Zip
|
The rant was not a gannon-bash, as much a reference to other posts. Believe me, refering to an array of structs with a passed int is certainly preferable to copy contructing each time for a struct return, and several other suggestions are 'correctness' rather than functionality based.
Certainly a few moar general libraries for useful common things would be nice - but I'm not sure how naming is important here. Surely a textbox struct for a library should be used only by the library funcs (abstracted data and all) that form the interface with the user? So why would the user need to worry about how it was named at all? Generic array sort is useful, but a similar thing with structs would be a big task in anyone's book.
Zip
Posted on 2004-08-08 02:44:59
|
gannon
|
I already made a general quicksort (see general shell sort thread) for ints expanding it to structs would not be hard.
ie:
quickSort(string arrayName, int arrayLen, string keyName, string swapList)//sorts the array using the keyName of the structs that have elements according to swapList
//swapList would have to be in a specific format to denote all the data types
What I am saying is that they should not have to know how the other persons code at all (either the users or the lib makers).
also rants are good in my book the verge community needs to improve itself and it can't do that by saying its all good.
Posted on 2004-08-08 04:13:58
|
Zip
|
Yes, I'm aware of (and impressed by) your sort, hence the mention. However, trying a swaping method like that with a large struct of multiple strings and ints would be a very bad idea. A more practical method would be to stack the data, and refer via and array index, then sort that with resect to one of the data types.
Zip
Posted on 2004-08-08 04:26:22
|
gannon
|
well if it was me I am more of a keep things sorted all the time so I would always insert into the correct array index.
but this is not about me this is about making things someone with little programming experience could just put in there code and call it.
plus if you had a ready made sorting by structs made for you, you wouldn't have to worry about parallel arrays so there would be less things to go wrong
maybe a good idea would be have the get/set cashe last one or two things that it did. that would make bind not needed
Posted on 2004-08-08 05:17:36 (last edited on 2004-08-08 05:18:21)
|
zaril
|
by the way, I was being silly.
Posted on 2004-08-09 11:16:05
|
andy
|
I don't suppose anybody cares, but I have written a relatively complete VC compiler that can handle arbitrary nesting of structs and arrays. (I went and made them reference types, so passing and returning them is easy too)
The main things that are missing are const and import. (better mechanisms for #define and #include, essentially)
Posted on 2004-08-09 18:28:52 (last edited on 2004-08-09 18:32:17)
|
gannon
|
I would love to see nesting of structs and passing of them let us see it. :)
Posted on 2004-08-09 18:40:05
|
andy
|
When my personal webspace provider comes back online (in a day or two) I'll upload a "stand-alone" demonstration package. Basically, a console-mode interpreter and compiler.
Integration with v3 is nontrivial because the VC bytecode format isn't documented, and even if it were, I'm somewhat skeptical that it can handle the sorts of things the compiler needs. (in particular, it pretty much requires a garbage collector)
(integration would mainly entail replacing the existing VC interpreter with the new one)
I've also thought about attaching it to v2.6, but it would probably be more work than it is worth. After all, nobody actually wants to use v2.6 anymore. :)
Posted on 2004-08-09 21:14:29 (last edited on 2004-08-09 21:15:10)
|
gannon
|
if it works as adveritised I would be very interested in it :)
Posted on 2004-08-10 01:10:31
|
cmdratz
|
People:
Well, my weapons post got thrown into the Comedy Goldmine forum on redorchestramod.com so I'll throw up a link to it from here. Why? No reason . . . . Watch me fall under the preasure of a dozen seasoned forum members, testers, and development team persons: each with a unique view on how full of shit I am. :)
Respectfully:
C. M. Dratz
Posted on 2004-08-10 02:01:22
|
Zip
|
Quote:Originally posted by some prick on the ro forum
People:
Also, for the love of god, please consider you're word usage when attempting to write anything.
AHAHA. Please consider you're use of apostrophes more like.
...but the guy later did have a point about the bold.
Zip
Posted on 2004-08-10 02:15:03
|
cmdratz
|
Zip:
Apostrophes in possesive voice do not apply to objects. I agree about the bold: I can't edit it now that its in the comedic archive.
Respectfully:
C. M. Dratz
Posted on 2004-08-10 02:56:22
|
Omni
|
Andy:
Yeah, Verge 2.6 is great. It's on its last legs, publicly, but it's still good.
cmdratz:
Its = possessive form of it
It's = contraction of "it is/tis"
Thus, "it's in the comedic archive"
:)
Respectfully:
Omni
Posted on 2004-08-10 07:12:44
|
Gayo
|
Quote:Originally posted by Omni
Andy:
cmdratz:
Its = possessive form of it
It's = contraction of "it is/tis"
Thus, "it's in the comedic archive"
:)
Respectfully:
Omni
Grue refuses to acknowledge this rule. It fills me with rage!
Posted on 2004-08-10 20:23:16
|
zaril
|
I love how some, who I thought were respectable and intelligent enough to write proper English, keep on writing "your dumb".
Posted on 2004-08-10 20:39:45
|
Gayo
|
Actually, most respectable and intelligent people can't write proper English!
But getting back on track...
here's one I just thought of that would save me some problems: it'd be neat if you could have a map event with the same name as a global function, and have the map event override the global while you're on the map. I have a specific, slightly peculiar reason for wanting to do this: there are some things I'd like to do in various games where a function always gets called at a certain time, but usually it does nothing. However, on certain maps you could define an overriding form of that function that WOULD do something. If I want to do this currently, I have to define the event as global and have a giant switch statement for the name of the map.
Posted on 2004-08-11 06:24:25
|
vecna
|
There's a much easier way to do this.
Make your map function SomeMapSpecificFunction().
Do not make a global function called SomeMapSpecificFunction().
When you want to call it, do CallFunction("SomeMapSpecificFunction");
If it exists it will be called. If it does not exist, nothing will happen.
Posted on 2004-08-11 21:52:09
|
gannon
|
I am thinking he wants a default function that if no map function exists.(you could put the same function in all the maps and change it in the one map but that seems messy)
Posted on 2004-08-11 21:56:34
|
Gayo
|
Quote:Originally posted by vecna
There's a much easier way to do this.
Make your map function SomeMapSpecificFunction().
Do not make a global function called SomeMapSpecificFunction().
When you want to call it, do CallFunction("SomeMapSpecificFunction");
If it exists it will be called. If it does not exist, nothing will happen.
That works? Zomg awesome.
Posted on 2004-08-12 03:40:27
|
Overkill
|
I request a built-in player variable. That way certain plugins with player manipulation (like the console I've made) will work on any verge3 game.
Posted on 2004-08-12 22:01:16
|