Hacker News new | past | comments | ask | show | jobs | submit login
Releasing fast Protocol Buffers for Python with lazy decoding support (connex.io)
13 points by atamyrat on Feb 1, 2011 | hide | past | favorite | 7 comments



Last week there was a discussion about Protocol Buffers for Python when Greplin released C++ wrappers. [http://news.ycombinator.com/item?id=2145534]

I mentioned that we were working on it too, so over the weekend I tried to make it usable by others.

Two main differences are:

1) Lazy decoding support: Message parts are decoded on the fly as they are accessed for first time.

2) No dependencies. C code is generated which compiles to Python module.


This is great news. The company I am currently working at is using the Google Protocol buffers within Python as well, and any and all speedups we can get is absolutely fantastic.

As of right now we are just getting off the ground and I don't think the protocol buffers are going to be our primary concern for a while, either way it is nice to know that in the future we can look forward to using these 3rd party libraries.


Does anyone have the TL;DR version of what PBs are or how they're used?


It is binary serialization format used and open sourced by Google. Unlike JSON, objects must adhere to a predefined scheme/structure, and hence, it is compact.


That was succinct and concise, thank you.


Is anyone pushing these updates back to google? Isn't protobufs open source?


I have just posted this to protobuf mailing list too, and license is the same so it can be integrated.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: