete HTML page with posts and pagination.
143
+
144
+
@param title Page title
145
+
@param posts List of pre-rendered post HTML strings
146
+
@param current_page Current page number (1-indexed)
147
+
@param total_pages Total number of pages
148
+
@param base_path Base path for pagination links (e.g., "" for root, "authors/" for author pages)
149
+
@param nav_current Which nav item is current ("posts", "authors", "categories", "links") *)
150
+
151
+
val page_template : title:string -> nav_current:string -> string -> string
152
+
(** [page_template ~title ~nav_current content] wraps content in the HTML page template.
153
+
154
+
@param title Page title
155
+
@param nav_current Which nav item is current
156
+
@param content The main content HTML *)
157
+
end
+9
-2
stack/river/lib/html_markdown.ml
+9
-2
stack/river/lib/html_markdown.ml
···
29
29
Soup.fold (fun acc link ->
30
30
match Soup.attribute "href" link with
31
31
| Some href ->
32
-
let text = Soup.texts link |> String.concat "" |> String.trim in
33
-
(href, text) :: acc
32
+
(* Filter out local anchors and only include absolute external URLs *)
33
+
let uri = Uri.of_string href in
34
+
let is_absolute = Uri.scheme uri <> None in
35
+
let is_local_anchor = String.starts_with ~prefix:"#" href in
36
+
if is_absolute && not is_local_anchor then
37
+
let text = Soup.texts link |> String.concat "" |> String.trim in
38
+
(href, text) :: acc
39
+
else
40
+
acc
34
41
| None -> acc
35
42
) [] links
36
43
|> List.rev
+19
stack/river/lib/river.mli
+19
stack/river/lib/river.mli
···
419
419
@param format Output format
420
420
@param limit Optional maximum number of entries *)
421
421
422
+
val export_html_site :
423
+
t ->
424
+
output_dir:Eio.Fs.dir_ty Eio.Path.t ->
425
+
title:string ->
426
+
?posts_per_page:int ->
427
+
unit ->
428
+
(unit, string) result
429
+
(** [export_html_site state ~output_dir ~title ()] exports a static HTML site.
430
+
431
+
Generates a complete static site with:
432
+
- Paginated post listings
433
+
- Author index and individual author pages
434
+
- Category index and individual category pages
435
+
- Links page showing all outgoing links from posts
436
+
437
+
@param output_dir Directory to write HTML files to
438
+
@param title Site title
439
+
@param posts_per_page Number of posts per page (default: 25) *)
440
+
422
441
(** {2 Analysis} *)
423
442
424
443
val analyze_user_quality :
+527
-5
stack/river/lib/state.ml
+527
-5
stack/river/lib/state.ml
···
207
207
let fetched_feeds =
208
208
Eio.Fiber.List.filter_map (fun source ->
209
209
try
210
-
Log.info (fun m -> m " Fetching %s (%s)..."
211
-
(Source.name source) (Source.url source));
210
+
Log.info (fun m -> m " [%s] Fetching %s (%s)..."
211
+
username (Source.name source) (Source.url source));
212
212
Some (Feed.fetch session source)
213
213
with e ->
214
-
Log.err (fun m -> m " Failed to fetch %s: %s"
215
-
(Source.name source) (Printexc.to_string e));
214
+
Log.err (fun m -> m " [%s] Failed to fetch %s: %s"
215
+
username (Source.name source) (Printexc.to_string e));
216
216
None
217
217
) (User.feeds user)
218
218
in
···
414
414
415
415
let export_merged_feed state ~title ~format ?limit () =
416
416
let all_posts = get_all_posts state ?limit () in
417
-
let entries = List.map snd all_posts in
417
+
418
+
(* Rewrite author metadata from Sortal user info *)
419
+
let rewrite_entry_author username (entry : Syndic.Atom.entry) =
420
+
match Storage.get_user state username with
421
+
| None -> entry
422
+
| Some user ->
423
+
(* Get user's full name and email from Sortal *)
424
+
let fullname = User.fullname user in
425
+
let email = User.email user in
426
+
let username = User.username user in
427
+
428
+
(* Create new author with Sortal information *)
429
+
let new_author =
430
+
match email with
431
+
| Some email_addr ->
432
+
Syndic.Atom.author ~email:email_addr ~uri:(Uri.of_string ("https://" ^ username)) fullname
433
+
| None ->
434
+
Syndic.Atom.author ~uri:(Uri.of_string ("https://" ^ username)) fullname
435
+
in
436
+
437
+
(* Update entry with new author, keeping existing contributors *)
438
+
let _, other_authors = entry.authors in
439
+
{ entry with authors = (new_author, other_authors) }
440
+
in
441
+
442
+
let entries = List.map (fun (username, entry) ->
443
+
rewrite_entry_author username entry
444
+
) all_posts in
418
445
419
446
match format with
420
447
| `Atom ->
···
429
456
| Error err -> Error (Printf.sprintf "Failed to serialize JSON Feed: %s" (Jsont.Error.to_string err))
430
457
else
431
458
Export.export_jsonfeed ~title entries
459
+
460
+
let export_html_site state ~output_dir ~title ?(posts_per_page = 25) () =
461
+
try
462
+
Log.info (fun m -> m "=== Starting HTML site generation ===");
463
+
Log.info (fun m -> m "Output directory: %s" (Eio.Path.native_exn output_dir));
464
+
Log.info (fun m -> m "Site title: %s" title);
465
+
Log.info (fun m -> m "Posts per page: %d" posts_per_page);
466
+
467
+
(* Sanitize a string for use in filenames - replace unsafe characters *)
468
+
let sanitize_filename s =
469
+
let buf = Buffer.create (String.length s) in
470
+
String.iter (fun c ->
471
+
match c with
472
+
| '/' | '\\' | ':' | '*' | '?' | '"' | '<' | '>' | '|' -> Buffer.add_char buf '-'
473
+
| ' ' -> Buffer.add_char buf '-'
474
+
| c -> Buffer.add_char buf c
475
+
) s;
476
+
Buffer.contents buf
477
+
in
478
+
479
+
(* Create directory structure *)
480
+
Log.info (fun m -> m "Creating directory structure");
481
+
let mkdir_if_not_exists dir =
482
+
try Eio.Path.mkdir ~perm:0o755 dir
483
+
with Eio.Io (Eio.Fs.E (Already_exists _), _) -> ()
484
+
in
485
+
mkdir_if_not_exists output_dir;
486
+
mkdir_if_not_exists Eio.Path.(output_dir / "authors");
487
+
mkdir_if_not_exists Eio.Path.(output_dir / "categories");
488
+
mkdir_if_not_exists Eio.Path.(output_dir / "thumbnails");
489
+
Log.info (fun m -> m "Directory structure created");
490
+
491
+
(* Helper to get and copy author thumbnail *)
492
+
let get_author_thumbnail username =
493
+
Log.debug (fun m -> m "Looking up thumbnail for username: %s" username);
494
+
match Sortal.lookup state.sortal username with
495
+
| Some contact ->
496
+
Log.debug (fun m -> m " Found Sortal contact for %s: %s" username (Sortal.Contact.name contact));
497
+
(match Sortal.thumbnail_path state.sortal contact with
498
+
| Some src_path ->
499
+
Log.info (fun m -> m " Copying thumbnail for %s from: %s" username (Eio.Path.native_exn src_path));
500
+
(* Copy thumbnail to output directory *)
501
+
let filename = Filename.basename (Eio.Path.native_exn src_path) in
502
+
let dest_path = Eio.Path.(output_dir / "thumbnails" / filename) in
503
+
(try
504
+
Log.debug (fun m -> m " Source path: %s" (Eio.Path.native_exn src_path));
505
+
Log.debug (fun m -> m " Destination path: %s" (Eio.Path.native_exn dest_path));
506
+
let content = Eio.Path.load src_path in
507
+
Eio.Path.save ~create:(`Or_truncate 0o644) dest_path content;
508
+
Log.info (fun m -> m " Successfully copied thumbnail to: thumbnails/%s" filename);
509
+
Some ("thumbnails/" ^ filename)
510
+
with e ->
511
+
Log.warn (fun m -> m " Failed to copy thumbnail for %s: %s" username (Printexc.to_string e));
512
+
None)
513
+
| None ->
514
+
Log.debug (fun m -> m " No thumbnail set for %s" username);
515
+
None)
516
+
| None ->
517
+
Log.warn (fun m -> m " No Sortal contact found for username: %s" username);
518
+
None
519
+
in
520
+
521
+
(* Helper to convert Atom entry to a simple record for HTML generation *)
522
+
let entry_to_html_data username (entry : Syndic.Atom.entry) =
523
+
let title = Text_extract.string_of_text_construct entry.title in
524
+
let link = List.find_opt (fun (l : Syndic.Atom.link) ->
525
+
l.rel = Syndic.Atom.Alternate
526
+
) entry.links in
527
+
let link_uri = match link with
528
+
| Some l -> Some l.href
529
+
| None -> if List.length entry.links > 0 then Some (List.hd entry.links).href else None
530
+
in
531
+
let content_html = match entry.content with
532
+
| Some (Syndic.Atom.Text s) -> s
533
+
| Some (Syndic.Atom.Html (_, s)) -> s
534
+
| Some (Syndic.Atom.Xhtml (_, nodes)) ->
535
+
String.concat "" (List.map Syndic.XML.to_string nodes)
536
+
| Some (Syndic.Atom.Mime _) | Some (Syndic.Atom.Src _) | None -> ""
537
+
in
538
+
let author, _ = entry.authors in
539
+
let tags = List.map (fun (c : Syndic.Atom.category) -> c.term) entry.categories in
540
+
(username, title, author.name, entry.updated, link_uri, content_html, tags)
541
+
in
542
+
543
+
(* Get all posts *)
544
+
Log.info (fun m -> m "Retrieving all posts from state");
545
+
let all_posts = get_all_posts state () in
546
+
let html_data = List.map (fun (username, entry) ->
547
+
entry_to_html_data username entry
548
+
) all_posts in
549
+
550
+
let unique_users = List.sort_uniq String.compare (List.map (fun (u, _, _, _, _, _, _) -> u) html_data) in
551
+
Log.info (fun m -> m "Retrieved %d posts from %d users" (List.length html_data) (List.length unique_users));
552
+
Log.info (fun m -> m "Users: %s" (String.concat ", " unique_users));
553
+
554
+
(* Generate main index pages with pagination *)
555
+
let total_posts = List.length html_data in
556
+
let total_pages = (total_posts + posts_per_page - 1) / posts_per_page in
557
+
Log.info (fun m -> m "Generating main index: %d posts across %d pages" total_posts total_pages);
558
+
559
+
for page = 1 to total_pages do
560
+
Log.info (fun m -> m " Generating index page %d/%d" page total_pages);
561
+
let start_idx = (page - 1) * posts_per_page in
562
+
let page_posts = List.filteri (fun i _ ->
563
+
i >= start_idx && i < start_idx + posts_per_page
564
+
) html_data in
565
+
566
+
let post_htmls = List.map (fun (username, title, author, date, link, content, tags) ->
567
+
Log.debug (fun m -> m " Processing post: %s by %s (@%s)" title author username);
568
+
(* Create a temporary Post-like structure for rendering *)
569
+
(* We'll need to adapt this since we're working with Atom entries *)
570
+
let post_html =
571
+
let date_str = Format.Html.format_date date in
572
+
let link_html = match link with
573
+
| Some uri ->
574
+
Printf.sprintf {|<a href="%s">%s</a>|}
575
+
(Format.Html.html_escape (Uri.to_string uri))
576
+
(Format.Html.html_escape title)
577
+
| None -> Format.Html.html_escape title
578
+
in
579
+
let excerpt = Format.Html.post_excerpt_from_html content ~max_length:300 in
580
+
let full_content = Format.Html.full_content_from_html content in
581
+
let tags_html =
582
+
match tags with
583
+
| [] -> ""
584
+
| _ ->
585
+
let tag_links = List.map (fun tag ->
586
+
Printf.sprintf {|<a href="categories/%s.html">%s</a>|}
587
+
(Format.Html.html_escape (sanitize_filename tag)) (Format.Html.html_escape tag)
588
+
) tags in
589
+
Printf.sprintf {|<div class="post-tags">%s</div>|}
590
+
(String.concat "" tag_links)
591
+
in
592
+
let thumbnail_html = match get_author_thumbnail username with
593
+
| Some thumb_path ->
594
+
Printf.sprintf {|<img src="%s" alt="%s" class="author-thumbnail">|}
595
+
(Format.Html.html_escape thumb_path)
596
+
(Format.Html.html_escape author)
597
+
| None -> ""
598
+
in
599
+
Printf.sprintf {|<article class="post">
600
+
<h2 class="post-title">%s</h2>
601
+
<div class="post-meta">
602
+
%s<div class="post-meta-text">By <a href="authors/%s.html">%s</a> on %s</div>
603
+
</div>
604
+
<div class="post-excerpt">
605
+
%s
606
+
</div>
607
+
<div class="post-full-content">
608
+
%s
609
+
</div>
610
+
<a href="#" class="read-more">Read more</a>
611
+
%s
612
+
</article>|}
613
+
link_html
614
+
thumbnail_html
615
+
(Format.Html.html_escape (sanitize_filename username))
616
+
(Format.Html.html_escape author)
617
+
date_str
618
+
excerpt
619
+
full_content
620
+
tags_html
621
+
in
622
+
post_html
623
+
) page_posts in
624
+
625
+
let page_html = Format.Html.render_posts_page
626
+
~title
627
+
~posts:post_htmls
628
+
~current_page:page
629
+
~total_pages
630
+
~base_path:""
631
+
~nav_current:"posts"
632
+
in
633
+
634
+
let filename = if page = 1 then "index.html"
635
+
else Printf.sprintf "page-%d.html" page in
636
+
Eio.Path.save ~create:(`Or_truncate 0o644)
637
+
Eio.Path.(output_dir / filename)
638
+
page_html
639
+
done;
640
+
641
+
(* Generate author index *)
642
+
Log.info (fun m -> m "Generating author index and pages");
643
+
let authors_map = Hashtbl.create 32 in
644
+
List.iter (fun (username, _, author, _, _, _, _) ->
645
+
let count = match Hashtbl.find_opt authors_map username with
646
+
| Some (_, c) -> c + 1
647
+
| None -> 1
648
+
in
649
+
Hashtbl.replace authors_map username (author, count)
650
+
) html_data;
651
+
652
+
let authors_list = Hashtbl.fold (fun username (author, count) acc ->
653
+
(username, author, count) :: acc
654
+
) authors_map [] |> List.sort (fun (_, a1, _) (_, a2, _) -> String.compare a1 a2) in
655
+
656
+
Log.info (fun m -> m "Found %d authors" (List.length authors_list));
657
+
658
+
let authors_index_content =
659
+
let items = List.map (fun (username, author, count) ->
660
+
Printf.sprintf {|<li><a href="%s.html">%s</a><span class="count">%d post%s</span></li>|}
661
+
(Format.Html.html_escape (sanitize_filename username))
662
+
(Format.Html.html_escape author)
663
+
count
664
+
(if count = 1 then "" else "s")
665
+
) authors_list in
666
+
Printf.sprintf "<ul class=\"author-list\">\n%s\n</ul>"
667
+
(String.concat "\n" items)
668
+
in
669
+
670
+
let authors_index_html = Format.Html.page_template
671
+
~title:(title ^ " - Authors")
672
+
~nav_current:"authors"
673
+
authors_index_content
674
+
in
675
+
Eio.Path.save ~create:(`Or_truncate 0o644)
676
+
Eio.Path.(output_dir / "authors" / "index.html")
677
+
authors_index_html;
678
+
679
+
(* Generate individual author pages *)
680
+
Hashtbl.iter (fun username (author, _) ->
681
+
let author_posts = List.filter (fun (u, _, _, _, _, _, _) -> u = username) html_data in
682
+
let author_total = List.length author_posts in
683
+
let author_pages = (author_total + posts_per_page - 1) / posts_per_page in
684
+
Log.info (fun m -> m " Author: %s (@%s) - %d posts, %d pages" author username author_total author_pages);
685
+
686
+
for page = 1 to author_pages do
687
+
let start_idx = (page - 1) * posts_per_page in
688
+
let page_posts = List.filteri (fun i _ ->
689
+
i >= start_idx && i < start_idx + posts_per_page
690
+
) author_posts in
691
+
692
+
let post_htmls = List.map (fun (_username, title, author, date, link, content, tags) ->
693
+
let date_str = Format.Html.format_date date in
694
+
let link_html = match link with
695
+
| Some uri ->
696
+
Printf.sprintf {|<a href="%s">%s</a>|}
697
+
(Format.Html.html_escape (Uri.to_string uri))
698
+
(Format.Html.html_escape title)
699
+
| None -> Format.Html.html_escape title
700
+
in
701
+
let excerpt = Format.Html.post_excerpt_from_html content ~max_length:300 in
702
+
let full_content = Format.Html.full_content_from_html content in
703
+
let tags_html =
704
+
match tags with
705
+
| [] -> ""
706
+
| _ ->
707
+
let tag_links = List.map (fun tag ->
708
+
Printf.sprintf {|<a href="../categories/%s.html">%s</a>|}
709
+
(Format.Html.html_escape (sanitize_filename tag)) (Format.Html.html_escape tag)
710
+
) tags in
711
+
Printf.sprintf {|<div class="post-tags">%s</div>|}
712
+
(String.concat "" tag_links)
713
+
in
714
+
Printf.sprintf {|<article class="post">
715
+
<h2 class="post-title">%s</h2>
716
+
<div class="post-meta">
717
+
By %s on %s
718
+
</div>
719
+
<div class="post-excerpt">
720
+
%s
721
+
</div>
722
+
<div class="post-full-content">
723
+
%s
724
+
</div>
725
+
<a href="#" class="read-more">Read more</a>
726
+
%s
727
+
</article>|}
728
+
link_html
729
+
(Format.Html.html_escape author)
730
+
date_str
731
+
excerpt
732
+
full_content
733
+
tags_html
734
+
) page_posts in
735
+
736
+
let page_html = Format.Html.render_posts_page
737
+
~title:(author ^ " - " ^ title)
738
+
~posts:post_htmls
739
+
~current_page:page
740
+
~total_pages:author_pages
741
+
~base_path:(sanitize_filename username ^ "-")
742
+
~nav_current:"authors"
743
+
in
744
+
745
+
let safe_username = sanitize_filename username in
746
+
let filename = if page = 1 then safe_username ^ ".html"
747
+
else Printf.sprintf "%s-%d.html" safe_username page in
748
+
Eio.Path.save ~create:(`Or_truncate 0o644)
749
+
Eio.Path.(output_dir / "authors" / filename)
750
+
page_html
751
+
done
752
+
) authors_map;
753
+
754
+
(* Generate category index and pages *)
755
+
Log.info (fun m -> m "Generating category index and pages");
756
+
let categories_map = Hashtbl.create 32 in
757
+
List.iter (fun (_, _, _, _, _, _, tags) ->
758
+
List.iter (fun tag ->
759
+
let count = match Hashtbl.find_opt categories_map tag with
760
+
| Some c -> c + 1
761
+
| None -> 1
762
+
in
763
+
Hashtbl.replace categories_map tag count
764
+
) tags
765
+
) html_data;
766
+
767
+
let categories_list = Hashtbl.fold (fun tag count acc ->
768
+
(tag, count) :: acc
769
+
) categories_map [] |> List.sort (fun (t1, _) (t2, _) -> String.compare t1 t2) in
770
+
771
+
Log.info (fun m -> m "Found %d categories" (List.length categories_list));
772
+
773
+
let categories_index_content =
774
+
let items = List.map (fun (tag, count) ->
775
+
Printf.sprintf {|<li><a href="%s.html">%s</a><span class="count">%d post%s</span></li>|}
776
+
(Format.Html.html_escape (sanitize_filename tag))
777
+
(Format.Html.html_escape tag)
778
+
count
779
+
(if count = 1 then "" else "s")
780
+
) categories_list in
781
+
Printf.sprintf "<ul class=\"category-list\">\n%s\n</ul>"
782
+
(String.concat "\n" items)
783
+
in
784
+
785
+
let categories_index_html = Format.Html.page_template
786
+
~title:(title ^ " - Categories")
787
+
~nav_current:"categories"
788
+
categories_index_content
789
+
in
790
+
Eio.Path.save ~create:(`Or_truncate 0o644)
791
+
Eio.Path.(output_dir / "categories" / "index.html")
792
+
categories_index_html;
793
+
794
+
(* Generate individual category pages *)
795
+
List.iter (fun (tag, count) ->
796
+
let tag_posts = List.filter (fun (_, _, _, _, _, _, tags) ->
797
+
List.mem tag tags
798
+
) html_data in
799
+
800
+
let tag_total = List.length tag_posts in
801
+
let tag_pages = (tag_total + posts_per_page - 1) / posts_per_page in
802
+
Log.info (fun m -> m " Category: %s - %d posts, %d pages" tag count tag_pages);
803
+
804
+
for page = 1 to tag_pages do
805
+
let start_idx = (page - 1) * posts_per_page in
806
+
let page_posts = List.filteri (fun i _ ->
807
+
i >= start_idx && i < start_idx + posts_per_page
808
+
) tag_posts in
809
+
810
+
let post_htmls = List.map (fun (username, title, author, date, link, content, tags) ->
811
+
let date_str = Format.Html.format_date date in
812
+
let link_html = match link with
813
+
| Some uri ->
814
+
Printf.sprintf {|<a href="%s">%s</a>|}
815
+
(Format.Html.html_escape (Uri.to_string uri))
816
+
(Format.Html.html_escape title)
817
+
| None -> Format.Html.html_escape title
818
+
in
819
+
let excerpt = Format.Html.post_excerpt_from_html content ~max_length:300 in
820
+
let full_content = Format.Html.full_content_from_html content in
821
+
let tags_html =
822
+
match tags with
823
+
| [] -> ""
824
+
| _ ->
825
+
let tag_links = List.map (fun t ->
826
+
Printf.sprintf {|<a href="%s.html">%s</a>|}
827
+
(Format.Html.html_escape (sanitize_filename t)) (Format.Html.html_escape t)
828
+
) tags in
829
+
Printf.sprintf {|<div class="post-tags">%s</div>|}
830
+
(String.concat "" tag_links)
831
+
in
832
+
Printf.sprintf {|<article class="post">
833
+
<h2 class="post-title">%s</h2>
834
+
<div class="post-meta">
835
+
By <a href="../authors/%s.html">%s</a> on %s
836
+
</div>
837
+
<div class="post-excerpt">
838
+
%s
839
+
</div>
840
+
<div class="post-full-content">
841
+
%s
842
+
</div>
843
+
<a href="#" class="read-more">Read more</a>
844
+
%s
845
+
</article>|}
846
+
link_html
847
+
(Format.Html.html_escape (sanitize_filename username))
848
+
(Format.Html.html_escape author)
849
+
date_str
850
+
excerpt
851
+
full_content
852
+
tags_html
853
+
) page_posts in
854
+
855
+
let page_html = Format.Html.render_posts_page
856
+
~title:(tag ^ " - " ^ title)
857
+
~posts:post_htmls
858
+
~current_page:page
859
+
~total_pages:tag_pages
860
+
~base_path:(sanitize_filename tag ^ "-")
861
+
~nav_current:"categories"
862
+
in
863
+
864
+
let safe_tag = sanitize_filename tag in
865
+
let filename = if page = 1 then safe_tag ^ ".html"
866
+
else Printf.sprintf "%s-%d.html" safe_tag page in
867
+
Eio.Path.save ~create:(`Or_truncate 0o644)
868
+
Eio.Path.(output_dir / "categories" / filename)
869
+
page_html
870
+
done
871
+
) categories_list;
872
+
873
+
(* Generate links page *)
874
+
Log.info (fun m -> m "Generating links page");
875
+
let all_links = List.concat_map (fun (username, title, author, date, post_link, content, _) ->
876
+
let links = Html_markdown.extract_links content in
877
+
List.map (fun (href, link_text) ->
878
+
(href, link_text, username, author, title, post_link, date)
879
+
) links
880
+
) html_data in
881
+
882
+
Log.info (fun m -> m " Extracted %d total links from all posts" (List.length all_links));
883
+
884
+
(* Group by URL and track most recent post date *)
885
+
let links_map = Hashtbl.create 256 in
886
+
List.iter (fun (href, link_text, username, author, post_title, post_link, date) ->
887
+
let existing = Hashtbl.find_opt links_map href in
888
+
let new_entry = (link_text, username, author, post_title, post_link, date) in
889
+
match existing with
890
+
| None -> Hashtbl.add links_map href [new_entry]
891
+
| Some entries ->
892
+
(* Add to list, will sort by date later *)
893
+
Hashtbl.replace links_map href (new_entry :: entries)
894
+
) all_links;
895
+
896
+
(* Sort links by most recent post date *)
897
+
let sorted_links = Hashtbl.fold (fun href entries acc ->
898
+
(* Get the most recent entry for this URL *)
899
+
let sorted_entries = List.sort (fun (_, _, _, _, _, d1) (_, _, _, _, _, d2) ->
900
+
Ptime.compare d2 d1
901
+
) entries in
902
+
let most_recent = List.hd sorted_entries in
903
+
(href, most_recent, entries) :: acc
904
+
) links_map [] |> List.sort (fun (_, (_, _, _, _, _, d1), _) (_, (_, _, _, _, _, d2), _) ->
905
+
Ptime.compare d2 d1
906
+
) in
907
+
908
+
Log.info (fun m -> m " Deduplicated to %d unique links" (List.length sorted_links));
909
+
910
+
let links_content =
911
+
let items = List.map (fun (href, (link_text, username, author, post_title, post_link, date), all_entries) ->
912
+
let date_str = Format.Html.format_date date in
913
+
let display_text = if link_text = "" || link_text = href then href else link_text in
914
+
let post_link_html = match post_link with
915
+
| Some uri ->
916
+
Printf.sprintf {|<a href="%s">%s</a>|}
917
+
(Format.Html.html_escape (Uri.to_string uri))
918
+
(Format.Html.html_escape post_title)
919
+
| None -> Format.Html.html_escape post_title
920
+
in
921
+
let count_str = if List.length all_entries > 1 then
922
+
Printf.sprintf " (mentioned in %d posts)" (List.length all_entries)
923
+
else ""
924
+
in
925
+
Printf.sprintf {|<div class="link-item">
926
+
<div class="link-url"><a href="%s">%s</a></div>
927
+
<div class="link-meta">From %s by <a href="authors/%s.html">%s</a> on %s%s</div>
928
+
</div>|}
929
+
(Format.Html.html_escape href)
930
+
(Format.Html.html_escape display_text)
931
+
post_link_html
932
+
(Format.Html.html_escape (sanitize_filename username))
933
+
(Format.Html.html_escape author)
934
+
date_str
935
+
count_str
936
+
) sorted_links in
937
+
String.concat "\n" items
938
+
in
939
+
940
+
let links_html = Format.Html.page_template
941
+
~title:(title ^ " - Links")
942
+
~nav_current:"links"
943
+
links_content
944
+
in
945
+
Eio.Path.save ~create:(`Or_truncate 0o644)
946
+
Eio.Path.(output_dir / "links.html")
947
+
links_html;
948
+
949
+
Log.info (fun m -> m "HTML site generated successfully in %s"
950
+
(Eio.Path.native_exn output_dir));
951
+
Ok ()
952
+
with e ->
953
+
Error (Printf.sprintf "Failed to generate HTML site: %s" (Printexc.to_string e))
432
954
433
955
let analyze_user_quality state ~username =
434
956
match Storage.get_user state username with
+19
stack/river/lib/state.mli
+19
stack/river/lib/state.mli
···
117
117
@param format Output format
118
118
@param limit Optional maximum number of entries *)
119
119
120
+
val export_html_site :
121
+
t ->
122
+
output_dir:Eio.Fs.dir_ty Eio.Path.t ->
123
+
title:string ->
124
+
?posts_per_page:int ->
125
+
unit ->
126
+
(unit, string) result
127
+
(** [export_html_site state ~output_dir ~title ()] exports a static HTML site.
128
+
129
+
Generates a complete static site with:
130
+
- Paginated post listings
131
+
- Author index and individual author pages
132
+
- Category index and individual category pages
133
+
- Links page showing all outgoing links from posts
134
+
135
+
@param output_dir Directory to write HTML files to
136
+
@param title Site title
137
+
@param posts_per_page Number of posts per page (default: 25) *)
138
+
120
139
(** {2 Analysis} *)
121
140
122
141
val analyze_user_quality :