From 86c1b56be11905df15d535c7c203c8ce08073950 Mon Sep 17 00:00:00 2001
From: Liyulingyue <852433440@qq.com>
Date: Tue, 16 Apr 2024 19:46:48 +0800
Subject: [PATCH 1/9] modify requestions

---
 ppstructure/recovery/requirements.txt | 5 ++++-
 requirements.txt                      | 4 ----
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/ppstructure/recovery/requirements.txt b/ppstructure/recovery/requirements.txt
index 761b9d7c3e..714f5aced2 100644
--- a/ppstructure/recovery/requirements.txt
+++ b/ppstructure/recovery/requirements.txt
@@ -2,4 +2,7 @@ python-docx
 beautifulsoup4
 fonttools>=4.24.0
 fire>=0.3.0
-pdf2docx
\ No newline at end of file
+pdf2docx
+lxml
+premailer
+openpyxl
\ No newline at end of file
diff --git a/requirements.txt b/requirements.txt
index d4c95b889f..3b430170c7 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -10,9 +10,5 @@ rapidfuzz
 opencv-python<=4.6.0.66
 opencv-contrib-python<=4.6.0.66
 cython
-lxml
-premailer
-openpyxl
-attrdict
 Pillow>=10.0.0
 pyyaml
\ No newline at end of file

From 5eb940a04b054f8f8ed50ccc2aa07b8c04dafcab Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E5=BC=A0=E6=98=A5=E4=B9=94?=
 <83450930+Liyulingyue@users.noreply.github.com>
Date: Tue, 16 Apr 2024 20:02:51 +0800
Subject: [PATCH 2/9] Update requirements.txt

---
 requirements.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/requirements.txt b/requirements.txt
index 3b430170c7..3af26908ab 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -11,4 +11,4 @@ opencv-python<=4.6.0.66
 opencv-contrib-python<=4.6.0.66
 cython
 Pillow>=10.0.0
-pyyaml
\ No newline at end of file
+pyyaml

From 55daba74188845aa3b01c19cddc50868227063ad Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E5=BC=A0=E6=98=A5=E4=B9=94?=
 <83450930+Liyulingyue@users.noreply.github.com>
Date: Tue, 16 Apr 2024 20:03:00 +0800
Subject: [PATCH 3/9] Update requirements.txt

---
 ppstructure/recovery/requirements.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/ppstructure/recovery/requirements.txt b/ppstructure/recovery/requirements.txt
index 714f5aced2..10d4ae4d66 100644
--- a/ppstructure/recovery/requirements.txt
+++ b/ppstructure/recovery/requirements.txt
@@ -5,4 +5,4 @@ fire>=0.3.0
 pdf2docx
 lxml
 premailer
-openpyxl
\ No newline at end of file
+openpyxl

From 858df6184519b45be9d202eda2f0436906b96196 Mon Sep 17 00:00:00 2001
From: Liyulingyue <852433440@qq.com>
Date: Thu, 25 Apr 2024 20:30:45 +0800
Subject: [PATCH 4/9] try import pdfconvert

---
 paddleocr.py                          | 2 ++
 ppstructure/pdf2word/pdf2word.py      | 4 +++-
 ppstructure/predict_system.py         | 2 ++
 ppstructure/recovery/requirements.txt | 1 -
 4 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/paddleocr.py b/paddleocr.py
index d03a6932e2..dcac47802e 100644
--- a/paddleocr.py
+++ b/paddleocr.py
@@ -19,6 +19,7 @@
 __dir__ = os.path.dirname(__file__)
 
 import paddle
+from paddle.utils import try_import
 
 sys.path.append(os.path.join(__dir__, ""))
 
@@ -910,6 +911,7 @@ def main():
                 img = cv2.imread(img_path)
 
             if args.recovery and args.use_pdf2docx_api and flag_pdf:
+                try_import("pdf2docx")
                 from pdf2docx.converter import Converter
 
                 docx_file = os.path.join(args.output, "{}.docx".format(img_name))
diff --git a/ppstructure/pdf2word/pdf2word.py b/ppstructure/pdf2word/pdf2word.py
index 06ae555598..c9e61eee8f 100644
--- a/ppstructure/pdf2word/pdf2word.py
+++ b/ppstructure/pdf2word/pdf2word.py
@@ -25,7 +25,6 @@
 
 fitz = try_import("fitz")
 from PIL import Image
-from pdf2docx.converter import Converter
 from qtpy.QtWidgets import (
     QApplication,
     QWidget,
@@ -209,6 +208,9 @@ def run(self):
                     break
                 # using use_pdf2docx_api for PDF parsing
                 if self.use_pdf2docx_api and os.path.basename(image_file)[-3:] == "pdf":
+                    try_import("pdf2docx")
+                    from pdf2docx.converter import Converter
+
                     self.totalPageCnt += 1
                     self.progressBarRange.emit(self.totalPageCnt)
                     print("===============using use_pdf2docx_api===============")
diff --git a/ppstructure/predict_system.py b/ppstructure/predict_system.py
index 9073e87ee1..6148cc8901 100644
--- a/ppstructure/predict_system.py
+++ b/ppstructure/predict_system.py
@@ -28,6 +28,7 @@
 import logging
 from copy import deepcopy
 
+from paddle.utils import try_import
 from ppocr.utils.utility import get_image_file_list, check_and_read
 from ppocr.utils.logging import get_logger
 from ppocr.utils.visual import draw_ser_results, draw_re_results
@@ -300,6 +301,7 @@ def main(args):
         img_name = os.path.basename(image_file).split(".")[0]
 
         if args.recovery and args.use_pdf2docx_api and flag_pdf:
+            try_import("pdf2docx")
             from pdf2docx.converter import Converter
 
             os.makedirs(args.output, exist_ok=True)
diff --git a/ppstructure/recovery/requirements.txt b/ppstructure/recovery/requirements.txt
index 10d4ae4d66..433f359e44 100644
--- a/ppstructure/recovery/requirements.txt
+++ b/ppstructure/recovery/requirements.txt
@@ -2,7 +2,6 @@ python-docx
 beautifulsoup4
 fonttools>=4.24.0
 fire>=0.3.0
-pdf2docx
 lxml
 premailer
 openpyxl

From 11cdbdb4bcdd5f6fe303a79eb911ae6ed7680dc3 Mon Sep 17 00:00:00 2001
From: Liyulingyue <852433440@qq.com>
Date: Thu, 25 Apr 2024 20:39:00 +0800
Subject: [PATCH 5/9] try import lxml

---
 ppstructure/table/table_metric/table_metric.py | 6 +++++-
 ppstructure/table/tablepyxl/tablepyxl.py       | 5 ++++-
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/ppstructure/table/table_metric/table_metric.py b/ppstructure/table/table_metric/table_metric.py
index d5ba6a0afb..0ec11f578d 100755
--- a/ppstructure/table/table_metric/table_metric.py
+++ b/ppstructure/table/table_metric/table_metric.py
@@ -12,10 +12,11 @@
 from rapidfuzz.distance import Levenshtein
 from apted import APTED, Config
 from apted.helpers import Tree
-from lxml import etree, html
+
 from collections import deque
 from .parallel import parallel_process
 from tqdm import tqdm
+from paddle.utils import try_import
 
 
 class TableTree(Tree):
@@ -161,6 +162,9 @@ def evaluate(self, pred, true):
         """Computes TEDS score between the prediction and the ground truth of a
         given sample
         """
+        try_import("lxml")
+        from lxml import etree, html
+
         if (not pred) or (not true):
             return 0.0
         parser = html.HTMLParser(remove_comments=True, encoding="utf-8")
diff --git a/ppstructure/table/tablepyxl/tablepyxl.py b/ppstructure/table/tablepyxl/tablepyxl.py
index 95b75d1ac7..3124ab66d4 100644
--- a/ppstructure/table/tablepyxl/tablepyxl.py
+++ b/ppstructure/table/tablepyxl/tablepyxl.py
@@ -1,11 +1,11 @@
 # Do imports like python3 so our package works for 2 and 3
 from __future__ import absolute_import
 
-from lxml import html
 from openpyxl import Workbook
 from openpyxl.utils import get_column_letter
 from premailer import Premailer
 from tablepyxl.style import Table
+from paddle.utils import try_import
 
 
 def string_to_int(s):
@@ -15,6 +15,9 @@ def string_to_int(s):
 
 
 def get_Tables(doc):
+    try_import("lxml")
+    from lxml import etree, html
+
     tree = html.fromstring(doc)
     comments = tree.xpath("//comment()")
     for comment in comments:

From 9ab3e71fbf7294fcb4b95fb3a16af3cd78db14da Mon Sep 17 00:00:00 2001
From: Liyulingyue <852433440@qq.com>
Date: Thu, 25 Apr 2024 20:41:06 +0800
Subject: [PATCH 6/9] try import lxml

---
 ppstructure/recovery/requirements.txt | 1 -
 1 file changed, 1 deletion(-)

diff --git a/ppstructure/recovery/requirements.txt b/ppstructure/recovery/requirements.txt
index 433f359e44..1003086628 100644
--- a/ppstructure/recovery/requirements.txt
+++ b/ppstructure/recovery/requirements.txt
@@ -2,6 +2,5 @@ python-docx
 beautifulsoup4
 fonttools>=4.24.0
 fire>=0.3.0
-lxml
 premailer
 openpyxl

From 63d9c78aede1f2abfcf1194ac6d1cc484124fe2e Mon Sep 17 00:00:00 2001
From: Liyulingyue <852433440@qq.com>
Date: Thu, 25 Apr 2024 20:44:03 +0800
Subject: [PATCH 7/9] try import premailer

---
 ppstructure/recovery/requirements.txt    | 1 -
 ppstructure/table/tablepyxl/tablepyxl.py | 5 ++++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/ppstructure/recovery/requirements.txt b/ppstructure/recovery/requirements.txt
index 1003086628..742ff3fde0 100644
--- a/ppstructure/recovery/requirements.txt
+++ b/ppstructure/recovery/requirements.txt
@@ -2,5 +2,4 @@ python-docx
 beautifulsoup4
 fonttools>=4.24.0
 fire>=0.3.0
-premailer
 openpyxl
diff --git a/ppstructure/table/tablepyxl/tablepyxl.py b/ppstructure/table/tablepyxl/tablepyxl.py
index 3124ab66d4..ac631ee333 100644
--- a/ppstructure/table/tablepyxl/tablepyxl.py
+++ b/ppstructure/table/tablepyxl/tablepyxl.py
@@ -3,7 +3,7 @@
 
 from openpyxl import Workbook
 from openpyxl.utils import get_column_letter
-from premailer import Premailer
+
 from tablepyxl.style import Table
 from paddle.utils import try_import
 
@@ -90,6 +90,9 @@ def document_to_workbook(doc, wb=None, base_url=None):
     every table in the document.
     The workbook is returned
     """
+    try_import("premailer")
+    from premailer import Premailer
+
     if not wb:
         wb = Workbook()
         wb.remove(wb.active)

From d4839b1709b00430764ea404957d0b03af6ba127 Mon Sep 17 00:00:00 2001
From: Liyulingyue <852433440@qq.com>
Date: Thu, 25 Apr 2024 21:06:27 +0800
Subject: [PATCH 8/9] try import openpyxl

---
 ppstructure/recovery/requirements.txt    |  1 -
 ppstructure/table/tablepyxl/style.py     | 33 ++++++++++++++----------
 ppstructure/table/tablepyxl/tablepyxl.py |  6 +++--
 3 files changed, 24 insertions(+), 16 deletions(-)

diff --git a/ppstructure/recovery/requirements.txt b/ppstructure/recovery/requirements.txt
index 742ff3fde0..bad600d728 100644
--- a/ppstructure/recovery/requirements.txt
+++ b/ppstructure/recovery/requirements.txt
@@ -2,4 +2,3 @@ python-docx
 beautifulsoup4
 fonttools>=4.24.0
 fire>=0.3.0
-openpyxl
diff --git a/ppstructure/table/tablepyxl/style.py b/ppstructure/table/tablepyxl/style.py
index 4787e7d377..dfd0f2478a 100644
--- a/ppstructure/table/tablepyxl/style.py
+++ b/ppstructure/table/tablepyxl/style.py
@@ -1,19 +1,26 @@
 # This is where we handle translating css styles into openpyxl styles
 # and cascading those from parent to child in the dom.
 
-from openpyxl.cell import cell
-from openpyxl.styles import (
-    Font,
-    Alignment,
-    PatternFill,
-    NamedStyle,
-    Border,
-    Side,
-    Color,
-)
-from openpyxl.styles.fills import FILL_SOLID
-from openpyxl.styles.numbers import FORMAT_CURRENCY_USD_SIMPLE, FORMAT_PERCENTAGE
-from openpyxl.styles.colors import BLACK
+try:
+    from openpyxl.cell import cell
+    from openpyxl.styles import (
+        Font,
+        Alignment,
+        PatternFill,
+        NamedStyle,
+        Border,
+        Side,
+        Color,
+    )
+    from openpyxl.styles.fills import FILL_SOLID
+    from openpyxl.styles.numbers import FORMAT_CURRENCY_USD_SIMPLE, FORMAT_PERCENTAGE
+    from openpyxl.styles.colors import BLACK
+except:
+    import warnings
+
+    warnings.warn(
+        "Can not import openpyxl, some functions in the ppstructure may not work. Please manually install openpyxl before using ppstructure."
+    )
 
 FORMAT_DATE_MMDDYYYY = "mm/dd/yyyy"
 
diff --git a/ppstructure/table/tablepyxl/tablepyxl.py b/ppstructure/table/tablepyxl/tablepyxl.py
index ac631ee333..92f6536052 100644
--- a/ppstructure/table/tablepyxl/tablepyxl.py
+++ b/ppstructure/table/tablepyxl/tablepyxl.py
@@ -1,8 +1,6 @@
 # Do imports like python3 so our package works for 2 and 3
 from __future__ import absolute_import
 
-from openpyxl import Workbook
-from openpyxl.utils import get_column_letter
 
 from tablepyxl.style import Table
 from paddle.utils import try_import
@@ -30,7 +28,9 @@ def write_rows(worksheet, elem, row, column=1):
     Writes every tr child element of elem to a row in the worksheet
     returns the next row after all rows are written
     """
+    try_import("openpyxl")
     from openpyxl.cell.cell import MergedCell
+    from openpyxl.utils import get_column_letter
 
     initial_column = column
     for table_row in elem.rows:
@@ -91,7 +91,9 @@ def document_to_workbook(doc, wb=None, base_url=None):
     The workbook is returned
     """
     try_import("premailer")
+    try_import("openpyxl")
     from premailer import Premailer
+    from openpyxl import Workbook
 
     if not wb:
         wb = Workbook()

From 3eee6bb8423e846b5953720ad988b6b9a3c196af Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E5=BC=A0=E6=98=A5=E4=B9=94?=
 <83450930+Liyulingyue@users.noreply.github.com>
Date: Thu, 25 Apr 2024 21:08:55 +0800
Subject: [PATCH 9/9] Apply suggestions from code review

---
 ppstructure/table/table_metric/table_metric.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/ppstructure/table/table_metric/table_metric.py b/ppstructure/table/table_metric/table_metric.py
index 0ec11f578d..327b87adbf 100755
--- a/ppstructure/table/table_metric/table_metric.py
+++ b/ppstructure/table/table_metric/table_metric.py
@@ -12,7 +12,6 @@
 from rapidfuzz.distance import Levenshtein
 from apted import APTED, Config
 from apted.helpers import Tree
-
 from collections import deque
 from .parallel import parallel_process
 from tqdm import tqdm