epub-generator 0.1.3__py3-none-any.whl

epub_generator/__init__.py

@@ -0,0 +1,41 @@
+from .generation import generate_epub
+from .options import LaTeXRender, TableRender
+from .types import (
+    BookMeta,
+    Chapter,
+    ChapterGetter,
+    ContentBlock,
+    EpubData,
+    Footnote,
+    Formula,
+    HTMLTag,
+    Image,
+    Mark,
+    Table,
+    TextBlock,
+    TextKind,
+    TocItem,
+)
+
+__all__ = [
+    # Main API function
+    "generate_epub",
+    # Options
+    "TableRender",
+    "LaTeXRender",
+    # Data types
+    "EpubData",
+    "BookMeta",
+    "TocItem",
+    "Chapter",
+    "ChapterGetter",
+    "ContentBlock",
+    "TextBlock",
+    "TextKind",
+    "Table",
+    "Formula",
+    "HTMLTag",
+    "Image",
+    "Footnote",
+    "Mark",
+]

epub_generator/context.py

@@ -0,0 +1,141 @@
+from dataclasses import dataclass
+from hashlib import sha256
+from importlib.resources import files
+from pathlib import Path
+from typing import cast
+from zipfile import ZipFile
+
+from jinja2 import Environment
+from jinja2 import Template as JinjaTemplate
+
+from .options import LaTeXRender, TableRender
+from .template import create_env
+
+
+@dataclass
+class _AssetNode:
+    file_name: str
+    media_type: str
+    content_hash: str
+
+class Context:
+    def __init__(
+        self,
+        file: ZipFile,
+        template: "Template",
+        table_render: TableRender,
+        latex_render: LaTeXRender,
+    ) -> None:
+        self._file: ZipFile = file
+        self._template: Template = template
+        self._table_render: TableRender = table_render
+        self._latex_render: LaTeXRender = latex_render
+        self._path_to_node: dict[Path, _AssetNode] = {}  # source_path -> node
+        self._hash_to_node: dict[str, _AssetNode] = {}  # content_hash -> node
+        self._chapters_with_mathml: set[str] = set()  # Track chapters containing MathML
+
+    @property
+    def file(self) -> ZipFile:
+        return self._file
+
+    @property
+    def template(self) -> "Template":
+        return self._template
+
+    @property
+    def table_render(self) -> TableRender:
+        return self._table_render
+
+    @property
+    def latex_render(self) -> LaTeXRender:
+        return self._latex_render
+
+    @property
+    def used_files(self) -> list[tuple[str, str]]:
+        nodes = list(self._hash_to_node.values())
+        nodes.sort(key=lambda node: node.file_name)
+        return [(node.file_name, node.media_type) for node in nodes]
+
+    def mark_chapter_has_mathml(self, chapter_file_name: str) -> None:
+        """Mark a chapter as containing MathML content for EPUB 3.0 manifest properties."""
+        self._chapters_with_mathml.add(chapter_file_name)
+
+    def chapter_has_mathml(self, chapter_file_name: str) -> bool:
+        """Check if a chapter contains MathML content."""
+        return chapter_file_name in self._chapters_with_mathml
+
+    def use_asset(
+        self,
+        source_path: Path,
+        media_type: str,
+        file_ext: str,
+    ) -> str:
+        if source_path in self._path_to_node:
+            return self._path_to_node[source_path].file_name
+
+        if not source_path.exists():
+            raise FileNotFoundError(f"Asset file not found: {source_path}")
+
+        with open(source_path, "rb") as f:
+            content = f.read()
+        content_hash = _sha256_hash(content)
+
+        if content_hash in self._hash_to_node:
+            node = self._hash_to_node[content_hash]
+            self._path_to_node[source_path] = node
+            return node.file_name
+
+        file_name = f"{content_hash}{file_ext}"
+        node = _AssetNode(
+            file_name=file_name,
+            media_type=media_type,
+            content_hash=content_hash,
+        )
+        self._path_to_node[source_path] = node
+        self._hash_to_node[content_hash] = node
+        self._file.write(
+            filename=source_path,
+            arcname="OEBPS/assets/" + file_name,
+        )
+        return file_name
+
+    def add_asset(self, data: bytes, media_type: str, file_ext: str) -> str:
+        content_hash = _sha256_hash(data)
+        if content_hash in self._hash_to_node:
+            return self._hash_to_node[content_hash].file_name
+
+        file_name = f"{content_hash}{file_ext}"
+        node = _AssetNode(
+            file_name=file_name,
+            media_type=media_type,
+            content_hash=content_hash,
+        )
+        self._hash_to_node[content_hash] = node
+
+        self._file.writestr(
+            zinfo_or_arcname="OEBPS/assets/" + file_name,
+            data=data,
+        )
+        return file_name
+
+class Template:
+    def __init__(self):
+        templates_path = cast(Path, files("epub_generator")) / "data"
+        self._env: Environment = create_env(templates_path)
+        self._templates: dict[str, JinjaTemplate] = {}
+
+    def render(self, template: str, **params) -> str:
+        jinja_template: JinjaTemplate = self._template(template)
+        return jinja_template.render(**params)
+
+    def _template(self, name: str) -> JinjaTemplate:
+        template = self._templates.get(name, None)
+        if template is None:
+            template = self._env.get_template(name)
+            self._templates[name] = template
+        return template
+
+def _sha256_hash(data: bytes) -> str:
+    hash256 = sha256()
+    hash256.update(data)
+    return hash256.hexdigest()

epub_generator/data/container.xml.jinja

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
+  <rootfiles>
+    <rootfile full-path="OEBPS/content.opf" media-type="application/oebps-package+xml" />
+  </rootfiles>
+</container>

epub_generator/data/content.opf.jinja

@@ -0,0 +1,68 @@
+<?xml version="1.0" encoding="utf-8"?>
+<package version="3.0" xmlns="http://www.idpf.org/2007/opf" unique-identifier="uid" xml:lang="zh">
+  <metadata xmlns:dc="http://purl.org/dc/elements/1.1/">
+    <dc:identifier id="uid">{{ ISBN }}</dc:identifier>
+    <dc:language>zh</dc:language>
+{% if meta and meta.title %}
+    <dc:title>{{ meta.title }}</dc:title>
+{% else %}
+    <dc:title>{{ i18n.unnamed }}</dc:title>
+{% endif %}
+{% if meta and meta.description %}
+    <dc:description>{{ meta.description }}</dc:description>
+{% endif %}
+{% if meta and meta.publisher %}
+    <dc:publisher>{{ meta.publisher }}</dc:publisher>
+{% endif %}
+{% if meta and meta.authors %}
+{% for author in meta.authors %}
+    <dc:creator id="creator{{ loop.index }}">{{ author }}</dc:creator>
+    <meta refines="#creator{{ loop.index }}" property="role" scheme="marc:relators">aut</meta>
+{% endfor %}
+{% endif %}
+{% if meta and meta.editors %}
+{% for editor in meta.editors %}
+    <dc:creator id="editor{{ loop.index }}">{{ editor }}</dc:creator>
+    <meta refines="#editor{{ loop.index }}" property="role" scheme="marc:relators">edt</meta>
+{% endfor %}
+{% endif %}
+{% if meta and meta.translators %}
+{% for translator in meta.translators %}
+    <dc:creator id="translator{{ loop.index }}">{{ translator }}</dc:creator>
+    <meta refines="#translator{{ loop.index }}" property="role" scheme="marc:relators">trl</meta>
+{% endfor %}
+{% endif %}
+    <meta property="dcterms:modified">{{ modified_timestamp }}</meta>
+{% if has_cover %}
+    <meta name="cover" content="cover-image" />
+{% endif %}
+  </metadata>
+  <manifest>
+    <item id="nav" properties="nav" media-type="application/xhtml+xml" href="nav.xhtml" />
+    <item id="a_css" media-type="text/css" href="styles/style.css" />
+{% for asset_file, media_type in asset_files %}
+    <item id="a_{{ asset_file|safe }}" media-type="{{ media_type|safe }}" href="assets/{{ asset_file|safe }}" />
+{% endfor %}
+{% if has_cover %}
+    <item id="cover-image" properties="cover-image" media-type="image/png" href="assets/cover.png" />
+    <item id="x_cover.xhtml" media-type="application/xhtml+xml" href="Text/cover.xhtml" />
+{% endif %}
+{% if has_head_chapter %}
+    <item id="x_head.xhtml" media-type="application/xhtml+xml" href="Text/head.xhtml" />
+{% endif %}
+{% for nav_point in nav_points %}
+    <item id="x_{{ nav_point.file_name|safe }}" media-type="application/xhtml+xml" href="Text/{{ nav_point.file_name|safe }}"{% if nav_point.file_name in chapters_with_mathml %} properties="mathml"{% endif %} />
+{% endfor %}
+  </manifest>
+  <spine>
+{% if has_cover %}
+    <itemref idref="x_cover.xhtml" linear="no" />
+{% endif %}
+{% if has_head_chapter %}
+    <itemref idref="x_head.xhtml" />
+{% endif %}
+{% for nav_point in nav_points %}
+    <itemref idref="x_{{ nav_point.file_name|safe }}" />
+{% endfor %}
+  </spine>
+</package>

epub_generator/data/cover.xhtml.jinja

@@ -0,0 +1,16 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" xml:lang="zh" lang="zh">
+
+<head>
+  <meta charset="utf-8"/>
+  <title>{{ i18n.cover }}</title>
+</head>
+
+<body>
+  <p>
+    <img src="../assets/cover.png" alt="Cover" />
+  </p>
+</body>
+
+</html>

epub_generator/data/mimetype.jinja

@@ -0,0 +1 @@
+application/epub+zip

epub_generator/data/nav.xhtml.jinja

@@ -0,0 +1,43 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" xml:lang="zh" lang="zh">
+<head>
+  <meta charset="utf-8"/>
+  <title>{{ i18n.table_of_contents }}</title>
+  <link href="styles/style.css" rel="stylesheet" type="text/css"/>
+</head>
+<body>
+  <nav epub:type="toc" id="toc" role="doc-toc">
+    <h1>{{ i18n.table_of_contents }}</h1>
+    <ol>
+{% if has_cover %}
+      <li>
+        <a href="Text/cover.xhtml">{{ i18n.cover }}</a>
+      </li>
+{% endif %}
+{% if has_head_chapter %}
+      <li>
+        <a href="Text/head.xhtml">{{ head_chapter_title }}</a>
+      </li>
+{% endif %}
+{{ toc_list|safe }}
+    </ol>
+  </nav>
+
+  <nav epub:type="landmarks" id="landmarks" hidden="hidden">
+    <h2>{{ i18n.landmarks }}</h2>
+    <ol>
+{% if has_cover %}
+      <li>
+        <a epub:type="cover" href="Text/cover.xhtml">{{ i18n.cover }}</a>
+      </li>
+{% endif %}
+{% if first_chapter_file %}
+      <li>
+        <a epub:type="bodymatter" href="Text/{{ first_chapter_file }}">{{ i18n.start_of_content }}</a>
+      </li>
+{% endif %}
+    </ol>
+  </nav>
+</body>
+</html>

epub_generator/data/part.xhtml.jinja

@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="utf-8"?>
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" xml:lang="zh" lang="zh">
+
+<head>
+  <meta charset="utf-8"/>
+  <title>Chapter</title>
+  <link href="../styles/style.css" rel="stylesheet" type="text/css"/>
+</head>
+
+<body>
+  {% for item in content %}
+  {{ item|safe }}
+  {% endfor %}
+  {% if citations %}
+  <section epub:type="footnotes" role="doc-endnotes">
+    {% for item in citations %}
+    {{ item|safe }}
+    {% endfor %}
+  </section>
+  {% endif %}
+</body>
+
+</html>

epub_generator/data/style.css.jinja

@@ -0,0 +1,68 @@
+blockquote {
+ 	border-left: 0.15em solid #c7c7c7;
+  margin: 0.75em;
+  padding: 1em;
+ 	text-indent:2em;
+  line-height:1.5em;
+}
+
+a.super {
+	font-size: 0.75em;
+	vertical-align: super;
+}
+
+a.citation {
+	font-size: 0.75em;
+  font-weight: bold;
+	vertical-align: super;
+  padding-right: 0.55em;
+}
+
+div.citation {
+  margin-top: 0.5em;
+  margin-left: 0.75em;
+  margin-bottom: 1.5em;
+}
+
+div.alt-wrapper {
+  display: flex;
+  flex-direction: column;
+  align-items: center;
+  padding-top: 0.8em;
+  padding-bottom: 3em;
+}
+
+div.alt-wrapper table {
+  width: 100%;
+  table-layout: fixed;
+  border-collapse: collapse;
+  border-spacing: 0;
+  margin: 0;
+}
+
+div.alt-wrapper th,
+.alt-wrapper td {
+  padding: 8px 12px;
+  border: 1px solid #e1e4e8;
+  text-align: left;
+  font-size: 14px;
+  line-height: 1.5;
+  word-wrap: break-word;
+}
+
+div.alt-wrapper th {
+  background-color: #f6f8fa;
+  font-weight: 600;
+}
+
+span.formula-inline {
+  display: inline;
+  vertical-align: middle;
+}
+
+span.formula-inline img {
+  display: inline-block;
+  vertical-align: middle;
+  margin: 0 0.2em;
+  max-height: 1.2em;
+}

epub_generator/generation/__init__.py

@@ -0,0 +1 @@
+from .gen_epub import generate_epub

epub_generator/generation/gen_asset.py

@@ -0,0 +1,156 @@
+import io
+import re
+from typing import Any, cast
+from xml.etree.ElementTree import Element, fromstring
+
+import matplotlib.pyplot as plt
+from latex2mathml.converter import convert
+
+from ..context import Context
+from ..options import LaTeXRender, TableRender
+from ..types import Formula, Image, Table
+
+_MEDIA_TYPE_MAP = {
+    ".png": "image/png",
+    ".jpg": "image/jpeg",
+    ".jpeg": "image/jpeg",
+    ".gif": "image/gif",
+    ".svg": "image/svg+xml",
+}
+
+def process_table(context: Context, table: Table) -> Element | None:
+    if context.table_render == TableRender.CLIPPING:
+        return None
+    try:
+        wrapped_html = f"<div>{table.html_content}</div>"
+        parsed = fromstring(wrapped_html)
+        wrapper = Element("div", attrib={"class": "alt-wrapper"})
+
+        for child in parsed:
+            wrapper.append(child)
+
+        return wrapper if len(wrapper) > 0 else None
+    except Exception:
+        return None
+
+
+def process_formula(
+        context: Context, 
+        formula: Formula, 
+        inline_mode: bool,
+    ) -> Element | None:
+
+    if context.latex_render == LaTeXRender.CLIPPING:
+        return None
+
+    latex_expr = _normalize_expression(formula.latex_expression)
+    if not latex_expr:
+        return None
+
+    if context.latex_render == LaTeXRender.MATHML:
+        return _latex2mathml(
+            latex=latex_expr, 
+            inline_mode=inline_mode,
+        )
+    elif context.latex_render == LaTeXRender.SVG:
+        svg_image = _latex_formula2svg(latex_expr)
+        if svg_image is None:
+            return None
+        file_name = context.add_asset(
+            data=svg_image,
+            media_type="image/svg+xml",
+            file_ext=".svg",
+        )
+        img_element = Element("img")
+        img_element.set("src", f"../assets/{file_name}")
+        img_element.set("alt", "formula")
+
+        if inline_mode:
+            wrapper = Element("span", attrib={"class": "formula-inline"})
+        else:
+            wrapper = Element("div", attrib={"class": "alt-wrapper"})
+
+        wrapper.append(img_element)
+        return wrapper
+
+    return None
+
+def process_image(context: Context, image: Image) -> Element | None:
+    file_ext = image.path.suffix or ".png"
+    file_name = context.use_asset(
+        source_path=image.path, 
+        media_type=_MEDIA_TYPE_MAP.get(file_ext.lower(), "image/png"), 
+        file_ext=file_ext,
+    )
+    img_element = Element("img")
+    img_element.set("src", f"../assets/{file_name}")
+    img_element.set("alt", image.alt_text)
+
+    wrapper = Element("div", attrib={"class": "alt-wrapper"})
+    wrapper.append(img_element)
+    return wrapper
+
+
+_ESCAPE_UNICODE_PATTERN = re.compile(r"&#x([0-9A-Fa-f]{5});")
+
+
+def _latex2mathml(latex: str, inline_mode: bool) -> None | Element:
+    try:
+        html_latex = convert(
+            latex=latex,
+            display="inline" if inline_mode else "block",
+        )
+    except Exception:
+        return None
+
+    # latex2mathml 转义会带上一个奇怪的 `&` 前缀，这显然是多余的
+    # 不得已，在这里用正则表达式处理以修正这个错误
+    def repl(match):
+        hex_code = match.group(1)
+        char = chr(int(hex_code, 16))
+        if char == "<":
+            return "&lt;"
+        elif char == ">":
+            return "&gt;"
+        else:
+            return char
+
+    mathml = re.sub(
+        pattern=_ESCAPE_UNICODE_PATTERN,
+        repl=repl,
+        string=html_latex,
+    )
+    try:
+        return fromstring(mathml)
+    except Exception:
+        return None
+
+
+def _latex_formula2svg(latex: str, font_size: int = 12):
+    # from https://www.cnblogs.com/qizhou/p/18170083
+    try:
+        output = io.BytesIO()
+        plt.rc("text", usetex=True)
+        plt.rc("font", size=font_size)
+        fig, ax = plt.subplots()
+        txt = ax.text(0.5, 0.5, f"${latex}$", ha="center", va="center", transform=ax.transAxes)
+        ax.axis("off")
+        fig.canvas.draw()
+        bbox = txt.get_window_extent(cast(Any, fig.canvas).get_renderer())
+        fig.set_size_inches(bbox.width / fig.dpi, bbox.height / fig.dpi)
+        plt.savefig(
+            output,
+            format="svg",
+            transparent=True,
+            bbox_inches="tight",
+            pad_inches=0,
+        )
+        return output.getvalue()
+    except Exception:
+        return None
+
+
+def _normalize_expression(expression: str) -> str:
+    expression = expression.replace("\n", "")
+    expression = expression.strip()
+    return expression