更新

difyPlugin/pdf/tools/pdf_toc.yaml

@@ -0,0 +1,79 @@
+identity:
+  name: "pdf_toc"
+  author: "yslg"
+  label:
+    en_US: "PDF TOC Parser"
+    zh_Hans: "PDF目录解析"
+    pt_BR: "Analisador de Sumário PDF"
+    ja_JP: "PDF目次解析"
+description:
+  human:
+    en_US: "Parse PDF table-of-contents text (from pdf_column_range) into structured JSON catalog via LLM"
+    zh_Hans: "通过LLM将PDF目录文本（来自目录页提取工具的输出）解析为结构化JSON目录"
+    pt_BR: "Analisar texto do sumário PDF em catálogo JSON estruturado via LLM"
+    ja_JP: "LLMを使用してPDF目次テキストを構造化JSONカタログに解析"
+  llm: "Parse PDF table-of-contents text into structured JSON with chapter names and page ranges. Input is the output of pdf_column_range tool (start/end/pages)."
+parameters:
+  - name: toc_start
+    type: number
+    required: true
+    label:
+      en_US: TOC Start Page
+      zh_Hans: 目录起始页
+      pt_BR: Página Inicial do Sumário
+      ja_JP: 目次開始ページ
+    human_description:
+      en_US: "Start page index of TOC (from pdf_column_range output)"
+      zh_Hans: "目录起始页码（来自目录页提取工具输出的 start）"
+      pt_BR: "Índice da página inicial do sumário"
+      ja_JP: "目次の開始ページ番号"
+    llm_description: "Start page index of TOC section, from pdf_column_range output field 'start'"
+    form: llm
+  - name: toc_end
+    type: number
+    required: true
+    label:
+      en_US: TOC End Page
+      zh_Hans: 目录结束页
+      pt_BR: Página Final do Sumário
+      ja_JP: 目次終了ページ
+    human_description:
+      en_US: "End page index of TOC (from pdf_column_range output)"
+      zh_Hans: "目录结束页码（来自目录页提取工具输出的 end）"
+      pt_BR: "Índice da página final do sumário"
+      ja_JP: "目次の終了ページ番号"
+    llm_description: "End page index of TOC section, from pdf_column_range output field 'end'"
+    form: llm
+  - name: toc_pages
+    type: string
+    required: true
+    label:
+      en_US: TOC Page Text
+      zh_Hans: 目录页文本
+      pt_BR: Texto das Páginas do Sumário
+      ja_JP: 目次ページテキスト
+    human_description:
+      en_US: "Raw text content of TOC pages (from pdf_column_range output 'pages' array, joined)"
+      zh_Hans: "目录页原始文本内容（来自目录页提取工具输出的 pages 数组）"
+      pt_BR: "Conteúdo de texto bruto das páginas do sumário"
+      ja_JP: "目次ページの生テキスト内容"
+    llm_description: "Raw text content extracted from TOC pages, from pdf_column_range output field 'pages'"
+    form: llm
+  - name: model
+    type: model-selector
+    scope: llm
+    required: true
+    label:
+      en_US: LLM Model
+      zh_Hans: LLM 模型
+      pt_BR: Modelo LLM
+      ja_JP: LLMモデル
+    human_description:
+      en_US: "LLM model for parsing TOC into structured JSON"
+      zh_Hans: "用于解析目录的 LLM 模型"
+      pt_BR: "Modelo LLM para análise do sumário"
+      ja_JP: "目次解析用のLLMモデル"
+    form: form
+extra:
+  python:
+    source: tools/pdf_toc.py