feat: Try to extract per-part PDF files

.github/workflows/ci.yml

@@ -50,7 +50,7 @@ jobs:
         uses: awalsh128/cache-apt-pkgs-action@latest
         with:
           packages: |
-            latexmk asymptote ghostscript
+            latexmk asymptote ghostscript qpdf
 
       - name: Restore cache
         uses: actions/cache/restore@v4
@@ -107,18 +107,24 @@ jobs:
       - name: Extract table of contents
         run: |
           python3 print-toc.py --parts --chapters --numbers > toc.txt
+      - name: Extract per-part PDF's
+        run: |
+          python3 print-toc.py --qpdf | bash
 
       - name: Create dist folder
         run: |
           mkdir -p dist
-          mkdir -p dist/asy
+          mkdir -p dist/figures
+          mkdir -p dist/cuts/
+
           cp Napkin.log dist/
           cp Napkin.pdf dist/
-          cp asy/*.asy dist/asy/
-          cp asy/*.pdf dist/asy/
+          cp asy/*.asy dist/figures/
+          cp asy/*.pdf dist/figures/
 
-          cp flowchart.png dist/
-          cp toc.txt dist/
+          mv flowchart.png dist/
+          mv toc.txt dist/
+          mv part-*.pdf dist/cuts
 
           cd dist
           tree -H '.' -I "index.html" -D --charset utf-8 --filesfirst -T "An Infinitely Large Napkin" >index.html

print-toc.py

@@ -1,41 +1,110 @@
 #!/usr/bin/env python3
 
 import argparse
+import re
 
 parser = argparse.ArgumentParser("print-toc.py")
+parser.add_argument(
+    "-q", "--qpdf", action="store_true", help="Instead produce commands for QPDF."
+)
 parser.add_argument("-p", "--parts", action="store_true", help="Show the parts.")
 parser.add_argument("-c", "--chapters", action="store_true", help="Show the chapters.")
 parser.add_argument(
     "-n", "--numbers", action="store_true", help="Include page numbers."
 )
+parser.add_argument(
+    "-l", "--links", action="store_true", help="Markdown hyperlinks for per-part links."
+)
 
 opts = parser.parse_args()
-if not (opts.parts or opts.chapters):
-    parser.error("Need to specify at least one of -p/--parts and -c/--chapters")
 
 part_chunk = r"\contentsline {part}{\numberline {"
 ch_chunk = r"\contentsline {chapter}{\numberline {"
 
+# only used for qpdf
+last_page_number = 1
+last_part_number = 0
+last_part_title = "Frontmatter"
+
+
+def slugify(s: str) -> str:
+    words = s.split(" ")[:3]
+    return "-".join(re.sub(r"[^a-z0-9 ]+", "", word.lower()) for word in words)
+
+
+def get_filename(part_number: int, part_title: str) -> str:
+    return f"part-{part_number:02d}-napkin-{slugify(part_title)}.pdf"
+
+
+def get_qpdf_invocation(
+    start_page: int,
+    end_page: int | None,
+    part_number: int,
+    part_title: str,
+) -> str:
+    return (
+        r"qpdf Napkin.pdf --pages . "
+        f"{start_page}-{end_page if end_page is not None else 'r1'} "
+        f"-- {get_filename(part_number, part_title)}"
+    )
+
+
+if opts.links:
+    filename = get_filename(last_part_number, last_part_title)
+    print(f"- [{last_part_title}](https://venhance.github.io/napkin/cuts/{filename})")
+
+
 has_started = False
 with open("Napkin.toc") as f:
     for line in f:
-        if line.startswith(part_chunk) and opts.parts:
+        if line.startswith(part_chunk):
             part_number, part_title, *junk = line.strip()[len(part_chunk) :].split("}")
             assert junk[0].startswith("{")
-            page_number = junk[0][1:]
-            if has_started is True and opts.chapters:
+            page_number = int(junk[0][1:])
+            if has_started is True and opts.chapters and not opts.qpdf:
                 print()
             else:
                 has_started = True
-            print(f"{part_number+'.':<6} {part_title}")
+            if opts.qpdf:
+                print(
+                    get_qpdf_invocation(
+                        last_page_number,
+                        page_number - 1,
+                        last_part_number,
+                        last_part_title,
+                    )
+                )
+            last_part_title = part_title
+            last_part_number += 1
+            last_page_number = page_number
+            if opts.links:
+                filename = get_filename(last_part_number, part_title)
+                print(
+                    f"- {part_number}. "
+                    f"[{part_title}](https://venhance.github.io/napkin/cuts/{filename})"
+                )
+
+            elif opts.parts:
+                print(f"{part_number+'.':<6} {part_title}")
         if line.startswith(ch_chunk) and opts.chapters:
             ch_number, ch_title, *junk = line.strip()[len(ch_chunk) :].split("}")
             assert junk[0].startswith("{")
-            page_number = junk[0][1:]
+            page_number = int(junk[0][1:])
             if ch_number.isdigit():
                 s = " " * 2 if opts.parts else ""
                 s += f"{ch_number}. {ch_title}"
                 if opts.numbers is True:
-                    s += "." * max(72 - len(s) - len(page_number), 3)
-                    s += page_number
+                    s += "." * max(72 - len(s) - len(str(page_number)), 3)
+                    s += str(page_number)
                 print(s)
+
+
+if opts.qpdf:
+    print(
+        get_qpdf_invocation(
+            last_page_number,
+            None,
+            last_part_number,
+            last_part_title,
+        )
+    )