From f68813eb2a1a3777be4ac3eb03a376d994746db9 Mon Sep 17 00:00:00 2001
From: user1018 <614803115@qq.com>
Date: Thu, 17 Nov 2022 16:18:05 +0800
Subject: [PATCH] optimize recovery (#8346)

* optimize recovery

* update
---
 ppstructure/predict_system.py           | 4 +++-
 ppstructure/recovery/recovery_to_doc.py | 2 +-
 ppstructure/recovery/requirements.txt   | 1 -
 ppstructure/recovery/table_process.py   | 2 --
 requirements.txt                        | 2 +-
 5 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/ppstructure/predict_system.py b/ppstructure/predict_system.py
index bb061c99..b32b7062 100644
--- a/ppstructure/predict_system.py
+++ b/ppstructure/predict_system.py
@@ -229,7 +229,9 @@ def main(args):
 
         if args.recovery and args.use_pdf2docx_api and flag_pdf:
             from pdf2docx.converter import Converter
-            docx_file = os.path.join(args.output, '{}.docx'.format(img_name))
+            os.makedirs(args.output, exist_ok=True)
+            docx_file = os.path.join(args.output,
+                                     '{}_api.docx'.format(img_name))
             cv = Converter(image_file)
             cv.convert(docx_file)
             cv.close()
diff --git a/ppstructure/recovery/recovery_to_doc.py b/ppstructure/recovery/recovery_to_doc.py
index 1d8f8d9d..05018120 100644
--- a/ppstructure/recovery/recovery_to_doc.py
+++ b/ppstructure/recovery/recovery_to_doc.py
@@ -73,7 +73,7 @@ def convert_info_docx(img, res, save_folder, img_name):
                 text_run.font.size = shared.Pt(10)
 
     # save to docx
-    docx_path = os.path.join(save_folder, '{}.docx'.format(img_name))
+    docx_path = os.path.join(save_folder, '{}_ocr.docx'.format(img_name))
     doc.save(docx_path)
     logger.info('docx save to {}'.format(docx_path))
 
diff --git a/ppstructure/recovery/requirements.txt b/ppstructure/recovery/requirements.txt
index ec08f9d0..761b9d7c 100644
--- a/ppstructure/recovery/requirements.txt
+++ b/ppstructure/recovery/requirements.txt
@@ -1,5 +1,4 @@
 python-docx
-PyMuPDF==1.19.0
 beautifulsoup4
 fonttools>=4.24.0
 fire>=0.3.0
diff --git a/ppstructure/recovery/table_process.py b/ppstructure/recovery/table_process.py
index 982e6b76..77a6ef76 100644
--- a/ppstructure/recovery/table_process.py
+++ b/ppstructure/recovery/table_process.py
@@ -278,8 +278,6 @@ class HtmlToDocx(HTMLParser):
                 cell_col += colspan
             cell_row += 1
 
-        doc.save('1.docx')
-
     def handle_data(self, data):
         if self.skip:
             return
diff --git a/requirements.txt b/requirements.txt
index f3d9ce89..b4f8b011 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -14,4 +14,4 @@ lxml
 premailer
 openpyxl
 attrdict
-PyMuPDF==1.19.0
\ No newline at end of file
+PyMuPDF<1.21.0
\ No newline at end of file
-- 
GitLab