Feed Sign in with OpenID OpenID

Simon Willison’s Weblog

Current State of Unladen Swallow (Towards a Faster Python). The Unladen Swallow team are now planning to merge their work back in to the mainline Python 3 branch, adding a powerful incentive for Python developers to port their old Python 2 code (since the official Python 2.x line is extremely unlikely to have Unladen Swallow merged in to it).

Tagged , ,

8 comments

  1. Now if the MySQL and Postgres drivers will only move to Python 3

    Josh - 7th January 2010 14:46 - #

  2. I agree Josh, when I read things like this:

    "Python programmers in general do not like SQL too much, they prefer to work with objects."
    http://www.python.org/doc/current/howto/webservers .html#data-persistence

    It makes me cringe. I guess PHP just spoiled me. I wanted to learn Python, 3 seemed like a good place to start since it's new, I couldn't even talk to my database, gave up on python.

    If relational database support isn't a core feature then I guess python isn't the language for me.

    Rob - 7th January 2010 15:24 - #

  3. @rob: You should have posted rest of the paragraph in your quote as well. What they are saying is they do have database suport, but python programmers prefer to use one of the ORMs. So, python do have rich database support, I would strongly encourage you to learn the language. Following is the full pararaph from the link.
    Often relational database engines like MySQL or PostgreSQL are used due to their good performance handling very large databases consisting of up to millions of entries. These are queried using a language called SQL. Python programmers in general do not like SQL too much, they prefer to work with objects. It is possible to save Python objects into a database using a technology called ORM. ORM translates all object-oriented access into SQL code under the hood, the user does not need to think about it. Most frameworks use ORMs and it works quite well.

    Senthil Vaiyapuri - 7th January 2010 17:21 - #

  4. Nice to see someone point out what "Python programmers in general" prefer. Even if there's an element of truth in it, it doesn't diminish the need for actually using the database system's features directly. I continually have the impression that many ORM users (I could write "ORM users in general"...) have narrow querying habits on small databases that seem to involve operations that can frequently be done in memory, yet seem like much wizardry was required. I'd like to see what ORMs can do for the entire data management scenario, because some optimisation tasks are certainly not magically resolved by a bit of technology that knows how to make objects out of joined tables, even if that is a nice achievement in itself.

    Paul Boddie - 7th January 2010 23:39 - #

  5. @Senthil: The 3rd party modules for mysql/postgresql support aren't (weren't) updated for version 3.

    Because of this, I have to decide if python is supported enough to make it worth my time to pickup. It seems the core language is missing elements I consider essential, and so I'll pass. I feel that there is too much risk that the 3rd party frameworks and modules required to make the language useful to me are not stable enough.

    Rob - 8th January 2010 19:43 - #

  6. Back on topic: an interesting look at Unladen Swallow's progress here:

    http://renesd.blogspot.com/2010/01/unladen-swallow -review-still-needs-work.html

    Paul Boddie - 9th January 2010 21:17 - #

  7. Am I the only one wanting to see the Unladen Swallow project post consistent, stable improvements _before_ this push to be included in the mainline begins?

    It's great that Google are investing this effort, but they've also repeatedly stated that the only speed improvements they're concerned with are the ones that benefit them directly.

    Why should the Python3 devs be expected to shoulder the ongoing maintenance of code that isn't aimed at being of benefit to everyone?

    alex dante - 11th January 2010 04:10 - #

  8. Super informative wirting; keep it up.

    Janet - 9th September 2011 07:10 - #

Comments are closed.
A django site