From 0ff074c24c3a96679a0555a98053c3277c9bdea8 Mon Sep 17 00:00:00 2001
From: CyMule <nhfranck@gmail.com>
Date: Tue, 8 Jul 2025 16:09:09 -0400
Subject: [PATCH 1/2] perf: add early page count check to prevent expensive
 PDFMiner processing

---
 unstructured/partition/pdf.py | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/unstructured/partition/pdf.py b/unstructured/partition/pdf.py
index d38658ed64..0d4063f5ae 100644
--- a/unstructured/partition/pdf.py
+++ b/unstructured/partition/pdf.py
@@ -290,6 +290,27 @@ def partition_pdf_or_image(
 
     validate_strategy(strategy, is_image)
 
+    # Early page count check for strategies that will use HI_RES
+    # This prevents expensive PDFMiner processing for documents that exceed page limits
+    pdf_hi_res_max_pages = kwargs.get("pdf_hi_res_max_pages")
+    if pdf_hi_res_max_pages is not None:
+        # Check if this strategy will result in HI_RES processing
+        will_use_hi_res = False
+        
+        if strategy == PartitionStrategy.HI_RES:
+            will_use_hi_res = True
+        elif strategy == PartitionStrategy.AUTO:
+            # AUTO resolves to HI_RES in these cases:
+            extract_element = extract_images_in_pdf or bool(extract_image_block_types)
+            will_use_hi_res = is_image or infer_table_structure or extract_element
+        
+        if will_use_hi_res:
+            check_pdf_hi_res_max_pages_exceeded(
+                filename=filename,
+                file=file,
+                pdf_hi_res_max_pages=pdf_hi_res_max_pages,
+            )
+
     last_modified = get_last_modified_date(filename) if filename else None
     pdfminer_config = PDFMinerConfig(
         line_margin=pdfminer_line_margin,

From b1b99029b08d7a0d12e1bd455bd87be7296043f2 Mon Sep 17 00:00:00 2001
From: CyMule <nhfranck@gmail.com>
Date: Tue, 8 Jul 2025 16:13:40 -0400
Subject: [PATCH 2/2] changelog

---
 CHANGELOG.md | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 7d45affb38..0af7658fca 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,3 +1,12 @@
+## 0.18.6
+
+### Enhancements
+- **Optimize PDF processing with early page count check** Prevents expensive PDFMiner processing for documents that exceed page limits by checking page count before strategy execution when `pdf_hi_res_max_pages` is set.
+
+### Features
+
+### Fixes
+
 ## 0.18.5-dev0
 
 ### Enhancements