Capture output of a DOS Window into AMS dynamically?

[Imagine Programming]

Yep, no sweat, IP. All in your own good time, mate.
And please, don't feel compelled or anything, yeh?
Cheers. :yes

It's gonna be there eventually ^^

[mystica]

I think patience would come in handy :o :yes I still have to do some work on something I'm doing for Dean

Yep, no sweat, IP. All in your own good time, mate.
And please, don't feel compelled or anything, yeh?
Cheers. :yes

[Imagine Programming]

thats the best explanation iv heard yet :yes

It's true isn't it?

lol, no rush mate, im not pushing

I know, I wish I could tell you when it's done though

[RizlaUK]

The thing I ask of people but rarely ever have myself

thats the best explanation iv heard yet :yes

still, not tomorrow though

lol, no rush mate, im not pushing

[Imagine Programming]

no, if you clear the top item while AMS is already doing something else then you clear the wrong item and cause a stack error in another function

anyhow, i think it will be ok as your pushing and not pulling, with pulling (as you know) you need to remove the stack items before pushing the return, but when pushing, your arguments will be used in lua and the engine takes care of clearing the stack (assigns the variables to a meta-table and pops the stack after calling lua_pcall)

only trial n error will find out i guess

Yeah ok, true. This would be a nice clean system for downloading, I'm looking forward to this
little project

not to seem pushy, just curious, hows it coming on

I'm still 'decoding' that what, 12K file? I've got the basics done, I copy all the neccesary stuff and in the end you'll have a 2K file Then I'll be adding the system, and a userfriendly interface and we're good to go :yes (still, not tomorrow though )

patience ??, what the heck is that ? lmao

The thing I ask of people but rarely ever have myself

[RizlaUK]

clearing the stacks top entries after calling the function would do the trick right?

no, if you clear the top item while AMS is already doing something else then you clear the wrong item and cause a stack error in another function

anyhow, i think it will be ok as your pushing and not pulling, with pulling (as you know) you need to remove the stack items before pushing the return, but when pushing, your arguments will be used in lua and the engine takes care of clearing the stack (assigns the variables to a meta-table and pops the stack after calling lua_pcall)

only trial n error will find out i guess

I still have to do some work on something I'm doing for Dean

not to seem pushy, just curious, hows it coming on

I think patience would come in handy

patience ??, what the heck is that ? lmao

[Imagine Programming]

Ah okay ... I see (sort of, lol). Sounds like v8.0 compatible is the only way to go then, yeh? Wow, this sounds quite exciting. Not to put on any pressure or anything ... hahaha! :lol

I think patience would come in handy :o :yes I still have to do some work on something I'm doing for Dean

[Imagine Programming]

yea, the AMS 8 version would directly call a callback function defined by you in your script, the AMS 7.5 version would need the memory plugin to subclass the AMS window and receive the events, both plugins would be completely different

Bas, go ahead, i aint got time anyway :yes

just a FYI, as a threaded function call only accepts 1 argument, send in a pointer to a structure, DO NOT use a linked list inside a thread, it will crash for sure, unless you add mutex protection, but thats rather a complex deal for a simple downloader

also, watch the stack, repeatedly calling a lua function while AMS is doing other stuff could really mess up the stack (lessons learned from PX, remember the timer issue i was having)

Im off to eat pizza.......god, i love Dominos!

Aye, I always use Structures when I want to pass something on to a thread, because I've tried linked lists once, and man that was a messed up error (Finding first instruction, editor points to keyword, remember me talking about that? )

The structure could contain a pointer to an Array, arrays are thankfully easy to reallocate in PB :yes
Aye, I'm going to have to find a way of staying out of the way of messing the stack up, anyhow, clearing the stacks top entries after calling the function would do the trick right? By the way, I could try to load the callback function in a seperate luastate... Well, That's all to be checked out :yes

Ooh Pizzaaaa, Man Hungry, Ding dong! Pizza!

[mystica]

Ah okay ... I see (sort of, lol). Sounds like v8.0 compatible is the only way to go then, yeh? Wow, this sounds quite exciting. Not to put on any pressure or anything ... hahaha! :lol

[RizlaUK]

yea, the AMS 8 version would directly call a callback function defined by you in your script, the AMS 7.5 version would need the memory plugin to subclass the AMS window and receive the events, both plugins would be completely different

Bas, go ahead, i aint got time anyway :yes

just a FYI, as a threaded function call only accepts 1 argument, send in a pointer to a structure, DO NOT use a linked list inside a thread, it will crash for sure, unless you add mutex protection, but thats rather a complex deal for a simple downloader

also, watch the stack, repeatedly calling a lua function while AMS is doing other stuff could really mess up the stack (lessons learned from PX, remember the timer issue i was having)

Im off to eat pizza.......god, i love Dominos!

[Imagine Programming]

IP, if you (or Riz, or you & Riz) do go ahead with this, would it be asking too much to compile both a v8.0 and a v7.5 compatible version? Or is this complicating things too much?

PB Plugins compiled for SDK1 (AMS7.5) in PureBasic have the Vista and Windows7 OnShutdown error.
So, sure I/we can do that, but you'd have to cope with the error. Allso, I wouldn't really wanna sell
a plugin containing an error like that

[mystica]

IP, if you (or Riz, or you & Riz) do go ahead with this, would it be asking too much to compile both a v8.0 and a v7.5 compatible version? Or is this complicating things too much?

[Imagine Programming]

Well yup, that'd be very cool. But don't knock yourself out on my account, IP. I actually spoke to Riz a couple weeks ago, asking whether he had any plans to further develop his Download Action Plugin in the near future. I understand this is something that'd take a fair bit of work, and if it were ever released as a 'reasonably-priced' commercial plugin, I'd definitely buy it.

I know, that's why I initially said 'If Dean is okay with it', since he wrote the first version of the multi
threaded downloader. However, if Dean is giving me the OK I'll do it for a reasonable price
So, Dean, whatta ya say? This would indeed be a 'fair' bit of work, not weeks of development, but
much code used. Also, I won't be implementing the Window message method Madsen pointed us to,
I'll be using a direct callback calling code without any related windows.

[mystica]

... Anyhow, the Windows Api provides you with several procedures used for downloading from external
addresses, also allowing you to parse a callback. I could, if it's okay with Dean (since he originally wrote the download plugin) investigate more and possibly write a plugin for that as well. (multi-threaded)

This would allow you to simultaniously download several files using custom callback procedures in
Autoplay Media Studio (8.0 probably).

Well yup, that'd be very cool. But don't knock yourself out on my account, IP. I actually spoke to Riz a couple weeks ago, asking whether he had any plans to further develop his Download Action Plugin in the near future. I understand this is something that'd take a fair bit of work, and if it were ever released as a 'reasonably-priced' commercial plugin, I'd definitely buy it.

@madsen,
Thanks for the tips there, madsen. In regards to the basic download plugin you referrred to, this sounds exactly like the Download Plugin Riz has already done ... same deal ... it doesn't have the bells and whistles in its current stage of development.

[Imagine Programming]

Madsen, well, a bit harsh, but no. Thanks anyway
I usuallly look things like these up on MSDN, and I have already used these procedures,
only not yet in a multi-threaded AMS plugin form (which actually takes a bit work )

It's not hard to do though, it'll be structurized procedure oriented multi-threaded programming, so it's not hard... ahum... No really, seriously though, it's doable so I might do this