Use *[path]/ so the PDF recipe follows nested directory indexes

*[file] cannot cross a slash, so *[file]/ only admits a single directory level and
silently drops PDFs linked under nested paths (dated archives, doc trees). *[path]/
spans slashes and follows directory-index pages at any depth, with no over-admission
(only trailing-slash URLs match, never files).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Signed-off-by: Xavier Roche <roche@httrack.com>

.github/workflows/ci.yml

@@ -618,31 +618,3 @@ jobs:
               echo "Fix locally with: git clang-format --binary clang-format-19 $base"
               exit 1 ;;
           esac
-
-  man-page-sync:
-    name: man page / html in sync
-    if: github.event_name == 'pull_request'
-    runs-on: ubuntu-24.04
-    steps:
-      - uses: actions/checkout@v6
-        with:
-          fetch-depth: 0
-
-      # html/httrack.man.html is groff-rendered from man/httrack.1 and committed.
-      # Rendering needs the full groff html device, so CI can't regenerate it;
-      # instead require the two to move together: a PR that touches httrack.1
-      # must also touch the html, catching the "regenerated roff, forgot html".
-      - name: httrack.1 changes must include html/httrack.man.html
-        run: |
-          set -euo pipefail
-          git fetch --no-tags origin \
-            "+refs/heads/${{ github.base_ref }}:refs/remotes/origin/${{ github.base_ref }}"
-          base="origin/${{ github.base_ref }}"
-          changed="$(git diff --name-only "$base"...HEAD)"
-          has() { printf '%s\n' "$changed" | grep -qx "$1"; }
-          if has man/httrack.1 && ! has html/httrack.man.html; then
-            echo "::error::man/httrack.1 changed but html/httrack.man.html did not."
-            echo "Regenerate it with: make -C man regen-man-html (needs the full groff package)."
-            exit 1
-          fi
-          echo "man/html sync OK."

html/cmdguide.html

@@ -404,14 +404,16 @@ host and stops the crawl; start from the final URL, or add
 rule wins.</small></p>
 
 <h4>Download the PDFs on a site</h4>
-<p><tt>httrack https://example.com/ --path mydir</tt><br>
-<small>There is no PDF-only crawl. HTTrack discovers PDF links by parsing the
-site's HTML pages, so a <tt>"-*" "+example.com/*.pdf"</tt> filter blocks the very
-pages that carry the links and grabs only the PDFs linked from the front page. Let
-it crawl the site: the HTML pages come along as the scaffolding, and every reachable
-PDF is saved with them. If some PDFs live on another host (a CDN or a docs
-subdomain), allow that host too, for example
-<tt>"+docs.example.com/*.pdf"</tt>.</small></p>
+<p><tt>httrack https://example.com/ "-*" "+https://example.com/*.html" "+https://example.com/*[path]/" "+https://example.com/*.pdf" --path mydir</tt><br>
+<small>HTTrack finds PDFs by parsing the site's HTML, so a plain
+<tt>"-*" "+example.com/*.pdf"</tt> is wrong: it prunes the pages that carry the
+links and keeps only PDFs reachable from the front page. Instead admit the HTML as
+scaffolding (<tt>*.html</tt> and <tt>*[path]/</tt> for directory-index pages at any
+depth, e.g. <tt>docs/</tt> or <tt>a/b/deep/</tt>; <tt>*[file]/</tt> would stop at one
+level), keep the PDFs, and let <tt>-*</tt> drop everything else (images,
+archives, off-site assets). PDFs on another host (a CDN or docs subdomain) are not
+included by default; allow that host too, e.g. <tt>"+docs.example.com/*.pdf"</tt>,
+or widen to <tt>"+*.pdf"</tt> for PDFs anywhere.</small></p>
 
 <h4>Keep page requisites, including off-host images</h4>
 <p><tt>httrack https://example.com/blog/ --near --path mydir</tt><br>

man/Makefile.am

@@ -16,11 +16,6 @@ regen-man: makeman.sh $(top_builddir)/src/httrack$(EXEEXT)
 
 # Render html/httrack.man.html from httrack.1. Needs the groff html device
 # (Debian: full "groff" package, not "groff-base"). Run by hand: make -C man regen-man-html
-# Strip groff's version-stamp and creation-date comments so the committed file
-# doesn't churn across groff versions or rebuilds; the ci man-page-sync guard
-# only requires this file to change whenever httrack.1 does.
 regen-man-html: httrack.1
-	groff -t -man -Thtml $(srcdir)/httrack.1 \
-		| sed -e 's/groff version [0-9][0-9.]*/groff/' -e '/^<!-- CreationDate:/d' \
-		> $(top_srcdir)/html/httrack.man.html
+	groff -t -man -Thtml $(srcdir)/httrack.1 > $(top_srcdir)/html/httrack.man.html
 .PHONY: regen-man-html