Optimize and prevent OOM in apply_png_predictor (#1247)

Co-authored-by: codeflash-ai[bot] <148906541+codeflash-ai[bot]@users.noreply.github.com>

Author: dhdaines

CHANGELOG.md

@@ -10,6 +10,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
 ### Fixed
 
 - Reproducibility issue when generating cmap `.json.gz` files ([#1242](https://github.com/pdfminer/pdfminer.six/pull/1242))
+- Switch to `bytearray` in `apply_png_predictor` to avoid out of memory error (and speed it up) ([#1247](https://github.com/pdfminer/pdfminer.six/pull/1247))
 - Check the type of `N` when creating an `ICCBased` color space ([#1252](http3://github.com/pdfminer/pdfminer.six/pull/1253))
 - Endless recursion issue with circular or corrupted `Prev` chains in cross-refeerence tables ([#1253](https://github.com/pdfminer/pdfminer.six/pull/1253))
 

pdfminer/utils.py

@@ -167,16 +167,16 @@ def apply_png_predictor(
 
     nbytes = colors * columns * bitspercomponent // 8
     bpp = colors * bitspercomponent // 8  # number of bytes per complete pixel
-    buf = []
-    line_above = list(b"\x00" * columns)
+    buf = bytearray()
+    line_above = bytearray(columns)
     for scanline_i in range(0, len(data), nbytes + 1):
         filter_type = data[scanline_i]
         line_encoded = data[scanline_i + 1 : scanline_i + 1 + nbytes]
-        raw = []
+        raw = bytearray()
 
         if filter_type == 0:
             # Filter type 0: None
-            raw = list(line_encoded)
+            raw = bytearray(line_encoded)
 
         elif filter_type == 1:
             # Filter type 1: Sub
@@ -186,7 +186,7 @@ def apply_png_predictor(
             # (computed mod 256), where Raw() refers to the bytes already
             #  decoded.
             for j, sub_x in enumerate(line_encoded):
-                raw_x_bpp = 0 if j - bpp < 0 else int(raw[j - bpp])
+                raw_x_bpp = 0 if j < bpp else raw[j - bpp]
                 raw_x = (sub_x + raw_x_bpp) & 255
                 raw.append(raw_x)
 
@@ -211,8 +211,8 @@ def apply_png_predictor(
             # bytes already decoded, and Prior() refers to the decoded bytes of
             # the prior scanline.
             for j, average_x in enumerate(line_encoded):
-                raw_x_bpp = 0 if j - bpp < 0 else int(raw[j - bpp])
-                prior_x = int(line_above[j])
+                raw_x_bpp = 0 if j < bpp else raw[j - bpp]
+                prior_x = line_above[j]
                 raw_x = (average_x + (raw_x_bpp + prior_x) // 2) & 255
                 raw.append(raw_x)
 
@@ -226,13 +226,13 @@ def apply_png_predictor(
             # already decoded. Exactly the same PaethPredictor() function is
             # used by both encoder and decoder.
             for j, paeth_x in enumerate(line_encoded):
-                if j - bpp < 0:
+                if j < bpp:
                     raw_x_bpp = 0
                     prior_x_bpp = 0
                 else:
-                    raw_x_bpp = int(raw[j - bpp])
-                    prior_x_bpp = int(line_above[j - bpp])
-                prior_x = int(line_above[j])
+                    raw_x_bpp = raw[j - bpp]
+                    prior_x_bpp = line_above[j - bpp]
+                prior_x = line_above[j]
                 paeth = paeth_predictor(raw_x_bpp, prior_x, prior_x_bpp)
                 raw_x = (paeth_x + paeth) & 255
                 raw.append(raw_x)