fix ch01

ocr/gendatav2.py

@@ -47,7 +47,7 @@ for raw_json in raw_jsons:
 currPageId = -1
 lastPageId = -1
 currPageText = ""
-# for js in paragraphs:
+
 for i, js in enumerate(paragraphs):
     startPageId = js["startPageId"]
     endPageId = js["endPageId"]
@@ -78,9 +78,10 @@ for i, js in enumerate(paragraphs):
 
     if i == len(paragraphs) - 1:
         if currPageText:  # 最后一个段落，保存上一页文本
-            text_pages[currPageId] = currPageText
-            currPageText = ""
-            currPageId = PageId
+            if currPageId not in text_pages:
+                text_pages[currPageId] = currPageText
+                currPageText = ""
+                currPageId = PageId
     else:
         currPageText += " "
         texts += " "