TST: Add Test for ASCII85Decode (#825)

Full credit to
6dc90b1

Co-authored-by: Acsor <[email protected]>

PyPDF2/filters.py

@@ -325,30 +325,30 @@ def decode(data, decodeParms=None):
         if version_info < ( 3, 0 ):
             retval = ""
             group = []
-            x = 0
-            hitEod = False
+            index = 0
+            hit_eod = False
             # remove all whitespace from data
             data = [y for y in data if y not in ' \n\r\t']
-            while not hitEod:
-                c = data[x]
-                if len(retval) == 0 and c == "<" and data[x+1] == "~":
-                    x += 2
+            while not hit_eod:
+                c = data[index]
+                if len(retval) == 0 and c == "<" and data[index+1] == "~":
+                    index += 2
                     continue
                 # elif c.isspace():
-                #    x += 1
+                #    index += 1
                 #    continue
                 elif c == 'z':
                     assert len(group) == 0
                     retval += '\x00\x00\x00\x00'
-                    x += 1
+                    index += 1
                     continue
-                elif c == "~" and data[x+1] == ">":
+                elif c == "~" and data[index+1] == ">":
                     if len(group) != 0:
                         # cannot have a final group of just 1 char
                         assert len(group) > 1
                         cnt = len(group) - 1
                         group += [ 85, 85, 85 ]
-                        hitEod = cnt
+                        hit_eod = cnt
                     else:
                         break
                 else:
@@ -361,37 +361,42 @@ def decode(data, decodeParms=None):
                         group[2] * (85**2) + \
                         group[3] * 85 + \
                         group[4]
+                    if b > (2**32 - 1):
+                        raise OverflowError(
+                            "The sum of a ASCII85-encoded 4-byte group shall "
+                            "not exceed 2 ^ 32 - 1. See ISO 32000, 2008, 7.4.3"
+                        )
                     assert b <= (2**32 - 1)
                     c4 = chr((b >> 0) % 256)
                     c3 = chr((b >> 8) % 256)
                     c2 = chr((b >> 16) % 256)
                     c1 = chr(b >> 24)
                     retval += (c1 + c2 + c3 + c4)
-                    if hitEod:
-                        retval = retval[:-4+hitEod]
+                    if hit_eod:
+                        retval = retval[:-4+hit_eod]
                     group = []
-                x += 1
+                index += 1
             return retval
         else:
             if isinstance(data, str):
                 data = data.encode('ascii')
-            n = b = 0
+            group_index = b = 0
             out = bytearray()
             for c in data:
                 if ord('!') <= c and c <= ord('u'):
-                    n += 1
+                    group_index += 1
                     b = b*85+(c-33)
-                    if n == 5:
+                    if group_index == 5:
                         out += struct.pack(b'>L',b)
-                        n = b = 0
+                        group_index = b = 0
                 elif c == ord('z'):
-                    assert n == 0
+                    assert group_index == 0
                     out += b'\0\0\0\0'
                 elif c == ord('~'):
-                    if n:
-                        for _ in range(5-n):
+                    if group_index:
+                        for _ in range(5-group_index):
                             b = b*85+84
-                        out += struct.pack(b'>L',b)[:n-1]
+                        out += struct.pack(b'>L',b)[:group_index-1]
                     break
             return bytes(out)
 

Tests/test_filters.py

@@ -1,10 +1,11 @@
+# -*- coding: utf-8 -*-
 import string
 from itertools import product as cartesian_product
 
 import pytest
 
 from PyPDF2.errors import PdfReadError, PdfStreamError
-from PyPDF2.filters import ASCIIHexDecode, FlateDecode
+from PyPDF2.filters import ASCIIHexDecode, FlateDecode, ASCII85Decode
 
 filter_inputs = (
     # "", '', """""",
@@ -97,3 +98,43 @@ def test_ASCIIHexDecode_no_eod():
     with pytest.raises(PdfStreamError) as exc:
         ASCIIHexDecode.decode("")
     assert exc.value.args[0] == "Unexpected EOD in ASCIIHexDecode"
+
+
+@pytest.mark.xfail
+def test_ASCII85Decode_with_overflow():
+    inputs = (
+        v + "~>"
+        for v in "\x00\x01\x02\x03\x04\x05\x06\x07\x08\x0e\x0f"
+        "\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a"
+        "\x1b\x1c\x1d\x1e\x1fvwxy{|}~\x7f\x80\x81\x82"
+        "\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d"
+        "\x8e\x8f\x90\x91\x92\x93\x94\x95\x96\x97\x98"
+        "\x99\x9a\x9b\x9c\x9d\x9e\x9f\xa0¡¢£¤¥¦§¨©ª«¬"
+        "\xad®¯°±²³´µ¶·¸¹º»¼½¾¿ÀÁÂÃÄÅÆÇ"
+    )
+
+    for i in inputs:
+        with pytest.raises(ValueError) as exc:
+            ASCII85Decode.decode(i)
+        assert exc.value.args[0] == ""
+
+
+@pytest.mark.no_py27
+def test_ASCII85Decode_five_zero_bytes():
+    """
+    From ISO 32000 (2008) §7.4.3:
+    «As a special case, if all five bytes are 0, they shall be represented
+    by the character with code 122 (z) instead of by five exclamation
+    points (!!!!!).»
+    """
+    inputs = ("z", "zz", "zzz")
+    exp_outputs = (
+        b"\x00\x00\x00\x00",
+        b"\x00\x00\x00\x00" * 2,
+        b"\x00\x00\x00\x00" * 3,
+    )
+
+    assert ASCII85Decode.decode("!!!!!") == ASCII85Decode.decode("z")
+
+    for expected, i in zip(exp_outputs, inputs):
+        assert ASCII85Decode.decode(i) == expected