Re: gEDA-user: help with pcb dsn plugin
On Sun, 2011-09-18 at 01:25 -0700, Jared Casper wrote: On Sun, Sep 18, 2011 at 1:00 AM, Bert Timmerman bert.timmer...@xs4all.nl wrote: PCB mostly has a consistent style, and we won't apply patches which don't follow that. This is the next patch on my todo list. Is the above the exact syntax ? No worries, I changed the style in the first place (when a good chunk of it was already in the standard style), I'll change it all to the standard when I put in some of the bug fixes I'm finding right now... Cool. Just for the record, I don't like PCB's code formatting style myself either ;) -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: help with pcb dsn plugin
On Fri, 2011-09-16 at 11:58 -0700, Jared Casper wrote:
-- extensive style changes for my own sanity. There was inconsistent
style throughout (like all of pcb's code), so I chose the one I
personally like best (linux kernel style with indent of 4 instead of
8). :)
PCB mostly has a consistent style, and we won't apply patches which
don't follow that.
Two space indents,
if (test)
{
statements (like, this);
}
else
{
even_if_they_ARE_horrid ();
}
A lot of the code then substitutes 8 spaces for a tab character (at the
beginnings of lines), but that is just the work of the devil IMO ;).
-- Brought all the handling of coordinates up to date with the new
Coord type and nm precision. There were a few places where dimensions
were being rounded to the nearest mil, etc. which would've been bad
for metric based boards. This has all been taken care of and appears
to be working in some simple tests. DSN files are now in mm units with
nm precision.
Nice.
-- Removed the somewhat dubious code for finding the rotation of the
part copied over from bom.c. We don't have the original footprint
anyway, so the rotation was not being used. The code is still there in
bom.c if it is needed in the future.
Good idea.
[snip]
Some things still not quite handled:
-- Existing polygons on the pcb don't make it into the dsn.
If you need any pointers on that, give me a shout.
-- As noted in the bug report, there is no copyright/license notice
from the original authors. Maybe some legal issues with the Specctra
file format as well (I doubt it though, the text-based file format
would be trivial to reverse-engineer even without the spec).
You might need to try and contact the original author, but I'd
personally not worry about implementing compatibility with the
file-format. The only issue we might have is what we call it - Specctra
might be a trademarked name.
-- Probably some other stuff.
It appears to be working with some early tests, and freerouting.net is
pretty awesome.
Great!
Best wishes,
--
Peter Clifton
Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA
Tel: +44 (0)7729 980173 - (No signal in the lab!)
Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me)
signature.asc
Description: This is a digitally signed message part
___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: unconnected tracks
On Fri, 2011-09-16 at 02:02 +0200, Kai-Martin Knaak wrote: here comes the forgotten patch for more intuitive select and unselect menu items. If there are corresponding items in pcb-menu.res, it would be good to keep those in sync as we change gpcb-menu.res, otherwise - looks good. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: windows testers wanted
On Fri, 2011-09-16 at 03:01 +0300, Hannu Vuolasaho wrote: With 10mil/px I got red screen and then used scrollbars to pan out of there. Did it zoom right? Nanometer bug? +/- 20% don't work. I believe all more views and zooms work from menu. Can you confirm that other menu items (e.g. File-Load layout) work correctly? As for the z / Z shortcuts, could you 1) Describe where those keys are on your keyboard? Are they the non-shifted, and shifted versions of the same physical button? Do you have to hold any other modifier keys? 2) Could you try turning off text-services in windows? From http://support.microsoft.com/kb/306993 To turn text services on or off (using Classic view in Control Panel): Click Start, click Control Panel, and then double-click Regional and Language Options. 1. On the Languages tab, under Text services and input languages, click Details. 2. Under Preferences, click Language Bar. 3. To turn text services off, select the Turn off advanced text services check box. 4. To turn text services on, clear the check box. 5. Click Yes if you are prompted to confirm your selection. 6. Notes: Use this procedure to turn off handwriting recognition, speech recognition, and some accessibility features temporarily. These services can affect performance, and if you are not using them with your current program, you can turn them off temporarily. * If you are not using a text service at all, you should remove it. * Turning off text services does not affect multiple languages or keyboards that you have added under Installed services in the Text Services and Input Languages dialog box. I'm looking to figure out whether something is blocking your short-cuts (like a hidden or transparent IME window or something like that). Do other keyboard shortcuts work? - For example, does typing : bring up a command window - either in the status bar area of PCB, or as a separate window? Best regards, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: windows testers wanted
On Fri, 2011-09-16 at 13:06 +0100, Peter Clifton wrote: As for the z / Z shortcuts, could you 1) Describe where those keys are on your keyboard? Are they the non-shifted, and shifted versions of the same physical button? Do you have to hold any other modifier keys? (Ignore this one... I just saw your previous email linking to a picture of the layout - thanks). -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: Zoom bug on Windows
I have an idea what might be the root cause of the zoom bug... I will take bets that all the locales where this was reported - Dutch, Finish, German - use , as a decimal separator rather than .. The action we execute on zoom is Zoom(+1.2) or Zoom(-1.2) which would not work if those numbers were miss-interpreted. Perhaps we are failing to set the appropriate LC_NUMERIC (or equivalent) on Win32. I recall a similar bug in gschem or gerbv which required us to use C not POSIX when setting the locale. A quick grep shows: src/hid/gtk/gui-top-window.c: setlocale (LC_NUMERIC, POSIX); /* use decimal point instead of comma */ FAIL ___^ The gschem commit in question was this: commit a78d166a1b57b80ff46e2ac98a14989b8af77c3e Author: Peter Clifton pc...@cam.ac.uk Date: Tue Jan 19 23:11:36 2010 + Set the LC_NUMERIC locale to C rather than POSIX This value is supported on Win32 platforms, whereas POSIX doesn't appear to have any effect. This is required in order to get correct postscript output in locales where , is used as the decimal point separator. It also affects the font strings passed to Pango, causing broken text rendering in gschem. Thanks to Cesar for testing this change indeed fixes the issue. Tested-by: Cesar Strauss cestra...@gmail.com (cherry picked from commit 5d130060e694cfd3b3be177f1fae4a576728ff25) A better solution would be to use locale agnostic string processing routines here. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: windows testers wanted
On Thu, 2011-09-15 at 14:51 -0600, asom...@gmail.com wrote: v key when board is flipped: doesn't work (bug #841547) The build you tested missed the fix for that (not by much), but it ought to be fixed in tomorrow's build. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Another pcb segfault
On Wed, 2011-09-14 at 21:46 +0200, Ivan Stankovic wrote: On Wed, Sep 14, 2011 at 02:25:54PM -0400, DJ Delorie wrote: Also, please file all bug reports in launchpad: https://bugs.launchpad.net/pcb Right. This bug is now tracked at https://bugs.launchpad.net/pcb/+bug/850311. Should be fixed now. Note that it appears to have been an assertion failure, not a segfault. I don't know exactly why that occurred, but with the fixed view-fit code, we no longer produce bad coordinates which upset PCB. Btw - PCB runs faster with asserts disabled - and that is the default state. Did you have any extra patches or configure options in place to get the asserts to be active? (I'm betting the fact that DJ and I didn't see the assert fail is only due to the fact that we're using the default build settings). Regards, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: trcing with pcb design
On Wed, 2011-09-14 at 16:41 -0400, DJ Delorie wrote: PCB's netlist is readily available within pcb's innards, and easily parsable out of the *.pcb file. So far nobody's needed it so there's no command to extract it, that's all. If you have a specific format you need it in, your best choice is to write an exporter for it, then you can extract it with a command line option. The netlist which is are available internally (and in the PCB file) is the forward-annotated netlist. If you wanted to extract the as-built connectivity from a layout (possibly one without a netlist specified in the PCB file), one would have to re-construct a new netlist by tracking connectivity as per the F key to find connected copper. Should be a simple enough algorithm if anyone cares to implement it. 1. Start with list of all package pins 2. Pick a pin from the list and remove it (this becomes our new net) 3. Explore track connectivity from that pin 4. Remove all found pins from the main pin list and add to the net created in 2. 5. Repeat from 2. until main list is empty. This should produce your traced (back-annotation) netlist. To be useful though, you would need either: 1. A schematic editor which cared, and could guide you how to update the schematic - if you wanted to bring it into sync. 2. A netlist diff tool 3. Something else useful to do with the extracted netlist ;) Anyway - this is just one form of back annotation. For many operations, change-by-change back-annotation would be easier to deal with than just throwing an extracted netlist back at the schematic editor. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: cygwin geda tools
On Tue, 2011-09-13 at 11:17 -0400, DJ Delorie wrote: Unless something has change that I'm unaware of, the Windows build of PCB does not print at all (you have to export PostScript to the That hasn't changed, but Peter has cairo-printing patches in the works. Specifically, I have a patch (which is in minipack), which drops all gschem's old printing code in favour for cairo printing - which works on Windows. PCB cannot print using cairo, and I have no immediate plans to work on that - sorry. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Wire-only Symbols - Netlist problems
On Sun, 2011-09-11 at 11:34 +0530, Abhijit Kshirsagar wrote: *Short summary of Bond Graph Technique: Do you have a simple reference as to where (and how) they are used? Wiki Page: http://en.wikipedia.org/wiki/Henry_Paynter Many tutorials on the web Textbook: http://www.amazon.com/System-Dynamics-Approach-Dean-Karnopp/dp/0471459402 Cool, that looks interesting. I look forward to hearing how you progress with this. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Wire-only Symbols - Netlist problems
On Sat, 2011-09-10 at 19:49 +0530, Abhijit Kshirsagar wrote: Hi all... The problem is solved with the latest gnetlist (1.7.1). Thanks so much! I will be documenting the library of bond graph elements (symbols and definitions). Just wanted to know if anyone uses bond graphs for simulation? Its not a concept I've come across before. Do you have a simple reference as to where (and how) they are used? -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: why some skip KiCAD and gEDA
On Sat, 2011-09-10 at 13:35 +0200, Stefan Salewski wrote: A lot of documentation can be bad. Consider the toys from the big company with the damaged fruit: A reason for the success of the toys is that documentations seems to be not needed. A lot of documentation can make people think that it is very complicated. This can be true if you don't manage the reveal well. IMO, what we could do with is a showcase of the kinds of things people have done with the tools - simple, complex and in-between, so people can gain confidence that the tools can do what they want them to do, before having even picked them up. Others have done good things with gEDA - haven't torn their hair out and given up... it _CAN_ be used to make a relatively complex design. There is a place for technical documentation on file-formats, but a getting started guide should not look like a Lord of the Rings style manuscript ;) -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Electromagnetic bike
On Fri, 2011-09-09 at 13:37 +1000, Geoff Swan wrote: So if I have an electromagnetic and I hold it next to the spinning metal disc as I increase the intensity of the magnetic field the metal disc should be harder to spin? Yes. Define conductive? The eddy current breaks says non-ferromagnetic which means to me not having any magnetic properties like aluminum. Conducts electricity. Doesn't magnetise. Eg aluminium, copper. NOT iron. If you're after significant resistance, I would go for a copper disk, about 5mm or thicker, with strong magnets - either an electromagnet on an iron core - placed quite close (within a few millimetres) of the spinning disk, OR - some neodymium hard-disk magnets (for example). You could use an aluminim disk (much cheaper, and easier to obtain I'd imagine) - but I would up the thickness. If you're fitting this onto a bike (e.g. to make an exercise bike), you might be able to do something clever with a bike designed to take a disk brake. Have your eddy brake disk machined with a matching hole and thread to the hub, and screw onto that. (Or, for hubs which have bolt-holes for the brake disk - bolt to it). Obviously this would mean you couldn't use the disk brake, so I hope the whole assembly is going to be stationary ;) -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: why some skip KiCAD and gEDA
On Fri, 2011-09-09 at 09:19 +0200, Gabriel Paubert wrote: For my money, we could kill ev and its menu item completely. I don't think it serves any useful purpose, and has caused me many a headache. Full agreement here. Goodness, I have poor memory. I went to see about this, looked at my gschem - then found I had ALREADY DONE THIS back in January! commit 1c531ec953bb3a7fe895eafc65c3d4f85c2603c6 Author: Peter Clifton pc...@cam.ac.uk Date: Sat Jan 15 13:45:08 2011 + gschem: Delete Edit-Make Inv Text Vis menu item This menu item trips up more people than it helps. Remove it and associated code. (By popular request on geda-user.) So it will be lovely and fixed in the latest revision. You guys need to try git HEAD more often ;) One feature I would like to see in gschem is to have two grids: - the default grid when moving objects or anything - a finer grid automatically used when moving a single attribute Its an often requested feature. Please file a feature request at http://launchpad.net/geda/+filebug However - it could well be a (long) while before anyone gets to it, as there isn't much focus on gschem UI at the moment. (More core scheme changes in libgeda are taking preference). If you wanted to code this up though - I'm confident you would find people willing to help you find the appropriate code. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Wire-only Symbols - Netlist problems
On Fri, 2011-09-09 at 11:32 +0530, Abhijit Kshirsagar wrote: Thank you for your inputs. Please tell us what version of gschem and gnetlist you're using. I'm using the latest stable release of gEDA (1.6.2-20110115) I compiled it from the source tarball here: http://geda.seul.org/wiki/geda:download (There is no --version switch for gnetlist, right?) I remember a similar bug has been fixed in gEDA 1.7.1: https://bugs.launchpad.net/geda/+bug/698395 I'm compiling 1.7.1 now. Will let you know as soon as i'm done. What you're doing looks legit, and but would certainly (without the 0-ohm resistors at least), have hit the above mentioned bug. Rather than build from a release tarball, you might prefer to grab git HEAD and test with that. git clone git://git.gpleda.org/gaf.git We've been a bit tardy with making releases of our latest and greatest code. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Wire-only Symbols - Netlist problems
On Thu, 2011-09-08 at 19:08 +0530, Abhijit Kshirsagar wrote: Hello everyone, I tried adding 0Volt sources for the series-creating device and 0Amp current sources for the paralleling device. But no avail. The nets are still disjoint in the second case. Would it help it I posted a schematic? Could anyone shed some light please? I think it is a known issue that you can't place nets (wires) inside symbols - only pins, attributes and graphical elements. Other than that - I'm not sure what issue you might be running into. Regards, -- Peter Clifton pc...@cam.ac.uk signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Wire-only Symbols - Netlist problems
On Thu, 2011-09-08 at 23:16 +0200, Krzysztof Kościuszkiewicz wrote: On Thu, Sep 08, 2011 at 07:08:45PM +0530, Abhijit Kshirsagar wrote: I tried adding 0Volt sources for the series-creating device and 0Amp current sources for the paralleling device. But no avail. The nets are still disjoint in the second case. Would it help it I posted a schematic? Could anyone shed some light please? Please tell us what version of gschem and gnetlist you're using. I remember a similar bug has been fixed in gEDA 1.7.1: https://bugs.launchpad.net/geda/+bug/698395 That is related, but referred to a short-circuit net in a schematic file. I'm not certain this is sufficient to allow nets within a symbol. (Although it might just work now). Posting the schematics / symbols in question would be a help for any further diagnostic, but it sounds like you might be trying to do something which isn't supported. -- Peter Clifton pc...@cam.ac.uk signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: why some skip KiCAD and gEDA
On Thu, 2011-09-08 at 13:43 -0600, Mark Rages wrote: On Thu, Sep 8, 2011 at 12:38 PM, Colin D Bennett co...@gibibit.com wrote: On Thu, 08 Sep 2011 17:50:33 +0200 Stefan Salewski m...@ssalewski.de wrote: For me, I never loved the many tool changes, and I was never able to remember all the key combinations. er is edit rotate, ve is view extend. For the later I am not really sure -- have not used gschem for a year. Don't forget, while “ve” is View Extents, “ev” alters all invisible text and attributes making them visible! It is not like “en” which just toggles the display mode to show invisible text, but “ev” actually changes the entities. For my money, we could kill ev and its menu item completely. I don't think it serves any useful purpose, and has caused me many a headache. -- Peter Clifton pc...@cam.ac.uk signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: why some skip KiCAD and gEDA
On Thu, 2011-09-08 at 13:10 -0500, John Griessen wrote: gschem is not as key binding configurable as PCB as far as I can tell. Adding that would be a fine goal. It is actually - its just not immediately obvious. Look in your $PREFIX/share/gEDA/system-gschemrc file and find: (define file-keymap '((w . file-new-window) (n . file-new) (o . file-open) (s . file-save) (e . page-close) ; yes this is okay; reusing page-close (a . file-save-as) (l . file-save-all) (p . file-print) (r . page-revert) ; yes this is okay; resuing page-revert (i . file-image) (t . file-script) (c . file-close-window) (q . file-quit))) (Those go with the second letter to 'f', which activates the file menu). Finally, the global keymap: ; All keys in the global-keymap *must* be unique (define global-keymap '((Escape . cancel) (a . add-keymap) (b . add-box-hotkey) (c . edit-copy-hotkey) (d . edit-delete) (e . edit-keymap) (f . file-keymap) (h . help-keymap) (i . add-component) (l . add-line-hotkey) (m . edit-move-hotkey) (n . add-net-hotkey) (o . options-keymap) (bracketright . options-scale-up-snap-size) (bracketleft . options-scale-down-snap-size) (p . page-keymap) (r . view-redraw) (s . edit-select) (t . attributes-keymap) (u . edit-undo) (v . view-keymap) (w . view-zoom-box-hotkey) (x . view-pan-hotkey) (Left . view-pan-left) (Right . view-pan-right) (Up . view-pan-up) (Down . view-pan-down) (y . buffer-keymap) (z . view-zoom-in-hotkey) (period . repeat-last-command) (Shift colon . edit-invoke-macro) (comma . misc-misc) (equal . misc-misc2) (Shift plus . misc-misc3) (Delete . edit-delete) (Shift greater . page-next) ; Deprecated; preserved for backward compat (Page_Down . page-next) (Shift less . page-prev) ; Deprecated; preserved for backward compat (Page_Up . page-prev) (Alt q . file-quit) (Shift B . add-bus-hotkey) (Shift H . hierarchy-keymap) (Shift U . edit-undo) (Shift R . edit-redo) (Shift Z . view-zoom-out-hotkey) (Control x . clipboard-cut) (Control c . clipboard-copy) (Control v . clipboard-paste-hotkey) (Control z . edit-undo) (Control y . edit-redo) (Control a . edit-select-all) (Control Shift A . edit-deselect))) ; finally set the keymap point to the newly created datastructure (define current-keymap global-keymap) Notice there are plenty of single-key bindings there already, for example, the group at the bottom. Hope that helps, -- Peter Clifton pc...@cam.ac.uk signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: time for a pcb release cycle?
On Wed, 2011-09-07 at 10:42 +0200, Kovacs Levente wrote: Each time I save my layout, a message appears: layout changed. Do you want to load it? or something like that. Brr well of couse it changed! I saved it! :-) Ok, I know it is a good feature... Yes - it turns out that my code here is very buggy. I'm working on a fix, but it turns out that there are some quite subtle things going on. Disconnecting the file-watch before save, then re-connecting it after doesn't even work (sometimes), as for some reason - after having used the GTK file-selector dialog to pick a file-name, the file-watch code will report events from changes on disk BEFORE you hooked it up. (Adding sync() and sleep(1) after the write to disk doesn't help, so I'm presuming it isn't just a race bug). Peter. signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: CERN goes for KiCAD
On Wed, 2011-09-07 at 15:01 +0200, Kai-Martin Knaak wrote: CERN is a missed opportunity. No reason to deny it, no reason to give up, either. But maybe an occasion to think about the reasons. I had wondered if part of the reasoning might be that KiCad feels a lot more local to them. (KiCad being a French originated project - CERN being on the French / Swiss border.) Although we do a lot of gEDA development from here in Europe, gEDA may still be viewed as more American, or less local. Not that I wish to imply that the CERN folks may be biased based on this - but it is a possibility. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: time for a pcb release cycle?
On Wed, 2011-09-07 at 10:56 -0700, Colin D Bennett wrote: On Wed, 07 Sep 2011 12:54:27 +0100 I've fixed it now by copying the gedit solution of checking file mtimes on mouse in to the main window. (Should cover most use-cases of the user switching between applications editing the same file). I did track down the root of my problems with the GIO stuff. Turns out to be a race / bug in GLib's inotify driver. Nothing we can fix that soon. How about this for a solution: 1. When saving to disk, don't disconnect/reconnect the file watch, just leave it connected. However, ignore any file modification events until the save is completed. Set a flag: ignore_changes := True Would have worked, apart from the File-Save As case. I'd have had to disconnect and re-connect the watch to point at the new file, BEFORE we save into it to be sure not to get (or miss) change notifications (depending on the GLib bug). Ok - we _could_ have made that work, but the error handling cases are nasty, as it would mean the core telling the GUI that its file-name has changed before we even tried to save into it. Currently, we only update the current board's filename if the SaveAs operation succeeds - which is nice. (I'm also thinking that we should be future-proof for any notional File-Save a copy operations which don't update the new board location. 2. Immediately after the save is completed, read the modification time on the file from disk. Store this time: save_time := file.mtime 3. Set ignore_changes := False to resume processing file modification events. The modification event handler should, however, ignore any modification event if the file modification time is equal to the previously captured save_time. I've done it purely on mtimes, and it seems to work. No doubt someone will find me a corner case where it is broken though :( Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: time for a pcb release cycle?
On Wed, 2011-09-07 at 22:47 +0100, Peter Clifton wrote: On Wed, 2011-09-07 at 10:56 -0700, Colin D Bennett wrote: How about this for a solution: 1. When saving to disk, don't disconnect/reconnect the file watch, just leave it connected. However, ignore any file modification events until the save is completed. Set a flag: ignore_changes := True Would have worked, apart from ... Also, the GFileMonitor works through the GLib main loop. You only get the events delivered _after_ the code finishes what its doing and returns to the main loop. That would be a cause of further problems. TBH, I think the GFileMonitor API is pretty broken for this basic use-case. It works lovely for the one which it was (probably) intended for - watching for changes in system config files, for new files popping up in directories etc.., but it is not something which we can (currently) synchronise w.r.t any other file operations we are performing ourselves. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: CERN goes for KiCAD
On Tue, 2011-09-06 at 20:37 -0400, DJ Delorie wrote: sad. Which part? The part where CERN found an open source app they liked, or the part where they're going to contribute to OSS? Sounds like a few spare cycles working on KiCad file-format import / export for our tools might be a wise move if we want them to reconsider after they have tried KiCAD. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: revert vs reload
On Mon, 2011-09-05 at 12:09 +0100, Richard Barlow wrote: On Wed, 2011-08-31 at 11:08 +0100, Rob Spanton wrote: On Tue, 2011-08-30 at 01:46 +0200, Kai-Martin Knaak wrote: With a number of recommended work-flows PCB and gschem do not return to the last saved state. Instead, they update to the current state of the source file. The gsch2pcb work-flow relies on the ability to change the *.pcb file in the back of PCB. Contrary to the meaning of the word revert the revert action does not go back but forward. When the files that evince or gedit have open are changed, they pop up a bar across the top of the window with a reload button in it. Perhaps this is a solution that would fit in better. I've started implementing this[1] with a little help from the gedit source. Unfortunately the info bar widget that I've used was introduced in GTK 2.18 and PCB currently depends on GTK 2.12. Is there any argument against upping PCBs GTK version dependency to 2.18? It was released almost two years ago now (23rd Sept 2009). Sorry Richard - I beat you to it - committed to git HEAD of PCB last night ;) I bumped PCB's GTK version requirement to 2.18. I'm also planning on implementing the same behavior in gschem which currently depends on GTK 2.10. Any arguments to upping this to 2.18 too? I probably ought not to do that unilaterally (like I just did for PCB), but _I_ would have no objection to it. Please feel free to take any of the code I wrote to help you on the way with a gschem / gattrib version. I would love to see this consistently across the tools, yet didn't have time to do the gEDA parts at the moment. Actually gattrib support might be tricky, since it opens multiple files and doesn't support reloading at the moment (?) - gschem support would be awesome. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: David Jones - EEVBlog on Open Hardware
On Mon, 2011-09-05 at 03:37 +0200, Kai-Martin Knaak wrote: Peter Clifton wrote: gEDA and KiCAD get a mention ;) http://www.eevblog.com/2011/08/12/eevblog-195-open-source-hardware-explained/ About what point in time? Sorry - can't recall. He mentions the opinions on open CAD tools with Open Hardware. Perhaps around half way in? -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: test repo
On Mon, 2011-09-05 at 15:14 +0100, Gareth Edwards wrote: In fact, I'd go further and say that if we also had a native OS X build (not macports/fink) we'd have a pretty killer proposition for the open hardware community. First person to send me a shiny Apple laptop bought themselves that development work ;) (Seriously). Best wishes. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: revert vs reload
On Mon, 2011-09-05 at 10:38 -0700, Steven Michalske wrote: On Sep 5, 2011, at 9:34 AM, DJ Delorie wrote: Er, what gtk do the Mac builds use? Macports is at 2.24.5 Fink is at 2.18.9 Homebrew is at 2.24.6 Do you happen to know if there are any builds of GTK 3.0 out there yet? -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: revert vs reload
On Mon, 2011-09-05 at 21:40 +0100, Richard Barlow wrote: Sorry Richard - I beat you to it - committed to git HEAD of PCB last night ;) I bumped PCB's GTK version requirement to 2.18. Haha, nice. I suspected you were working on this after seeing your commits cleaning up gui-top-window.c. I see you've made a load more modifications today bringing the code up to date, nice work. Yes - mostly taking advantage of the GTK 2.18 version bump to get rid of some deprecated APIs and other issues. (One day we'll be able to compile with GTK 3.0d!) I've made a couple of small modifications to the message that's displayed to the user in the info bar[1]. Note that if you hit the revert button, and PCB will still prompt you for further confirmation if you had unsaved changes. I deliberately copied the message text straight from gedit in order to try and keep it consistent with the rest of the desktop, but I'm not completely against changing it, so long as the new message still fits on the small-ish screens without undue wrapping/ It would also be good if when reloading/reverting the file the state of the UI wasn't reset too, it's quite annoying for all of the layers to be re-enabled if you're in the middle of routing. I'll look into that later. That would be handy. Looking at where I hooked up my code, I have hooked it up in a stupid place - the file monitoring is cancelled and re-wired every time PCB handles a menu action! It probably needs to hook up so it is only called upon the PCBChanged action. We could consider splitting that action into PCBChanged and PCBReverted actions (or add arguments to the former), so the GUI can do different levels of UI resetting for load and revert. I'm also planning on implementing the same behavior in gschem which currently depends on GTK 2.10. ... Please feel free to take any of the code I wrote to help you on the way with a gschem / gattrib version. I would love to see this consistently across the tools, yet didn't have time to do the gEDA parts at the moment. Cool, thanks. I asked Ales and DJ on IRC earlier, and we have a consensus that bumping the version requirement to GTK 2.18 is OK. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: revert vs reload
On Mon, 2011-09-05 at 21:55 +0100, Peter Clifton wrote: Looking at where I hooked up my code, I have hooked it up in a stupid place - the file monitoring is cancelled and re-wired every time PCB handles a menu action! It probably needs to hook up so it is only called upon the PCBChanged action. Gah - this causes fun and games. I turns out that we only missed our own save action modifications to the PCB file because of the stupid place I hooked up the code. (It cancelled and re-attached the file-monitor right after any menu action - including File-Save of course). I've pushed a few preliminaries (HID API) to git HEAD to allow the core to notify us around a SavePCB() call, and to tell us if the filename of the PCB has changed (e.g. File-SaveAs). Unfortunately, I can't make it work for the SaveAs case. I'm sure I must be missing something simple, but I just can't get the file monitors to ignore changes to the new PCB file despite tracing the fact that we delete the old file monitor, save the PCB (in the new filename), THEN hook up the new file change monitor. I'm _guessing_ that either the saved changes don't hit the disk immediately, or GLib / GIO is helpfully caching the directory state from before I hooked up the file monitor - then falsely detecting a change (as we hook up the new file-monitor right after we wrote to the file). Anyway, I'm not quite sure how to fix it. It might come down to not disconnecting the file-change monitor, but adding a flag somewhere to make the callback ignore the next change we get notified about (after a save). Anyway - if you fancy figuring out just what I'm doing wrong, the patch (work in progress) against git HEAD is attached. We could consider splitting that action into PCBChanged and PCBReverted actions (or add arguments to the former), so the GUI can do different levels of UI resetting for load and revert. I added a revert parameter to the PCBChanged action, so that should be a good start for you looking into this. It appears there will be some special casing in the HID to do, and possibly a bit of work in the core to make sure it doesn't wipe out more state than it needs as it reloads the board. Obviously we need to be careful that we don't get PCB into an inconsistent state reloading a changed board from disk whilst trying to keep some view settings. Remember that we could notionally change _anything_ in the PCB file we revert to. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) From b3bf8f798e7daf4d9bbf4dac9d0c5725ac6ac911 Mon Sep 17 00:00:00 2001 From: Peter Clifton pc...@cam.ac.uk Date: Mon, 5 Sep 2011 22:18:16 +0100 Subject: [PATCH] hid/gtk: Connect up the file-change monitor in a more sensible place I must have had a brain-fail when I hooked it up in the call which sets the window title. NB: That also gets called after every menu operation! Hook up the connection in the ghid_sync_with_new_layout() function, which looks to be a much more appropriate place, however changing this alone revealed another issue - we now get notified for changes WE make to the files. We were avoiding those events as the file-monitor was being reset before it could pop up, at the end of the menu action which invoked the save). Hook up to the new HID API (added in the previous commit) which allows us to temporarily disconnect the file-monitor when PCB's core is saving to the PCB file. --- src/file.c |1 + src/hid/gtk/gtkhid-main.c|3 + src/hid/gtk/gui-top-window.c | 271 -- src/hid/gtk/gui.h|2 + 4 files changed, 159 insertions(+), 118 deletions(-) diff --git a/src/file.c b/src/file.c index be7f6a5..d31b157 100644 --- a/src/file.c +++ b/src/file.c @@ -358,6 +358,7 @@ SavePCB (char *file) gui-notify_save_pcb (file, false); retcode = WritePipe (file, true); + printf (File changes hitting disk now\n); gui-notify_save_pcb (file, true); return retcode; diff --git a/src/hid/gtk/gtkhid-main.c b/src/hid/gtk/gtkhid-main.c index 89d50be..a855aa7 100644 --- a/src/hid/gtk/gtkhid-main.c +++ b/src/hid/gtk/gtkhid-main.c @@ -2170,6 +2170,9 @@ hid_gtk_init () ghid_hid.flush_debug_draw = ghid_flush_debug_draw; ghid_hid.finish_debug_draw= ghid_finish_debug_draw; + ghid_hid.notify_save_pcb = ghid_notify_save_pcb; + ghid_hid.notify_filename_changed = ghid_notify_filename_changed; + hid_register_hid (ghid_hid); #include gtk_lists.h } diff --git a/src/hid/gtk/gui-top-window.c b/src/hid/gtk/gui-top-window.c index 73bada5..1277da0 100644 --- a/src/hid/gtk/gui-top-window.c +++ b/src/hid/gtk/gui-top-window.c @@ -262,6 +262,130 @@ void ghid_hotkey_cb (int which) (gpointer) ghid_hotkey_actions[which].node
gEDA-user: David Jones - EEVBlog on Open Hardware
gEDA and KiCAD get a mention ;) http://www.eevblog.com/2011/08/12/eevblog-195-open-source-hardware-explained/ -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Tearoffs on popup menu in pcb?
On Mon, 2011-08-29 at 02:28 +0200, Kai-Martin Knaak wrote: The list of tools reminds me of the pecularity of the main tool button area. Its 3x4-1 placement is quite un-ergonomic. The group of tools does not lend itself to any natural order. When an item is added, many buttons change position. What was on the bottom right might end up in the bottom left, etc. We had this recently, when the hole tool was added. This caused me to do quite a number of mis-clicks. Have you ever tried the mode Preferences-Alternate window layout to allow smaller vertical size? That mode moves the tool buttons underneath the menu-bar in a row, perhaps that is closer to what you were looking for. I just happened to commit a fix to the theming of that area, and vaguely recalled someone (might have been you?) asking about having the tool buttons on a tool-bar. One of the better(?) hidden features in PCB ;) -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Fwd: Re: [OH Updates] How can you help solve the proprietary tool problem?
On Fri, 2011-09-02 at 18:08 +0200, Stefan Salewski wrote: On Fri, 2011-09-02 at 11:36 -0400, DJ Delorie wrote: Does the category low end bother you? Nope. Dave comes from Altium so everything is relative to that ;-) Besides, I always thought geda/pcb's competition was Eagle, And Target3001 and DesignSpark now. And of course KiCAD. gEDA/PCB is really not very popular here in Germany. I suppose there is no accounting for taste. I hear DesignSpark is rubbish (comment from someone I know who tried it once). The term suite and collection of tools makes people think it must be difficult to use, that is one point. But now we have a moderated mailing list, that may attract some people from old eastern Germany (GDR), some of these really miss the berlin wall. I'm not sure who ought to be more offended at that - the ex-East-Germans you mention, or the list administrators. Comparing a moderated mailing list to the kind of oppression and freedom limitations imposed during that era is a bit of an exaggeration. A bit like calling someone who is overly strict or prescriptive a Nazi. (Ok - I found the comment funny - just don't tell anyone ;)). And to say it clear: There may be a (good?) reason for moderation, but when post to this list needs many hours or days to become visible: Really bad for beginners. Or is this only for Kai-Martin and me. Just for the record, I'm not sure what our moderation policy is. Are any of your posts being rejected by moderation? (If so - just what kind of emails are you sending? ;)). -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: pcb gtk: Toggle buttons for route styles?
On Thu, 2011-09-01 at 20:26 -0700, Russell Dill wrote: Throwing out a crazy idea.. in word-processors etc.., styles are found in a drop-down combo-box. I know that doesn't fit so well with where we have space for the route-style selector, but just a crazy though. One day, when we support an unlimited number of route styles. ;) Yes, 4 hardly seems like enough. Especially since many designs require a different trace width based on the layer for the same signal. For impedance reasons? I was also aware that you could also have different requirements for current carrying capacity - if different layers used different copper weights. There are also different DRC rules on certain objects on inner layers for some fabs IIRC. Perhaps this ought the be encompassed in a smart route-style which has different settings for different layers. That would make assigning one route-style to one net much more practical. You (and the auto-router) wouldn't have to switch styles when switching layers etc.. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: gpcb-menu.res
On Thu, 2011-09-01 at 09:13 +0200, Kovacs Levente wrote: On Thu, 01 Sep 2011 02:01:34 +0100 Peter Clifton pc...@cam.ac.uk wrote: Not a complete fix Peter, thanks for fixing it! Sorry I broke it in the first place! I was almost certain it was broken before I wrote that patch (where I definately broke it more), but I couldn't pinpoint where. Either my (fixed) patch did more to the logic than I thought when I made the function more clear, or perhaps I was mistaken about when the feature broke. I guess I might have been carrying some patch in my PCB +GL branch which broke it earlier. Oh well - its working again now, that is the main thing. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: pcb gtk: Toggle buttons for route styles?
On Thu, 2011-09-01 at 10:55 -0700, Andrew Poelstra wrote: Hey all, Can we use toggle buttons instead of radio buttons for the route style selector? It would look like so: http://wpsoftware.net/andrew/dump/toggle.png This gives more clickable area and a cleaner look, I think. I don't know, but my personal feeling is that toggle buttons are a PITA, and they don't look that natural in this application. Pretty much for the same reason as we didn't like them in the layer selector - which you've just done a great job of re-writing ;). Perhaps I'm just used to the old way, so am resistant to change. Throwing out a crazy idea.. in word-processors etc.., styles are found in a drop-down combo-box. I know that doesn't fit so well with where we have space for the route-style selector, but just a crazy though. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New gtk menu system
On Wed, 2011-08-31 at 01:12 +0200, Kai-Martin Knaak wrote: Two more issues: * A left-mouse-click on the square toggles the visibility of the whole layer group on the canvas. But it toggles only the visual state of this layer, not the state of the squares that correspond to the other layers in the group. As a result, the state of the squaere is not necessarily in unison with the visibility of the layers on the canvas. If this is the case, it would seem that the selector's data model may reside in the wrong place. A quick skim of the new code suggests that the click action is directly updating the row, where it (arguably) should be poking PCB's core (using an API such as ChangeGroupVisibility) then reading back the changed states. While it probably doesn't, PCB could quite merrily change the layer on / off status from (say), an action in the core, or a plugin we have no control of. Unfortunately, we don't yet have a way to keep track of this happening in the GUI, but perhaps we should grow one. Whilst the definitive source of the underlying data is in PCB's core, the widget ought ideally to be working from that. Since PCB doesn't do signals on changed data, we might have to improve the HID API a little here. The graphics API work I'm doing could remove part of the the need to have the layer on / off status in the core (for rendering at least), but the autorouter(s) use it to determine which layers to route on, and various other bits of code (including object selection) seem to care. Perhaps the easier option would be to contractually ban the core from assigning layer visibilities. That would mean banning or moving some of the misc.c APIs which touch that. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: revert vs reload
On Wed, 2011-08-31 at 11:08 +0100, Rob Spanton wrote: On Tue, 2011-08-30 at 01:46 +0200, Kai-Martin Knaak wrote: With a number of recommended work-flows PCB and gschem do not return to the last saved state. Instead, they update to the current state of the source file. The gsch2pcb work-flow relies on the ability to change the *.pcb file in the back of PCB. Contrary to the meaning of the word revert the revert action does not go back but forward. When the files that evince or gedit have open are changed, they pop up a bar across the top of the window with a reload button in it. Perhaps this is a solution that would fit in better. Would be nice to have. Perhaps file a feature-request on http://launchpad.net/pcb/+bugs We would probably need the g_file_monitor_file API in GLib/GIO. This is present (I think), since GLib 2.18. This corresponds to approximately GTK 2.16 and onwards. (From a quick scan of the git repository, GTK 2.16 requires at least GLIB 2.19.7). Regards, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: gpcb-menu.res
On Wed, 2011-08-31 at 14:00 +0200, Kovacs Levente wrote:
On Tue, 30 Aug 2011 12:01:00 -0700
Andrew Poelstra as...@sfu.ca wrote:
Fixed in git head.
Thanks! Maybe I can see why the automatic layer (group) change doesn't work.
Can you give me any clue where it is located in the code?
src/hid/gtkhid-main.c, SwapSides (...)
Approx line 1408:
if ((active_group == comp_groupcomp_on!solder_on) ||
(active_group == solder_group solder_on !comp_on))
{
bool new_comp_vis = Settings.ShowSolderSide active_group == comp_group;
ChangeGroupVisibility (PCB-LayerGroups.Entries[comp_group][0],
new_comp_vis, new_comp_vis);
ChangeGroupVisibility (PCB-LayerGroups.Entries[solder_group][0],
!new_comp_vis, !new_comp_vis);
}
The failure is probably in picking the correct groups..
at the top of that function:
int active_group = GetLayerGroupNumberByNumber (LayerStack[0]);
-- Probably ok
int comp_group = GetLayerGroupNumberByNumber (component_silk_layer);
-- Probably ok
int solder_group = GetLayerGroupNumberByNumber (solder_silk_layer);
-- Probably ok
bool comp_on = LAYER_PTR (PCB-LayerGroups.Entries[comp_group][0])-On;
-- PROBABLY BAD
bool solder_on = LAYER_PTR (PCB-LayerGroups.Entries[solder_group][0])-On;
-- PROBABLY BAD
IIRC, there is no guarantee that the 0'th index in the group is the one
we should be testing here. The (PCB+GL) rendering code tests group
on/off-ness like this:
int idx = group; /* idx is a group number here */
if (idx = 0 idx max_group)
{
int n = PCB-LayerGroups.Number[group];
for (idx = 0; idx n-1; idx ++)
{ /* idx is a group entry index here */
int ni = PCB-LayerGroups.Entries[group][idx]; /* ni is a layer
number */
if (ni = 0 ni max_copper_layer + 2
PCB-Data-Layer[ni].On)
break;
}
idx = PCB-LayerGroups.Entries[group][idx]; /* idx is now a layer
number! */
}
if (idx = 0 idx max_copper_layer + 2)
{
group_visible = PCB-Data-Layer[idx].On;
... /* idx is a layer number in this section */
}
else if (idx 0)
... /* idx is a magic group number, created with SL() macros */
What that is doing, is looking for a copper (or silk) layer in the
appropriate group (which gets its number set in idx, then testing if
that layer is on.
Note that the for-loop stops one-short, as the value left in idx can be
one greater than the termination criteria of the for loop (which IS the
last in the layer group).
idx is abused as both a group index, and a layer index. These are
completely different (boy this code sucks).
(Oh yea, that is ugly!)
I felt sure I'd re-written this because it was ugly before now. I really
shudder to think the above was what _I_ came up with to fix it being
ugly.
--
Peter Clifton
Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA
Tel: +44 (0)7729 980173 - (No signal in the lab!)
Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me)
signature.asc
Description: This is a digitally signed message part
___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: gpcb-menu.res
On Thu, 2011-09-01 at 00:56 +0100, Peter Clifton wrote: On Wed, 2011-08-31 at 14:00 +0200, Kovacs Levente wrote: On Tue, 30 Aug 2011 12:01:00 -0700 Andrew Poelstra as...@sfu.ca wrote: Fixed in git head. Thanks! Maybe I can see why the automatic layer (group) change doesn't work. Can you give me any clue where it is located in the code? FWIW, a little skim and a vague recollection suggests this commit might have been related... I can't recall whether it was that it broke some behaviour in PCB+GL, or broke the tab-changes side feature you're talking about. I think it broke PCB+GL actually (at the time it hit). commit 9768e060fad7bc3dfc366da76ea1db8154005018 Author: DJ Delorie d...@delorie.com Date: Tue Sep 7 23:29:59 2010 -0400 Fix layer visibility logic for some boards. If the Groups() line listed the c/s layer before the copper layer, disabling silk would also disable the copper. This change searches the entire layer group looking for any layer that might need to be drawn, and uses that as the exemplar instead of blindly choosing the first layer in the list. (Turns out it is the origin of that complex code I was whining about in my last email ;) - Sorry DJ!). Perhaps a git bisect will help us pin down when the tab feature first stopped working... I'm really quite curious now. This commit looks like it fixes the issue for Lesstif: commit d7be5538da245ccdbd5e2c50d0231fb2d154e9ac Author: DJ Delorie d...@delorie.com Date: Fri Sep 10 01:53:31 2010 -0400 Fix lesstif SwapSides magic layer option. If the first layer in the layer group is silk, the swap sides key won't auto-toggle the solder and component groups, nor properly activate them if it does. Add code to find a visible *copper* layer in the layer group and use that. Finally, it was broken before my commit: commit f903b4be6b85efc110852f7be40edf8245f0a513 Author: Peter Clifton pc...@cam.ac.uk Date: Sat Aug 20 15:45:20 2011 +0100 hid/gtk: Some NOOP and whitespace changes to the SwapSides() function Split from of a later patch which reworks the view flipping APIs. Hopefully this makes the function a little simpler. However, it might be worth checking the logic there, as having re-read my NOOP changes, I wonder if I might have made a mistake about the setting of my bool new_comp_vis variable. Looking back, I'm 99% certain that I made a mistake on this. Furthermore, the assumed [0] index on the group entry is also present in this code. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: revert vs reload
On Wed, 2011-08-31 at 19:44 -0400, DJ Delorie wrote: We would probably need the g_file_monitor_file API in GLib/GIO. Given the nature of PCB, I'd suggest tying it to the changed flag of the loaded PCB - so if you haven't edited it yet, you see reload but if you have unsaved changes, you see revert. Gah, PLEASE NO. Just stick with the Gnome HIG specification and be done with it. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: gpcb-menu.res
On Wed, 2011-08-31 at 20:41 -0400, DJ Delorie wrote: The original bug for my layer visibility patch, IIRC, is a groups line like c,1;s,2 - note that the c/s are listed before the 1/2. The max_layer+1 trick is used for both the c/s magic layers *and* the silk layers. I'm sure it used to work on fairly stock boards - perhaps those created with gsch2pcb, or a default PCB stack. I got the impression that the feature had broken recently. (I'm beginning to think I broke it further when re-writing parts of that code for clarity), but that was only last week or so. :( Yes - I did break it (at least partially).. I'll fix the bit I broke. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: revert vs reload
On Wed, 2011-08-31 at 20:43 -0400, DJ Delorie wrote: Given the nature of PCB, I'd suggest tying it to the changed flag of the loaded PCB - so if you haven't edited it yet, you see reload but if you have unsaved changes, you see revert. Gah, PLEASE NO. Well, yeah, I wouldn't recommend magically changing menu buttons at all, but it's better than trying to see if someone edited the file out from under you when you're the one with the editor. Its a common practice for a lot of editors now, including vim (or at least gvim). I've lost count the number of times I've had the same schematic accidentally open twice by accident - for example. A warning when the file changes underneath me could be a huge bonus to know I need to _think_ before I save. More usefully, it would also make working between gschem and gattrib easier. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: gpcb-menu.res
On Thu, 2011-09-01 at 01:46 +0100, Peter Clifton wrote:
Yes - I did break it (at least partially).. I'll fix the bit I broke.
Not a complete fix, but:
commit acbf10c2bf1814c5ffbe13dbcfd03fc9ffcaca89
Author: Peter Clifton pc...@cam.ac.uk
Commit: Peter Clifton pc...@cam.ac.uk
hid/gtk: Attempt to fix logic to flip component / solder group visibility
on flip
Should get back to better behaviour. I probably broke this somewhat with
commit f903b4be6b85efc110852f7be40edf8245f0a513, which attempted to
re-state the previous logic in a clearer fashon.
The logic now should:
If flipping sides, and only ONE of the solder / component layers (groups)
is visible, and that layer (group) is _active_, then swap the visibilities
of the component / solder layers (groups), and make the newly visible
layer (group) active.
There are still bugs in this code relating to the assumption that the
first entry in the layer group is the one which is being toggled. This
breaks if the first entry in the group is that corresponding to the silk
for that side of the board.
diff --git a/src/hid/gtk/gtkhid-main.c b/src/hid/gtk/gtkhid-main.c
index 4d9fdbe..7a6b8a8 100644
--- a/src/hid/gtk/gtkhid-main.c
+++ b/src/hid/gtk/gtkhid-main.c
@@ -1408,12 +1408,12 @@ SwapSides (int argc, char **argv, Coord x, Coord y)
if ((active_group == comp_groupcomp_on!solder_on) ||
(active_group == solder_group solder_on !comp_on))
{
- bool new_comp_vis = Settings.ShowSolderSide active_group ==
comp_group;
+ bool new_solder_vis = Settings.ShowSolderSide;
ChangeGroupVisibility (PCB-LayerGroups.Entries[comp_group][0],
- new_comp_vis, new_comp_vis);
+ !new_solder_vis, !new_solder_vis);
ChangeGroupVisibility (PCB-LayerGroups.Entries[solder_group][0],
- !new_comp_vis, !new_comp_vis);
+ new_solder_vis, new_solder_vis);
}
return 0;
--
Peter Clifton
Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA
Tel: +44 (0)7729 980173 - (No signal in the lab!)
Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me)
signature.asc
Description: This is a digitally signed message part
___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: timestamps in pcb.git
On Mon, 2011-08-29 at 00:04 -0700, Andrew Poelstra wrote: Looks like 728f3502. The timestamps shown by git are only for human use; they often end up out of order when one dev pushes old commits from his machine after another pushs recent ones. Andrew, I sometimes get emails from you with a bad time-stamp. Is it possible your computer clock is (or was) set wrong? -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: revert vs reload
On Mon, 2011-08-29 at 19:30 +0200, Kai-Martin Knaak wrote: I like doing tricks with scripts or with the text editor. Some features of gschem are not even accesssible via the GUI. The revert action in the file to propagate the changes to the GUI. menu. However, under these circumstances this is not a revert in the original meaning of the term. Proposal: Rename the menu item revert to the more neutral reload. This applies to both, PCB and gschem. Would this be appreciated? I tend to look at a couple of popular GNOME (/ GTK) apps as well as the GNOME HIG as guides to how GTK applications ought to look to be consistent on the desktop. Picking GIMP and gedit as examples of popular, well written applications, they both use Revert. This is specified in the HIG: http://developer.gnome.org/hig-book/3.0/menus-standard.html.en So my preference is that we stick with Revert here. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New gtk menu system
On Mon, 2011-08-29 at 14:50 -0700, Andrew Poelstra wrote: Hey all, I have pushed the new menu system to git head. This lets the new layer selector manage its own menu entries and accelerators. src/Makefile.am |6 +- src/hid/gtk/ghid-layer-selector.c | 432 -- src/hid/gtk/ghid-layer-selector.h | 10 +- src/hid/gtk/ghid-main-menu.c | 559 + src/hid/gtk/ghid-main-menu.h | 42 ++ src/hid/gtk/gtkhid-main.c | 18 +- src/hid/gtk/gui-command-window.c |8 +- src/hid/gtk/gui-output-events.c |8 +- src/hid/gtk/gui-top-window.c | 1235 - src/hid/gtk/gui.h | 14 +- 10 files changed, 1001 insertions(+), 1331 deletions(-) That is my kind of diffstat ;) New functionality, less lines. Awesome. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New layer selector to play with (git preview)
On Fri, 2011-08-26 at 22:32 -0700, Andrew Poelstra wrote: Fixed. The new behavior should now be the same as the old. Awesome. I've pushed my PCB+GL branch(s) rebased onto git HEAD now. There are a few bugs lurking in my branch regarding postscript export of solder mask layers though, and I've not updated the pours branch in a while. Both of those need work on the drawing APIs, which I've not sorted out yet. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New layer selector to play with (git preview)
On Sat, 2011-08-27 at 11:32 -0400, DJ Delorie wrote: I've pushed my PCB+GL branch(s) rebased onto git HEAD now. There are a few bugs lurking in my branch regarding postscript export of solder mask layers though, and I've not updated the pours branch in a while. Both of those need work on the drawing APIs, which I've not sorted out yet. Is it isolated enough that I could try adding it to the lesstif hid? There was that patch I came up with to prove it can be done. I've not revamped that recently though. There is a lot of code which would be duplicated if we were to put it into the lesstif HID in its current state. Assuming that you wish to keep the X11 rendering as well as GL, we will have to figure out how to manage two renderers in Lesstif as well as GTK. I'm still trying to juggle how to manage two different rendering engines in the GTK HID - it gets more complex when you go 3D, as even the rendering agnostic parts of the GTK HID are a little tangled with the coordinate systems being used by the renderer. (I'm working on that now). Realistically, I would suggest waiting a little until I've split the drawing API out a more. Currently it is only the very low level GL routines which are separate from the GTK HID. Everything apart from the toolkit specific GL setup should in theory be shareable. Finally, the more advanced rendering techniques required to get high-speed rendering (my branches, not git HEAD) require work to the drawing APIs though. I'm procrastinating that, as finding the right way is hard. The idea is that any changes here would make it easier to slot in the new renderer. It is side-effects of those changes which are currently breaking some aspects of postscript (and perhaps other) output in my PCB+GL branches though. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New layer selector to play with (git preview)
On Wed, 2011-08-24 at 03:21 -0700, Andrew Poelstra wrote: Hey all, The new gtk layer selector is finished for most uses now. If you want to play with it before it's ready to be published (which is a few days away still), you can get it here: git://wpsoftware.net/pcb-andrew.git I see this is committed now, but perhaps we can rename some of the new stuff at some point? We should never use the gtk_ prefix for our own widgets. That belongs to GTK's API only. Our GTK HID usually uses ghid_..., and we should probably stick to that for new code. The new widget looks really good.. when I get some time to finish off the viewport / zoom patches I've been working on, I'll rebase PCB+GL on top of it. (My own patches will currently clash with my PCB+GL stuff, so best I figure out what shape they will land in before rebasing, or I'll end up resolving conflicts more than once). Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New layer selector to play with (git preview)
On Fri, 2011-08-26 at 06:23 +0100, Peter Clifton wrote: On Wed, 2011-08-24 at 03:21 -0700, Andrew Poelstra wrote: Hey all, The new gtk layer selector is finished for most uses now. If you want to play with it before it's ready to be published (which is a few days away still), you can get it here: git://wpsoftware.net/pcb-andrew.git I see this is committed now, but perhaps we can rename some of the new stuff at some point? A few nits when I tested it. (And not just looked at the screenshots).. 1. PCB crashes immediately on startup for me: UPDATE COLORS Program received signal SIGSEGV, Segmentation fault. 0x004d6883 in gtk_pcb_layer_selector_update_colors (ls=0x0, callback=0x4ca630 get_layer_color) at hid/gtk/gtk-pcb-layer-selector.c:694 694 gtk_tree_model_get_iter_first (GTK_TREE_MODEL (ls-list_store), iter); 2. It should not be printing debug text like UPDATE COLORS to the console ;) If I'm seeming brief, it might be because its now 6:30AM and I've not slept. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New layer selector to play with (git preview)
On Fri, 2011-08-26 at 05:52 -0700, Andrew Poelstra wrote:
On Fri, Aug 26, 2011 at 06:28:09AM +0100, Peter Clifton wrote:
A few nits when I tested it. (And not just looked at the screenshots)..
1. PCB crashes immediately on startup for me:
UPDATE COLORS
Program received signal SIGSEGV, Segmentation fault.
0x004d6883 in gtk_pcb_layer_selector_update_colors (ls=0x0,
callback=0x4ca630 get_layer_color)
at hid/gtk/gtk-pcb-layer-selector.c:694
694 gtk_tree_model_get_iter_first (GTK_TREE_MODEL (ls-list_store),
iter);
Cool. I have never seen this. I also don't get the UPDATE COLORS
text on startup. How have you configured your colors?
Nothing special.. but I do have a colour scheme:
~/.pcb/preferences contains the line:
color-file = /home/pcjc2/.pcb/colors/foo
(Very imaginatively named ;).
I've attached the colour scheme file just in case you're curious.
2. It should not be printing debug text like UPDATE COLORS to the
console ;)
Oops. But it seems to have helped here. ;)
It is approximately the right place.
This is a bandaid to fix the crash:
diff --git a/src/hid/gtk/gui-top-window.c b/src/hid/gtk/gui-top-window.c
index 001d2ca..d6d061c 100644
--- a/src/hid/gtk/gui-top-window.c
+++ b/src/hid/gtk/gui-top-window.c
@@ -1033,6 +1033,8 @@ get_layer_color (gint layer)
void
ghid_layer_buttons_color_update (void)
{
+ if (PCB == NULL)
+return;
printf (UPDATE COLORS\n);
gtk_pcb_layer_selector_update_colors
(GTK_PCB_LAYER_SELECTOR (ghidgui-layer_selector), get_layer_color);
But perhaps there is a nicer way.
--
Peter Clifton
Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA
Tel: +44 (0)7729 980173 - (No signal in the lab!)
Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me)
black-color = #00
white-color = #ff
background-color = #00
crosshair-color = #ff
cross-color = #cdcd00
via-color = #8c8c8c
via-selected-color =#00
pin-color = #99
pin-selected-color =#00
pin-name-color =#ff
element-color = #e6e6e6
rat-color = #b8860b
invisible-objects-color = #4c4c4c
invisible-mark-color = #66
element-selected-color =#00
rat-selected-color =#00
connected-color = #00ff00
off-limit-color = #66
grid-color =#ff
layer-color-1 = #cd3700
layer-color-2 = #395ecc
layer-color-3 = #cfcf00
layer-color-4 = #8b2323
layer-color-5 = #548b54
layer-color-6 = #8b7355
layer-color-7 = #00868b
layer-color-8 = #228b22
layer-color-9 = #8b2323
layer-color-10 =#3a5fcd
layer-color-11 =#104e8b
layer-color-12 =#cd3700
layer-color-13 =#548b54
layer-color-14 =#8b7355
layer-color-15 =#00868b
layer-color-16 =#228b22
layer-selected-color-1 =#00
layer-selected-color-2 =#00
layer-selected-color-3 =#00
layer-selected-color-4 =#00
layer-selected-color-5 =#00
layer-selected-color-6 =#00
layer-selected-color-7 =#00
layer-selected-color-8 =#00
layer-selected-color-9 =#00
layer-selected-color-10 = #00
layer-selected-color-11 = #00
layer-selected-color-12 = #00
layer-selected-color-13 = #00
layer-selected-color-14 = #00
layer-selected-color-15 = #00
layer-selected-color-16 = #00
warn-color =#ff8000
mask-color =#009900
### PCB configuration file. ###
gui-compact-horizontal = 1
gui-compact-vertical = 0
gui-title-window = 1
use-command-window = 0
save-in-tmp = 0
grid-units = mil
history-size = 5
auto-pan-speed = 3
top-window-width = 1680
top-window-height = 1026
log-window-width = 1680
log-window-height = 1026
drc-window-width = 795
drc-window-height = 500
library-window-width = 917
library-window-height = 657
netlist-window-height = 577
keyref-window-width = 0
keyref-window-height = 0
text-scale = 296
backup-interval = 60
groups = 1,c:2:3:4:5:6,s:7:8
route-styles = Signal,1000,3600,2000,1000:Power,2500,6000,3500,1000:Fat,4000,6000,3500,1000:Skinny,600,2402,1181,600
library-newlib =
color-file = /home/pcjc2/.pcb/colors/foo
layer-name-1 = top
layer-name-2 = ground
layer-name-3 = signal2
layer-name-4 = signal3
layer-name-5 = power
layer-name-6 = bottom
layer-name-7 = outline
layer-name-8 = spare
signature.asc
Description: This is a digitally signed message part
___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New layer selector to play with (git preview)
On Fri, 2011-08-26 at 06:01 -0700, Andrew Poelstra wrote: On Fri, Aug 26, 2011 at 06:28:09AM +0100, Peter Clifton wrote: A few nits when I tested it. (And not just looked at the screenshots).. 1. PCB crashes immediately on startup for me: 2. It should not be printing debug text like UPDATE COLORS to the console ;) Both fixed and pushed. Thanks for noticing so quickly! That was quick! (Ignore my last email with the bandaid patch). -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Foss-pcb Proposed plan from CERN
On Wed, 2011-08-24 at 01:42 +0200, Kai-Martin Knaak wrote: Peter Clifton wrote: Even conversion of old legacy Altium designs could be done given access ^ Is this a serious restriction? Would it be possible for a user of a current altium license to export to this old legacy format? Sorry, I was meaning legacy as in their old designs. It was a current file-format I was working on. I didn't bother with their old ACSII format, as not so many designs / versions can export it. A transition path from altium to geda would be huge! I think so. Actually, a path back from gEDA to Altium would probably give as much of a boost in confidence for people to try gEDA it as well. (Assuming an incumbent Altium user). The main sticking points for people converting so far, seem to be: 1. Confidence in the tool reliability and feature set 2. Taking their existing footprint / symbol libraries with them to gEDA 3. (Can't recall a No. 3). Certainly migration of existing designs was not a huge concern to the people I've spoken too, since at the early transition stage, I would expect people to keep a copy of Altium around. For converting legacy designs, an import from gerber might be more generally useful to our user base, and doesn't require decyphering unknown file formats. Admittedly this looses footprint information, but perhaps that isn't always an issue. How far did you advance on this road? Unpacking the container format, separating out individual footprints from a library, identifying most of the binary record types within those footprints. It would take time and examples to verify we can know the meaning of each field in the format, but that is only a time + effort job. I had similar tentative progress with the PCB format its self, and verified that it would be a similar job to parse the schematic and schematic symbol contents. I'll get hold of the guys I was working for and see if I can make a preliminary release of the findings made. (This was all done in the support of gEDA anyway). ---)kaimartin(--- PS: Why did none of my todays posts hit the list, yet? (While others seem to have no problem to get through within minutes) -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Foss-pcb Proposed plan from CERN
On Wed, 2011-08-24 at 15:01 +0200, Kai-Martin Knaak wrote: Peter Clifton wrote: Unpacking the container format, separating out individual footprints from a library, identifying most of the binary record types within those footprints. It would take time and examples to verify we can know the meaning of each field in the format, but that is only a time + effort job. Cool. Any chance, you can fan out part of the job to volunteeers? Possibly. The caveats is that it requires access to a copy of Altium, and a LOT of patience ;). I'll get hold of the guys I was working for and see if I can make a preliminary release of the findings made. (This was all done in the support of gEDA anyway). What was the plan for the final product? Any license strings attached, that would prevent distribution by super correct distros like Debian? It was being done in general support of gEDA, sponsored by an outfit who use Altium to make open hardware, but wanted a path to start making things with gEDA as well. The intention was to release the work under an appropriate open license, probably some version of GPL / LGPL, or both, depending on how the final code was structured. ---)kaimartin(--- PS: Why did none of my todays posts hit the list, yet? (While others seem to have no problem to get through within minutes) Any answers to this one? Not a clue. Mailman fail? -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Foss-pcb Proposed plan from CERN
On Wed, 2011-08-24 at 22:14 +0100, Peter Clifton wrote: On Wed, 2011-08-24 at 15:01 +0200, Kai-Martin Knaak wrote: Peter Clifton wrote: Unpacking the container format, separating out individual footprints from a library, identifying most of the binary record types within those footprints. It would take time and examples to verify we can know the meaning of each field in the format, but that is only a time + effort job. Cool. Any chance, you can fan out part of the job to volunteeers? Possibly. The caveats is that it requires access to a copy of Altium, and a LOT of patience ;). And - for the record, I do not have access to Altium myself. I've never actually sat in front of it and driven either, but I do have some guys helping who can produce me test files to play with. This is probably best from a clean room POV anyway. I have never signed any Altium EULA, so if there were any conditions there about not reverse engineering file-formats, I am not bound by them. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: pcb HID GUI options: gtk, lesstif?
On Wed, 2011-08-24 at 20:28 +0200, Stefan Salewski wrote: On Wed, 2011-08-24 at 13:53 -0400, DJ Delorie wrote: eh, one of the reasons I really like having two monitors is so I can I have four monitors (monitors are cheap!). layout on the main monitor, menus and dialogs on the left, pdfs on the right, schematic on the upper. More or less related: Gimp seems to prefer (default to) one large window now, as reported by heise yesterday. http://www.h-online.com/open/news/item/GIMP-2-7-3-arrives-with-single-window-mode-1328585.html Oh dear goodness do I _love_ this new mode in GIMP. I've been using it since it was first out in the gimp devel PPA, and it is superb. You can still undock things as you want, but it stops the damned floaty tool-palettes getting all over the place and either lost, or in your way. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: pcb HID GUI options: gtk, lesstif?
On Wed, 2011-08-24 at 13:53 -0400, DJ Delorie wrote: eh, one of the reasons I really like having two monitors is so I can I have four monitors (monitors are cheap!). layout on the main monitor, menus and dialogs on the left, pdfs on the right, schematic on the upper. A decent graphics card capable of driving them is not though.. and it needs to have working GL drivers under Linux too ;) (I'm limited by my laptop, so just 1 or 2 monitors here). -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Foss-pcb Proposed plan from CERN
On Mon, 2011-08-22 at 18:29 +0200, Kai-Martin Knaak wrote: I subscribed to the mailing list, too. And so did Larry Doolittle, who already wrote to the list. :-) The response by Javier Serrano was interesting: ( http://lists.ohwr.org/sympa/arc/foss-pcb/2011-08/msg3.html ) The commercial package he wants to keep up with seems to be altium designer. I used protel99 (the predecessor of altium designer) from 2000 to 2005. And I must say, that I too missed most of the features Javier asks for, since I started with geda. I think many of the issues he raises can be addressed if there is a serious motivation to do it. Even conversion of old legacy Altium designs could be done given access to known sample files and developer time (e.g. money). I was working on a funded project to reverse engineer those file-format at while back, and the only reason it has stalled so far is a lack of my time. The formats aren't so bad to understand once you've had some luck figuring out the binary compression scheme. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: viewing side vs. layers
On Mon, 2011-08-22 at 22:54 +0200, Levente Kovacs wrote: On Mon, 22 Aug 2011 16:51:03 -0400 DJ Delorie d...@delorie.com wrote: Which GUI ? GTK DJ, It broke when you committed a fix for a different issue, I can't recall exactly which though. I did have a note of the commit which caused it, but I can't seem to find it now. I never did figure out whether the bug was in the original logic or not, but I know roughly where to look to start fixing it. (I came across that code recently when looking at refactoring the view-port change APIs in the GTK HID). -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Layer button backgrounds - summary
On Tue, 2011-08-23 at 10:36 +1000, Stephen Ecob wrote: A few hours in the future, Andrew Poelstra as...@sfu.ca wrote: On Tue, Aug 23, 2011 at 09:52:19AM +1000, Stephen Ecob wrote: A few hours in the future, Andrew Poelstra as...@sfu.ca wrote: * Hide all layers button * Show all layers button These would be easy enough to add to the widget. Just tack an `All Layers' entry below the other non-selectable entries. Whether there would be demand for this, I don't know. Would it affect the pins, pads and via layers? Would it affect the solder mask layer? Why would you want to hide everything? Often I want to see one layer alone, so it would be convenient to be able to hide all and then show a single layer, using just 2 clicks. In your example artwork there are 14 layers, going from all visible to just one visible would take 13 clicks. Right-click, Show only might be nice here. Thinking of some nice signal displaying software I use at work, it has Hide all waveforms below Hide all waveforms above as well. Those might also be useful options for a layer to have. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Layer button backgrounds - summary
On Mon, 2011-08-22 at 23:45 -0700, Andrew Poelstra wrote: On Sat, Aug 20, 2011 at 04:41:05PM +0200, Kai-Martin Knaak wrote: Andrew Poelstra wrote: Relatedly, why do we let the user select the 'rats' layer? Can you actually draw with it? You can draw new connections between components and have them add to the rats nest. There are a few limitations to the mode, and I don't expect many people use it though. This said, when I first started using PCB, I designed some moderately complex boards with it (including ones with micro-controllers and discretely constructed switched mode regulators). I came to gschem and gnetlist later only after I had grown to love PCB's ease of use when drawing boards. My netlists were hand-extracted from a breadboard prototype and pencil + paper drawings! This was back in the day where gschem would crash no sooner look at it - often (and very ironically as it turned out), caused by the page auto-save code! -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: How to do PCB Autorouting with non-plated holes
On Tue, 2011-08-23 at 13:52 -0400, Cory Papenfuss wrote:
I thought about that... making different footprints that don't
have copper on the component side of the pins. Since that would require
making new footprints for pretty much everything, I was hoping for a
different solution... :) It seems like it would be a relatively common
thing for hobbyists to want (whether it's a milled or home-etched board),
so I thought there might be a setting on the autorouter config to ignore
component hold plating or something.
Its not a setting, but it ought to be possible to hack something in
autoroute.c to persuade it.
A quick glance (but I'm by no means familiar with this quite complex
code), suggests that you might take a look at AddPin where pins from
the net-list are added to the auto-routing structures.
...
/* a pin cuts through every layer group */
for (i = 0; i max_group; i++)
{
...
}
...
(Fix that up with an appropriate kludge to suit your group numbering for
a quick test)
e.g., a minimal delta which might work is changing the for loop to:
for (i = my_group_number; i == my_group_number; i++)
{
...
}
Or:
do
{
i = my_group_number;
...
}
while (0); /* NB: The loop body will still execute once */
Regards,
--
Peter Clifton
Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA
Tel: +44 (0)7729 980173 - (No signal in the lab!)
Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me)
signature.asc
Description: This is a digitally signed message part
___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: gschem crash on object drag
On Tue, 2011-08-23 at 23:22 +0530, Abhijit Kshirsagar wrote: Thanks. I will try with the newer version. As of now I think the problem lies in one of the custom components I'm using - but that should still not cause a segfault right?` No - it should not. If you can still reproduce it in a newer gEDA version (or can find instructions to reproduce it reliably at your end with a minimal test-case), perhaps you could send us a copy of component causing the problem and we can investigate the problem further. Also, I'd have liked some inputs on how to debug the problem. Is there some way of enabling debug logging or something? Not that I recall. Running under gdb gdb --args pcb path_to_your_pcb_file.pcb then getting a backtrace with the btENTER command when it crashes might be useful. Running under valgrind valgrind pcb path_to_your_pcb_file.pcb Would be very slow, but is excellent at nailing the root cause of memory corruption problems which lead to crashes. Again - checking if you can reproduce the issue under a later gschem version is probably the most efficient way to proceed. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: pcb HID GUI options: gtk, lesstif?
On Tue, 2011-08-23 at 16:06 -0400, Mark Anderson wrote: I'm still planning on an OSX Cocoa HID. I haven't gotten very far, but I do have the very, very beginning. If any one else is interested, let me know. I'd like to do the same for gschem, but that isn't as modular just yet. Mark OSX supports GL rendering quite well doesn't it... I guess I'll have to make sure I split out as much of the platform / toolkit independent parts of the GTK HID's GL rendering setup code as I can so other HIDs can share. (Lesstif can use GL rendering too at some point if DJ wants - I already did a quick proof-of concept test to make sure it was feasible, but at this stage the code isn't perfectly factored out). Regards, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Layer button backgrounds - summary
On Sat, 2011-08-20 at 08:09 -0700, Andrew Poelstra wrote: On Fri, Aug 19, 2011 at 01:46:24PM -0700, Steven Michalske wrote: On Aug 19, 2011, at 1:39 PM, Vanessa Ezekowitz wrote: On Fri, Aug 19, 2011 at 12:41:58PM -0400, DJ Delorie wrote: [...] Though, I am still unsure how to indicate that the last 4 layers can't be used for drawing. Put the item names in parenthesis. A little padlock icon is universal as a locked thingy. Also, try italicised text. That might make them appear distinct. Bear in mind that the current layer chooser doesn't make any distinction here, so it shouldn't trip people up _that_ much if we don't differentiate them. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Layer button backgrounds
On Thu, 2011-08-18 at 18:57 -0700, Andrew Poelstra wrote: That is how I would like it to work. But clicking the swatch also selects the layer, which is undesirable behavior. There is no clean way to prevent this with a GtkTreeView widget. Is there any dirty way to prevent it? ;) I note that what you are trying to achieve is similar to the GIMP layers dialogue (with different icons), and you _can_ toggle visibility in GIMP without changing the active layer selection. Hmm - looks at source.. gimp/app/widgets/gimplayertreeview.c 14k-line re-implementation of gtk_tree_view. Given GTK was originally the GIMP TOOL KIT, it seems sad that they have to go to such lengths to implement basic functionality. If we are really keen on the behaviour described - perhaps we could implement our own widget for it too. Regards, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: PCB+GL rebased to git HEAD (with nanometers)
Hi everyone, My PCB+GL branches are now rebased against git HEAD, with all the excellent work from Andrew Poelstra on the internal unit conversion to nanometers. Please test and let me know if you find any bugs. I've given it cursory testing, but haven't actually used it much myself yet. Best regards, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Layer button backgrounds
On Fri, 2011-08-19 at 16:32 -0700, Andrew Poelstra wrote: Yes, I found one -- and it is not actually that dirty. You can intercept the button-press-event signal, use the function gtk_tree_view_get_path_at_pos () to figure out what cell is clicked on, then prevent the propagation of the button-press-event if it was a swatch. (In this case, you have to toggle the visibility in the tree model yourself.) Cool. Here's a fun idea - had you considered using the tree view's _tree_ feature to show layer groups? That might let us allow switching off entire layer groups, or individual layers within them. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Nanometer conversion pushed to git head
On Mon, 2011-08-15 at 18:34 -0700, Andrew Poelstra wrote: Exciting news everyone! I have just pushed the nanometer conversion patches to git HEAD. I'm really glad to see this work merged. When I get a moment, I will rebase my PCB+GL branches against it. (I might do it in a few gentle steps, rather than all 56 commits at once - depending on how brave I feel!). I see a lot of people think this will help with (hiding) the rotating a square pad bug. I'm still uneasy about that though. We should perhaps make our next order of business fixing that properly. I guess we should modify / extending the pad definition line to specify an orientation, or allow specification of an overall rotation for a package. I have some ideas about how pad-stacks could be made more general though, so a square pad (like many others) is just a special case of a polygon - which can, of course.. be rotated arbitrarily, either at the pad level, or on the component level. Please test and let me know how things are working. I have compiled all 56 commits (by script) to confirm that they can compile so git bisect will work. That is really good. Perhaps you can lend me your script to run over my PCB+GL stuff ;) I do manual compile tests for each commit before pushing stuff to git HEAD, but don't always test everything after I rebase. I have done a few save/load tests. There does not appear to be any breaks in the file format. Thank you all for your support and testing as I have worked on this over the last month or two. Your efforts are much appreciated. As soon as I get a chance, I'll try running it on some real production board data to verify things look the same. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: personal component library frustration-HELP/suggestions please?
On Mon, 2011-08-15 at 13:14 -0700, Colin D Bennett wrote: On Mon, 15 Aug 2011 22:02:07 +0200 I can't see any reason NOT to use the skip-m4 option. Every footprint in the default pcb library is available as a newlib-format footprint, right? Actually (and sadly), no.. there are some naming clashes with the way the M4 macro names were converted, so some are not in the newlib format. 200.fp such a wonderfully descriptive name.. ;) It might be an example of a clash (I recall there were with similar naming), but I can't remember weactly which one. Advantages of using skip-m4: + Avoids that nasty bug that occurs when a footprint name contains a hyphen character (-). Which is now fixed anyway. Disadvantages of using skip-m4: - Can't use your own custom m4 footprints directly? -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: PCB opengl?
On Mon, 2011-08-15 at 00:06 +0200, Kai-Martin Knaak wrote: Peter Clifton wrote: If the problem only applies to me then could You please give me some tips what to check? I can reproduce it. For some reason I can't. I get silk on the far side displayed in grey as expected. Last time I updated PCB to git was august 6th. It was broken in my repository, not git HEAD - perhaps you are using that? -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: PCB opengl?
On Sun, 2011-08-14 at 08:44 +0200, michalwd1979 wrote: Normally components that are on the far side are drawn with grey colour, but now I can see only pins, SMD pads (drown as expected in grey) but no silk screen. The layout looks the same when I use TAB to flip the board or 3D control window. I've tried to fix it modifying src/draw.c but no success, now far side is drawn on top :-). If I can reproduce it, I'll try to get it fixed. I've not noticed it myself, but then I probably haven't been working on boards which had back-silk recently. Dear Peter, If the problem only applies to me then could You please give me some tips what to check? I can reproduce it. I'm not quite sure why I never noticed it before! I'll see if I can get it fixed and ping the list back when I have done so. Thanks! -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: PCB opengl?
On Sun, 2011-08-14 at 00:50 -0700, Andrew Poelstra wrote: Assuming there are no more major bugs, I will push my metric base-unit changes to git head Monday morning. Cool ;) I would love to see a new version released in 2011. I think we should aim to set a rough date for a release, and work towards it. Perhaps, say, end of September or October? This has worked in the past, for example, times when I was keen to release in time to get our changes into a particular Ubuntu release. I'm not so fussed about Ubuntu releases at the moment, but in those cases, having a deadline to work to helped focus the development effort. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: PCB opengl?
On Sun, 2011-08-14 at 22:57 +0100, Peter Clifton wrote: On Sun, 2011-08-14 at 08:44 +0200, michalwd1979 wrote: Normally components that are on the far side are drawn with grey colour, but now I can see only pins, SMD pads (drown as expected in grey) but no silk screen. The layout looks the same when I use TAB to flip the board or 3D control window. I've tried to fix it modifying src/draw.c but no success, now far side is drawn on top :-). If I can reproduce it, I'll try to get it fixed. I've not noticed it myself, but then I probably haven't been working on boards which had back-silk recently. Dear Peter, If the problem only applies to me then could You please give me some tips what to check? I can reproduce it. I'm not quite sure why I never noticed it before! I'll see if I can get it fixed and ping the list back when I have done so. Now fixed.. I was calling some draw.c API from src/hid/gtkhid-gl.c, and that API had been split up into two parts (a gathering part, and an actual drawing part). Looks like I missed updating my GL HID code to call the drawing part directly. I've changed the GL HID now to use DrawSilk() to draw the back-side silk, rather than home-brewing its own code for it. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: PCB opengl?
On Sat, 2011-08-13 at 22:19 +0200, michalwd1979 wrote: Yes. Since about two months clone git://git.gpleda.org/pcb.git will give you transparent tracks and polygons :-) BTW, when can we eyxpect an official release, so that openGL trickles down to distro packages? Thanks for reply. I've tried git://git.gpleda.org/pcb.git and compare it to http://repo.or.cz/w/geda-pcb/pcjc2.git (Peter Clifton's branch). The main version is a bit slower then Peter's (but I expect that this depends on PC), while Peter's has some problems with rendering silk screen on far side. My branches still have many of speed-improvements (and rendering algorithm changes) which are not yet in git HEAD. They need further clean-up and QA before merging, and I'm hoping to get around to it when I have some free time. There is also the 3D board view stuff and SpaceNavigator 6-DOF controller support, which has not made it to git HEAD. Normally components that are on the far side are drawn with grey colour, but now I can see only pins, SMD pads (drown as expected in grey) but no silk screen. The layout looks the same when I use TAB to flip the board or 3D control window. I've tried to fix it modifying src/draw.c but no success, now far side is drawn on top :-). If I can reproduce it, I'll try to get it fixed. I've not noticed it myself, but then I probably haven't been working on boards which had back-silk recently. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Constraint-based PCB footprint design
On Mon, 2011-08-08 at 03:47 +0100, Rob Spanton wrote: The tool is by no means complete... it's the result of just a few hours work right now. Anyway, I hope either the tool or the concept of the tool might be of use to at least some in the gEDA community. Certainly a feature like this will be a work of awesome when it is complete. I've wanted such a feature (to use) for a LONG while, and never got around to quantifying just what it would look like. I've been thinking about grid-less board routing recently (which is in some ways similar), and 'm really glad to see you've started this tool. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New line-snapping feature for PCB - please try it!
On Mon, 2011-08-01 at 03:09 +0200, Kai-Martin Knaak wrote: Kai-Martin Knaak wrote: Now, that I tried a bit more, I see room for improvement :-) Currently the cross hair snaps to lines on all layers. I have tried it for the last touches on my big layout and I have to make the point more pointy. With this densely routed four layer board the cursor seems to snap to every object and theirt aunt, but not where I want it to. This could only be avoided by making most layers invisible. But then, I don't see where I move things... So, please revert the feature in the current state. It creates more grief than joy. Agreed (having used it myself for a while now) - it needs revision before it is ready for prime-time. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Warp pointer nonesense
On Tue, 2011-07-26 at 10:15 -0700, Colin D Bennett wrote: Another method of drawing attention to a part of the display that could be helpful is a “tunnel vision” effect: fade most of the display, leaving a small area non-faded around the feature of interest. For instance, when Ubuntu pops up a “Administrator Access Required” dialog requesting password entry. That is something I'd like to try. It might take a bit of fiddling with to get right for all different colour schemes. Personally I always work on a black background for board design. Regards, Colin ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Warp pointer nonesense
On Mon, 2011-07-25 at 09:04 +0200, Gabriel Paubert wrote: On Mon, Jul 25, 2011 at 01:15:17AM +0200, Kai-Martin Knaak wrote: Peter Clifton wrote on the other list where mere users are not allowed to post: If no-one has any objection, I will remove the code which warps the mouse pointer when selecting nets in the netlist window. Speaking of pointer warping: why is the pointer not warped when swapping sides (with TAB for example) ? Probably an omission (assuming things ought to be at least consistent). There was a bug recently where the GTK HID would not flip about the correct location, which I fixed yesterday. (I probably broke it recently, but didn't bisect to check). The idea here is that the board flips around the mouse pointer. The only problem with that is that the GTK HID still rigidly enforces that the board can't be panned further than the edges of the screen, so sometimes it can't land the board in the right place. The fix I aim to apply is to expand the allowed panning region, but an alternative would be to (ick) add a pointer warp when the board is un-able to flip exactly under the pointer. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Warp pointer nonesense
On Mon, 2011-07-25 at 02:44 +0200, Kai-Martin Knaak wrote: Peter Clifton wrote in the other mailinglist where mere users are not allowed to post: How about the attached patch? GTK HID, GL implementation only at the moment (GDK renderer won't build with this patch - sorry). Please give it a try... done. It kind of works. On a large, crammed full board, where the feature is needed the most, the cross hair is almost invisible in when zoomed out. Highlighted SMD pads does not stand out on their own. They are just too small. You see the great big animated circles homing in on the part? Are they not large enough? Animated hypnotic goodness for those who do ;). I'd like a green one, please! :-P Since it is XOR drawn, that would depend on what colour your board is ;) -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New line-snapping feature for PCB - please try it!
On Sun, 2011-07-24 at 01:37 +0200, Kai-Martin Knaak wrote: Peter Clifton wrote: adding a feature for snapping to off-grid points along the body of a track. Sounds good, but I don't quite understand the details. Where exactly are these points? Try it and you will see ;) Basically, when you're over the line, the snap code picks the nearest grid point to where your mouse pointer is, then casts out from it in vertical, horizontal and both 45 degree angles in between, until it hits the line your mouse pointer is over. Any of the intersections between the cast lines and the line you're over will snap - the closest to your mouse pointer is choesen. From the commit message: commit 9e33678b20433f571c54009c704e75d114a3ea10 Author: Peter Clifton pc...@cam.ac.uk Date: Sat Jul 23 20:47:47 2011 +0100 crosshair.c: Snap to points along off-grid lines when drawing tracks This should greatly easy making tidy layouts where some lines have (perhaps by necessity) ended up off-grid. This patch adds code to snap onto the center of a line. It finds the nearest grid point to the cursor, then will allow snapping at the intersections between the line in question and the lines of an imaginary X and + centered on the nearest grid-point to the cursor. This allows neat drawing of horizontal, vertical and 45 degree lines which will land correctly on the existing line. signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Recap of remaining metric-conversion work
On Sun, 2011-07-24 at 08:33 -0700, Andrew Poelstra wrote: This causes no regressions, though if you are playing with grids you may need commit #2 below, which does cause regressions. I've not been working with the grid really - just the FitCrosshairIntoGrid function in crosshair.c. (Incidentally - this doesn't really do what it says in the function name.. evaluating GRIDFIT_X and GRIDFIT_Y on the mouse pointer position is only a tiny aspect of what it does, the rest involves snapping to various different kinds of objects. The Gtk GUI uses _(mm) and _(mil) in its current code. I did not want to change this, so I internationalized all the units. Yuck - never mind. We could change it to not translate, but since you've gone to the trouble of making it work, I don't see why we should. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Recap of remaining metric-conversion work
On Sun, 2011-07-24 at 18:37 -0700, Andrew Poelstra wrote: I have pushed my first commit. Rebasing my work causes merge failures in three files (action.c, crosshair.c, fontmode.c). Probably mostly just some fixes I made for compiler warnings. I am plodding away at refactoring the event coordinate conversion for the GTK HID though, and that could potentially conflict a bit. I'm trying to use the new types where applicable though - so I hope resolving the conflicts should just be a matter of checking whether I did anything stupid with my changes. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New line-snapping feature for PCB - please try it!
On Sun, 2011-07-24 at 20:23 +0200, Kai-Martin Knaak wrote: Peter Clifton wrote: Try it and you will see ;) Now, that I tried a bit more, I see room for improvement :-) Currently the cross hair snaps to lines on all layers. It should perhaps be a stronger snap on the layer you are on perhaps. Placing vias is layer agnostic of course, and I can imagine cases where I want to line a track up against one on a different layer _before_ placing the via. Perhaps a corner case though.. I'll leave it to others to experiment with the heuristic.. I've got other patches to cook at the moment. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New line-snapping feature for PCB - please try it!
On Sun, 2011-07-24 at 20:09 -0400, DJ Delorie wrote: Also - it seems to snap to the lines you're moving, when you move an intersection. Not sure if that makes sense. I think that makes sense.. I'll probably have to fix that. Having said that.. does it help you shorten an off-grid line without changing its angle? -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Warp pointer nonesense
On Mon, 2011-07-25 at 01:15 +0200, Kai-Martin Knaak wrote: Peter Clifton wrote on the other list where mere users are not allowed to post: My objection: I use regularily this feature to locate nets. It would be a regression if it were removed without a replacement. Yes, it is less than perfect now. A better way to achieve the same goal would be: 1) center the canvas on the feature to be warped to 2) move the cursor right on top of the feature 3) adjust zoom so that the diameter of the feature is about a tenth of the canvas. 4) make the main window of PCB active Do all this when explicitely asked for with a click on a warp-button. Try the patch on geda-dev, see how that works as a first cut. PS: Please let users have a say, when it comes to GUI behavior. That is, discuss such topics in geda.user, not in geda devel. Alternatively, ask Ales nicely, subscribe to geda-dev and discuss there. There are a huge number of users on _this_ list who don't want to get bogged down in details of an ongoing development discussion. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Recap of remaining metric-conversion work
On Sat, 2011-07-23 at 19:38 -0700, Andrew Poelstra wrote: Hey all, On a recent bug report, Peter C. asked for a summary of the remaining work needed for the metric conversion. Here you go: It has been a busy week for me so no progress in the last eight days or so. Even so, it's been a long time since my last update... On my personal branch I have made a fair bit of progress that has not been pushed. I don't want to publish these changes yet because I'm still reordering and revising them. Another reason not to push is that I have converted PCB.Grid and Settings.Grid from double to Coord (an integer unit). This causes -severe- rounding errors for people using metric grids with cmil base units. Anyway, here are the commits I have made but not pushed: 1. Introduce Coord and Angle units, set a bunch of BDimension, LocationType and int variables in global.h to one of these. Convert pcb-printf to use these (including adding %ma spec to print angles). Oops, I've just been playing with the grid snap code and no doubt introducing additional uses of LocationType which you will need to fix up. Any chance you could push this one ASAP, or does it cause regressions? 2. Convert grid units to Coord. Set mm/mil autodetect code to look for multiples of 127nm (see bug lp-811393), though I doubt this will work properly with cmil base units. Unlikely I would imagine. 3. Add scale-factor lookup functions to pcb-printf and make GetValue use this instead of its own table. Now there is only one table of units, in pcb-printf.c, and everything should be using it. 4. Upgrade unit structures in pcb-printf. Now units have a real suffix that pcb uses internally and in file.c, and an internationalized copy of the same suffix. Is there any use-case for I18N suffixes? I would imagine units are pretty much universal and language agnostic - unless you are spelling them out, e.g. mètre (French) vs. metre (English) / meter (American). In principle, just audits. The ones I have done so far have been quick and easy, essentially just replacing LocationType/BDimension with Coord, using Angle, using NormalizeAngle() and Distance() instead of repeating code. Nothing deep. I may even have been done by now except that when auditing search.c, I found the arc-intersection problems that lead to a recent long thread on this board. When I get a free moment (probably not till tomorrow) I will just post a bug report and move on, fixing the unit handling without fixing the actual algorithm. Always *BLEEPING* arcs which cause grief. If you could describe a perfect circle with B-Splines, I would be up for killing the arc primitive and replacing it with something a little more robust (unique) in how it may be defined. I expect that aside from that, everything left will be easy. I just need to spend some time and git'er'done. After all the source has been audited, I can commit the small changes to const.h that actually do the conversion. Then I'll push and you guys can jump on me for all the bugs I've caused. I'm looking forward to opening my mm designs with a mm grid and not chasing lost of off-grid gremlins ;) Let me know if I can be of any help. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: New line-snapping feature for PCB - please try it!
Hi guys, Having been dogfooding with PCB a bit recently, I decided to look at adding a feature for snapping to off-grid points along the body of a track. I've just committed a patch which implements that in crosshair.c. It should make it a lot less painful to keep neat layout when you have parts on different grids. It might be possible to implement this more intelligently in a mode-specific way, but for now, the solution I came up with seems to be fairly effective. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: New line-snapping feature for PCB - please try it!
Hi guys, Having been dogfooding with PCB a bit recently, I decided to look at adding a feature for snapping to off-grid points along the body of a track. I've just committed a patch which implements that in crosshair.c. It should make it a lot less painful to keep neat layout when you have parts on different grids. It might be possible to implement this more intelligently in a mode-specific way, but for now, the solution I came up with seems to be fairly effective. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: gEDA-dev: Draft documentation for Scheme API
On Sat, 2011-06-25 at 07:45 +0100, Peter TB Brett wrote: On Friday 24 June 2011 22:44:29 Peter TB Brett wrote: Hi folks, I've now written the documentation for the Scheme API so far (in TexInfo, simply because that was the quickest way to get it done and looking nice). I've put an HTML version on the web temporarily here: http://peter-b.co.uk/geda/geda-scheme/ That URL will change soon(ish), and I'll let you know when it does. Now at http://geda.peter-b.co.uk/geda-scheme/. That look amazing... The manual is pretty darn comprehensive, and from what I've read, the API is really nice too. I've not had much time for doing any gEDA stuff recently, but I look forward to seeing more of this when it lands. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: trcing with pcb design
On Tue, 2011-06-07 at 12:56 -0400, DJ Delorie wrote: Yes, you need to specify a background PPM image. http://www.delorie.com/pcb/bg-image.html It should load PNG fine as well... (certainly the GTK HID can), and probably other formats supported by gdkpixbuf, such as JPEG. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Blog post: Porting gEDA to Guile 2.0, and future plans
On Sun, 2011-06-05 at 21:53 -0400, Dan McMahill wrote: has anyone tried to compile against gtk3? Yes - and the answer is we cannot ;) I can revive my attempt at some point, and propose we proceed as follows: Identify widgets / UI elements which use deprecated APIs or direct structure access forbidden in GTK 3.0. For each identified area, do a brief code and UI design review to see whether the quality of the existing code and user interface is worth spending effort on, or needs to be redesigned. If the UI and basic code is solid, fix up the GTK 2.0 / 3.0 incompatibilities as directly as possible. (No need to re-write these as fully developed widgets with data access via GObject properties and notification via signals if they aren't already so). If the UI is bad, redesign a replacement (coding in line with best practice GTK 3.0 styles), basically from scratch. If the UI is ok but depends on completely deprecated widgets or is generally hateful - write it again from scratch. The aim should not really be to hack hack hack until we just compile with GTK 3.0, but to do a complete UI review - focusing on the most legacy code first. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: PCB head fails with ***MEMORY-ERROR***
On Sat, 2011-06-04 at 10:51 -0700, Andrew Poelstra wrote: Another leak can be seen when zooming in and out. Typing Zz on the board takes another 12Mb. It is only a leak if repeated cycles continue to cause this extra memory to be allocated. It is quite possible that hitting different zoom levels causes the rendering code to evaluate and cache polygon geometry which wasn't displayed at the previous zoom level. PS.. Many thanks for fixing that other leak I introduced. I completely forgot the possibility of breaking out of the loops via other means. That commit was a quick (but intrusive) change required to fix drawing whilst auto-routing, and has caused a couple of teething problems so far. I still think it is the right way to go (moving other types of memory allocation over to it in the future), but we do need to be careful about iterating over lists we might modify. I think the macro loops should ideally be used for read-only access to the data-structures only. I thought I was being clever (and more robust against problems) by adding the g_list_copy (), but obviously I had not taken care of the clean-up correctly. Best wishes, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Two things ... or actually, three
On Mon, 2011-05-30 at 23:34 +0200, Richard Rasker wrote: It compiled OK after installing a bunch of dependencies, but I can't recall at all whether dbus was enabled or not. By the looks of it, I didn't use git, so I guess dbus is switched off. Is there a way to check if it is enabled? xgsch2pcb will probably just hang whilst trying to update your design if PCB isn't built with DBUS support. Crappy I know, but we didn't finish all the timeouts and error messages properly yet. Since I recall you were having issues with xgsch2pcb, I wondered if your PCB might be missing DBUS support. From a terminal, run: dbus-monitor | grep pcb Then open and close PCB. (From a different terminal) I get on my machine (with PCB built with DBUS support): string org.seul.geda.pcb string org.seul.geda.pcb string org.seul.geda.pcb string pcb string Unnamed (Unsaved.pcb) - PCB string Unnamed (Unsaved.pcb) - PCB string org.seul.geda.pcb There is also a new File - Import Schematics option which DJ produced, which you might like to look at as an alternative to xgsch2pcb. I tried this (see another reply of mine to the mailing list), but right now, it doesn't seem all that user-friendly, to put it mildly. As Peter Brett and I developed xgsch2pcb, I'm pretty happy with using that myself - so I'll confess that I've not used the new Import schematics functionality on an actual project. That said.. I've not started an actual project since that functionality was introduced. Either way, I can't really comment - other than it looked ok when I gave it a cursory testing. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Two things ... or actually, three
On Mon, 2011-05-30 at 22:59 +0200, Richard Rasker wrote: Um, OK ... but somehow, my older xgsch2pcb (or perhaps just the older PCB build) doesn't recognize the layout file edited with the newer PCB version any more. As a result, I haven't succeeded in updating any changes in my schematic into the newer PCB version (also see below). When I try updating the schematic now with xgsch2pcb, I get a completely new PCB window with all the elements in the upper left corner, and not the 99% completed and routed layout I have. Hmm, if you have updated gEDA / gnetlist etc.. from git this year, but not recently - you might want to try rebuilding that from a current git version. I'm not sure if it fits your symptoms or not, but there was a nasty bug which sometimes caused problems in gsch2pcb, which was fixed recently. (xgsch2pcb uses gsch2pcb internally). If you're still stuck (and are able / willing to do so), send me a tarball with the design you're having problems with and I'll see what I can do. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: pcjc2 tessellation
On Mon, 2011-05-30 at 09:37 +0200, Gabriel Paubert wrote: Think i know now. Polygons can't have edges that cross other edges. You got it.. PCB doesn't allow self intersecting polygons, so I removed support for rendering them in an attempt to manage the complexity (and hopefully improve speed of) the tessellator. So, what happens if one directly tinkers with the PCB file with an editor or script and creates a self interscting polygon by mistake? Bad things (TM). Certainly the polygon computations will give bogus results, because one of the key underlying assumptions of the polygon algebra routines is violated. If PCB is built with asserts enabled, it will just quit immediately when loading your file. It's not so far fetched, I sometimes edit coordinates in the file to match mechanical design. I can make a mistake or, if it is a script (which also happens), it can have a bug. Its not ideal, but PCB doesn't report these problems unless asserts are enabled (which makes PCB slow). We should probably add a validation pass when loading the board and refuse to display any damaged polygon. That should at least help to pick any problems up with hand (or script) edited boards sooner rather than later. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: pcjc2 tessellation
On Sun, 2011-05-29 at 11:37 +1000, Russell Shaw wrote: Hi, In _borast_bentley_ottmann_tessellate_bo_edges() which is from _cairo_bentley_ottmann_tessellate_bo_edges(), i see you deleted case CAIRO_BO_EVENT_TYPE_INTERSECTION: What effect does this have on the types of polygons that can be rasterized? Think i know now. Polygons can't have edges that cross other edges. You got it.. PCB doesn't allow self intersecting polygons, so I removed support for rendering them in an attempt to manage the complexity (and hopefully improve speed of) the tessellator. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Two things ... or actually, three
On Thu, 2011-05-26 at 10:03 +0200, Richard Rasker wrote: - Work flow of newer gschem/PCB version: until recently, I worked with an older PCB version (20080202), but I finally got round to compiling and installing the latest version -- and it's quite an improvement (no more disappearing crosshairs, outlines and other GUI imperfections). Now I was used to the work flow of gschem - [update pcb in] xgsch2pcb - PCB, but this appears to have changed. Now I think I can find out by myself, but being a lazy person, I wondered if someone could explain this in a few words. Do I also need a newer gschem version? And has anything changed with regard to custom created symbols and footprints (as in: how do I make the new gschem and PCB versions aware of their existence?). Did you build the version from git HEAD, or a recent release? I can't quite recall when --enable-dbus became a default-on option, but if it is not switched on, then xgsch2pcb will not work. git HEAD defaults to --enable-dbus being on. There is also a new File - Import Schematics option which DJ produced, which you might like to look at as an alternative to xgsch2pcb. -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me) signature.asc Description: This is a digitally signed message part ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
