Skip to content
View huuufu's full-sized avatar

Block or report huuufu

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
huuufu/README.md

huuufu

🧩 Products

Netflix 挂载本地双语字幕 — Chrome 拡張 · v2.0.2

  • Netflix のプレイヤーにローカル字幕(SRT / VTT)を重ねて二言語表示。20 言語のリアルタイム翻訳、位置・サイズ・色・影のカスタマイズ。UI は日本語 / 英語 / 中国語に対応。
  • Overlay local SRT/VTT subtitles on the Netflix player for bilingual viewing; real-time translation into 20 languages; customizable position, size, color and shadow. UI in Japanese, English and Chinese.

🌱 Open Source Contributions

Unstructured-IO/unstructured (15k ★) — document parsing / ETL library for LLM and RAG pipelines

PR #4484 · merged 2026-09 — fix(html): recognize single-block list items

  • Markdown の空行区切りリスト(loose list)は <li><p>…</p></li> にコンパイルされ、partition_md() が ListItem ではなく Text として返していた。<li> が単一の段落ブロックだけを含む場合にその内容を採用するよう HTML パーサを修正し、インライン注釈・入れ子の深さ・ページ番号を維持。崩してはいけない 14 ケースをテストで固定。2024 年からの Issue #3499 を解決。
  • Markdown loose lists compile to <li><p>…</p></li>, which partition_md() returned as Text instead of ListItem. A <li> whose only child is a single plain text block now adopts that block's content, preserving inline annotations, nesting depth and page numbers; 14 must-not-change cases are pinned by tests. Resolves #3499 (open since 2024).

PR #4465 · merged 2026-09 — fix: tolerate non-UTF-8 soffice output in doc/ppt conversion

  • partition_doc() / partition_ppt() が Windows 上でマルチバイト文字を含むパスの文書を変換すると、soffice の出力を UTF-8 で厳密デコードして UnicodeDecodeError で失敗する問題を修正。3 箇所のデコードを 1 つのヘルパーに集約して backslashreplace で寛容化し、テストを追加。2024 年からの Issue #3652 を解決。
  • Fixed convert_office_doc() crashing with UnicodeDecodeError on non-UTF-8 soffice output (Windows locale-codepage paths) during DOC/PPT partitioning; the three decode sites now share one backslashreplace helper, with tests. Resolves #3652 (open since 2024).

Issue #4466 · open — detect_file_encoding() returns mojibake

  • 文字コード自動判定が、Shift_JIS の日本語テキストを johab、Latin-1 のフランス語を windows-1250 として読み、文字化けのまま要素テキストにしてしまう問題を報告。再現コード、原因(未使用の validate_encoding()、iso_8859_1 が候補リストの残り 18 件を到達不能にしている)、試した修正案とそれぞれが回帰させる反例(cp1254 のトルコ語)を添付。
  • Reported silent mojibake from encoding auto-detection (Shift_JIS Japanese read as johab, Latin-1 French as windows-1250), with a repro, root-cause analysis (unused validate_encoding(), iso_8859_1 shadowing the other 18 candidates) and the trade-offs of the candidate fixes, including the regression each one introduces.

Pinned Loading

  1. ass-danmaku-analysis ass-danmaku-analysis Public

    Danmaku (bullet-comment) density chart generator for Bilibili live recordings — parses ASS/XML danmaku files, plots hotness over time, click-to-seek playback in PotPlayer

    Python 5

  2. video-watermark-generator video-watermark-generator Public

    Python

  3. Unstructured-IO/unstructured Unstructured-IO/unstructured Public

    Convert documents to structured data effortlessly. Unstructured is open-source ETL solution for transforming complex documents into clean, structured formats for language models. Visit our website …

    HTML 15.6k 1.4k