Replace PyPDF2 with pymupdf for PDF text extraction

pymupdf produces cleaner text (e.g. no concatenated words in headers),
handles encrypted and malformed PDFs that PyPDF2 failed on, and
supports non-Latin scripts.

modules/chat.py

@@ -657,15 +657,13 @@ def add_message_attachment(history, row_idx, file_path, is_user=True):
 
 def extract_pdf_text(pdf_path):
     """Extract text from a PDF file"""
-    import PyPDF2
+    import pymupdf
 
     text = ""
     try:
-        with open(pdf_path, 'rb') as file:
-            pdf_reader = PyPDF2.PdfReader(file)
-            for page_num in range(len(pdf_reader.pages)):
-                page = pdf_reader.pages[page_num]
-                text += page.extract_text() + "\n\n"
+        with pymupdf.open(pdf_path) as doc:
+            for page in doc:
+                text += page.get_text() + "\n\n"
 
         return text.strip()
     except Exception as e:

requirements/full/requirements.txt

@@ -17,7 +17,7 @@ peft==0.18.*
 Pillow>=9.5.0
 psutil
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/full/requirements_amd.txt

@@ -15,7 +15,7 @@ peft==0.18.*
 Pillow>=9.5.0
 psutil
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/full/requirements_apple_intel.txt

@@ -15,7 +15,7 @@ peft==0.18.*
 Pillow>=9.5.0
 psutil
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/full/requirements_apple_silicon.txt

@@ -15,7 +15,7 @@ peft==0.18.*
 Pillow>=9.5.0
 psutil
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/full/requirements_cpu_only.txt

@@ -15,7 +15,7 @@ peft==0.18.*
 Pillow>=9.5.0
 psutil
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/full/requirements_nowheels.txt

@@ -15,7 +15,7 @@ peft==0.18.*
 Pillow>=9.5.0
 psutil
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/portable/requirements.txt

@@ -6,7 +6,7 @@ jinja2==3.1.6
 markdown
 numpy==2.2.*
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/portable/requirements_amd.txt

@@ -6,7 +6,7 @@ jinja2==3.1.6
 markdown
 numpy==2.2.*
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/portable/requirements_apple_intel.txt

@@ -6,7 +6,7 @@ jinja2==3.1.6
 markdown
 numpy==2.2.*
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/portable/requirements_apple_silicon.txt

@@ -6,7 +6,7 @@ jinja2==3.1.6
 markdown
 numpy==2.2.*
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/portable/requirements_cpu_only.txt

@@ -6,7 +6,7 @@ jinja2==3.1.6
 markdown
 numpy==2.2.*
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/portable/requirements_cuda131.txt

@@ -6,7 +6,7 @@ jinja2==3.1.6
 markdown
 numpy==2.2.*
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/portable/requirements_nowheels.txt

@@ -6,7 +6,7 @@ jinja2==3.1.6
 markdown
 numpy==2.2.*
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests

requirements/portable/requirements_vulkan.txt

@@ -6,7 +6,7 @@ jinja2==3.1.6
 markdown
 numpy==2.2.*
 pydantic==2.11.0
-PyPDF2==3.0.1
+pymupdf==1.27.1
 python-docx==1.1.2
 pyyaml
 requests