pystream-protobuf 1.5.1

Python implementation of stream library

pyStream

Python implementation of stream library. It enables stream processing of protobuf messages; i.e. multiple protobuf messages can be written (read) into (from) a stream by using this library. It can be used for parsing all files encoded by stream library and writing protobuf instances into a file by the same encoding. Refer to the library GitHub page for more information about formatting.

Installation

You can install pyStream using pip:

pip install pystream-protobuf

Usage

Reading

Here is a sample code to read a file containing a set of protobuf messages (here is a set of VG's Alignment objects, so-called GAM file, defined here). The Alignment class is just an example and it can be any protobuf message. It yields the protobuf objects stored in the file:

import stream
import vg_pb2

alns = [a for a in stream.parse('test.gam', vg_pb2.Alignment)]

Instead of file path, an input stream can be passed to the method parse:

import stream
import vg_pb2

# ... an already existing file-like object `f` as input stream.
alns = [a for a in stream.parse(f, vg_pb2.Alignment)]

In order to have more control over opening the stream and reading data, the lower-level method open can be used:

import stream
import vg_pb2

alns_list = []
with stream.open('test.gam', 'rb') as istream:
    for data in istream:
        aln = vg_pb2.Alignment()
        aln.ParseFromString(data)
        alns_list.append(aln)

The stream can be closed by calling close method explicitly, in which case the stream is opened without using with statement (see more examples in the test package).

The method open is not restricted to files, as it can be used for any binary stream. It can be done by passing file object rather than file name to method open or directly to Stream class:

# ... an already existing file-like object `f` as input stream.
with stream.open(fileobj=f, mode='rb') as istream:
# ... continue using istream

Writing

Multiple protobuf objects can be written into a file or any output stream by calling dump function. An example of writing a list of Alignment objects to a file named test.gam:

import stream

stream.dump('test.gam', *objects_list, buffer_size=10)

If writing to an existing output stream is desired, the dump method accepts any file-like object as output stream:

import stream

# ... an already existing file-like object `f` as input stream.
stream.dump('test.gam', *objects_list, buffer_size=10)

Or using open method for lower-level control. This example appends a set of messages to the output stream:

import stream

with stream.open('test.gam', 'ab') as ostream:
    ostream.write(*objects_list)
    ostream.write(*another_objects_list)

Similar to reading, open method accepts fileobj argument for any output stream and the stream can be closed by explicitly calling close; particularly when the stream is opened without using with statement.

More features

Optional GZip compression

The streams encoded by Stream library is GZip compressed. The compression can be disabled by passing gzip=False when opening an stream.

Buffered write

By default, all protobuf message objects provided on each call are written in a group of messages (see Stream library for encoding details). The messages can be buffered and write to the stream in a group of fixed size whenever possible. The size of the buffer can be set by keyword argument buffer_size to open, dump methods or when Stream class is constructed (default size is 0 --- means no buffer).

Grouping message

Messages can be grouped in varied size when writing to a stream by setting buffer size sufficiently large or infinity (-1) and calling flush method of Stream class whenever desired.

Group delimiter

Group of objects can be separated by a delimiter of the choice (or by default None) when reading from a stream. Sometimes, it can help to identify the end of a group which is hidden from the library user by default. This feature can be enable by setting group_delimiter to True when constructing a Stream instance or opening a stream. The delimiter class can also be specified by delimiter_cls.

Development

In case, you work with the source code and need to build the package:

python setup.py build

The proto file in the test module required to be compiled before running test cases. To do so, it is required to have Google protobuf compiler (>=3.0.2) installed. After installing protobuf compiler, run:

make init

to compile proto files required for test module. Then, use nosetests command of the setup script to execute test cases:

python setup.py nosetests