calling all Linux hackers...

[Silvan]

We don't have a cutlist program. My shop is all but closed for the winter, and I'm just not finding a lot of things to do on the Rosegarden project anymore, owing to the rather meager skillset I can bring to the table, and the rather involved nature of the outstanding bugs.

I'm looking for something to distract me until spring, and a recent thread about using a cutlist program got me to thinking...

So let's think about this... I'm not terribly gifted, but QT/KDE isn't rocket science. I could probably knock out a basic framework in a couple of months. Especially if I use QT Designer to prototype and machine code the GUI elements. I usually don't, but I've been thinking how much easier it would be if I did. (Three weeks for one fscking dialog box. Gack! I'm not eager to do that again!)

I'm a mathematical retard (seriously, a total brain damaged how did he get a college degree retard), and I haven't done any sort of graphic work (drawing lines, boxes, that sort of hands on, computer-calculated vector thing) since Turbo Basic. I can knock out a GUI framework for the thing, but I will need real help with the math and graphics to make the thing perform a useful function. I'm very weak on OOP design too. Getting better, but I still have a fundamentally procedural mindset to trip over.

It would help if I knew what a cutlist program is supposed to look like too. I don't own any copies of Windows, so even if there's a free one or a demo out there, it doesn't do me much good. (I got rid of Wine a long time ago, and am not particularly interested in fooling with it. It never worked worth a damn anyway.)

So, we all know there are about a half million projects out there just languishing in vapor land, with bold promises of features to come last updated in 1999. The chances of our getting this off the ground and turning out something useful are very slim. But it *could* happen.

If we have a user base for this, it will likely be centered right here on the Wreck. Woodworking Linux users are a pretty specialized subset of both the woodworking and computer using breeds. If there's any interest to be drummed up, much of it will likely be right here.

So how about it? If this sounds like something you want to have, speak up. If you want to hack on it too, speak up loudly. If you're a Linux hacking god and my meager skills would just stand in your way, then speak up very loudly. I'm much better at testing/debugging and writing docs than I am at coding. I only open a source file out of necessity, because no one else is getting around to whatever needs doing fast enough to suit me, and about half the time I don't get it fixed before someone with real skill gets a chance to take a crack at it.

(If you think "hacker" means someone who breaks into computer systems illegally, you probably have no idea what the hell I was just talking about anyway. Fear not, gentle Wrecker, there is no evil conspiracy afoot here. We're talking about creating a cutlist program to run on our operating system. Nothing more insidious than that, I assure you.)

[EvoDawg]

Silvan count me in. But like you my computer skills are limited. I can probably do something in Qt3 but I will have to look into it again, I havent messed with code either for some time. I'm more into scripting at this point.

Rich

[Silvan]

I *vastly* prefer scripting too, but some things have to be done the hard way, I'm afraid. I hate GUI stuff, but QT/KDE is pretty easy to get the hang of. If you know what you want to do, it's just a matter of grazing the ample docs long enough to find the example code you need.

I'll think on this hard and long on my trip. I'm heading for Savannah, GA tonight, back sometime around 24 hours from now.

If I think I can really do something, I'll go ahead and register a project with SourceForge, so we can get a devel list (to get all this traffic off the Wreck) and a CVS repository, then we can go from there. (If nothing comes of it in the end, we won't be the only useless project on SourceForge... :)

So, important things first, what should we call this thing? The first thing popping into my mind is "Kutlist."

What, *exactly* should it do? KISS. Get something working, then make it pretty. What's the minimum it needs to do in order to be useful?

I'll come back with my answer to that last question after I've had a day to ponder it.

[Steve Dunbar]

I'd be interested in a program like this, but I'm afraid it may prove rather difficult to write it. Unfortunately I have neither the expertise nor the time to be of much help with the program.

I did a Google search and found a number of links dealing with the "stock-cutting problem" or "2-D bin packing problem." This seems to be an active area of research. See, for instance, .

There is a GPL'd program called Linpacker which deals with the problem of fitting rectangles into a rectangular region of fixed width. This might be a good starting point. See .

-- Steve

[LP]

I'm not a linux user but I applaud your thoughts on this.

If you go to

there are features and screen shots to help you get an idea of what a cut list program should do.

Also check

for additional ideas.

Good luck on this. I'm all for anything that will help bring linux to the mainstream. I know at least one linux user who would be grateful for something like this and I'll certainly bring your post to his attention.

[LP]

Just another thought; if you and whoever need a way to discuss this project in real time, feel free to use the #woodworking channel on IRC.

I dont know what the linux IRC client is called, but I do know there's one out there. If you can locate it, just point it to the AccessIRC network, and join the channel #woodworking.

A ready-to-go windows client can be downloaded from

there is some additional info about the channel/irc that might also be helpful to you.

[EvoDawg]

It's called KSirc (IRC Client) I haven't used it only because I don't do IRC. But maybe that would be a great resource.

Great idea, thanks.

Rich

[EvoDawg]

Just testing it would be of great help.

Rich

[EvoDawg]

Brilliant name!!!!

Check some of the links that have been mentioned in the post.

I'm off to check links, talk with you when you return from your trip.

Rich

[Bigpole]

I sure can USE it. I have LignumCad on my shop computer and then have to come out of the basement to use the cutlist program on the other computer. Please include boards in the program, not just plywood. Thanks for your efforts, Ted

[Doug Winterburn]

xchat is another Linux IRC client.

A start on this project that doesn't require any coding skills would be to write a requirements document, followed by a design document.

-Doug

[EvoDawg]

That would be the first step, indeed.

Rich

[Doug Winterburn]

This ain't a requirements doc, but here's a start at a list of requirements:

- should work with sheetgoods and other wood products of various type and sizes

- should allow for saw kerfs

- should take as input part#/material/thickness/width/length in a common order. This is necessary for proper grain orientation, even with sheetgoods.

- should be able to import parts list from design applications such as lignumcad. Since lignumcad is open source, if it does not contain an export facility, one could be contributed using a standards based approach such as XML.

- should not produce any stopped cuts, i.e. full rips and then crosscuts

- open source development model would be desirable as speed of development and bug fixes/enhancements could draw on a large pool of developers, as well as using existing open source package(s) as a jump start.

- should provide a project database for storing results and be able to recall and edit the parts list.

- should be able to produce postscript or other printable output along with fully qualified/sorted parts list.

I'm sure there are many more points, so feel free to rip into it and produce a real requirements doc.

-Doug

[Stan Graves]

You might want to consider writing the tool in "python."

is an object oriented, interpreted language that is supported on a wide variety of platforms. Python is compatable with the Tk GUI toolkit. Python is easy to learn, and comes with a huge library of built in functions.

Python has a number of advantages for project like this. Python source is easy to read and understand, there are a lot of very helpful people in the python community, and more than a lot of other languages it is easy to think about your problem instead of thinking about the language.

The best part? When you are done with your program, it won't just be for linux.

--Stan Graves [e-mail]

[Swingman]

May have missed it in your list - Ability to maintain a material inventory for rough, dimensioned, wood type and dimension, sheet goods, and cost based on board foot, lineal foot, square foot, and sheet, etc.

[Iraxl Enb]

i have hacked linux for a while, but have no idea what a 'cutlist' prog is supposed to do. anybody care to post a not-too-brief functional description?

is this supposed to handle non-rectilinear shapes also? from the description for cutlistplus it looks like only len*wid needs to be specified for the pieces...

irax.

[Mark & Juanita]

Maybe have some standard sizes as selectable inputs (like standard melamine, baltic birch (half sheets), standard ply.

Perhaps with an option to turn off orientation in order to allow more efficient selection for sheet goods where orientation is not important. i.e. melamine

[Mark & Juanita]

Would it be more effective to do this development in phases? First phase concentrates on sheet goods, second on dimension lumber? Otherwise, price inputs and costs is a great idea.

[Doug Winterburn]

Good stuff!

All we need now is a volunteer for the requirements doc, and after that, the design doc.

I'm a coder and don't do "docs" ;-) But I know where to get templates and so forth...

BTW, I've been trying to install/build lignumCAD on RH9 with no joy (but haven't really gotten into the source yet). If it really does what is documented, this whole project could be a feature enhancement.

-Doug

[Doug Winterburn]

Probably - the Saturn V wasn't the first rocket :-)

-Doug