fix: should keep namespace structure of toc & opf

[Moskize91]

No description provided.

[coderabbitai]

Summary by CodeRabbit

• Refactor

  • Metadata and TOC flows now return and propagate context objects through read/write and translation task flows for safer in-place updates.

• Bug Fixes

  • Namespace-agnostic XML parsing with resilient handling when elements are missing.
  • OPF is no longer treated as a self-closing/void element.
  • Serialization fixes avoid incorrect prefixing of standard HTML attributes on link elements.

• Chores

  • New context types exposed in the package public API; tests updated accordingly.

Walkthrough

The PR exposes MetadataContext and TocContext from the epub package initializer and adds corresponding dataclasses. read_metadata/read_toc now return (items, context) tuples and write_metadata/write_toc accept those contexts to perform in-place updates using XMLLikeNode. translator.py threads toc and metadata contexts through task generation and conditionally invokes write-back when contexts exist. xml/self_closing.py removes "meta" from the void tag set. xml/xml_like.py applies regex-based fixes to revert epub-prefixed link attributes to standard HTML attribute names. Tests updated to match new signatures and behaviors.

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~45 minutes

Possibly related PRs

• feat: support metadata & toc translation #57 — Overlaps changes to translator flow and TOC/metadata handling, touching the same modules.
• feat: implement new toc & spines logic #43 — Makes related edits to epub TOC/spine APIs and package exports that intersect with this PR.
• feat: use XML transcode logic & merge toc, metadatas into XML serials #90 — Earlier work on metadata/TOC read/write surfaces that this context-based refactor builds upon.

🚥 Pre-merge checks | ✅ 1 | ❌ 1

❌ Failed checks (1 inconclusive)

Check name	Status	Explanation	Resolution
Description check	❓ Inconclusive	No description was provided by the author, making it impossible to assess whether the description relates to the changeset.	Add a pull request description explaining the changes, why they were necessary, and how they address the namespace structure preservation for TOC and OPF files.

✅ Passed checks (1 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title follows the required format <type>(<scope>): <subject> with type 'fix' and subject describing keeping namespace structure of TOC and OPF files.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@epub_translator/epub/metadata.py`:
- Around line 15-18: The comment for the MetadataContext dataclass uses a
fullwidth comma causing RUF003; update the inline comment for the xml_node field
to replace the fullwidth comma with a standard ASCII comma (i.e., change
"XMLLikeNode 对象，保留原始文件信息" to use a normal comma) so the comment for the opf_path
and xml_node fields is unambiguous and Ruff-compliant; locate the
MetadataContext dataclass and edit the comment on the xml_node field
accordingly.

[coderabbitai]

Actionable comments posted: 0

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (2)

epub_translator/xml/self_closing.py (2)

165-168: ⚠️ Potential issue | 🟡 Minor

Docstring example still mentions <meta> despite exclusion.

unclose_void_elements no longer processes <meta> because it’s not in _VOID_TAGS, so the example is now misleading.

📝 Suggested docstring fix

-        <meta charset="utf-8" /> → <meta charset="utf-8">
         <br /> → <br>
         <img src="test.png" /> → <img src="test.png">

---

6-45: ⚠️ Potential issue | 🟠 Major

Inconsistency: <meta> should be self-closed for XML parsing, but it's excluded from _VOID_TAGS.

The tests expect <meta charset="utf-8"> → <meta charset="utf-8" /> (lines 16, 113, 372-374), but self_close_void_elements won't process <meta> tags because they're missing from _VOID_TAGS. When this function is called before XML parsing in xml_like.py:49, non-self-closed <meta> tags will cause parsing failures.

The comment states "meta is excluded because OPF files have <meta property="...">content</meta>" (a non-void element), but this conflates two different contexts: OPF metadata documents vs. HTML/XHTML content. Consider making <meta> handling context-aware or restoring it to _VOID_TAGS if only XHTML content is processed here.

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@epub_translator/xml/xml_like.py`:
- Around line 224-226: The current unconditional loop that applies
_STANDARD_HTML_ATTRS to xml_string (the pattern.sub(...) block) removes
epub:type/epub:rel from all <link> elements; restrict this rewrite so it only
runs for Package Documents/container.xml contexts or when a caller-provided flag
indicates it's safe. Update the code that invokes the xml_string normalization
(or the function containing the loop) to accept a context parameter (e.g.,
"package" vs "content") or a boolean (e.g., allow_epub_prefixes) and only
execute the for pattern, replacement in _STANDARD_HTML_ATTRS: xml_string =
pattern.sub(replacement, xml_string) when the context is package/container or
the flag is true; alternatively, preserve attributes by tracking original
attribute namespaces and skip replacing attributes that originated from EPUB
Content Documents. Ensure you reference the _STANDARD_HTML_ATTRS list and the
xml_string normalization loop when making the change.

🧹 Nitpick comments (1)

epub_translator/xml/xml_like.py (1)

35-52: Add a regression test for the <link> attribute workaround.

Given the regex-based post‑processing in Line 35-52, a small test covering <link type="text/css" rel="stylesheet"> (and a control element like <nav epub:type="toc">) would lock in the intended behavior and reduce regressions.

[coderabbitai]

Actionable comments posted: 0

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

tests/test_metadata.py (1)

16-77: ⚠️ Potential issue | 🟠 Major

Fix unused context variables to clear lint failures.
Pylint flags unused-variable at Line 22, Line 58, and Line 69. Rename to _context to keep CI green.

🔧 Proposed fix

-            metadata, context = read_metadata(zip_file)
+            metadata, _context = read_metadata(zip_file)
...
-            metadata, context = read_metadata(zip_file)
+            metadata, _context = read_metadata(zip_file)
...
-            metadata, context = read_metadata(zip_file)
+            metadata, _context = read_metadata(zip_file)