How are people progressing?

[clutch110]

Actually I was thinking of writing a data file for the new Dark Eldar and Necrons codices. I play DE and my son, the Necrons. I'll definitely make sure to check in on a more regular basis.

As for the project to take a month, I spend maybe 4 hours a week on it at most. Programming without a debugger is very hard work at times. Right now we have the first hardware prototypes done and I am working on the code to drive it all. Very fulfilling to have a physical device in your hand that reacts in a predictable and repeatable manner. For what its worth I am doing a lasertag game. Feel free to check it out, http://blog.kulinski.net. I haven't posted there in a while as I am beating my head against some code I can't seem to get to work correctly.

I have ideas on making a turn counter for 40k that would be integrated into the landscape, not sure how much I am going to go through with that.

Clutch

[snowblizz]

Actually I was thinking of writing a data file for the new Dark Eldar

I've already done that.

and Necrons codices. I play DE and my son, the Necrons. I'll definitely make sure to check in on a more regular basis.

Not got the codex but am going to sometime soon, it's academic paper deadlien season for me now so I've been busy too. But if someone else would do some files that not be bad.
I've left it a bit pending when I realised how complicated Grey Knights were with their retinue type units. And most new codices have been doing similar stuff too, Necrons have court as well now. That dang containedUnit functionality is increasing in importance.

Of course I've been eagerly awaiting the requirements functionality too, so I've been waiting for that so I start using ti too.

[IBBoard]

We can bump the Contained Unit up the priority list if needs be. Hopefully the nightly builds will make the requirement functionality easier to test now. One of the latest commits should also have joined the right pieces of code up, so that as well as having classes that are capable of parsing the data, they will now actually be called on to parse it!

A turn counter in a bit of scenery (like the side of a hill, or maybe strategically) sounds like a gamer geek bit of kit. We did the logic for turning an 8-bit binary number into a 7-bar LED number in the first year of University. Quite simple, but it depends how it interacts with the button and how much you constrain its numbers, etc.

[snowblizz]

I'm going to test that nightly build thing too. I always have issues with compiling so that probably is good thing for me

[IBBoard]

Sorry, everything seems to have gone really quiet again. I was planning for dev time over christmas, but a combination of the in-laws and important work on my wife's online store many I didn't get a chance.

I had expected to get something done this weekend, as the store updates are now live. However, when trying to add a spare hdd to my machine to rebuild its raid array I must have put them to close in their racks and the circuit board of on arced to the drive below, complete with a nice puff of smoke. I now have two hdds that prevent the machine even booting and one that will boot but isn't recognised.

Moral of the story: don't put hdds in adjacent slots in a computer case, especially if the middle one is a stiff fit.

Luckily I made a backup first, in case the raid rebuild went wrong, but I currently don't have an OS drive in that machine. Hopefully I'll get something soon and get back to work.

[IBBoard]

I think quite a few things have happened since that last update. The latest update includes support for "M units per N models in parent unit" (e.g. "3 beasts per packmaster", if you split the units up that way) and nesting in the GTK# interface.

As usual, please download the nightly builds and report any problems.

[snowblizz]

Back from a trip to the UsofA and ready to test the new stuff.

Can't see that the stuff is included in nightly#51?

Having massive problems with getting to run dev builds through VBStudio so I've been relying on the nightly builds.

[IBBoard]

I wondered where you'd got to WinForms Nightly #51 should include the new work - for some reason it seems that the Change Log isn't working properly on the build page (if it ever did).

[snowblizz]

Totally not in there. I can see in the codebase that eg you have updated the Empire datafile, but when I d/l #51 the sample file does not contain the updates with the cannon unit. Also I get a crash on start that I ticketed.

[IBBoard]

I've just checked and realised that the files are embedded now (so that the libraries will work with Android), so the data files shouldn't exist any more. I'll see why it is bundling them and what I can do about your crash.

[snowblizz]

mmm 'kay... I see what it is now. There's something off with the validation trigger which is why I couldn't see it doing anything.

If I edit the unit size of "cannon battery" it won't test for validity. Only when I add "engineers" will it test to see if the correct number is used.

Also curious, if I change the "maxSzie" of the engineer it still won't let me add more engineers for that unit? Why?

Has there been any changes to handling datafiles? I throw everything in the /data folder where I work so I'd like to se eit update datafiels form there whenever I reload them. That hasn't changed right?

All this of course pending that the IO failure on start isn't throwing everything off subsequently.

[IBBoard]

Files are still loaded from the data folder (or should be), but we also support embedding them at compile time now because Android (and presumably other non-PC platforms) don't have the "same directory as the executable" context. The sample files should now be embedded, so I don't know why they're also being zipped, unless it just isn't clearing its folders before build. Ditto for schemas - they're now embedded rather than loaded locally.

I'll try to fix the error this morning. It isn't actually an IO failure (which I thought it was at first glance as well). Instead, it is a mismatch caused by a change of API. I changed the return type of one of the methods, but for some reason then the apps are still trying to find the old method.