Language:
switch to room list switch to menu My folders
Go to page: First ... 65 66 67 68 [69] 70 71 72 73 ... Last
[#] Thu Aug 26 2010 18:29:32 EDT from LoanShark @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Seriously. Ford you should just code in Scheme, clearly you'd be
happier...

And your code would compress really well, since 90% of it is just two characters...

[#] Thu Aug 26 2010 19:28:01 EDT from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

So "A" == "A" returns true only by accident, because the compiler
optimized it such that there is only one object instance created.

I didn't see the precedence problem at first so I thought it was probably in the java spec that the compiler will optimize string constants to the same object. I mean it should do it anyway, but being so obvious I thought they might put it in the spec.
the == was obvious, I thought the point was about what the == was operating on.
If you had used .equals it would have been obvious what was being compared to what.


[#] Fri Aug 27 2010 05:20:13 EDT from johnnyl @ Uncensored

Subject: Re:

[Reply] [ReplyQuoted] [Headers] [Print]

 

Thu Aug 26 2010 02:00:10 PM EDT from Spell Binder @ Uncensored
Hey, at least he's honest. :)

Said as in what point was said. foo.



[#] Fri Aug 27 2010 05:21:19 EDT from johnnyl @ Uncensored

Subject: Re:

[Reply] [ReplyQuoted] [Headers] [Print]

 

Thu Aug 26 2010 03:24:14 PM EDT from Ford II @ Uncensored Subject: Re:
I don't know what you just said, but ok.

Don't worry, a lot of people have that problem, it's part of the rite of passage of being able to play ball in this room.
I understand maybe 95-98% of what he says most of the time and Ihave to read 2 paragraphs of wikipedia to get the rest. :-)

Yeah sounds like a real sausage contest.



[#] Fri Aug 27 2010 12:06:54 EDT from LoanShark @ Uncensored

Subject: Re:

[Reply] [ReplyQuoted] [Headers] [Print]

Yeah sounds like a real sausage contest.

I really prefer wiener schnitzel, myself.

[#] Mon Nov 29 2010 16:19:02 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

I've almost given up with gtkmm. Good documentation or bad, it just shouldn't be this hard.
Details for anybody who cares.
They include a very powerful treeview that can do simple lists as well.
So I got it all set up, using glade, you can define the list and the list data, you can even put data in the xml file to render on startup.
Brilliant. But nothing comes up.
Why? Near as I can tell, there's a cell renderer, a type of which gets assigned to your column as you add it (or in this case as gtkbuilder adds it) to the treeview. And when you define your list you can tell it what type of data is in the list so gtk can pick the best kind of cellrenderer to use to draw your data.
Well, your options for what classes of data are allowed are volumnious and useless. You do get a gchararray (which for other reasons is the only one tha works.)
But when you call treeview->get_column_cell_renderer() you get NULL which means nothing was defined for ytou.
And I can't find any way to add one once a column has been created.
Not to mention that glade lets you add types to your list store that it knows damn well it won't be able to render.
Gargh.

[#] Tue Nov 30 2010 12:37:32 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

So with the help of the wonderful people at stackoverflow I got it working.


http://stackoverflow.com/questions/4307786/glade-treeview-liststore-data-not-v
isible-cant-imagine-why

And what I realize is this: once again somebody has missed the boat.


GTK is all powerful and lets you do anything. The treeview class is infinitely configurable and customizable so that you can have variable width and height and user-drawable items in it.

But the fact is nobody's going to use any of that shit, and mostly people just want a list box with text items in it.
so rather than design it for the 99% use case of adding text items to a list, they design it so that you have to go through ALL the hell just to get a listbox.
why not instead design for the simple case and allow for the complex case by making that part hard.


But at least I can add to my list now.


[#] Tue Nov 30 2010 12:39:52 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Oh so I didn't explain the solution.

It turns out that you have to define the data store as having n columns of X type of data which I was doing.

But you ALSO have to define the tree view as having n number of columns of X type of data. And the menu option to set it, isn't all that obvious, so I didn't know you could even set it let alone that it had to be set at all.

Why couldn't the treeview just take its design from the data model? Too fucking simple for the developer obviously.

[#] Tue Nov 30 2010 20:24:47 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

I love being behind the curve, I got hello world to run on my g1 in about 5 minutes.
Now... what to do next....

[#] Wed Dec 01 2010 20:08:18 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Subject: Haskell Researchers Announce Discovery of Industry Programmer Who
Gives a Shit
Comments


http://steve-yegge.blogspot.com/2010/12/haskell-researchers-announce-d

iscovery.html


Stolen right out of the onion.

[#] Thu Dec 02 2010 13:05:38 EST from Spell Binder @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

I haven't played around with Haskell too much, yet, but I've been researching on ways to incorporate certain features of Haskell into TCL. Mainly because TCL is what we use for the bulk of our automated test scripts where I work.

I should see if there's an expect library for Haskell, though I get the feeling I'd want to have myself committed if I saw how it was implemented. If there is one, though, it would make an interesting experiment to port a couple of our test scripts into Haskell.
Spell

[#] Sat Dec 04 2010 12:47:36 EST from IGnatius T Foobar @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

(Spell is still waiting for a computing environment implemented completely in Smalltalk.)

[#] Sat Dec 04 2010 13:34:58 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

I'll let ls figure this out and explain it to me:

So I'm having problems linking my C++ program with the android cross compiler gcc. Apparently they have a slightly modified version of gcc? I dunno.
Anyway, I narrowed it down to virtual destructors.
Then I found this : http://zedcode.blogspot.com/2007/02/gcc-c-link-problems-on-small-embedded.html
guy has exact same problems.
His solution is to supply an overriding delete operator because he doesn't use new and delete. Gets rid of his linker problem.
Doesn'ty solve my problem though because I need to delete things.
What I'm actually going to do is remove my inhereted class and just make the class I need (listen to me, i'm listening to myself, all that reuse stuff is bullshit!)
One of the things I read said basically that the android c++ compiler supports just enough to compile webkit and that's it.

Which is a) dumb, but b) cool because it again proves my argument that if you keep things simple you'll be better off. Almost all my code compiles in this crippled compiler, except for the one thing I started to get complex with.

There's at least one guy having a hell of a time trying to get boost to build. Good luck to him. Probably shouldn't be running boost on a phone anyway (at least not yet)


So perhaps I answered my own question: basically the crippled compiler doesn't support inheretence.

So I can either find a better cross compiler and fight to the death to get a build kit that works on android, or the much more simple option: get rid of my inheretence.

[#] Sat Dec 04 2010 14:02:04 EST from dothebart @ Uncensored

Subject: Re:

[Reply] [ReplyQuoted] [Headers] [Print]

19:49 @dothebart: cirrus: wo hastn du deinen android cross compiler toolchain her? aus embedian?
19:50 @cirrus: das offizielle ndk, bzw. der fork von crystax mit aktivierter libstdc++

-> using the NDK / the fork of crystax with activated libstdc+

which is probably what you need too, ford.



[#] Sat Dec 04 2010 17:59:45 EST from LoanShark @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]


Ford, all I know is that used to be handled in collect2. If they didn't test the feature in their version of collect2, it's probably broken...

[#] Sat Dec 04 2010 21:21:55 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

never heard of crystax, but I'll chjeck it out, thanks.

[#] Sat Dec 04 2010 21:22:15 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

ls: you can't dissapoint me... you know EVERYthing. :-)

[#] Wed Dec 08 2010 13:12:21 EST from LoanShark @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]


Everybody follows
Speedy bits exchange
Stars await to glow

(stolen without permission)

[#] Wed Dec 08 2010 15:25:23 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

Everybody follows
Speedy bits exchange
Stars await to glow

So that's why it takes so long to make an oracle connection.

[#] Wed Dec 08 2010 17:49:34 EST from Ford II @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

today I learned that on android an int a long and a pointer are all 4 bytes.
Go figure.

Go to page: First ... 65 66 67 68 [69] 70 71 72 73 ... Last