Add TODOs to gzipstreamfile & minor fix in README

Author: Smerity

README.md

@@ -11,10 +11,11 @@ This is highly inefficient as (a) a gzipped WARC file is composed of multiple in
 
 For detailed usage, see the examples folder, but minimally...
 
-    :::python
-    from gzipstream import GzipStreamFile
-    f = open('huge_file.gz') # Any streaming file object that supports `read`
-    gz = GzipStreamFile(f)
+```python
+from gzipstream import GzipStreamFile
+f = open('huge_file.gz') # Any streaming file object that supports `read`
+gz = GzipStreamFile(f)
+```
 
 # License
 

gzipstream/gzipstreamfile.py

@@ -19,6 +19,8 @@ def restart_decoder(self):
       self.unused_buffer += self.decoder.decompress(unused_raw)
 
   def read(self, size):
+    # TODO: Update this to use unconsumed_tail and a StringIO buffer
+    # http://docs.python.org/2/library/zlib.html#zlib.Decompress.unconsumed_tail
     # Check if we need to start a new decoder
     if self.decoder and self.decoder.unused_data:
       self.restart_decoder()
@@ -41,6 +43,7 @@ def read(self, size):
     return self.read(size)
 
   def readline(self):
+    # TODO: This should work in large chunks rather than a byte at a time
     chars = []
     c = self.read(1)
     while c != '\n':