更新

2026-03-06 14:50:43 +08:00
parent 843146cdd7
commit 91ff28bdcf
18 changed files with 1316 additions and 100 deletions
--- a/difyPlugin/pdf/tools/pdf_extract_range.yaml
+++ b/difyPlugin/pdf/tools/pdf_extract_range.yaml
@@ -0,0 +1,68 @@
+identity:
+  name: "pdf_extract_range"
+  author: "yslg"
+  label:
+    en_US: "Extract Page Range Text"
+    zh_Hans: "提取页面范围文本"
+    pt_BR: "Extrair Texto do Intervalo de Páginas"
+    ja_JP: "ページ範囲テキスト抽出"
+description:
+  human:
+    en_US: "Extract plain text from a specified page range of a PDF file"
+    zh_Hans: "从PDF文件的指定页码范围提取纯文本"
+    pt_BR: "Extrair texto simples de um intervalo de páginas especificado de um arquivo PDF"
+    ja_JP: "PDFファイルの指定ページ範囲からプレーンテキストを抽出"
+  llm: "Extract plain text from PDF pages in the given start-end range. Returns concatenated text of all pages in range."
+parameters:
+  - name: file
+    type: file
+    required: true
+    label:
+      en_US: PDF File
+      zh_Hans: PDF 文件
+      pt_BR: Arquivo PDF
+      ja_JP: PDFファイル
+    human_description:
+      en_US: "PDF file to extract text from"
+      zh_Hans: "要提取文本的 PDF 文件"
+      pt_BR: "Arquivo PDF para extrair texto"
+      ja_JP: "テキストを抽出するPDFファイル"
+    llm_description: "PDF file to extract page range text from"
+    form: llm
+    fileTypes:
+      - "pdf"
+  - name: start_page
+    type: number
+    required: true
+    label:
+      en_US: Start Page
+      zh_Hans: 起始页码
+      pt_BR: Página Inicial
+      ja_JP: 開始ページ
+    human_description:
+      en_US: "Start page index (0-based)"
+      zh_Hans: "起始页码（从0开始）"
+      pt_BR: "Índice da página inicial (base 0)"
+      ja_JP: "開始ページ番号（0始まり）"
+    llm_description: "Start page index (0-based)"
+    form: llm
+    default: 0
+  - name: end_page
+    type: number
+    required: true
+    label:
+      en_US: End Page
+      zh_Hans: 结束页码
+      pt_BR: Página Final
+      ja_JP: 終了ページ
+    human_description:
+      en_US: "End page index (0-based, inclusive)"
+      zh_Hans: "结束页码（从0开始，包含该页）"
+      pt_BR: "Índice da página final (base 0, inclusivo)"
+      ja_JP: "終了ページ番号（0始まり、含む）"
+    llm_description: "End page index (0-based, inclusive)"
+    form: llm
+    default: 0
+extra:
+  python:
+    source: tools/pdf_extract_range.py