#emc-devel | Logs for 2006-04-26

[01:32:46] <rayh> When you get a chance, try out the picker.
[01:33:00] <jmkasunich> will do
[01:33:12] <jmkasunich> I want to ask you some questions, you gonna be around for a bit?
[01:36:09] <jmkasunich> rayh: you there?
[01:36:23] <jmkasunich> something funky with pickconfig
[01:36:24] <rayh> yep
[01:36:32] <rayh> What?
[01:36:39] <jmkasunich> it highlighted the correct config
[01:36:49] <jmkasunich> but the entire tree is shifted to the left
[01:36:53] <rayh> Darn don't you hate it when that happens
[01:37:02] <jmkasunich> so only the lowest level is completely visible
[01:37:08] <rayh> Alex did a sort on it.
[01:37:22] <rayh> Oh but I'll be that's mine.
[01:37:40] <rayh> I use $tree see
[01:37:54] <rayh> and maybe it shifts so the whole leaf is visible.
[01:38:45] <jmkasunich> on my screen nothing is long enough to require a left/right shift
[01:38:59] <jmkasunich> and if such a shift does happen there should be a scrollbar at the bottom
[01:39:14] <rayh> I don't see any of that here.
[01:39:46] <jmkasunich> stand by, I'll post a screenshot
[01:39:52] <rayh> k
[01:41:37] <jmkasunich> http://home.att.net/~jmkasunich/Pics/picker.png
[01:43:11] <rayh> Not like that at all here.
[01:43:51] <rayh> Try commenting line 334
[01:44:09] <rayh> $tree see $last_ini
[01:44:55] <jmkasunich> dat's mo betta!
[01:45:05] <rayh> That is really odd.
[01:45:38] <rayh> What I'll have to do is change the reference to the directory from which the selected config comes.
[01:45:50] <jmkasunich> why?
[01:46:08] <jmkasunich> just do "$tree selection set $last_ini" and it works
[01:46:30] <rayh> That should be what's in there.
[01:46:50] <jmkasunich> that is immediately prior to "$tree see $last_ini"
[01:46:54] <rayh> Not for ini's below the displayed area.
[01:47:03] <rayh> That was the problem I had.
[01:47:05] <jmkasunich> oh, lemme try one
[01:47:36] <jmkasunich> I see what you mean
[01:48:13] <jmkasunich> we want to position it vertically but not horizontally
[01:48:29] <rayh> I'll work on that in the morning. Unless you want to now.
[01:48:46] <jmkasunich> I might futz with it a little
[01:48:59] <jmkasunich> I have the advantage of being able to reproduce the problem
[01:49:10] <rayh> Sure.
[01:49:21] <jmkasunich> anyway, before you go I wanted to ask you some questions about documentation
[01:49:28] <jmkasunich> I'm working on HAL_Intro.lyx
[01:49:34] <rayh> you bet.
[01:49:48] <jmkasunich> (which by the way could stand to be renamed, only part of it is intro now)
[01:49:58] <rayh> okay.
[01:50:03] <jmkasunich> one thing I did was add "Part" headers
[01:50:36] <jmkasunich> chapters 1 & 2 are now "Part I: Introduction and Tutorial"
[01:50:50] <jmkasunich> chapters 3 and up are "Part II: HAL Reference"
[01:51:01] <rayh> I don't remember how parts come out. Do they look okay?
[01:51:18] <jmkasunich> eventually I think there will be "Part III: Guide for Programmers writing modules" or something like that
[01:51:26] <jmkasunich> they look OK on PDF output
[01:51:30] <rayh> Ah sounds great.
[01:52:04] <jmkasunich> I was wondering if some of the reference info (which pins a component exports) would be better in a man page
[01:52:20] <jmkasunich> (or maybe in both places, even tho that doubles the work)
[01:52:42] <rayh> Could be.
[01:53:01] <jmkasunich> man hal_parport is easier (at least for some types of people) than going to page 67 in a 100 page document
[01:53:02] <rayh> Problem with man pages is you got to print them out.
[01:53:24] <jmkasunich> ?
[01:53:49] <jmkasunich> if I was going to print I would print the book (HAL Intro, it's becoming a book)
[01:54:02] <rayh> I can see the advantage if the person is working on a pc with the stuff there.
[01:54:28] <jmkasunich> speaking of printing, I thought about making copies of HAL Intro for the fest
[01:54:31] <rayh> Let me look and see if there is a troff2 lyx.
[01:54:41] <rayh> Great minds.
[01:54:57] <rayh> I want it to be an assignment for the evenings.
[01:55:00] <jmkasunich> we have a fast 2-sided printer at work
[01:55:08] <rayh> Oh that would be great.
[01:55:19] <jmkasunich> but fast is relative
[01:55:25] <rayh> That way I can concentrate on getting handouts and worksheets.
[01:55:30] <jmkasunich> 100 pages or more will still be slow for multiple copies
[01:55:59] <rayh> Roland told me the other day that he has access to a fast, collating, stapling printer at his church.
[01:56:13] <jmkasunich> theres also kinkos, etc
[01:56:22] <rayh> But he has to "contribute" to the cause.
[01:57:11] <jmkasunich> I dunno what kinkos costs, I would be willing to pay them or Roland if its not too high
[01:57:16] <rayh> It's online so I could make a copy here or there and have him get the folk to do the work.
[01:57:26] <jmkasunich> probalby kinkos is faster tho
[01:57:40] <rayh> Nah. A day to turn it around.
[01:57:49] <jmkasunich> I could print one copy at work, then go to kinkos with paper and pdf on disk
[01:57:58] <rayh> If you want.
[01:58:19] <rayh> You city guys have all the resources.
[01:58:23] <jmkasunich> heh
[01:58:40] <jmkasunich> I actually have kinkos, staples and office max all within 1 mile
[01:58:51] <rayh> It's easy to cut the trees here but making the paper takes time.
[01:58:54] <jmkasunich> staples and office max both have printing/copy centers
[01:59:15] <jmkasunich> oooh, they can bind too
[01:59:19] <jmkasunich> spiral, etc
[01:59:21] <rayh> Go for lowest price. If we need more at fest we can tap roland for them.
[01:59:52] <rayh> I'd even kick in a few bucks for nice copies.
[02:00:19] <jmkasunich> I'm gonna look into the options and pricing
[02:00:35] <rayh> Good, let me know.
[02:01:36] <jmkasunich> fsckers
[02:01:52] <rayh> ??
[02:02:07] <jmkasunich> they have an online print/bind service, you download their app, it pretends to be a printer, lets you preview and price things
[02:02:12] <jmkasunich> windoze only
[02:02:16] <rayh> You could put those pins and params in apendex.
[02:02:26] <rayh> shit
[02:03:12] <jmkasunich> I could install it at work, take the pdf there
[02:03:17] <rayh> You'd think sombody as big as that would use javascript.
[02:03:17] <jmkasunich> but I hoped to get pricing today
[02:12:51] <jmkasunich> looks like Staples has online pricing, I'm gonna price out a few options
[02:12:59] <jmkasunich> care to guess at quantity?
[02:13:04] <jmkasunich> 20? 50?
[02:14:39] <rayh> roland said he has a few more registered this year than last.
[02:14:59] <rayh> We must have had 20 folk at the morning sessions.
[02:17:06] <jmkasunich> for 50 copies, black & white on white paper, spiral bound, its $3.72 each
[02:17:26] <jmkasunich> (thats the existing HAL_Intro, it will be longer by the time fest rolls around)
[02:18:33] <jmkasunich> 5 copies, same configuration, $4.54
[02:18:37] <jmkasunich> (each)
[02:19:31] <rayh> So 50 of em is about $200
[02:20:06] <jmkasunich> stapled instead of bound is $2.57 each for 5, $1.75 each for 50
[02:20:14] <jmkasunich> binding is expensive
[02:21:00] <jmkasunich> I'd be tempted to have 50 stapled copies of Part I, for handouts
[02:21:24] <jmkasunich> and 10-20 bound copies of the entire thing including the reference info, sell em for $5 each
[02:22:01] <jmkasunich> part I would probably only cost a buck or so each, can afford to give them away
[02:22:36] <jmkasunich> people who might actually want to use it probably wouldn't object to forking over $5 for something they'll use
[02:24:00] <jmkasunich> looks like comb bound and spiral bound cost the same
[02:24:59] <rayh> Sure or even a bit more for the whole thing.
[02:25:29] <jmkasunich> money is a great filter for seperating the curious from the serious
[02:25:29] <rayh> $tree see $last_ini
[02:25:30] <rayh> $tree see [file dirname $last_ini]
[02:25:46] <jmkasunich> replace the first with the second?
[02:25:46] <rayh> Try that extra line and see if you still get it shifted off.
[02:25:53] <rayh> use both.
[02:26:33] <jmkasunich> no joy
[02:27:12] <rayh> Okay. I'll keep looking
[02:27:30] <jmkasunich> is there a url for bwidget docs?
[02:27:38] <jmkasunich> * jmkasunich is too lazy to google
[02:28:12] <jmkasunich> just for the heck of it I priced color printing - $28 each!
[02:29:57] <rayh> I found em. Let me look.
[02:31:08] <rayh> /usr/share/doc/bwidget/html
[02:31:55] <rayh> there is a contents.html that links them all
[02:39:13] <jmkasunich> I found a partial fix
[02:39:53] <jmkasunich> change line 235 from:
[02:39:59] <jmkasunich> set s1 [ SW $f2.f3 -scrollbar vertical]
[02:40:01] <jmkasunich> to:
[02:40:07] <jmkasunich> set s1 [ SW $f2.f3 -auto both]
[02:40:18] <jmkasunich> it still winds up scrolled to the left, but you can scroll it back
[02:40:40] <jmkasunich> I bet its possible to tell the scrollbar to go to a desired position in the code
[02:46:06] <rayh> Is this the ubuntu version of bwidget? I think it's 1.7.0
[02:46:24] <jmkasunich> I'm running ubuntu, so I assume so
[02:46:41] <jmkasunich> checking
[02:47:17] <jmkasunich> 1.7.0-1
[02:47:43] <rayh> same here. I wonder what makes the difference.
[02:48:03] <rayh> fonts, screen size, display size.
[02:48:51] <jmkasunich> could be any of those things
[02:51:16] <jmkasunich> I'd put that change I suggested in there anyway, the existing code forces vertical (and only vertical) scrollbars
[02:51:37] <jmkasunich> the -auto both automaticially puts scrollbars if needed
[02:55:26] <rayh> Okay. Gotta hang it up for tonight. Thanks for the help.
[02:55:33] <rayh> Yep that let's it scroll.
[02:56:01] <jmkasunich> fixed it!
[02:56:21] <rayh> What.
[02:56:37] <jmkasunich> add " $tree xview moveto 0.0
[02:56:37] <jmkasunich> " right after the $tree see $last_ini
[02:57:07] <jmkasunich> xview moveto 0.0 means scroll to show the left end if the line
[02:57:42] <jmkasunich> keep the scrollbars, that allows the user to do whatever he wants
[02:57:46] <rayh> Good.
[02:57:58] <jmkasunich> the moveto is the eqivalent of the user doing a manual scroll
[02:58:04] <jmkasunich> do you want to commit, or should I?
[02:58:16] <jmkasunich> (I already have the mods made)
[02:59:34] <rayh> just did.
[02:59:55] <jmkasunich> * jmkasunich kicks CIA in the head
[03:00:08] <jmkasunich> heh, it just needed a swift kick
[03:00:27] <rayh> Ah I used a passwd for my stuff and keep forgetting to wait.
[03:00:36] <jmkasunich> ok, back to lyx for me
[03:00:43] <rayh> See you later.
[03:00:51] <jmkasunich> I'll talk to you about printing again when fest is closer and the doc is more complete
[03:01:15] <rayh> okay.
[03:45:56] <jmkasunich> I really wish the documents were in the same cvs module as the code
[03:46:11] <jmkasunich> HAL_Intro describes a specific version of HAL, which changes over time
[03:46:32] <jmkasunich> there really should be a version of the document that describes emc-2.0, stored in the V2.0 branch
[03:46:50] <jmkasunich> and a version in HEAD that describes the latest code
[11:20:33] <rayh> jmk and I got .emcrc running with pickconfig last evening.
[11:21:16] <rayh> Odd problem with tree shoving stuff to the left on his machine.
[11:27:39] <cradek_> cradek_ is now known as cradek
[12:40:52] <jepler> + $tree see $last_ini
[12:40:53] <jepler> + $tree xview moveto 0.0
[12:41:27] <jepler> is this really what you want? Why shouldn't it horizontally scroll to show as much of the auto-selected config as possible, when the path is long?
[12:41:58] <rayh> jmk was having a problem with "see."
[12:42:22] <rayh> It shoved the left edge behind the logo.
[12:42:30] <rayh> I didn't have the problem here.
[12:42:46] <rayh> But we made the sliders work for both directions.
[12:43:10] <rayh> If needed.
[12:43:45] <rayh> I put the see in there so the selected ini would always display.
[12:44:59] <rayh> I'd favor collapsing at least the last + level of nodes.
[12:45:38] <rayh> and show only the selected, if a previous selection had been made.
[12:46:24] <rayh> Is there a better way than the show, moveto combination?
[12:46:37] <jepler> What's wrong with the "see" alone?
[12:47:11] <rayh> His box shoved 10 letters or so of the left edge out of sight.
[12:47:57] <rayh> He commented the see and it displayed like it should.
[12:48:12] <rayh> except that the stepper selection was below the screen.
[12:50:48] <jepler> * jepler looks at what bwidget tree actually does, and groans
[12:51:15] <jepler> What "see" should do is make as much of the item visible as possible (and it should do the same thing when you cursor up and down the list). what it is doing .. is something else entirely.
[12:51:15] <rayh> A mess indeed.
[12:52:26] <jepler> Here's something I don't expect: I cursor to a new item and then hit escape. I would expect my "last config" not to be changed, but instead it changes to the item I was on when I hit ESC.
[12:53:17] <rayh> Didn't try that.
[12:53:37] <jepler> It looks like all the 'write emcrc' code should be conditional on if { $choice == "OK" } {
[12:54:22] <rayh> True.
[12:54:38] <jepler> do you want me to fix it, or do you want to fix it?
[12:55:34] <rayh> Go ahead if you wish.
[12:56:14] <jepler> done
[12:56:26] <jepler> bbl
[12:56:50] <rayh> Thanks
[13:04:22] <alex_joni> hello
[13:06:20] <rayh> Hi alex.
[13:06:49] <rayh> I was just looking at your download page and didn't see bdi-4.42?
[13:07:33] <rayh> btw I kinda liked the stats on the front... but we do what we must.
[13:10:45] <alex_joni> heh .. yup
[13:10:52] <alex_joni> I didn't have time to mirror 4.42
[13:10:58] <alex_joni> and I didn't know it's out..
[13:20:26] <rayh> Just saw it on cncgear.
[13:20:50] <rayh> Trying to get my s&^t together for fest.
[13:20:59] <alex_joni> oh cool ;)
[13:37:14] <jepler> * jepler burns ubuntu on a bunch of cds and labels them "bdi 4.43"
[13:37:18] <jepler> * jepler packs it in with his stuff for fest
[13:37:38] <jepler> (ok, maybe that wasn't funny)
[13:37:39] <alex_joni> lol
[13:37:39] <rayh> Okay. Sounds great.
[13:37:48] <cradek> maybe you should label it 4.48 or so just in case
[13:37:50] <rayh> I love it.
[13:37:56] <alex_joni> jepler: try 5.x
[13:37:56] <alex_joni> ;)
[13:38:16] <alex_joni> although bdi might be a trademark
[13:38:25] <alex_joni> hi chris & jeff, btw
[13:38:33] <cradek> hi alex
[13:38:40] <jepler> Nah, it stands for "breezy debian installer"
[14:57:39] <rayh> rayh is now known as rayh-away
[16:13:43] <rayh-away> rayh-away is now known as rayh
[17:10:21] <alex_joni> bbl
[19:24:27] <Lerneaen_Hydra_> Lerneaen_Hydra_ is now known as Lerneaen_Hydra
[20:20:12] <alex_joni> rayh: seen you added some stuff to the wiki
[20:22:50] <rayh> just a bit about machine logic.
[20:23:00] <alex_joni> yup..
[20:23:10] <alex_joni> how do we want to go on with the lyx stuff?
[20:23:17] <rayh> We might want to put it ahead of tuning.
[20:23:22] <alex_joni> and emc2 folder inside documents/lyx/ ?
[20:23:40] <alex_joni> or add the files to emc2/docs/ ?
[20:23:48] <rayh> With our control of the cvs we can change our minds down the line.
[20:23:55] <alex_joni> right :)
[20:24:21] <rayh> I'd rather not clutter the emc2/docs with all of it.
[20:24:38] <rayh> That should rather be a place where the final docs might reside.
[20:24:41] <alex_joni> maybe only pdf versions of it
[20:24:44] <alex_joni> yeah
[20:24:56] <rayh> Right. Pdf would be good in there.
[20:24:57] <alex_joni> ok, I'll start working on this if that's ok
[20:25:11] <rayh> Although we're talking about 2-3 meg
[20:25:32] <alex_joni> we'll see, when we get that far
[20:25:38] <rayh> Lets make an emc2 directory in documents
[20:26:03] <rayh> That way we could use the same images and all, if you think that is a good idea.
[20:26:27] <rayh> Or documents -->emc2 and both lyx and images in there.
[20:26:47] <alex_joni> jmkasunich: guess you'll be around later, I've seen your info (encoder -> Hal_Intro), I think I understood part of the stuff you decided (by reading the code), but I'll wait for the docs (or maybe till we can discuss it a bit)
[20:26:58] <alex_joni> I would add: documents/lyx/emc2/
[20:27:37] <rayh> How would you refer to the images then?
[20:28:22] <alex_joni> ../../images/ ?
[20:29:20] <alex_joni> right now I see ../images/, so adding another ../ wouldn't be problematic I think
[20:29:27] <rayh> Okay. Then if we get really ambitious we can move the existing stuff into emc1
[20:29:54] <rayh> Shouldn't be a problem.
[20:30:01] <alex_joni> right
[20:30:25] <rayh> And the web converters move all the stuff, text and images, into a new directory anyway.
[20:31:22] <rayh> I'd like to use descriptive file names but not int or dev or usr.
[20:31:43] <rayh> That way it is easier to move stuff between for different books.
[20:31:55] <alex_joni> user_1_introduction.lyx ?
[20:32:03] <alex_joni> ahh.. I see
[20:32:07] <alex_joni> introduction.lyx then
[20:32:17] <rayh> yep.
[20:32:24] <alex_joni> and referenced from User, Integrator and Developer
[20:32:34] <rayh> canned_cycles.lyx
[20:33:00] <rayh> gcode_lathe.lyx
[20:33:13] <rayh> Wherever we need them.
[20:33:21] <alex_joni> right
[20:33:28] <rayh> In fact these need not be chapters. They could be sections.
[20:34:31] <alex_joni> yeah, whatever.. shouldn't matter.. should it?
[20:34:53] <rayh> If we get them to small, there may be a few issues with references between.
[20:35:08] <rayh> Other than that there would be no issues.
[20:39:53] <rayh> You adding the directories now?
[20:39:58] <alex_joni> I added emc2
[20:40:11] <rayh> Okay.
[20:40:14] <alex_joni> but for dirs there are no commit messages
[20:42:07] <rayh> I wonder if we should identify the toplevel used in lyx in the filename.
[20:42:18] <alex_joni> how do you mean that?
[20:42:45] <rayh> like history_sect.lyx
[20:42:58] <rayh> versus history_chap.lyx
[20:43:08] <alex_joni> nah.. don't think that's needed
[20:43:31] <rayh> Okay. When we put together documents, the editor does that work.
[20:44:14] <rayh> cradek, How much work is it to remove the documents/html and it's directories?
[20:44:59] <cradek> removing is easy
[20:45:10] <rayh> temp_html and down
[20:45:11] <cradek> is it junk?
[20:45:16] <rayh> Yep.
[20:45:22] <rayh> just empty stuff.
[20:45:27] <alex_joni> cradek: how about moving stuff?
[20:45:40] <cradek> alex_joni: depends
[20:45:42] <alex_joni> if we ever change our mind.. not that it will become habit
[20:45:51] <alex_joni> a file to a different dir or such
[20:45:56] <rayh> there isn't anything useful in there now.
[20:47:00] <rayh> nothing there at all.
[20:47:19] <rayh> We had it all removed at sf. only the directories remain.
[20:48:58] <cradek> I deleted it
[20:49:10] <alex_joni> coo
[20:49:10] <cradek> not sure how your repositories will handle it, maybe you'll have to delete it too
[20:49:16] <rayh> that was quick.
[20:49:18] <alex_joni> cradek: phillipe asking for you in #emc
[20:49:48] <cradek> brb
[20:50:12] <rayh> works here.
[20:50:18] <rayh> thanks
[22:33:00] <rayh> # linksp Xindex => axis.0.index-pulse-in doesn't work in m5i20 any more.
[22:46:30] <rayh> bbl