homepage

GzipFile claims to implement BufferedIOBase but doesn't have peek().

Here is a first patch, tests still need to be written.

Same patch with tests.

Committed in r85100.

Hi Antoine,
BufferedIOBase is not documented to have peek():
http://docs.python.org/dev/py3k/library/io.html
Small note about patch:
1) IOError string says "read() on write-only...", should be "peek() on write-only..." ?
2) Should be min() in self._read(max(self.max_read_chunk, n))

Hir Nir,
> BufferedIOBase is not documented to have peek():
> http://docs.python.org/dev/py3k/library/io.html
Ah, you're right.
> Small note about patch:
> 1) IOError string says "read() on write-only...", should be "peek() on write-only..." ?
Indeed.
> 2) Should be min() in self._read(max(self.max_read_chunk, n))
Actually, I think I should have reproduced the algorithm in read(),
where there's a read_size distinct from the size requested by the user.
Thanks for the review, it looks like I should have waited a bit before
committing :)

Here is a patch fixing these issues.

Should be min(n, 1024) instead of max(...)

> Should be min(n, 1024) instead of max(...)
Well, no, because we want to buffer a non-trivial amount of bytes for
the next accesses. So, if n < 1024, buffer at least 1024 bytes.

Right, I missed the change from self.max_read_chunk to 1024 (read_size). Should not peek() limit to self.max_read_chunk as read() does?

> Right, I missed the change from self.max_read_chunk to 1024
> (read_size). Should not peek() limit to self.max_read_chunk as read()
> does?
This is used for the chunking of huge reads, but for peek():
1) there is no chunking (peek() should do at most one raw read)
2) huge reads are not really the use case peek() is intended for

I've committed the improvements in r85221. Thank you!