Closed Gasman2014 closed 6 years ago
I developped this on a debian testing system and it worked. When ported to Ubuntu, it failed with the same error. I'll try fixing it, stay tuned.
Thank you - I was also trying an alternative (eeshow https://neo900.org/stuff/eeshow/ )but that just spirals into an annoying dependency hell related to problems building for x-windows on macOS. Would prefer not to have to run everything in a virtual environment!
Did you have any further thoughts on this? I have been writing a visual diff for Kicad pcbnew with a web output https://github.com/Gasman2014/KiCad-Diff but would quite like to get something similar working for schematics. Your notes on smudge/clean were very helpful to me.
The export to svg of already existing schematic files is still working (plotkicadsch), but the git library for extracting the files at a given reference has undergone a lot of changes and I hadn't much time to try to fix it.
OK, thanks for the hint - I will have a look and see if I can bend it to my needs. Actually, I don’t want to use git anyway - I would quite like to get the versioning aspect to pull out Fossil scm versions (https://www.fossil-scm.org/index.html/doc/trunk/www/index.wiki) as it has more integrated functionality for my plans.
BW
John
On 13 Sep 2017, at 10:15, Jean-Noël Avila notifications@github.com wrote:
The export to svg of already existing schematic files is still working (plotkicadsch), but the git library for extracting the files at a given reference has undergone a lot of changes and I hadn't much time to try to fix it.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/jnavila/plotkicadsch/issues/2#issuecomment-329108407, or mute the thread https://github.com/notifications/unsubscribe-auth/AIAyueBKo2BnkRXZ6wx1e6gEetwcEJ14ks5sh51EgaJpZM4OMNyv.
JFTR, I fixed the ocaml-git interface. I had a look at your graphical comparison script and it seems your image magick scripts are much more advanced than mine. I will maybe borrow you some ideas. Thanks.
Thank you - I will have a go again tomorrow with the ocaml-git interface - just been trying to sort out some python related issues in a new Kicad install :(….
Not sure about ‘advanced’ but hope they work for you - you are very welcome to borrow (as I did from you!)
One of the issues I have struggled with is the seeming random area that pcbnew plots. I am hitting a problem that if you add something OUTSIDE the current board outline, the plot area is adjusted - so it throws off any visual diff. I will have another play with it in a week or two when I get time.
BW
John
On 15 Sep 2017, at 12:35, Jean-Noël Avila notifications@github.com wrote:
JFTR, I fixed the ocaml-git interface. I had a look at your graphical comparison script and it seems your image magick scripts are much more advanced than mine. I will maybe borrow you some ideas. Thanks.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/jnavila/plotkicadsch/issues/2#issuecomment-329757166, or mute the thread https://github.com/notifications/unsubscribe-auth/AIAyuaAeIoBgjTERnuPjIzQadWTQo2zuks5simDrgaJpZM4OMNyv.
Managed to get this to compile and run now on macOS - Thank you!. The main issues now are that the font is fixed on TimesNew Roman and I am having problems with hierarchical sheets but I will keep having a play with it (had a bit of a major laptop disaster last week involving a new keyboard replacement in my MacBook Pro - LOTS of small screws). BW
John
On 17 Sep 2017, at 23:56, John Pateman johnpateman@me.com wrote:
Thank you - I will have a go again tomorrow with the ocaml-git interface - just been trying to sort out some python related issues in a new Kicad install :(….
Not sure about ‘advanced’ but hope they work for you - you are very welcome to borrow (as I did from you!)
One of the issues I have struggled with is the seeming random area that pcbnew plots. I am hitting a problem that if you add something OUTSIDE the current board outline, the plot area is adjusted - so it throws off any visual diff. I will have another play with it in a week or two when I get time.
BW
John
On 15 Sep 2017, at 12:35, Jean-Noël Avila <notifications@github.com mailto:notifications@github.com> wrote:
JFTR, I fixed the ocaml-git interface. I had a look at your graphical comparison script and it seems your image magick scripts are much more advanced than mine. I will maybe borrow you some ideas. Thanks.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/jnavila/plotkicadsch/issues/2#issuecomment-329757166, or mute the thread https://github.com/notifications/unsubscribe-auth/AIAyuaAeIoBgjTERnuPjIzQadWTQo2zuks5simDrgaJpZM4OMNyv.
Hi @Gasman2014
I have a few questions concerning your last comment:
Jean-Noël
The font thing is not a major problem but here are a couple of pairs of screenshots;
There are some minor positioning errors - the input label, for instance, is on the other side. Ignore the hierarchical sheet comment - I was failing to load in the necessary library.
Text block alignments are not correctly managed when there are rotations or mirrors.
Hi Jean-Noël,
I am pretty sure that the text alignment is OK for the rotated components - I was meaning the 'Input' is the wrong side of the line (see circled item).Providing it is consistent, it isn't a problem, I guess.
Any idea on the font issue? How are fonts specified? I think Kicad uses its own internal font and renders it in a nonstandard way.
BW
John
On Oct 15, 2017, at 01:35 PM, Jean-Noël Avila notifications@github.com wrote:
Text block alignments are not correctly managed when there are rotations or mirrors. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.
Just a heads up to say that the texts should now be correctly handled.
Relying on the default font is not an option. Which font would you prefer?
Hiya
That sounds good!
The kicad default font is a monospaced sans serif. It is also a single line font - like an engraving font. I guess that something similar as possible would be best. I tried out a few samples on various font matching sites and think that, of the ttf fonts, something like Ubuntu mono is a pretty good match (and has a suitable licence).
Alternatively, there are a number of Hershey engraving fonts which are ’single stroke’ (http://climserv.ipsl.polytechnique.fr/documentation/idl_help/Hershey_Vector_Font_Samples.html http://climserv.ipsl.polytechnique.fr/documentation/idl_help/Hershey_Vector_Font_Samples.html) These are available in a off .cxf format - I have used these on my cnc machine- they are installed with inkscape, I think. one of these might also be appropriate? I don’t think there are any licensing issues with these either.
BW
John
On 27 Oct 2017, at 20:11, Jean-Noël Avila notifications@github.com wrote:
Just a heads up to say that the texts should now be correctly handled.
Relying on the default font is not an option. Which font would you prefer?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/jnavila/plotkicadsch/issues/2#issuecomment-340045217, or mute the thread https://github.com/notifications/unsubscribe-auth/AIAyuUYyhZj-ctk7vTeHhOc3olWuptRCks5swhzTgaJpZM4OMNyv.
For the fonts, I just chose a mono font. The one used in Kicad cannot be found as a system one. Did you notice any difference?
Hi JN,
I have now reinstalled the latest version but I am still getting everything in ‘Times'. From what you say, should now be in mono so not sure what is happening.
To make things ‘cleaner’ I did a fresh ‘opam install plotkicadsch’ in a completely new repo and I am on version 1.4 according to the install log.
The text positioning issue I mentioned previously had been resolved but the font issue unfortunately has not.
BW
John
On 9 Nov 2017, at 21:43, Jean-Noël Avila notifications@github.com wrote:
For the fonts, I just chose a mono font. The one used in Kicad cannot be found as a system one. Did you notice any difference?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/jnavila/plotkicadsch/issues/2#issuecomment-343301091, or mute the thread https://github.com/notifications/unsubscribe-auth/AIAyudstnQEmOOq2Fw6NkAYiwMcgWYVKks5s03HogaJpZM4OMNyv.
Version 1.4 does not include the fix for the font, but the latest Git does. The SVG viewer might be picky about font specifications.
OK will have play with that (but I have got to get my appraisal and tax return done :(
BW
John
On 14 Nov 2017, at 21:19, Jean-Noël Avila notifications@github.com wrote:
Version 1.4 does not include the fix for the font, but the latest Git does. The SVG viewer might be picky about font specifications.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/jnavila/plotkicadsch/issues/2#issuecomment-344402041, or mute the thread https://github.com/notifications/unsubscribe-auth/AIAyuY_y1aP0J08s8pHwsomgtMV1S964ks5s2gPmgaJpZM4OMNyv.
I was able to reproduce the serif-font issue with Firefox. The latest version forces a sans-serif font which seems to be understood by Firefox. So it might also help for you.
Yes, now working. Also like your visual diff - hope that my KiDiff was helpful!
In fact, the screenshot of visual diff is from the new diffing strategy : diff at the level of drawing primitives and compute a diff in vector image. This allows to zoom at any level. You can try plotgitsch
for two directories also (if it can be of use in the case of Fossil).
I have managed to compile this on macOS although I have had a few problems. :(
The initial compile failed with a
git.unix
dependency issue although I havegit-unix
installed and updated. I changedgit.unix
togit-unix
in the _oasis file and I could then compile and install OK. However, I am now gettingFatal error: exception Plotgitsch.InternalGitError("path not found: /")
when I runplotgitsch
from within a valid git checkout containing a Kicad project .pro & schematic .sch with a committed revision history.Any advice?