programmierung - 12.2.2010 - 12.5.2010

sethtrain's beget - alternatively to leiningen-war, you could also use this base project and simply adapt it. The Google AppEngine Tools are also fetched as a dependency here.

Marak's JSLINQ at master - GitHub - a nice small JavaScript library that offers a query language for JSON data. It is oriented towards Microsoft's LINQ, but currently only has simple queries implemented. Nevertheless, it might be quite interesting to make JavaScript code more flexible and readable when working with larger amounts of JSON data.

parsedatetime - a very practical library that converts "normal" date specifications (unfortunately only in English as far as I can see) into Python datetime objects.

PyPy Status Blog: Running wxPython on top of pypy - PyPy is really making huge strides towards being usable. It's already faster than CPython in some cases and now even larger C extensions like wxPython are running. Cool.

Zoolander - a small Python library that allows you to use Python as a DSL for generating CSS. Sounds silly at first, but if you want or need to produce CSS dynamically and then embed it in a web framework, it can be quite practical.

The Brads – How to Alienate a Fanbase - if anyone needs a short summary of what Adobe stands for.

Thoughts on Flash - is of course again dismissed as blah-blah by all Apple opponents, but well - the reasons are compelling. And sorry, but it's really true: Flash stinks.

django-pagination - I need to take a closer look at this, it looks interesting. Pagination is not really difficult, but it's annoying to build it yourself every time - and Django's built-in tools are not always optimal for this (especially with large amounts of data).

Henry's EuLisp - someone has revived EuLisp and gathered the sources, as well as the specification. At least historically interesting, because EuLisp was one of the standard efforts for a more modern Lisp with quite good object-oriented support. But the implementation itself also has some interesting features.

jcotton - Build animations and graphics with JavaScript and Canvas. Looks quite interesting.

XML in Postgres – The Game Changer « Flex and Specs() - I should really take a closer look at the new PostgreSQL features. Especially since the XML support in PostgreSQL brings some of the advantages of document-oriented databases to the relational world, without needing extra middleware.

Archives of the Caml Mailing list: O'Caml for DOS - because I just stumbled upon it again. Wow, 96, that's a long time ago. Why is OCaml always listed as such a modern language? It's already 14 years old ... (and the language on which OCaml is based - Caml Light - is even older)

Daring Fireball: New iPhone Developer Agreement Bans the Use of Adobe's Flash-to-iPhone Compiler - well, of course Apple has the right to set the terms themselves. And I have the right to find iPhone programming completely uninteresting now - sorry, but I'm not going to deal with such low-level programming languages anymore.

django-ajax-filtered-fields - I need to take a closer look at this, it could be quite interesting in the Admin for larger amounts of sentences in relations.

My experience with using MongoDB for great science. - NoSQL is, after all, in many cases a playground for people who want to try out how databases actually work. With many of these projects, I already wonder what possessed them when they built it. I'd rather rely on solid and proven tools like PostgreSQL and SQLite. And if a NoSQL database, then better one that has been in productive use in larger installations for a longer time. Cassandra comes to mind, for example.

twitter's gizzard - could become interesting, a framework for distributing and replicating data across various backends. Gizzard deals exclusively with sharding and replication, the datastore itself is treated separately, making it interesting for various scenarios.

Writing a non-relational Django backend - Django nonrel / NoSQL blog - All buttons pressed - I'm not a big fan of NoSQL (in my opinion, many NoSQL approaches reflect a lack of understanding of relational databases rather than actual shortcomings or weaknesses of relational databases), but if you're going to use NoSQL, I'd prefer to do it through the Django ORM, which I quite like. And here's how you can build a Django ORM wrapper for NoSQL databases with relatively little effort.

Perfection kills » What’s wrong with extending the DOM - because I keep discussing with colleagues why JQuery is better than Prototype: Prototype heavily uses the extension of prototypes, while JQuery hangs almost everything on its own JQuery object and is therefore much more cooperative in interaction with other JavaScript.

Oracle Announces Latest Release of Oracle® Berkeley DB - Berkeley DB now has a SQL API based on SQLite. Source code compatibility with SQLite, so programmers can switch if they prefer the much more unstable and vulnerable storage of Berkeley DB and like to repair their databases. Sorry, Oracle, but that's ridiculous. BDB is only interesting for those who have to work with it by force - anyone who wants to switch to BDB today must be crazy. If I'm already programming against the SQLite API, I'd rather use the right tool right away. Yes, of course, SQLite has some bottlenecks when you want to access it in parallel with multiple processes. But I'll let Oracle in on a little secret here: SQLite has such a tolerant SQL parser because you can then write source code whose SQL works seamlessly with both SQLite and PostgreSQL. So if you hit the limits of SQLite - just switch to PostgreSQL and you're good to go.

NLTK Home (Natural Language Toolkit) - and if you want something more powerful and flexible, this is so to speak the grab bag for parsers. Focus is on the analysis of natural languages, hence also things like stemmers (stem finding for word forms) are included. Could be overkill for simple embedded languages, though.

Python Package Index : Esrapy 0.5 - a parser and lexer toolkit completely in Python. Might become interesting in some projects later, at least for smaller configuration languages.

Building Skills in Python - Online book about Python for programmers who simply don't know the language yet. Looks very well made at first glance.

Bottle: Python Web Framework - super-simple Python web framework that comes as a single Python file. No dependencies other than the standard library. No built-in ORM, but very lean and perhaps interesting for projects where you don't need or want a database (or use the file system as a database).

clojure-python - an interesting project that aims to simplify interoperability between Jython and Clojure and raise it to a similar level as it already is between Clojure and Java. Particularly interesting for me, as it would allow me to rely more on Clojure as an alternative - Jython is already a planned component of the toolkit, but has some performance issues that Clojure does not have through more direct Java integration. Moreover, I prefer to write compact Lisp code rather than verbose Java ...

hugoduncan's clj-ssh at master - GitHub - quite an interesting library that enables ssh access in Clojure scripts. For example, very interesting for server automation. Uses Jsch, a native Java ssh library (so no detour via shell-pipes or similar).

Scala: Post-Functional, Post-Modern, or Just Perl++? - interesting post that addresses some of the points that also bother me when looking at Scala. I particularly like the designation as Perl++, as that is exactly the impression that comes to me whenever I delve deeper into Scala. Perl has always fascinated me, but by the time I built larger projects with it and used the advanced features of Perl more intensively, I had some doubts about the maintainability of the result - especially with regard to handing over the work to one of my colleagues for further maintenance. At that time, I switched to Python because it offered me many of the features in a much cleaner language concept. I think this could also explain why I just can't warm up to Scala, even though many aspects of it fascinate me.

digg's lazyboy at master - GitHub - because key-value datastores are currently all the hype (and because they are really more practical for some things than classic databases), I will probably take a look at Cassandra. Simply because reports on the web suggest it offers the best scaling possibilities. And because it is used in some large websites - specifically, for example, at Digg (which I find as a site to be stupid, but hey, they have a lot of traffic and run relatively stably) with lazyboy as the Python binding.

17.6. multiprocessing - much better than external modules for process communication are the tools included with Python since version 2.6 in multiprocessing.

rfc1437 / lazypy / source — bitbucket.org - and another project of mine (again) online. Lazypy is a small library that makes lazy evaluation and futures (thread and process based) available for Python. Very practical for simple concurrent programming. Ok, you can do everything by hand, but I prefer the more functional approach. It's actually from 2004, but I modernized it (the process-based futures to bypass the GIL) and uploaded it again.

Semanchuk.com - Python IPC Modules - inter-process communication with Python.

A simple web application in Clojure using ring and enlive « LShift Ltd. - and here is a small example of how to actually work with ring and Clojure. Looks quite interesting, could be particularly interesting for me for web services in Clojure.

Dynamic Web Development with Seaside - those who want to get started with Seaside might find an approach here. Free online book (also available as a paid PDF or print-on-demand via Lulu) about a quite impressive web framework for Smalltalk. And since it now also runs with GNU Smalltalk, operating it as a headless server on your own root box is no longer a big problem.

Heroku | Ruby Cloud Platform as a Service - also quite interesting: a Ruby service that enables easy website hosting in Ruby within a cloud structure. In principle, something like Google App Engine, but with Ruby. The approach is quite interesting, you generate a base app and then pull it to your own computer with Git, change and update with Git. There are various addons and plugins that you can use, Rails is of course also supported. And since you keep your app as a normal Ruby app locally, you are also relatively independent of the provider and can switch to self-hosting if necessary.

inessential.com: On switching away from Core Data - scary read. Really - sure, ORMs are nice. And practical. But somehow it scares me when programmers like Brent Simmons (the NetNewswire guy) so openly demonstrate that they actually have no idea what they are doing. Just because you use an ORM to walk through lists of objects and edit individual objects and then wonder about poor performance? And only at the end of the optimization sessions ask the question whether an ORDB is actually the right way? Hello, are you still there? As soon as large amounts of data are involved, the question of mass data processing automatically arises, and if the ORM does not provide usable abstractions, then it is out ... (one reason why I like the Django-ORM, it cooperates well with handcrafted SQL and offers a lot of helper tools via introspection to create these own SQLs as model-abstract as possible). For me, the linked post is on a similar level as Guido van Rossum's "what do you actually use continuations for, I don't get it".

Johnny Cache v0.1 documentation - definitely try this out with a project at the company, as the model is quite powerful and I could elegantly solve some of the problems for which I currently have special solutions. It's also quite similar to my first approach to this problem (although I have now solved the biggest performance problems through redundant data storage and automatic updates to objects, also quite elegantly).

Kotka : Projects : Clojure : VimClojure - and anyone like me who is a VIM fan might be pleased with this Clojure integration. Many of the features already come close to the performance class of IDEs like Netbeans or Eclipse. (although the Clojure plugins for Eclipse and Netbeans also make a very good impression).

LinuxTuples - a Tuple-Space Server for Linux, written in C but with a Python API. I should take a closer look, could be interesting for distributed apps. Although I would prefer a Python-local implementation based on standard process communication tools to work better with multiprocessing in Python. For simple tools or web apps, it would be easier to fork some things directly from there and then communicate via TupleSpaces. But to start an extra server for that, it's not quite right either.

mmcgrana's ring at master - GitHub - nice small lib at the level of Python WSGI. So absolutely minimal HTTP bindings for Clojure with the possibility to then operate the whole thing via a whole range of different techniques. Particularly interesting for the cases where you don't want to be put into the corset of a finished framework like Compojure.

PiCloud | Cloud Computing. Simplified. - very interesting service: trivial distribution of Python code (with access to C/C++ libraries for number crunching and other things, e.g. also image processing, even your own C/C++ libraries are possible) on a provider-managed EC2 grid. The programmer only writes his Python code, tests locally, if everything works well with small sets, upload base data, import, function call and wait until the results are there - payment is made according to usage time. Definitely keep an eye on it, in case you need to process larger amounts of data - this can indeed be cheaper than providing the necessary resources yourself.

rfc1437 / django-standalone / overview — bitbucket.org - As I often refer to Bitbucket, GitHub, or Google Code, here's a link to my own small package on Bitbucket: django-standalone. It was created because I often need an ORM for small scripts and tools, but I want to keep the hassle to a minimum - not set up an entire Django project, but simply define a few models in my script and initialize the DB via parameters and then use it. Preferably with sqlite3. This library makes it wonderfully easy, and I can finally cross one of my long-standing projects - "write a simple ORM for simple scripts myself" - off my to-do list.

dajaxproject.com - easy to use ajax library for django - maybe I should take a look at that, the current project might use quite a bit of Ajax and if you can reduce the amount of JavaScript that would be quite desirable.

Squeryl — Introduction - I should also take a look at that, as I was not so enthusiastic about the previous persistence layers for Scala. And for initial experiments, I actually don't want to build a web application with Lift right away, but perhaps just rewrite a few tools that I have solved differently with Scala.

IronPython 2.0 and Jython 2.5 performance compared to Python 2.5 - word of warning: both Jython and IronPython are significantly (and I mean significantly significantly) slower than CPython. The overhead for Jython with very large data structures is eventually better than with CPython, but for normal use it doesn't look so great.

IronPython hammers CPython when not mutating class attributes - more information about the performance issue. Here related to IronPython - apparently class variables can be problematic in some cases, as the classes themselves change through them and thus JIT compiler information must be discarded (due to the rather static structure of the VM, both in the JVM and in the CLI, probably a very similar problem), which means the JIT compiler then has to rework everything and thus not only are performance advantages lost, but they can potentially even be counterproductive.

bpython interpreter - definitely try this out at the company, at home this alternative Python shell made a really good impression. In some points better than ipython and that is already very good (but in my opinion too focused on its own features and away from the Python philosophy, while bpython seems more pythonic)

DreamPie: The Python shell you've always dreamed about! - another interesting alternative Python shell, this one as a GTK window. This opens up completely new possibilities, such as real popups as small graphical windows and direct graphical output. However, py-gtk for OS X is still rather shaky (like all GTK stuff currently, it's still alpha) and actually I prefer similar environments under OS X and Linux.

ZODB - a native object database for Python — ZODB v3.9.0 documentation - because I always forget: ZODB is also available standalone, without the Zope monster on top. And with direct access from Python, ZODB offers some very interesting features. Still one of the most developed real object databases for Python (but unfortunately still no general query language for the database to handle object sets more efficiently).

django-piston - I should also take a closer look at this, as it is supposed to facilitate the building of Web APIs with Django. And some of my company projects could benefit from it.

Murky - a nice small GUI client for Mercurial for OS X. It already looks quite useful, you can easily navigate through the history of a project, display differences, etc. - of course, all of this can also be done with the shell, but sometimes it's just simpler to work with a GUI.

Front Range Pythoneering: Realizing Jython 2.5 - it's written further down there. Jython has a GIL as a funny Easter egg in the future module (which makes future language features available as "Beta"). So no GIL, just a joke. I would have been quite surprised otherwise.