Hey! "Release early, release often!"

That's the Open Source credo for rapid development!

Cheers

Hijack already pauses the player software while doing the EXEC lines.. so no problem there.

As for the menu entries, programs which show up on the menu are ACTIVE in memory, robbing cache space. The best way at present is to use Bitt's menu thingie, except you can now start his menu thingie with an EXEC_ONCE line rather than fiddling too much with the preinit stuff, I think (I don't use any of that stuff myself..).

Eventually, we'll be reversing the whole menu thing, so that items on the menu are not actually started or in-memory until the menu item is selected..

Not there yet.

-ml

As a point of clarification, it's not my menu thingie. It's Canuck's.

I believe it's actually CanukinLA's menu thingy, which can now be started with an ;@EXEC_ONCE rather than bitt's preinit.

Hopefully the menu starting won't be too bad since you already have the excecuting code there. The config line parsing might get a little ugly though. Perhaps: ;@MENU_EXEC "Hijack Menu Title" command, or something like that.

Of course all the application writers will need to fix their app's so that they no longer daemonize or bind to the menu.

Edit: Wow, I type slowly...

Okay it might be simple for you guys but put it this way to remove a piston from a sulzer you just take the head off, undo the cross head, get the stuffing box out, take the wear ridge off then lift out the piston piece of piss. That is how to do it but ain't enough detail for you guys could do it.

So could some one please write a blow by blow account on how to do it like we have for displayserver

Thanks folk

In reply to:

---

Mark, when playing around with this, I loaded emptris, empacman, empsoko, and GPSapp. All would show on the hijack menu, but as I ran them, they would eventually drop off the hijack menu. This is running one app, then another, etc... Emptris always stayed, but the rest disappeared. This was using the ;@EXEC_ONCE, and I also tried using just ;@EXEC on all of the program calls. Is this how it is "supposed" to function?

---

I had exactly the same problem about 3 months ago using the old style init program. I was running emptris, empacman, empsoko and emptriv.

It's not a Hijack problem, it's one of those that is the problem. I removed emptris, and the rest worked fine.

I'm not saying emptris is incorrectly coded - one of the others may be wrong, instead. But removing emptris made the rest work OK.

A.

now that the dust has settled...mirror updated to v309

Hi Mark,
Thanks for the info, this is how i setup my empeg. I removed preinit, and setup the launcher script to run via exec. It works great. Only loading 1 program into memory untill i pick to run it. It will be great after you get this intergrated right into hijack. If you need any help testing anything. just let me know.
Once again, thanks mark for all your hard work!

Sorry Bitt,
I just thought all my preinit, and launcher where programmed by you. That why i put the question mark when i said it was your launcher Actually, i guess i should be saying sorry to Canuck's

to remove a piston from a sulzer you just take the head off, undo the cross head, get the stuffing box out, take the wear ridge off then lift out the piston piece of piss.

that is a great post.

there has been enough interest shown in a easy-to-use app installer built into JEmplode that it will get implemented by someone, soon enough. when that gets done we'll put a step by step process into an FAQ that will have something like

part 1) procedure to install JEmplode (already written)
part 2) procedure to install hijack (already written)
part 3) procedure to download prepackaged apps from riocar.org
part 4) procedure to to click "install third-party application" and browse to the downloaded file
part 5) procedure to answer the really simple questions that JEmplode would then pose to you that would already have default choices.

it can't get too much simpler than that

John

Was the visual + ir translate function broken in 302+ or has that function been obsoleted by something else? I didn't see anything about it so I thought I would ask. I get an ir_translate error with it in my config now.

This feature will only work with the next version of the empeg software. Its currently in alpha, i can't wait to see what that feature is just like you. It shouldn't be much longer untill they release the next version. I was wishing rob would have released it by now

Isnt that for something different? The Visual + has worked fine up until 302 or so. It's listed in the Jan 2002 Symbolic Button Names section of the Hjack web page and it's been in my config for over 9 months now with no problems until I upgraded from 301 to 303.

Oliver was referring to Visual -. I think that's what he thought you meant.

(+ has always been there, it's - that's been added in the next release)

opps i guess i shouldn't commen't i things i don't use. my fault

My apologies.. I neglected to point this out in the release note:

Visual+ has been renamed to NextVisual
Visual- has been renamed to PrevVisual

And an even better method now exists: VisualSeek

Cheers

Hi Mark,
I just thought of this. Maybe you are already planning it, but i thought i would mention it.
Could you setup a button that if we hold it down when the empeg is starting up, would ignore the exec and exec_once lines, giving us a stock hijack bootup?
I think it would be a good idea, if it’s possible. Anyone else?

Edit: yay! i'm now an enthusiast!

Or I suppose a menu (and a bit from the flash savearea) to select EXEC ON/OFF operation on the next reboot.

yes, but what if that app is causing the empeg to not boot?

Hijack's menu is active regardless.

Unless the app is broken, in which case it shouldn't be installed to begin with. But with the new savearea bit for tracking this, hijack could prevent repeated failures -- if the system locks up during the EXEC calls, then the EXEC calls simply don't happen on subsequent boots until reenabled. Nice and simple that way.

Cheers

I just got my empacman to work for the first time, after tooling around with the linux shim etc, I was never able to mangle it enough to get it to work. Now, I got it to work, and in under twenty minutes! You rock! thanks for all the hard work!

v309 is out, and fixes some nasty interactions from v307 and also from the brief existance of v308.

v309 ext3 flavo(u)r now available.

Will emptiv load with hijack? I have: ;@DC ;@EXEC_ONCE /drive0/var/games/emptriv and It wont load. Everything else works great. Much easier then the preinit setup.

-Thanks,
Greg

I bet the problem is that emptriv expects that the current working directory is the emptriv directory when it looks for fonts. and other files. You could confirm this (probably unnecessary) by looking at the serial output while emptive is starting (on boot). Last summer, I hacked the necessary code into emptriv to let it start from anywhere, but I hesitate to post it here as i'm sure it's changed in other ways since then.

One way you might be able to get around the problem until emptriv is fixed by using the line:

;@DC ;@EXEC_ONCE cd /drive0/var/games && ./emptriv

instead of what you had before and see what happens. let us know if you still can't get it to work.

John

That won't work.

But this version MIGHT work:

;@DC ;@EXEC_ONCE "cd /drive0/var/games && ./emptriv"

Or possibly this, but it is more wasteful of memory:

;@DC ;@EXEC_ONCE /bin/bash -c "cd /drive0/var/games && ./emptriv"

How about:

;@DC ;@EXEC_ONCE /bin/bash -c "cd /drive0/var/games && exec ./emptriv"

That could save on memory.

Yup, that's it. Or even incrementally better:

;@DC ;@EXEC_ONCE /bin/bash -c "cd /drive0/var/games && exec -c ./emptriv"

Argh. I feel like such a dope for making you guys hack around a stupid bug in emptriv. I will fix this soon.

I feel like such a dope for making you guys hack around a stupid bug in emptriv. I will fix this soon.

I'm anxious to see this. I've never tried emptriv and I'd like to. Let us know when you've gotten that working.

Well I'm not sure I'll get to completely automating the install... I just want to fix the bug which forces you to change the current working directory before you run it. I want to see where this empeg packaging format thing goes (johnmcd3's idea) before spending a lot of time on a flawless install procedure. Emptriv installation is tricky because it generally needs to be installed on a non-root partition (most people use the music partition) and automating that isn't "trivial."