From 0d59ab2be9727d86cfe5217dab4862b277fd407a Mon Sep 17 00:00:00 2001
From: CaptainB <bin@fit2cloud.com>
Date: Fri, 16 Aug 2024 10:36:27 +0800
Subject: [PATCH] =?UTF-8?q?refactor:=20=E4=BD=BF=E7=94=A8lazy=5Fload?=
 =?UTF-8?q?=E6=96=B9=E5=BC=8F=E5=8A=A0=E8=BD=BDpdf?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 apps/common/handle/impl/pdf_split_handle.py | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/apps/common/handle/impl/pdf_split_handle.py b/apps/common/handle/impl/pdf_split_handle.py
index 6adc59ea5..e4bac8d08 100644
--- a/apps/common/handle/impl/pdf_split_handle.py
+++ b/apps/common/handle/impl/pdf_split_handle.py
@@ -26,10 +26,6 @@ default_pattern_list = [re.compile('(?<=^)# .*|(?<=\\n)# .*'),
                         re.compile("(?<!\n)\n\n+")]
 
 
-def number_to_text(pdf_document, page_number):
-    return pdf_document[page_number].page_content
-
-
 def check_pdf_is_image(pdf_path):
     try:
         # 打开PDF文件
@@ -64,8 +60,9 @@ class PdfSplitHandle(BaseSplitHandle):
                 loader = PyPDFLoader(temp_file_path, extract_images=True)
             else:
                 loader = PyPDFLoader(temp_file_path, extract_images=False)
-            pdf_document = loader.load()
-            content = "\n".join([number_to_text(pdf_document, page_number) for page_number in range(len(pdf_document))])
+
+            content = "\n".join([page.page_content for page in loader.lazy_load()])
+
             if pattern_list is not None and len(pattern_list) > 0:
                 split_model = SplitModel(pattern_list, with_filter, limit)
             else: