CL-Twitter is here, for (some of) your lispish twittery needs. Or it will be, once the code is in a usable state.
Right now you can load twitter.lisp into a REPL and make, delete, list posts, and so on, but it’s broken as an asdf package. I’ll fix that later – right now this is more proof-of-concept stuff while I learn how CL works.
It’s tested in Clozure, but the heavy lifting is all done in drakma and cl-json.
It does. I’m not entirely it’s an active project these days, but I’ve been using it to get up and running with clozure. Getting clbuild to work with clozure involved a bit of jiggery-pokery, however. Basically what I needed to do was:
implementations to point to an architecture-specific release./clbuild update cclclbuild.conf:
DEFAULT_LISP_IMPLEMENTATION=ccl
CCL=/home/alex/clbuild/source/ccl/lx86cl64...and that's it. ./clbuild slime does what I want, and I haven't found anything that breaks ./clbuild install yet.
I'm cutting my teeth on a twitter client library (because yes, the world does need another one, damnit) which I will either a) post to github, or b) bury under the blissful outpourings of /dev/rand.
Well, all I can say is that I hope I’m not treading on anyone else’s namespace with my little effort.
I finally got bored with rolling and re-rolling my own command line processors in Python, so I’ve knocked together a little something to make things simpler for myself.
It’s loosely based on the marvellously prolific Ara T. Howard’s main gem, but it doesn’t have nearly the feature set. Yet. I guess I’ve given myself something to aim for there.
Anyhow, here’s how to use it, nicked straight from the README:
import main
import sys
def callback(params):
print "Called with " + params['foo'].value
m = main.mode("run", callback)
m.option("foo")
m.description="How not to be seen"
main.process(sys.argv)
Called as:
$ python b.pyOutputs:
Usage:
# How not to be seen
b.py run --foo=<foo>Called as:
$ python b.py run --foo=barOutputs:
Called with barThere’s lots left to do (decent error handling, for instance – it makes a modicum of effort, but largely you can expect mistakes to blow up in your face), but it’s a start.
My last post (from ages ago) dealt with creating a parser wrapped around celerity. That approach has since stopped working because of API changes, so I figured I might as well see about doing it properly by directly talking to NekoHTML, which is the parser underlying HtmlUnit. If you install celerity, you get the nekohtml jar anyway, so I figured I might as well try to make some sort of use of it.
Because I’m not so familiar with the Java XML APIs, this took me more hunting than I expected, but I’ve wrapped it up in a gem for posterity.
In other words: ew. This code is icky but (possibly) useful.
require 'nekohtml'
html = "<html><head><title>Title of Majesty</title></head></html>"
Nekohtml.parse(html).at("//TITLE").text
# => "Title of Majesty"