A combination of persistently distracting $WORK and and initial lack of overview of the problem at hand has delayed this blog post more than an extra week. I had a clear goal, but no clear way of getting there.
My clear goal was this: get the Lobster running locally on the browser. For this, I would use the HTTP::Daemon module written by mberends++. I had significant problems using that code, most of which later turned out to be due to my own inability to follow mberends' given instructions. Scant progress and plenty of distractions are a bad combination.
When I finally managed to hook the Lobster up to HTTP::Daemon, things went quickly. Strangely enough, nothing happened when clicking the 'flip!' link... a moment's reflection gave the cause of this: a few days ago, TimToady has syntactically separated reversal of List (.reverse) from the reversal of Str (.flip); my code was like three weeks old and was still using .reverse for strings, which in the newest Rakudo got interpreted as one-elem lists, and list-reversed with no visible effect whatsoever. I fixed that, and found the fact that the link in the Lobster app was already called 'flip' an indication of the appropriateness of the name change. Perl 6 just keeps getting nicer and nicer.
You could head over to github and look at the Lobster in its functioning form. (By the way, I'm considering changing the Lobster to a Nibbler, so as to retain some originality wrt Rack. Anyone who masters the ancient art of ASCII images is welcome to help me out, thus being immortalized as the artist of the example webapp of Web.pm.)
According to the plan, this week should go to "Changing November to run on top of the framework". I've chosen to postpone this a bit, because there's lower-hanging fruit to be picked before that, namely Druid. So I spent this morning hooking up Druid to HTTP::Daemon the same way I did Lobster. It's nice to see the Druid board in the browser — things are really ugly so far (a pre with the ASCII board, followed by a list of links with possible moves), but I have high hopes for improvements for the following weeks.
As a bonus, I got to test Tene++'s Tags module when creating the Druid web output. My use of it looked like this. I can definitely learn to live with that.
A strange thing happened though, when I rebuilt after including Tags in the webapp: the new method in Druid::Game broke! WTF? To repeat, I included one module, and another, completely unrelated one, stopped working. Not good.
After a bit of the usual detective work, I found the culprit: the Tags module exported a sub called map, and Druid::Game.new tried to use the vanilla map that we all know and love. ☺ I'm glad I was the one to find that... feature. (I've since patched the module so that it doesn't do that.)
What else? Oh, just a lot of small details that I've skipped over. Things that don't jump out as plain wrong until you try to run them. Fixing those is tedious, but the advantage is that my piece of Web.pm is no longer a theoretical construct, it's practiaclly useable (and used!).
I will spend the next week solidifying this groundwork, as well as giving Druid::Webapp some love, and starting to look at porting November over to use Web.pm. I also want to spend some time getting back to the overall plan for Web.pm — what we have so far, how the pieces fit together, and where we're heading.
I wish to thank The Perl Foundation for sponsoring the Web.pm effort.
+ ?:
. M..
7
... $
MMO
8MM.
~MM.
MM.
+MMMM~.
MMMMMMMMMO
=MMMMMMMMMMMMMMMMM..
..MMMMMMMMMMMMMMMMMMMMM7MMMMMM .
MMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM?
=MMMMMMMMMMMMMMMMMMMMMMMMMMN.. M,
$ :MMMMMMMMMMMMMMMM .
. MM MMMMMMMMMMMMMM. MMM.,.
I..Z? ~MMMMMMMMMMMMM ...
: .MMMMMMMMMMMMM =
N. .NMMMMM88DMMMMMM. .
Z =MM,,:MMMMMI,:MMM ..D.
,. .MMMMM::$MMMM::~$MMMMMDDMD.
..MMMMMD:,,,,,:,:,,:,,:,,,:::MMM
.MM,,,,,,,,,,,,,,,,,,,,,,,,:,,$ .
=:,,::::,,:?$O88OZZZZON88O$Z8?
.M::,,:N.I,,,,,,,,,,,,. ,.
..Z+,::.+::,,,,,,,,,,:M M
...I~::,:,,,,,,,,8..
Re:might be copyrighted, but here it is
masak on 2009-04-13T15:10:08
Wow, thanks!
I don't want to seem ungrateful, but the likeness to Nibbler isn't striking, I think. Perhaps it's the particular font in my browser, but my eyes first spot a five-pointed star of M's in the middle of the image, and only then identify a face. If my magic wishing-wand worked, I'd wish for something smaller, and preferably full-body.
Still, it's better to use this one and be original than just to parrot Rack's lobster.
:)
I can't wait to give Web.pm and Nibbler a run
Re:Enjoy reading the updates
masak on 2009-04-15T05:18:49
Thank you for the encouragement. I appreciate it.
