update

app.go

@@ -822,13 +822,17 @@ func (a *App) WriteBookmarkByFile(inFile string, outFile string, tocFile string,
 	return nil
 }
 
-func (a *App) WriteBookmarkByGap(inFile string, outFile string, gap int, format string) error {
+func (a *App) WriteBookmarkByGap(inFile string, outFile string, gap int, format string, startNumber int, pages string) error {
 	logger.Printf("inFile: %s, outFile: %s, gap: %d\n", inFile, outFile, gap)
 	if _, err := os.Stat(inFile); os.IsNotExist(err) {
 		logger.Println(err)
 		return err
 	}
 	args := []string{"bookmark", "add", "--method", "gap"}
+	if pages != "" {
+		args = append(args, "--page_range", pages)
+	}
+	args = append(args, "--start-number", fmt.Sprintf("%d", startNumber))
 	args = append(args, "--gap", fmt.Sprintf("%d", gap))
 	if format != "" {
 		args = append(args, "--format", format)

frontend/src/components/Forms/BookmarkForm.vue

@@ -33,9 +33,16 @@
                 <a-form-item name="write_offset" label="页码偏移量" v-if="formState.write_type == 'file'">
                     <a-input-number v-model:value="formState.write_offset" />
                 </a-form-item>
+                <a-form-item name="page" label="页码范围" hasFeedback :validateStatus="validateStatus.page"
+                    :help="validateHelp.page">
+                    <a-input v-model:value="formState.page" placeholder="e.g. 3-N (留空表示全部页面)" allow-clear />
+                </a-form-item>
                 <a-form-item name="write_gap" label="间隔页数" v-if="formState.write_type == 'page'">
                     <a-input-number v-model:value="formState.write_gap" />
                 </a-form-item>
+                <a-form-item name="start_number" label="起始编号" v-if="formState.write_type == 'page'">
+                    <a-input-number v-model:value="formState.start_number"/>
+                </a-form-item>
                 <a-form-item name="bookmark.write_format" label="命名格式" v-if="formState.write_type == 'page'">
                     <a-input v-model:value="formState.write_format" placeholder="e.g. 第%p页(%p表示页码)" allow-clear />
                 </a-form-item>
@@ -208,6 +215,7 @@ export default defineComponent({
             default_level: 1,
             remove_blank_lines: true,
             recognize_type: "font",
+            start_number: 1,
         });
 
         const indentItems = reactive<{ items: IndentItem[] }>({
@@ -330,7 +338,7 @@ export default defineComponent({
                             break;
                         }
                         case "page": {
-                            await handleOps(WriteBookmarkByGap, [formState.input, formState.output, formState.write_gap, formState.write_format]);
+                            await handleOps(WriteBookmarkByGap, [formState.input, formState.output, formState.write_gap, formState.write_format, formState.start_number, formState.page]);
                             break;
                         }
                     }

frontend/src/components/data.tsx

@@ -154,6 +154,7 @@ interface BookmarkState {
     default_level: number;
     remove_blank_lines: boolean;
     recognize_type: string;
+    start_number: number;
 }
 
 interface OcrState {

frontend/wailsjs/go/main/App.d.ts

@@ -126,4 +126,4 @@ export function WatermarkPDFByText(arg1:string,arg2:string,arg3:string,arg4:stri
 
 export function WriteBookmarkByFile(arg1:string,arg2:string,arg3:string,arg4:number):Promise<void>;
 
-export function WriteBookmarkByGap(arg1:string,arg2:string,arg3:number,arg4:string):Promise<void>;
+export function WriteBookmarkByGap(arg1:string,arg2:string,arg3:number,arg4:string,arg5:number,arg6:string):Promise<void>;

frontend/wailsjs/go/main/App.js

@@ -250,6 +250,6 @@ export function WriteBookmarkByFile(arg1, arg2, arg3, arg4) {
   return window['go']['main']['App']['WriteBookmarkByFile'](arg1, arg2, arg3, arg4);
 }
 
-export function WriteBookmarkByGap(arg1, arg2, arg3, arg4) {
-  return window['go']['main']['App']['WriteBookmarkByGap'](arg1, arg2, arg3, arg4);
+export function WriteBookmarkByGap(arg1, arg2, arg3, arg4, arg5, arg6) {
+  return window['go']['main']['App']['WriteBookmarkByGap'](arg1, arg2, arg3, arg4, arg5, arg6);
 }

thirdparty/ocr.py

@@ -12,7 +12,9 @@
 import fitz
 import matplotlib.pyplot as plt
 import numpy as np
-from paddleocr import PaddleOCR, PPStructure, draw_ocr
+from paddleocr import PaddleOCR, PPStructure, draw_ocr, save_structure_res
+from paddleocr.ppstructure.recovery.recovery_to_doc import (
+    convert_info_docx, sorted_layout_boxes)
 from PIL import Image
 from tqdm import tqdm
 
@@ -423,6 +425,30 @@ def extract_item_from_pdf(doc_path: str, page_range: str = 'all', type: str = "f
         logger.error(traceback.format_exc())
         dump_json(cmd_output_path, {"status": "error", "message": traceback.format_exc()})
 
+def convert_pdf2docx(doc_path: str, lang: str = "ch", dpi: int = 300, page_range: str = "all", output_path: str = None):
+    try:
+        doc: fitz.Document = fitz.open(doc_path)
+        table_engine = PPStructure(recovery=True, lang=lang)
+        roi_indicies = parse_range(page_range, doc.page_count)
+        p = Path(doc_path)
+        if output_path is None:
+            output_dir = p.parent / "docx"
+            output_dir.mkdir(parents=True, exist_ok=True)
+        else:
+            output_dir = Path(output_path)
+            output_dir.mkdir(parents=True, exist_ok=True)
+        for page_index in roi_indicies:
+            page = doc[page_index]
+            pix = page.get_pixmap(matrix=fitz.Matrix(dpi/72, dpi/72))
+            img = np.frombuffer(buffer=pix.samples, dtype=np.uint8).reshape((pix.height, pix.width, -1))
+            result = table_engine(img)
+            h, w, _ = img.shape
+            res = sorted_layout_boxes(result, w)
+            convert_info_docx(img, res, output_dir, f"{p.stem}-{page_index+1}")
+    except:
+        logger.error(traceback.format_exc())
+        dump_json(cmd_output_path, {"status": "error", "message": traceback.format_exc()})
+
 def main():
     parser = argparse.ArgumentParser()
     sub_parsers = parser.add_subparsers()
@@ -470,4 +496,5 @@ def main():
         extract_item_from_pdf(doc_path=args.input_path, page_range=args.page_range, type=args.type, output_dir=args.output)
 
 if __name__ == "__main__":
-    main()
+    # main()
+    convert_pdf2docx(r"C:\Users\kevin\Desktop\书签测试\计算机网络-目录_提取.pdf")

thirdparty/pdf.py

@@ -868,14 +868,17 @@ def add_toc_from_file(toc_path: str, doc_path: str, offset: int, output_path: st
         dump_json(cmd_output_path, {"status": "error", "message": traceback.format_exc()})
 
 @batch_process
-def add_toc_by_gap(doc_path: str, gap: int = 1, format: str = "第%p页", output_path: str = None):
+def add_toc_by_gap(doc_path: str, gap: int = 1, format: str = "第%p页", start_number: int = 1, page_range: str = "all", output_path: str = None):
     try:
         doc: fitz.Document = fitz.open(doc_path)
         p = Path(doc_path)
         toc = []
-        for i in range(0, doc.page_count, gap):
-            toc.append([1, format.replace("%p", str(i+1)), i+1])
-        toc.append([1, format.replace("%p", str(doc.page_count)), doc.page_count])
+        roi_indicies = parse_range(page_range, doc.page_count)
+        n = len(roi_indicies)
+        for i in range(0, n, gap):
+            toc.append([1, format.replace("%p", str(start_number)), roi_indicies[i]+1])
+            start_number += gap
+        # toc.append([1, format.replace("%p", str(doc.page_count)), doc.page_count])
         doc.set_toc(toc)
         if output_path is None:
             output_path = str(p.parent / f"{p.stem}-[页码书签版].pdf")
@@ -2370,6 +2373,8 @@ def main():
     ### 页码书签
     bookmark_add_parser.add_argument("--gap", type=int, default=1, help="页码间隔")
     bookmark_add_parser.add_argument("--format", type=str, default="第%p页", help="页码格式")
+    bookmark_add_parser.add_argument("--page_range", type=str, default="all", help="页码范围")
+    bookmark_add_parser.add_argument("--start-number", type=int, default=1, help="起始编号")
 
     ## 提取书签
     bookmark_extract_parser = bookmark_sub_parsers.add_parser("extract", help="提取书签")
@@ -2630,7 +2635,7 @@ def main():
             if args.method == "file":
                 add_toc_from_file(toc_path=args.toc, doc_path=args.input_path, offset=args.offset, output_path=args.output)
             elif args.method == "gap":
-                add_toc_by_gap(doc_path=args.input_path, gap=args.gap, format=args.format, output_path=args.output)
+                add_toc_by_gap(doc_path=args.input_path, gap=args.gap, format=args.format, start_number=args.start_number, page_range=args.page_range, output_path=args.output)
         elif args.bookmark_which == "extract":
             extract_toc(doc_path=args.input_path, format=args.format, output_path=args.output)
         elif args.bookmark_which == "transform":
@@ -2726,11 +2731,3 @@ def main():
 
 if __name__ == "__main__":
     main()
-    # find_title_by_rect_annot(r"C:\Users\kevin\Desktop\书签测试\2023考研英语一真题-去水印版.pdf")
-    # find_title_by_rect_annot(r"C:\Users\kevin\Desktop\书签测试\迅捷PDF编辑器v2.0使用手册-去水印版.pdf", "3-N")
-    # find_title_by_rect_annot(r"C:\Users\kevin\Desktop\书签测试\2022-中国计算机学会推荐国际学术会议和期刊目录.pdf")
-    # find_title_by_rect_annot(r"C:\Users\kevin\Desktop\书签测试\Computer Networking_ A Top-Down Approach, Global Edition, 8th Edition.pdf", "33-N")
-    # find_title_by_rect_annot(r"C:\Users\kevin\Desktop\书签测试\项目任务书(最终签字版).pdf", "11-13")
-    # find_title_by_rect_annot(r"C:\Users\kevin\Desktop\书签测试\SQL必知必会（第5版）.pdf", "18-N")
-    # find_title_by_rect_annot(r"C:\Users\kevin\Desktop\书签测试\汤书操作系统课本.pdf", "10-N")
-    # extract_encrypt_pdf_hash(r"C:\Users\kevin\Downloads\Detecting Ponzi Schemes on Ethereum Proceedings of the 2018 World Wide Web-加密.pdf")