Re: [Kicad-developers] Git transition
On 08/10/2016 04:49 PM, Chris Pavlina wrote: [snip] > Any particular reason for having (fixes lp:) in the first > line? Typically the first line of the commit message is kept very > short and limited to things one might need when quickly browsing a > log. The usual convention in git is to write something like this at > the bottom of the message: > > Fixes: lp: > > It's very easy to search for those if you want, using git log --grep: > > git log --grep="^Fixes: lp:12345678" Currently the webhook accepts a few formats: "fix(es)?[ ]+(lp:|#)?([0-9]+)", but it can be easily changed. I would not mind extending it to catch the mentioned phrase. It also accepts the formula anywhere in the commit message, not only the first line, so it is up to Wayne to set the rules. Regards, Orson signature.asc Description: OpenPGP digital signature ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Unable to login to launchpad.net
I can log in now. On Wed, Aug 10, 2016 at 9:52 AM, Wayne Stambaugh wrote: > > > On 8/10/2016 11:52 AM, jp charras wrote: > > Le 10/08/2016 à 17:28, Chris Pavlina a écrit : > >> Same here, just wait a while. It's just being Launchpad. > > > > Or something else... > > I have no problem. > > Me neither but I was already logged. > > > > >> > >> On Wed, Aug 10, 2016 at 11:26 AM, Duane Johnson < > duane.john...@gmail.com> wrote: > >>> When I try to login at launchpad, I get a very long "Waiting for > >>> launchpad.net" message followed by "OpenID Provider is Unavailable at > This > >>> Time". > >>> -- > > > > Jean-Pierre CHARRAS > > > > ___ > > Mailing list: https://launchpad.net/~kicad-developers > > Post to : kicad-developers@lists.launchpad.net > > Unsubscribe : https://launchpad.net/~kicad-developers > > More help : https://help.launchpad.net/ListHelp > > > > ___ > Mailing list: https://launchpad.net/~kicad-developers > Post to : kicad-developers@lists.launchpad.net > Unsubscribe : https://launchpad.net/~kicad-developers > More help : https://help.launchpad.net/ListHelp > ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Unable to login to launchpad.net
On 8/10/2016 11:52 AM, jp charras wrote: > Le 10/08/2016 à 17:28, Chris Pavlina a écrit : >> Same here, just wait a while. It's just being Launchpad. > > Or something else... > I have no problem. Me neither but I was already logged. > >> >> On Wed, Aug 10, 2016 at 11:26 AM, Duane Johnson >> wrote: >>> When I try to login at launchpad, I get a very long "Waiting for >>> launchpad.net" message followed by "OpenID Provider is Unavailable at This >>> Time". >>> -- > > Jean-Pierre CHARRAS > > ___ > Mailing list: https://launchpad.net/~kicad-developers > Post to : kicad-developers@lists.launchpad.net > Unsubscribe : https://launchpad.net/~kicad-developers > More help : https://help.launchpad.net/ListHelp > ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Unable to login to launchpad.net
It was resolved quickly. Still, it was noticed independently by at least three people, so def lp :) On Wed, Aug 10, 2016 at 11:52 AM, jp charras wrote: > Le 10/08/2016 à 17:28, Chris Pavlina a écrit : >> Same here, just wait a while. It's just being Launchpad. > > Or something else... > I have no problem. > >> >> On Wed, Aug 10, 2016 at 11:26 AM, Duane Johnson >> wrote: >>> When I try to login at launchpad, I get a very long "Waiting for >>> launchpad.net" message followed by "OpenID Provider is Unavailable at This >>> Time". >>>-- > > Jean-Pierre CHARRAS > > ___ > Mailing list: https://launchpad.net/~kicad-developers > Post to : kicad-developers@lists.launchpad.net > Unsubscribe : https://launchpad.net/~kicad-developers > More help : https://help.launchpad.net/ListHelp ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Unable to login to launchpad.net
Le 10/08/2016 à 17:28, Chris Pavlina a écrit : > Same here, just wait a while. It's just being Launchpad. Or something else... I have no problem. > > On Wed, Aug 10, 2016 at 11:26 AM, Duane Johnson > wrote: >> When I try to login at launchpad, I get a very long "Waiting for >> launchpad.net" message followed by "OpenID Provider is Unavailable at This >> Time". >>-- Jean-Pierre CHARRAS ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Unable to login to launchpad.net
Same here, just wait a while. It's just being Launchpad. On Wed, Aug 10, 2016 at 11:26 AM, Duane Johnson wrote: > When I try to login at launchpad, I get a very long "Waiting for > launchpad.net" message followed by "OpenID Provider is Unavailable at This > Time". > > ___ > Mailing list: https://launchpad.net/~kicad-developers > Post to : kicad-developers@lists.launchpad.net > Unsubscribe : https://launchpad.net/~kicad-developers > More help : https://help.launchpad.net/ListHelp > ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
[Kicad-developers] Unable to login to launchpad.net
When I try to login at launchpad, I get a very long "Waiting for launchpad.net" message followed by "OpenID Provider is Unavailable at This Time". ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Ideas about improving of the *.kicad_mod filestructure and *.pretty directories
Le 10/08/2016 à 16:36, Wayne Stambaugh a écrit : > On 8/6/2016 11:03 AM, Thomas Pointhuber wrote: >> Hy, >> >> as Library maintainer I got some insights about the various file formats and >> as well their current >> limits. Inspired by the current eeschema file format rework I would propose >> some changes for the >> footprint files as well. >> >> >> Adding features which are planned for eeschema (or already implemented): >> >> * bezier function > > Could be done. We have had file and object support for them in Eeschema > since I've been with the project (10yrs) but we still do not have anyway > to edit them in either the schematic or symbol editors yet. That should > tell you something about any implementation time even if I add support > for them in the file format. > >> * properties > > We already support them. Once again no editing support. > >> * datasheet urls > > This would be done using a field just like in Eeschema. > >> >> >> Other ideas: >> >> * adding something like a group attribute or allow folders inside *.pretty >> folders. > > I don't think there is anything preventing you from adding sub-folders > inside folders. You will just have to adjust your fp-lib-table > accordingly. I'm not thrilled with the idea of automatically parsing > subfolders in the pcb footprint library plugin. The current parser only > looks for .kicad_mod files in the current folder and will ignore subfolders. I am not sure it works for libs on Github. > >> >>Currently our *.pretty folders are getting bigger and bigger and adding >> an additional level >>in the tree selector would likely improve UX. (For example we have now a >> *.pretty folder >>with >500 footprints of a single manufacturer) > > I would break these into subfolders. It seems to me that this is more > of a library management issue than a file format issue. > >> >> * From what I learned, *.pretty folders are currently specified to only >> contain *.kicad_mod files. >>From a logical view (As well as for maintainability and consistency) we >> should add the 3d files >>into those directories as well. > > I don't think this is a good idea because you would have same 3D model > in multiple footprint libraries which would make model maintenance a > nightmare. Once again, there is nothing from preventing you for doing > this now. AFAIK you can put your 3D models anywhere. All you have to > do is make sure your 3D model path is correct. Once again, I see this > is library management issue. These files are usually very large. They could also create download time issues for *.pretty libs on Github. > >> >> * there are cases where footprints need cuttings inside the pcb, which are >> not round or oval. >>It's possible to use Edge.Cuts inside the footprint (with external >> editing) which works, but >>it would be a good idea to specify this case. (Which would also allow us >> to display those >>footprints correctly inside the 3d viewer) > > This is also one of those things that is already supported in the file > format but there is no support for in the board or footprint library > editors. > >> >> >> Regards, >> >> Thomas Pointhuber Please, don't see *.pretty folders as usual folders, but as libraries. This is not exactly the same. -- Jean-Pierre CHARRAS ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Git transition
On Wed, Aug 10, 2016 at 9:34 AM, Wayne Stambaugh wrote:
>
> On 8/10/2016 5:02 AM, Maciej Sumiński wrote:
> > On 08/08/2016 06:09 PM, Wayne Stambaugh wrote:
> >> The last time I looked, notifications of repo commits still were not
> >> implemented. This is a show stopper for me. I don't want to have to
> >> constantly grep the git commit log to see what changed. If change
> >> notifications are working correctly, then I'm OK with moving forward on
> >> this if you can get the bug fix linking working. We definitely should
> >> do some testing before we go live with this.
> >
> > I see there is an option to set notifications, in the same way as for
> > the bazaar branches ("Edit your subscriptions" on the right side pane).
> > I could not verify it, as likely I cannot receive notifications for the
> > changes I introduce. Even if it does not work, I can implement it in my
> > webhook.
>
> I spent some time yesterday creating my own git clone of kicad on LP and
> I noticed that the subscriptions that I need appear to be available for
> git repos so we shouldn't need any webhooks in for that unless they do
> not work.
>
> >
> > The webhook has reached beta stage. I have created a dummy project for
> > testing purposes, where you can see a bug report [1] and a commit [2]
> > with message that includes a "fix(es)?[ ]+(lp:|#)?([0-9]+)" regex match.
> > When it is detected, it automatically adds a message, changes the bug
> > status and assignee. One thing that is not possible right now is linking
> > with git branches, as apparently launchpad does not handle this at the
> > moment (or I could not find the right format to specify a branch).
>
> Bug report linking is very important to me since I am responsible for
> the stable branch. If there is no support for this yet, I'm OK with
> adding the bug report number to the first line of the commit message and
> the URL somewhere in the commit message body. If I give the OK to use
> git, I will expect all developers that have commit privileges to the
> product repo to follow this without exception. The commit message for
> bug report fixes must have this format:
>
> Description of bug report fix. (fixes lp:)
>
> * https://bugs.launchpad.net/kicad/+bug/
>
> If this is not acceptable, then the git transition will have to wait
> until Canonical gets git bug report linking implemented or Orson beats
> them to it.
Any particular reason for having (fixes lp:) in the first
line? Typically the first line of the commit message is kept very
short and limited to things one might need when quickly browsing a
log. The usual convention in git is to write something like this at
the bottom of the message:
Fixes: lp:
It's very easy to search for those if you want, using git log --grep:
git log --grep="^Fixes: lp:12345678"
>
>
>
>
> >
> > All we need to do is to set a webhook pointing to my script [3]. If it
> > is accepted, then I am going to create a separate lp account for the
> > automated changes.
> >
> > Currently the webhook works on my home server which has a high uptime,
> > but still it is not as reliable as dedicated servers. If there is
> > someone willing to host it on a better machine, I will be pleased to help.
> >
> > If you are curious about the source code, then I can put it in the KiCad
> > github (once I get a repository there) or just post it somewhere.
>
> I can create a repo on github or you can create a repo on launchpad.
> Either way is fine by me. If you want to use github, let me know what
> name you want for the repo and your github user name and I will set up
> the repo and give you admin rights.
>
> Thanks for working on this.
>
> Cheers,
>
> Wayne
>
> >
> > Regards,
> > Orson
> >
> > 1. https://bugs.launchpad.net/kicad-git-test/+bug/1611664
> > 2.
> > https://git.launchpad.net/kicad-git-test/commit/?id=3d29b9be29346fdfaa87cdf8abf6957bf46bb5cd
> > 3. https://orson.net.pl/kicad_git_hook
> >
> >> Before every starts beating the GitHub drum, I have one major issue with
> >> GitHub and that is control. There is no way that I know of to moderate
> >> a project on github. Anyone with a github account can submit a pull
> >> requests at anytime even if they are not part of the dev team. As
> >> project leader, this is an issue. I'm already a my limit with the
> >> development team we have in place and I really don't want to deal with a
> >> wide open code hosting. I also have no way of removing someone from the
> >> list should I need to. I know it hasn't happened yet but I am not naive
> >> enough to think that it wont happen. At this time, I am more
> >> comfortable with LP until something better comes along or we take full
> >> control a provide our own hosting.
> >>
> >> On 8/8/2016 3:58 AM, Maciej Sumiński wrote:
> >>> Apparently we have not discussed git transition for a long time now, so
> >>> I felt it is the right time to raise the subject again to keep our
> >>> routine.
> >>>
> >>> Launchpad now supp
Re: [Kicad-developers] Ideas about improving of the *.kicad_mod filestructure and *.pretty directories
On 8/6/2016 11:03 AM, Thomas Pointhuber wrote: > Hy, > > as Library maintainer I got some insights about the various file formats and > as well their current > limits. Inspired by the current eeschema file format rework I would propose > some changes for the > footprint files as well. > > > Adding features which are planned for eeschema (or already implemented): > > * bezier function Could be done. We have had file and object support for them in Eeschema since I've been with the project (10yrs) but we still do not have anyway to edit them in either the schematic or symbol editors yet. That should tell you something about any implementation time even if I add support for them in the file format. > * properties We already support them. Once again no editing support. > * datasheet urls This would be done using a field just like in Eeschema. > > > Other ideas: > > * adding something like a group attribute or allow folders inside *.pretty > folders. I don't think there is anything preventing you from adding sub-folders inside folders. You will just have to adjust your fp-lib-table accordingly. I'm not thrilled with the idea of automatically parsing subfolders in the pcb footprint library plugin. The current parser only looks for .kicad_mod files in the current folder and will ignore subfolders. > >Currently our *.pretty folders are getting bigger and bigger and adding an > additional level >in the tree selector would likely improve UX. (For example we have now a > *.pretty folder >with >500 footprints of a single manufacturer) I would break these into subfolders. It seems to me that this is more of a library management issue than a file format issue. > > * From what I learned, *.pretty folders are currently specified to only > contain *.kicad_mod files. >From a logical view (As well as for maintainability and consistency) we > should add the 3d files >into those directories as well. I don't think this is a good idea because you would have same 3D model in multiple footprint libraries which would make model maintenance a nightmare. Once again, there is nothing from preventing you for doing this now. AFAIK you can put your 3D models anywhere. All you have to do is make sure your 3D model path is correct. Once again, I see this is library management issue. > > * there are cases where footprints need cuttings inside the pcb, which are > not round or oval. >It's possible to use Edge.Cuts inside the footprint (with external > editing) which works, but >it would be a good idea to specify this case. (Which would also allow us > to display those >footprints correctly inside the 3d viewer) This is also one of those things that is already supported in the file format but there is no support for in the board or footprint library editors. > > > Regards, > > Thomas Pointhuber > > > > ___ > Mailing list: https://launchpad.net/~kicad-developers > Post to : kicad-developers@lists.launchpad.net > Unsubscribe : https://launchpad.net/~kicad-developers > More help : https://help.launchpad.net/ListHelp > ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] [PATCH] Memoize SHAPE_LINE_CHAIN bounding box computation
Orson & Tom, Please look over this patch when you get a chance to see if it makes sense to commit it. Thanks, Wayne On 8/6/2016 9:36 PM, Chris Pavlina wrote: > The board I'm working on is quite complex and makes KiCad _very_ slow, > so I'm working on a bit of profiling and optimizing. The attached patch > memoizes SHAPE_LINE_CHAIN bounding box computation, which is a hot spot > during netlist sync (presumably when trace/net connections are > calculated). For my specific case, it results in a 38% speedup, from 21 > seconds down to 13 seconds. > > Next hotspot to tackle is NETLIST_OBJECT_LIST::BuildNetListInfo :) > > > > ___ > Mailing list: https://launchpad.net/~kicad-developers > Post to : kicad-developers@lists.launchpad.net > Unsubscribe : https://launchpad.net/~kicad-developers > More help : https://help.launchpad.net/ListHelp > ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Git transition
On 8/10/2016 5:02 AM, Maciej Sumiński wrote:
> On 08/08/2016 06:09 PM, Wayne Stambaugh wrote:
>> The last time I looked, notifications of repo commits still were not
>> implemented. This is a show stopper for me. I don't want to have to
>> constantly grep the git commit log to see what changed. If change
>> notifications are working correctly, then I'm OK with moving forward on
>> this if you can get the bug fix linking working. We definitely should
>> do some testing before we go live with this.
>
> I see there is an option to set notifications, in the same way as for
> the bazaar branches ("Edit your subscriptions" on the right side pane).
> I could not verify it, as likely I cannot receive notifications for the
> changes I introduce. Even if it does not work, I can implement it in my
> webhook.
I spent some time yesterday creating my own git clone of kicad on LP and
I noticed that the subscriptions that I need appear to be available for
git repos so we shouldn't need any webhooks in for that unless they do
not work.
>
> The webhook has reached beta stage. I have created a dummy project for
> testing purposes, where you can see a bug report [1] and a commit [2]
> with message that includes a "fix(es)?[ ]+(lp:|#)?([0-9]+)" regex match.
> When it is detected, it automatically adds a message, changes the bug
> status and assignee. One thing that is not possible right now is linking
> with git branches, as apparently launchpad does not handle this at the
> moment (or I could not find the right format to specify a branch).
Bug report linking is very important to me since I am responsible for
the stable branch. If there is no support for this yet, I'm OK with
adding the bug report number to the first line of the commit message and
the URL somewhere in the commit message body. If I give the OK to use
git, I will expect all developers that have commit privileges to the
product repo to follow this without exception. The commit message for
bug report fixes must have this format:
Description of bug report fix. (fixes lp:)
* https://bugs.launchpad.net/kicad/+bug/
If this is not acceptable, then the git transition will have to wait
until Canonical gets git bug report linking implemented or Orson beats
them to it.
>
> All we need to do is to set a webhook pointing to my script [3]. If it
> is accepted, then I am going to create a separate lp account for the
> automated changes.
>
> Currently the webhook works on my home server which has a high uptime,
> but still it is not as reliable as dedicated servers. If there is
> someone willing to host it on a better machine, I will be pleased to help.
>
> If you are curious about the source code, then I can put it in the KiCad
> github (once I get a repository there) or just post it somewhere.
I can create a repo on github or you can create a repo on launchpad.
Either way is fine by me. If you want to use github, let me know what
name you want for the repo and your github user name and I will set up
the repo and give you admin rights.
Thanks for working on this.
Cheers,
Wayne
>
> Regards,
> Orson
>
> 1. https://bugs.launchpad.net/kicad-git-test/+bug/1611664
> 2.
> https://git.launchpad.net/kicad-git-test/commit/?id=3d29b9be29346fdfaa87cdf8abf6957bf46bb5cd
> 3. https://orson.net.pl/kicad_git_hook
>
>> Before every starts beating the GitHub drum, I have one major issue with
>> GitHub and that is control. There is no way that I know of to moderate
>> a project on github. Anyone with a github account can submit a pull
>> requests at anytime even if they are not part of the dev team. As
>> project leader, this is an issue. I'm already a my limit with the
>> development team we have in place and I really don't want to deal with a
>> wide open code hosting. I also have no way of removing someone from the
>> list should I need to. I know it hasn't happened yet but I am not naive
>> enough to think that it wont happen. At this time, I am more
>> comfortable with LP until something better comes along or we take full
>> control a provide our own hosting.
>>
>> On 8/8/2016 3:58 AM, Maciej Sumiński wrote:
>>> Apparently we have not discussed git transition for a long time now, so
>>> I felt it is the right time to raise the subject again to keep our routine.
>>>
>>> Launchpad now supports merge proposals [1] and it looks like they are
>>> still improving git integration. The only missing feature are links
>>> between bug fixing commits and respective bug reports (bzr commit ...
>>> --fixes=lp:123456). It is not supported by git itself, but might be
>>> resolved using webhooks [2] and appropriate keywords in commit messages
>>> (e.g. "Fixed a memory leak [fixes #123456]"). If this is the only
>>> obstacle, then I volunteer to provide code for the hook.
>>>
>>> What do you think? Is there anything else that prevents transition?
>>>
>>> Regards,
>>> Orson
>>>
>>> 1. https://help.launchpad.net/Code/Git
>>> 2. https://help.launchpad.net/API/We
Re: [Kicad-developers] Integrated Simulator
Le 10/08/2016 à 14:47, Maciej Sumiński a écrit : > On 08/10/2016 02:43 PM, jp charras wrote: >> Le 10/08/2016 à 10:21, jp charras a écrit : >>> Le 10/08/2016 à 09:43, jp charras a écrit : Le 10/08/2016 à 09:03, Maciej Sumiński a écrit : > Hi Jean-Pierre, > > This is strange. I had the problem you describe (Windows 7 x86_64) and > fixed it in commit ee24275f [1]. > 1. > https://github.com/orsonmmz/kicad-source-mirror/commit/ee24275fc8fd78fe47428d6b45a41684dae9f6f2 > >>> >>> Hi Orson, >>> >>> I am trying your spice branch, but I cannot compile your code: a file is >>> missing: >>> in eeschema/sim/ngspice.h: >>> "#include <../share/ngspice/include/ngspice/sharedspice.h>" >>> fails (no such file). >>> >>> Sorry. >>> >> >> Hi Orson, >> >> After copying the missing file sharedspice.h from >> https://github.com/twlostow/kicad-dev (ngspice branch) >> all compiles and works fine. >> Thanks. >> > > Hi Jean-Pierre, > > I removed the header file intentionally, as it is supposed to be > delivered with (lib)ngspice package. How did you obtain the libngspice > library? I built it from sources, with your explanations in a previous mail. > > I have just tested a package generated with ngspice mingw64 PKGBUILD [1] > and the new version of get_libngspice_so.sh [2] and they both installed > the header file. > > Regards, > Orson > > 1. > https://github.com/Alexpux/MINGW-packages/blob/master/mingw-w64-ngspice/PKGBUILD > 2. https://orson.net.pl/pub/libngspice/get_libngspice_so.sh -- Jean-Pierre CHARRAS ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Integrated Simulator
On 08/10/2016 02:43 PM, jp charras wrote: > Le 10/08/2016 à 10:21, jp charras a écrit : >> Le 10/08/2016 à 09:43, jp charras a écrit : >>> Le 10/08/2016 à 09:03, Maciej Sumiński a écrit : Hi Jean-Pierre, This is strange. I had the problem you describe (Windows 7 x86_64) and fixed it in commit ee24275f [1]. 1. https://github.com/orsonmmz/kicad-source-mirror/commit/ee24275fc8fd78fe47428d6b45a41684dae9f6f2 >> >> Hi Orson, >> >> I am trying your spice branch, but I cannot compile your code: a file is >> missing: >> in eeschema/sim/ngspice.h: >> "#include <../share/ngspice/include/ngspice/sharedspice.h>" >> fails (no such file). >> >> Sorry. >> > > Hi Orson, > > After copying the missing file sharedspice.h from > https://github.com/twlostow/kicad-dev (ngspice branch) > all compiles and works fine. > Thanks. > Hi Jean-Pierre, I removed the header file intentionally, as it is supposed to be delivered with (lib)ngspice package. How did you obtain the libngspice library? I have just tested a package generated with ngspice mingw64 PKGBUILD [1] and the new version of get_libngspice_so.sh [2] and they both installed the header file. Regards, Orson 1. https://github.com/Alexpux/MINGW-packages/blob/master/mingw-w64-ngspice/PKGBUILD 2. https://orson.net.pl/pub/libngspice/get_libngspice_so.sh signature.asc Description: OpenPGP digital signature ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Integrated Simulator
Le 10/08/2016 à 10:21, jp charras a écrit : > Le 10/08/2016 à 09:43, jp charras a écrit : >> Le 10/08/2016 à 09:03, Maciej Sumiński a écrit : >>> Hi Jean-Pierre, >>> >>> This is strange. I had the problem you describe (Windows 7 x86_64) and >>> fixed it in commit ee24275f [1]. >>> 1. >>> https://github.com/orsonmmz/kicad-source-mirror/commit/ee24275fc8fd78fe47428d6b45a41684dae9f6f2 >>> > > Hi Orson, > > I am trying your spice branch, but I cannot compile your code: a file is > missing: > in eeschema/sim/ngspice.h: > "#include <../share/ngspice/include/ngspice/sharedspice.h>" > fails (no such file). > > Sorry. > Hi Orson, After copying the missing file sharedspice.h from https://github.com/twlostow/kicad-dev (ngspice branch) all compiles and works fine. Thanks. -- Jean-Pierre CHARRAS ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Git transition
On 08/08/2016 06:09 PM, Wayne Stambaugh wrote:
> The last time I looked, notifications of repo commits still were not
> implemented. This is a show stopper for me. I don't want to have to
> constantly grep the git commit log to see what changed. If change
> notifications are working correctly, then I'm OK with moving forward on
> this if you can get the bug fix linking working. We definitely should
> do some testing before we go live with this.
I see there is an option to set notifications, in the same way as for
the bazaar branches ("Edit your subscriptions" on the right side pane).
I could not verify it, as likely I cannot receive notifications for the
changes I introduce. Even if it does not work, I can implement it in my
webhook.
The webhook has reached beta stage. I have created a dummy project for
testing purposes, where you can see a bug report [1] and a commit [2]
with message that includes a "fix(es)?[ ]+(lp:|#)?([0-9]+)" regex match.
When it is detected, it automatically adds a message, changes the bug
status and assignee. One thing that is not possible right now is linking
with git branches, as apparently launchpad does not handle this at the
moment (or I could not find the right format to specify a branch).
All we need to do is to set a webhook pointing to my script [3]. If it
is accepted, then I am going to create a separate lp account for the
automated changes.
Currently the webhook works on my home server which has a high uptime,
but still it is not as reliable as dedicated servers. If there is
someone willing to host it on a better machine, I will be pleased to help.
If you are curious about the source code, then I can put it in the KiCad
github (once I get a repository there) or just post it somewhere.
Regards,
Orson
1. https://bugs.launchpad.net/kicad-git-test/+bug/1611664
2.
https://git.launchpad.net/kicad-git-test/commit/?id=3d29b9be29346fdfaa87cdf8abf6957bf46bb5cd
3. https://orson.net.pl/kicad_git_hook
> Before every starts beating the GitHub drum, I have one major issue with
> GitHub and that is control. There is no way that I know of to moderate
> a project on github. Anyone with a github account can submit a pull
> requests at anytime even if they are not part of the dev team. As
> project leader, this is an issue. I'm already a my limit with the
> development team we have in place and I really don't want to deal with a
> wide open code hosting. I also have no way of removing someone from the
> list should I need to. I know it hasn't happened yet but I am not naive
> enough to think that it wont happen. At this time, I am more
> comfortable with LP until something better comes along or we take full
> control a provide our own hosting.
>
> On 8/8/2016 3:58 AM, Maciej Sumiński wrote:
>> Apparently we have not discussed git transition for a long time now, so
>> I felt it is the right time to raise the subject again to keep our routine.
>>
>> Launchpad now supports merge proposals [1] and it looks like they are
>> still improving git integration. The only missing feature are links
>> between bug fixing commits and respective bug reports (bzr commit ...
>> --fixes=lp:123456). It is not supported by git itself, but might be
>> resolved using webhooks [2] and appropriate keywords in commit messages
>> (e.g. "Fixed a memory leak [fixes #123456]"). If this is the only
>> obstacle, then I volunteer to provide code for the hook.
>>
>> What do you think? Is there anything else that prevents transition?
>>
>> Regards,
>> Orson
>>
>> 1. https://help.launchpad.net/Code/Git
>> 2. https://help.launchpad.net/API/Webhooks
>>
>>
>>
>> ___
>> Mailing list: https://launchpad.net/~kicad-developers
>> Post to : kicad-developers@lists.launchpad.net
>> Unsubscribe : https://launchpad.net/~kicad-developers
>> More help : https://help.launchpad.net/ListHelp
>>
>
> ___
> Mailing list: https://launchpad.net/~kicad-developers
> Post to : kicad-developers@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~kicad-developers
> More help : https://help.launchpad.net/ListHelp
>
signature.asc
Description: OpenPGP digital signature
___
Mailing list: https://launchpad.net/~kicad-developers
Post to : kicad-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kicad-developers
More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Integrated Simulator
Le 10/08/2016 à 09:43, jp charras a écrit : > Le 10/08/2016 à 09:03, Maciej Sumiński a écrit : >> Hi Jean-Pierre, >> >> This is strange. I had the problem you describe (Windows 7 x86_64) and >> fixed it in commit ee24275f [1]. >> 1. >> https://github.com/orsonmmz/kicad-source-mirror/commit/ee24275fc8fd78fe47428d6b45a41684dae9f6f2 >> Hi Orson, I am trying your spice branch, but I cannot compile your code: a file is missing: in eeschema/sim/ngspice.h: "#include <../share/ngspice/include/ngspice/sharedspice.h>" fails (no such file). Sorry. -- Jean-Pierre CHARRAS ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Integrated Simulator
Le 10/08/2016 à 09:03, Maciej Sumiński a écrit : > Hi Jean-Pierre, > > This is strange. I had the problem you describe (Windows 7 x86_64) and > fixed it in commit ee24275f [1]. Does the cursor look fine if you revert > the commit? What wxWidgets & Windows versions do you use? Yes, the cursor looks fine, just with your code. Thanks. I am using W7 32 bits, wxWidgets 3.1 (and sometimes 3.02) With initial code, the problem was on both wxWidgets versions. (code from https://github.com/twlostow/kicad-dev) > > Regards, > Orson > > 1. > https://github.com/orsonmmz/kicad-source-mirror/commit/ee24275fc8fd78fe47428d6b45a41684dae9f6f2 > > On 08/09/2016 12:10 PM, jp charras wrote: >> Hi Orson, >> >> At least on Windows (and perhaps on others OS) the cursor mask looks like it >> is inverted (0 instead >> of 1 and 1 instead of 0) >> Therefore the probes are not visible on a white background, and are a >> negative shape on a black >> background. >> >> Here is a patch to use the right (inverted) mask. >> (sorry, the initial mask is not modified, due to my lazyness, the right mask >> is built on the fly). >> Not tested on OSX. >> >> Thanks, -- Jean-Pierre CHARRAS ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
Re: [Kicad-developers] Integrated Simulator
Hi Jean-Pierre, This is strange. I had the problem you describe (Windows 7 x86_64) and fixed it in commit ee24275f [1]. Does the cursor look fine if you revert the commit? What wxWidgets & Windows versions do you use? Regards, Orson 1. https://github.com/orsonmmz/kicad-source-mirror/commit/ee24275fc8fd78fe47428d6b45a41684dae9f6f2 On 08/09/2016 12:10 PM, jp charras wrote: > Hi Orson, > > At least on Windows (and perhaps on others OS) the cursor mask looks like it > is inverted (0 instead > of 1 and 1 instead of 0) > Therefore the probes are not visible on a white background, and are a > negative shape on a black > background. > > Here is a patch to use the right (inverted) mask. > (sorry, the initial mask is not modified, due to my lazyness, the right mask > is built on the fly). > Not tested on OSX. > > Thanks, > > > > ___ > Mailing list: https://launchpad.net/~kicad-developers > Post to : kicad-developers@lists.launchpad.net > Unsubscribe : https://launchpad.net/~kicad-developers > More help : https://help.launchpad.net/ListHelp > signature.asc Description: OpenPGP digital signature ___ Mailing list: https://launchpad.net/~kicad-developers Post to : kicad-developers@lists.launchpad.net Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
