From 60e329fdb4ee1debc49ca2eae5e351b6b2004d51 Mon Sep 17 00:00:00 2001
From: Jamie Hardt <jamiehardt@me.com>
Date: Thu, 2 Jan 2020 10:50:36 -0800
Subject: [PATCH] Improved exceptions in certain EOF cases

Pursuant to #4
---
 wavinfo/__init__.py    |  3 ++-
 wavinfo/riff_parser.py | 46 +++++++++++++++++++++++-------------------
 2 files changed, 27 insertions(+), 22 deletions(-)

diff --git a/wavinfo/__init__.py b/wavinfo/__init__.py
index 61726f2..51fd41f 100644
--- a/wavinfo/__init__.py
+++ b/wavinfo/__init__.py
@@ -5,7 +5,8 @@ Go to the documentation for wavinfo.WavInfoReader for more information.
 """
 
 from .wave_reader import WavInfoReader
+from .riff_parser import WavInfoEOFError
 
-__version__ = '1.3'
+__version__ = '1.3.1'
 __author__ = 'Jamie Hardt <jamiehardt@gmail.com>'
 __license__ = "MIT"
\ No newline at end of file
diff --git a/wavinfo/riff_parser.py b/wavinfo/riff_parser.py
index 290fa76..21833a7 100644
--- a/wavinfo/riff_parser.py
+++ b/wavinfo/riff_parser.py
@@ -1,52 +1,60 @@
 
 import struct
-import pdb
 from collections import namedtuple
 from .rf64_parser import parse_rf64
 
-class ListChunkDescriptor(namedtuple('ListChunkDescriptor' , 'signature children')):
 
-    def find(chunk_path):
+class WavInfoEOFError(EOFError):
+    def __init__(self, identifier, chunk_start):
+        self.identifier = identifier
+        self.chunk_start = chunk_start
+
+
+class ListChunkDescriptor(namedtuple('ListChunkDescriptor', 'signature children')):
+    def find(self, chunk_path):
         if len(chunk_path) > 1:
             for chunk in self.children:
                 if type(chunk) is ListChunkDescriptor and \
                         chunk.signature is chunk_path[0]:
-                            return chunk.find(chunk_path[1:])
+                    return chunk.find(chunk_path[1:])
         else:
             for chunk in self.children:
                 if type(chunk) is ChunkDescriptor and \
                         chunk.ident is chunk_path[0]:
-                            return chunk
+                    return chunk
 
 
-class ChunkDescriptor(namedtuple('ChunkDescriptor', 'ident start length rf64_context') ):
+class ChunkDescriptor(namedtuple('ChunkDescriptor', 'ident start length rf64_context')):
     def read_data(self, from_stream):
         from_stream.seek(self.start)
         return from_stream.read(self.length)
 
-def parse_list_chunk(stream, length, rf64_context =None):
+
+def parse_list_chunk(stream, length, rf64_context=None):
     start = stream.tell()
 
     signature = stream.read(4)
 
-    #print("Parsing list chunk with siganture: ", signature)
     children = []
     while (stream.tell() - start) < length:
-        child_chunk = parse_chunk(stream, rf64_context= rf64_context)
+        child_chunk = parse_chunk(stream, rf64_context=rf64_context)
         if child_chunk:
-             children.append(child_chunk)
+            children.append(child_chunk)
         else: 
-             break
+            break
 
     return ListChunkDescriptor(signature=signature, children=children)
 
+
 def parse_chunk(stream, rf64_context=None):
     ident = stream.read(4)
-    if len(ident) != 4:
-        return
+    size_bytes = stream.read(4)
 
-    sizeb = stream.read(4)
-    size  = struct.unpack('<I',sizeb)[0]
+    if len(ident) != 4 or len(size_bytes) < 4:
+        raise WavInfoEOFError(identifier=ident, chunk_start=stream.tell() - 4)
+
+    size_bytes = stream.read(4)
+    size = struct.unpack('<I', size_bytes)[0]
     
     if size == 0xFFFFFFFF:
         if rf64_context is None and ident == b'RF64':
@@ -58,13 +66,9 @@ def parse_chunk(stream, rf64_context=None):
     if displacement % 2 is not 0:
         displacement = displacement + 1
 
-    if ident in [b'RIFF',b'LIST', b'RF64']:
-        #print("Parsing list chunk with ident: ", ident)
+    if ident in [b'RIFF', b'LIST', b'RF64']:
         return parse_list_chunk(stream=stream, length=size, rf64_context=rf64_context)
     else:
         start = stream.tell()
-        stream.seek(displacement,1)
-        #print("Parsing chunk with start=%i, ident=%s" % (start, ident))
+        stream.seek(displacement, 1)
         return ChunkDescriptor(ident=ident, start=start, length=size, rf64_context=rf64_context)
-
-