I'm used to C++, and I build my data handling classes/functions to handle stream objects instead of files. I'd like to know how I might modify the following code, so that it can handle a stream of binary data in memory, rather than a file handle.
def get_count(self):
curr = self.file.tell()
self.file.seek(0, 0)
count, = struct.unpack('I', self.file.read(c_uint32_size))
self.file.seek(curr, 0)
return count
In this case, the code assumes self.file
is a file, opened like so:
file = open('somefile.data, 'r+b')
How might I use the same code, yet instead do something like this:
file = get_binary_data()
Where get_binary_data()
returns a string of binary data. Although the code doesn't show it, I also need to write to the stream (I didn't think it was worth posting the code for that).
Also, if possible, I'd like the new code to handle files as well.
You can use an instance of StringIO.StringIO (or cStringIO.StringIO, faster) to give a file-like interface to in-memory data.
Take a look at Python's StringIO module, docs here, which could be pretty much what you're after.
I like the timing of the answer. (except mine)
We can see response time in milliseconds ?
Use StringIO.
Have a look at 'StringIO' (Read and write strings as files)