Spaces:

Brettapps
/

FairDinkumPublishing

Runtime error

App Files Files Community

Brettapps commited on Aug 12

Commit

e05c607

verified ·

1 Parent(s): ab1e0b4

Update FairDinkumPublishing ebook pipeline

Browse files

Files changed (3) hide show

requirements.txt +0 -1
src/__pycache__/pipeline_runner.cpython-313.pyc +0 -0
src/pipeline_runner.py +162 -80

requirements.txt CHANGED Viewed

@@ -1,4 +1,3 @@
 gradio>=5.0
-ebooklib>=0.20
 fpdf2>=2.8
 Pillow>=10.0

 gradio>=5.0
 fpdf2>=2.8
 Pillow>=10.0

src/__pycache__/pipeline_runner.cpython-313.pyc ADDED Viewed

Binary file (74.8 kB). View file

src/pipeline_runner.py CHANGED Viewed

@@ -8,7 +8,7 @@ Agents:
   2. Outline Architect — part/chapter/section structure
   3. Chapter Writer  — full long-form chapters
   4. Cover Generator — Pillow front/back covers
-  5. Publisher       — EPUB (ebooklib) + PDF (fpdf2) + Obsidian vault
 Usage:
   python3 pipeline_runner.py "Port Noarlunga jetty fishing South Australia"
@@ -19,14 +19,15 @@ from pathlib import Path
 # ── pip install inline (Space env) ──────────────────────────────────────────
 def ensure_deps():
     try:
-        import ebooklib, fpdf, PIL  # noqa: F401
     except ImportError:
         subprocess.check_call([sys.executable, "-m", "pip", "install", "--quiet",
-                               "ebooklib", "fpdf2", "Pillow"])
 ensure_deps()
-import ebooklib
-from ebooklib import epub
 from fpdf import FPDF
 from PIL import Image, ImageDraw, ImageFont
@@ -527,116 +528,197 @@ def _md_to_xhtml(md_text: str) -> str:
 def build_epub(concept: dict, outline: dict, out_path: str) -> str:
-    book = epub.EpubBook()
-    book.set_identifier(concept["id"])
-    book.set_title(concept["title"])
-    book.set_language("en-AU")
-    book.add_author(concept.get("author", "Brett Anthony Sjoberg"))
-    book.add_metadata("DC", "description", concept["description"])
-    book.add_metadata("DC", "subject", ", ".join([concept["primary_keyword"]] + concept["secondary_keywords"]))
-    book.add_metadata("DC", "publisher", concept.get("publisher", "Brett Anthony Sjoberg"))
-    book.add_metadata("DC", "date", concept.get("created_at", "")[:10])
-    book.add_metadata("DC", "rights",
-        f"© {datetime.date.today().year} {concept.get('author', 'Brett Anthony Sjoberg')} — ABN {concept.get('abn', '')}")
-    book.add_metadata("DC", "coverage", "Australia")
-    book.add_metadata("DC", "type", "Text")
-    book.add_metadata(None, "meta", "", {"name": "geo", "content": "Australia"})
-    book.add_metadata(None, "meta", "", {"name": "keywords", "content": ", ".join(concept["geo_keywords"])})
-    nav_css = epub.EpubItem(uid="style_nav", file_name="style/nav.css",
-                             media_type="text/css", content=EPUB_CSS)
-    book.add_item(nav_css)
-    spine = ["nav"]
-    toc_links = []
-    def make_ch(ch_num: int, title: str, body_xhtml: str):
-        fname = f"text/ch{ch_num:02d}_{slugify(title)}.xhtml"
-        item = epub.EpubHtml(title=title, file_name=fname, lang="en-AU")
-        item.content = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>{title}</title><link rel="stylesheet" href="../style/nav.css"/></head>
 <body>{body_xhtml}</body></html>"""
-        item.add_item(nav_css)
-        book.add_item(item)
-        spine.append(fname)
-        toc_links.append(item)
     # Title page
-    tp = epub.EpubHtml(title="Title Page", file_name="text/title_page.xhtml", lang="en-AU")
-    tp.content = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>Title</title><link rel="stylesheet" href="../style/nav.css"/></head>
 <body><div class="cover-page">
 <h1>{concept['title']}</h1><h2>{concept['subtitle']}</h2>
-<p><em>by {concept.get('author', 'Authored Guide Australia')}</em></p>
 <p class='noindent'>{concept['primary_keyword']} — {', '.join(concept['geo_keywords'])}</p>
 </div></body></html>"""
-    tp.add_item(nav_css); book.add_item(tp); spine.append("text/title_page.xhtml")
-    toc_links.insert(0, tp)
-    # Cover image
-    front_jpg = COVERS / "front_cover.jpg"
-    if front_jpg.exists():
-        book.add_item(epub.EpubItem(uid="cover_img", file_name="images/cover.jpg",
-                                    media_type="image/jpeg", content=front_jpg.read_bytes()))
-        cov = epub.EpubHtml(title="Cover", file_name="text/cover.xhtml", lang="en-AU")
-        cov.content = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>Cover</title><link rel="stylesheet" href="../style/nav.css"/></head>
 <body><div class="cover-page"><img class="cover" src="../images/cover.jpg" alt="Cover"/></div></body></html>"""
-        cov.add_item(nav_css); book.add_item(cov); spine.insert(1, "text/cover.xhtml")
-    # TOC page (hyperlinked)
-    toc_x = epub.EpubHtml(title="Contents", file_name="text/toc_page.xhtml", lang="en-AU")
-    toc_html = ["<nav epub:type='toc'><h2>Table of Contents</h2><ul>"]
     for part in outline["parts"]:
-        toc_html.append(f"<li><span class='toc-part'>{part['name']}</span><ul>")
         for ch in part["chapters"]:
-            toc_html.append(f"<li><a class='toc-entry' href='text/ch{int(ch['id'][2:]):02d}_{slugify(ch['title'])}.xhtml'>{ch['title']}</a></li>")
-        toc_html.append("</ul></li>")
-    toc_html.append("</ul></nav>")
-    toc_x.content = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>Contents</title><link rel="stylesheet" href="../style/nav.css"/></head>
-<body>{chr(10).join(toc_html)}</body></html>"""
-    toc_x.add_item(nav_css); book.add_item(toc_x); spine.append("text/toc_page.xhtml")
     # Chapters
     for part in outline["parts"]:
         for ch in part["chapters"]:
-            md_path = CHAPTERS / f"{ch['id']}_{slugify(ch['title'])}.md"
             if md_path.exists():
                 body = _md_to_xhtml(md_path.read_text(encoding="utf-8"))
             else:
                 body = f"<h1>{ch['title']}</h1><p>Content coming soon.</p>"
-            make_ch(int(ch["id"][2:]), ch["title"], body)
     # Back cover
-    back = epub.EpubHtml(title="Back Cover", file_name="text/back_cover.xhtml", lang="en-AU")
-    back_jpg = COVERS / "back_cover.jpg"
-    if back_jpg.exists():
-        book.add_item(epub.EpubItem(uid="back_img", file_name="images/back_cover.jpg",
-                                    media_type="image/jpeg", content=back_jpg.read_bytes()))
-        back.content = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>Back Cover</title><link rel="stylesheet" href="../style/nav.css"/></head>
-<body><div class="back-cover"><img class="cover" src="../images/back_cover.jpg" alt="Back Cover"/></div></body></html>"""
-    else:
-        back.content = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>Back Cover</title><link rel="stylesheet" href="../style/nav.css"/></head>
-<body><div class="back-cover"><h2>About This Book</h2><p>{concept['description']}</p></div></body></html>"""
-    back.add_item(nav_css); book.add_item(back); spine.append("text/back_cover.xhtml")
-    book.toc = toc_links
-    book.add_item(epub.EpubNcx())
-    book.spine = spine
-    out_path_p = Path(out_path)
-    out_path_p.parent.mkdir(parents=True, exist_ok=True)
-    epub.write_epub(str(out_path_p), book, {"compress": True})
-    return str(out_path_p)
 # ────────────────────────────────────────────────────────────────────────────

   2. Outline Architect — part/chapter/section structure
   3. Chapter Writer  — full long-form chapters
   4. Cover Generator — Pillow front/back covers
+  5. Publisher       — EPUB (zipfile) + PDF (fpdf2) + Obsidian vault
 Usage:
   python3 pipeline_runner.py "Port Noarlunga jetty fishing South Australia"
 # ── pip install inline (Space env) ──────────────────────────────────────────
 def ensure_deps():
     try:
+        import fpdf, PIL  # noqa: F401
     except ImportError:
         subprocess.check_call([sys.executable, "-m", "pip", "install", "--quiet",
+                               "fpdf2", "Pillow"])
 ensure_deps()
+import zipfile
+from fpdf import FPDF
+from PIL import Image, ImageDraw, ImageFont
 from fpdf import FPDF
 from PIL import Image, ImageDraw, ImageFont
 def build_epub(concept: dict, outline: dict, out_path: str) -> str:
+    """Build a valid EPUB 3 package using zipfile (no fpdf2 dependency)."""
+    out = Path(out_path)
+    out.parent.mkdir(parents=True, exist_ok=True)
+    def slugify_toc(text):
+        text = text.lower()
+        text = re.sub(r"[^a-z0-9]+", "_", text)
+        text = re.sub(r"_+", "_", text).strip("_")
+        return text[:60]
+    chapters = []
+    for part in outline["parts"]:
+        for ch in part["chapters"]:
+            chapters.append(ch)
+    # ── OPF manifest entries ────────────────────────────────────────────────
+    manifest_items = [
+        '<item id="style" href="style/nav.css" media-type="text/css"/>',
+    ]
+    if (COVERS / "front_cover.jpg").exists():
+        manifest_items.append('<item id="cover" href="images/cover.jpg" media-type="image/jpeg"/>')
+    if (COVERS / "back_cover.jpg").exists():
+        manifest_items.append('<item id="back" href="images/back_cover.jpg" media-type="image/jpeg"/>')
+    spine_ids = ["titlepage"]
+    toc_entries = []
+    def add_chapter_xhtml(ch_num, title, body_xhtml, spine_id):
+        fname = f"ch{ch_num:02d}_{slugify_toc(title)}.xhtml"
+        xhtml = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>{title}</title><link rel="stylesheet" href="../style/nav.css"/></head>
 <body>{body_xhtml}</body></html>"""
+        files[f"text/{fname}"] = xhtml.encode("utf-8")
+        manifest_items.append(f'<item id="{spine_id}" href="text/{fname}" media-type="application/xhtml+xml"/>')
+        spine_ids.append(spine_id)
+        toc_entries.append((spine_id, title, f"text/{fname}"))
     # Title page
+    tp_xhtml = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>Title</title><link rel="stylesheet" href="../style/nav.css"/></head>
 <body><div class="cover-page">
 <h1>{concept['title']}</h1><h2>{concept['subtitle']}</h2>
+<p><em>by {concept.get('author', 'Brett Anthony Sjoberg')}</em></p>
 <p class='noindent'>{concept['primary_keyword']} — {', '.join(concept['geo_keywords'])}</p>
 </div></body></html>"""
+    files["text/title_page.xhtml"] = tp_xhtml.encode("utf-8")
+    manifest_items.append('<item id="titlepage" href="text/title_page.xhtml" media-type="application/xhtml+xml"/>')
+    # Cover page
+    if (COVERS / "front_cover.jpg").exists():
+        files["images/cover.jpg"] = (COVERS / "front_cover.jpg").read_bytes()
+        cov_xhtml = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>Cover</title><link rel="stylesheet" href="../style/nav.css"/></head>
 <body><div class="cover-page"><img class="cover" src="../images/cover.jpg" alt="Cover"/></div></body></html>"""
+        files["text/cover.xhtml"] = cov_xhtml.encode("utf-8")
+        manifest_items.append('<item id="coverpage" href="text/cover.xhtml" media-type="application/xhtml+xml"/>')
+        spine_ids.insert(1, "coverpage")
+    # TOC page
+    toc_xhtml_lines = ["<nav epub:type='toc'><h2>Table of Contents</h2><ul>"]
     for part in outline["parts"]:
+        toc_xhtml_lines.append(f"<li><span class='toc-part'>{part['name']}</span><ul>")
         for ch in part["chapters"]:
+            ch_num = int(ch["id"][2:])
+            href = f"text/ch{ch_num:02d}_{slugify_toc(ch['title'])}.xhtml"
+            toc_xhtml_lines.append(f"<li><a class='toc-entry' href='../{href}'>{ch['title']}</a></li>")
+        toc_xhtml_lines.append("</ul></li>")
+    toc_xhtml_lines.append("</ul></nav>")
+    toc_xhtml = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>Contents</title><link rel="stylesheet" href="../style/nav.css"/></head>
+<body>{chr(10).join(toc_xhtml_lines)}</body></html>"""
+    files["text/toc_page.xhtml"] = toc_xhtml.encode("utf-8")
+    manifest_items.append('<item id="tocpage" href="text/toc_page.xhtml" media-type="application/xhtml+xml"/>')
+    spine_ids.append("tocpage")
     # Chapters
     for part in outline["parts"]:
         for ch in part["chapters"]:
+            ch_num = int(ch["id"][2:])
+            md_path = CHAPTERS / f"{ch['id']}_{slugify_toc(ch['title'])}.md"
             if md_path.exists():
                 body = _md_to_xhtml(md_path.read_text(encoding="utf-8"))
             else:
                 body = f"<h1>{ch['title']}</h1><p>Content coming soon.</p>"
+            add_chapter_xhtml(ch_num, ch["title"], body, f"ch{ch_num:02d}")
     # Back cover
+    back_xhtml = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>Back Cover</title><link rel="stylesheet" href="../style/nav.css"/></head>
+<body><div class="back-cover">
+<h2>About This Book</h2><p>{concept['description']}</p></div></body></html>"""
+    if (COVERS / "back_cover.jpg").exists():
+        files["images/back_cover.jpg"] = (COVERS / "back_cover.jpg").read_bytes()
+        back_xhtml = f"""<?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
 <title>Back Cover</title><link rel="stylesheet" href="../style/nav.css"/></head>
+<body><div class="back-cover"><img class="cover" src="../images/back_cover.jpg" alt="Back Cover"/></div></body></html>"""
+    files["text/back_cover.xhtml"] = back_xhtml.encode("utf-8")
+    manifest_items.append('<item id="backcover" href="text/back_cover.xhtml" media-type="application/xhtml+xml"/>')
+    spine_ids.append("backcover")
+    # ── NCX ─────────────────────────────────────────────────────────────────
+    ncx_nav_points = []
+    for spine_id, title, href in toc_entries:
+        ncx_nav_points.append(f"""<navPoint id="{spine_id}" playOrder="{spine_ids.index(spine_id) + 1}">
+<navLabel><text>{title}</text></navLabel>
+<content src="{href}"/>
+</navPoint>""")
+    toc_ncx = f"""<?xml version="1.0" encoding="UTF-8"?>
+<ncx xmlns="http://www.daisy.org/z3986/2005/ncx/" version="2005-1">
+<head><meta name="dtb:uid" content="{concept['id']}"/>
+<meta name="dtb:depth" content="1"/><meta name="dtb:totalPageCount" content="0"/>
+<meta name="dtb:maxPageNumber" content="0"/></head>
+<docTitle><text>{concept['title']}</text></docTitle>
+<navMap>{chr(10).join(ncx_nav_points)}</navMap>
+</ncx>"""
+    # ── Container ────────────────────────────────────────────────────────────
+    container_xml = """<?xml version="1.0" encoding="UTF-8"?>
+<container xmlns="urn:oasis:names:tc:opendocument:xmlns:container" version="1.0">
+<rootfiles><rootfile full-path="OEBPS/content.opf" media-type="application/oebps-package+xml"/></rootfiles>
+</container>"""
+    # ── OPF ──────────────────────────────────────────────────────────────────
+    opf = f"""<?xml version="1.0" encoding="UTF-8"?>
+<package xmlns="http://www.idpf.org/2007/opf" version="3.0" unique-identifier="uid">
+<metadata xmlns:dc="http://purl.org/dc/elements/1.1/">
+<dc:identifier id="uid">{concept['id']}</dc:identifier>
+<dc:title>{concept['title']}</dc:title>
+<dc:language>en-AU</dc:language>
+<dc:creator>{concept.get('author', 'Brett Anthony Sjoberg')}</dc:creator>
+<dc:publisher>{concept.get('publisher', 'Brett Anthony Sjoberg')}</dc:publisher>
+<dc:description>{concept['description']}</dc:description>
+<dc:subject>{', '.join([concept['primary_keyword']] + concept['secondary_keywords'])}</dc:subject>
+<dc:date>{concept.get('created_at', '')[:10]}</dc:date>
+<dc:rights>© {datetime.date.today().year} {concept.get('author', 'Brett Anthony Sjoberg')} — ABN {concept.get('abn', '')}</dc:rights>
+<meta name="geo" content="Australia"/>
+<meta name="keywords" content="{', '.join(concept['geo_keywords'])}"/>
+</metadata>
+<manifest>
+{chr(10).join(manifest_items)}
+</manifest>
+<spine toc="toc">
+{chr(10).join(f'<itemref idref="{sid}"/>' for sid in spine_ids)}
+</spine>
+</package>"""
+    # ── Assemble ZIP ─────────────────────────────────────────────────────────
+    files = {
+        "mimetype": b"application/epub+zip",
+        "META-INF/container.xml": container_xml.encode("utf-8"),
+        "OEBPS/content.opf": opf.encode("utf-8"),
+        "OEBPS/toc.ncx": toc_ncx.encode("utf-8"),
+        "OEBPS/style/nav.css": EPUB_CSS.encode("utf-8"),
+        "OEBPS/text/title_page.xhtml": tp_xhtml.encode("utf-8"),
+        "OEBPS/text/toc_page.xhtml": toc_xhtml.encode("utf-8"),
+        "OEBPS/text/back_cover.xhtml": back_xhtml.encode("utf-8"),
+    }
+    if (COVERS / "front_cover.jpg").exists():
+        files["OEBPS/images/cover.jpg"] = (COVERS / "front_cover.jpg").read_bytes()
+    if (COVERS / "back_cover.jpg").exists():
+        files["OEBPS/images/back_cover.jpg"] = (COVERS / "back_cover.jpg").read_bytes()
+    # Add chapters
+    for part in outline["parts"]:
+        for ch in part["chapters"]:
+            ch_num = int(ch["id"][2:])
+            md_path = CHAPTERS / f"{ch['id']}_{slugify_toc(ch['title'])}.md"
+            if md_path.exists():
+                body = _md_to_xhtml(md_path.read_text(encoding="utf-8"))
+            else:
+                body = f"<h1>{ch['title']}</h1><p>Content coming soon.</p>"
+            fname = f"OEBPS/text/ch{ch_num:02d}_{slugify_toc(ch['title'])}.xhtml"
+            xhtml = f"""<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE html><html xmlns="http://www.w3.org/1999/xhtml"><head>
+<title>{ch['title']}</title><link rel="stylesheet" href="../style/nav.css"/></head>
+<body>{body}</body></html>"""
+            files[fname] = xhtml.encode("utf-8")
+    with zipfile.ZipFile(out, "w", zipfile.ZIP_DEFLATED) as zf:
+        # mimetype must be first and uncompressed
+        zf.writestr("mimetype", files.pop("mimetype"), zipfile.ZIP_STORED)
+        for arcname, data in sorted(files.items()):
+            zf.writestr(arcname, data)
+    return str(out)
 # ────────────────────────────────────────────────────────────────────────────