[Kicad-developers] Request to allow text rotation/mirroring in any position/angle, silk screen

Hi , Importing from Eagle to KiCad I'm hitting the problem of KiCad not allowing text in any position, this make's for a big manual clean up. Of course there is too side's to this story. Not allowing the user to make easy mistakes. But from the import of side of things. From Eagle and maybe

Re: [Kicad-developers] [PATCH] Rework OpenSSL mutex workaround

Hi Wayne, On 05.05.2016 22:51, Wayne Stambaugh wrote: > Is this patch for the stable branch, the product branch, or both? Currently, for the product branch, and ideally it should be trivial to backport to the stable branch afterwards (it removes a lot of code that doesn't exist in the stable

Re: [Kicad-developers] [PATCH] Rework OpenSSL mutex workaround

Is this patch for the stable branch, the product branch, or both? On 5/5/2016 3:25 PM, Simon Richter wrote: > > This allows compiling KiCad without OpenSSL. If CURL uses OpenSSL in the > background, we serialize requests to avoid a bug there, which degrades > performance. Given that no one

[Kicad-developers] [PATCH] Rework OpenSSL mutex workaround

This allows compiling KiCad without OpenSSL. If CURL uses OpenSSL in the background, we serialize requests to avoid a bug there, which degrades performance. Given that no one should link KiCad against OpenSSL really because of licence conflicts, this is not so bad. --- common/CMakeLists.txt

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

Hi, On 05.05.2016 20:01, Bernhard Stegmaier wrote: > Just a note… I may be wrong, but I guess for OS X this is only true if you > build against the curl supplied by OS X natively? Would it make sense to serialize the network accesses if libcurl tells us that it is using the OpenSSL backend (at

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

Just a note… I may be wrong, but I guess for OS X this is only true if you build against the curl supplied by OS X natively? My builds per default always pick the one from MacPorts, which links against OpenSSL or GnuTLS (depending on MacPorts config)… in this case I guess you will see the same

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

You don't need openssl locking for OSX >= 10.9. curl uses Apple Common Crypto. Other platforms may need it depending on Linux distro (some have dropped openssl for curl, some haven't). On Thu, May 5, 2016 at 12:17 PM, Adam Wolf wrote: > Simon, > > Is the OpenSSL

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

Simon, Is the OpenSSL locking workaround something we know we'll need for the libcurl backport? If so, do we have someone who can work on it? I can set up a build for this whenever it's needed. Adam Wolf On Thu, May 5, 2016 at 10:26 AM, Simon Richter wrote: > Hi

Re: [Kicad-developers] PATCH: OS X copy/close bug fix

No, it is not yet fixed. Bug ticket is still open http://trac.wxwidgets.org/ticket/15678 If this workaround works, maybe it would be better instead of working around any single instance in KiCad to patch the wxButton(…) function in wxWidgets to avoid

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

On 5/5/2016 11:26 AM, Simon Richter wrote: > Hi Wayne, > > On 05.05.2016 16:15, Wayne Stambaugh wrote: > >> Thanks Simon! Nice work. I just tested this on windows and it appears >> to work fine. I'm assuming you tested this on Linux. > > Yes, briefly. These are just the two "switch to CURL"

Re: [Kicad-developers] PATCH: OS X Beep Fix

Thank you! I pushed your patch in revision 6775 with coding style fixes. On Thu, May 05, 2016 at 01:44:54AM -0600, Collin Anderson wrote: > Hi, I have attached a patch that fixes an annoyance that is present in the > footprint editor and pcbnew, but only under OS X. In either of these >

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

Hi Wayne, On 05.05.2016 16:15, Wayne Stambaugh wrote: > Thanks Simon! Nice work. I just tested this on windows and it appears > to work fine. I'm assuming you tested this on Linux. Yes, briefly. These are just the two "switch to CURL" patches from the devel branch, and one small fix, so not

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

Please excuse my fat fingers. That should be 6440. On 5/5/2016 11:04 AM, Chris Pavlina wrote: > I do hope you didn't intent to credit Dick for 6640, that one's mine! ;) > > On Thu, May 05, 2016 at 10:57:34AM -0400, Wayne Stambaugh wrote: >> On 5/5/2016 10:47 AM, Chris Pavlina wrote: >>> On Thu,

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

I do hope you didn't intent to credit Dick for 6640, that one's mine! ;) On Thu, May 05, 2016 at 10:57:34AM -0400, Wayne Stambaugh wrote: > On 5/5/2016 10:47 AM, Chris Pavlina wrote: > > On Thu, May 05, 2016 at 10:15:02AM -0400, Wayne Stambaugh wrote: > >> [snip] > >> > >> I didn't see any credit

Re: [Kicad-developers] PATCH: OS X copy/close bug fix

Thanks for your attention to this, Collin. Another annoyance on OS X, and I'm glad you're looking for solutions to these (like I should have!) Duane On Thu, May 5, 2016 at 4:48 AM, jp charras wrote: > Le 05/05/2016 à 10:38, Collin Anderson a écrit : > > Another little OS

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

On Thu, May 05, 2016 at 10:15:02AM -0400, Wayne Stambaugh wrote: > [snip] > > I didn't see any credit to Dick for the work he did on libcurl > transition so that will need to be added to commit message. Hm, did he write any of the code in these patches? > > On 5/5/2016 4:39 AM, Simon Richter

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

On 5/5/2016 10:47 AM, Chris Pavlina wrote: > On Thu, May 05, 2016 at 10:15:02AM -0400, Wayne Stambaugh wrote: >> [snip] >> >> I didn't see any credit to Dick for the work he did on libcurl >> transition so that will need to be added to commit message. > > Hm, did he write any of the code in these

Re: [Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

Thanks Simon! Nice work. I just tested this on windows and it appears to work fine. I'm assuming you tested this on Linux. Would one of our osx devs please apply these patches to the 4 stable branch and make sure that it builds and works correctly on osx? Once I finish vetting Chris` file

Re: [Kicad-developers] PATCH: OS X Beep Fix

On Thu, May 5, 2016 at 1:44 AM, Collin Anderson wrote: > I can definitely confirm that it has been an ever-present problem on OS X, > to the point I'm in the habit of turning off my speakers as soon as I open > pcbnew :). It's minor but annoying. It was never fixed as

Re: [Kicad-developers] PATCH: OS X Beep Fix

On Thu, May 5, 2016 at 1:44 AM, Collin Anderson wrote: > I can definitely confirm that it has been an ever-present problem on OS X, > to the point I'm in the habit of turning off my speakers as soon as I open > pcbnew :). It's minor but annoying. It was never fixed as

Re: [Kicad-developers] PATCH: OS X Beep Fix

On 5/5/2016 6:52 AM, Chris Pavlina wrote: > Yes, please don't go interleaving preprocessor statements with fragments of > code like that, it's ugly and very prone to causing mistakes and > misunderstandings. > > #ifdef __APPLE__ > if( type != wxEVT_CHAR ) > aEvent.Skip(); > #else >

Re: [Kicad-developers] PATCH: OS X Beep Fix

Yes, please don't go interleaving preprocessor statements with fragments of code like that, it's ugly and very prone to causing mistakes and misunderstandings. #ifdef __APPLE__ if( type != wxEVT_CHAR ) aEvent.Skip(); #else aEvent.Skip(); #endif ^ That is much clearer to me, it actually

Re: [Kicad-developers] PATCH: OS X copy/close bug fix

Le 05/05/2016 à 10:38, Collin Anderson a écrit : > Another little OS X fix. Most, though not all, of the dialogs in KiCad that > have cancel buttons > break copy of text on OS X. If you highlight text (for example, the net of a > pad, an operation I > find myself doing fairly often) and hit

Re: [Kicad-developers] PATCH: OS X Beep Fix

In its current state i am not a fan of this patch, as if someone comes along and adds a line between the #ifdev and aEvent.skip() is will break, Ideally it should probably be #ifdef __APPLE__ if ( type != wxEVT_CHAR ) aEvent.Skip(); #else aEvent.Skip(); #endif the comment can still be

Re: [Kicad-developers] PATCH: OS X copy/close bug fix

the only issue i see with this patch is it seems to be working around the problem rather than fixing it. Has this been fixed in wxwidgets 3.1 if anyone knows? Simon On Thu, May 5, 2016 at 8:38 PM, Collin Anderson wrote: > Another little OS X fix. Most, though not all,

[Kicad-developers] [PATCH 4/4] Remove unused FindOpenSSL.cmake

--- CMakeModules/FindOpenSSL.cmake | 342 - 1 file changed, 342 deletions(-) delete mode 100644 CMakeModules/FindOpenSSL.cmake diff --git a/CMakeModules/FindOpenSSL.cmake b/CMakeModules/FindOpenSSL.cmake deleted file mode 100644 index de91787..000 ---

[Kicad-developers] [PATCH 3/4] Replace avhttp with libcurl: Some fixes: 1. Fixed an assumption somebody originally made in the plugin that std::string had contiguous storage. This is not specced behav

From: unknown --- Documentation/development/compiling.md | 12 +--- common/basicframe.cpp | 3 +++ common/kicad_curl/kicad_curl.cpp | 26 ++ common/kicad_curl/kicad_curl_easy.cpp | 26 +-

[Kicad-developers] [PATCH 2/4] Add missing dependency github_plugin -> pcbcommon

The GitHub plugin needs the generated PCB parser code. --- pcbnew/CMakeLists.txt | 2 ++ 1 file changed, 2 insertions(+) diff --git a/pcbnew/CMakeLists.txt b/pcbnew/CMakeLists.txt index 1249069..6fa6ede 100644 --- a/pcbnew/CMakeLists.txt +++ b/pcbnew/CMakeLists.txt @@ -536,6 +536,8 @@

[Kicad-developers] [PATCH 1/4] Replace the avhttp library used by the github plugin with libcurl.

From: Mark Roszko --- CMakeLists.txt| 5 + common/CMakeLists.txt | 6 +- common/kicad_curl/kicad_curl.cpp | 54 +++ common/kicad_curl/kicad_curl_easy.cpp | 163 +

[Kicad-developers] [PATCH 0/4] Backport libcurl code from development branch

Hi, this backports Mark's avhttp -> libcurl transition to the stable series. With these, I've been able to build a stable version of KiCad with the GitHub plugin enabled that does not link against OpenSSL, which should allow everyone to ship binaries with the Github plugin enabled. Simon

[Kicad-developers] PATCH: OS X copy/close bug fix

Another little OS X fix. Most, though not all, of the dialogs in KiCad that have cancel buttons break copy of text on OS X. If you highlight text (for example, the net of a pad, an operation I find myself doing fairly often) and hit 'Command-C', the dialog is closed and the text is not

Re: [Kicad-developers] Debian and Ubuntu packaging.

Hi Wayne, On 02.05.2016 21:09, Wayne Stambaugh wrote: > Do we have any Debian developers on our mailing list? I though I saw a > Simon Richter on Planet Debian. Is that you Simon? Yes. > Can someone tell me > why the Debian package devs decided to build KiCad with the GitHub > plugin

Re: [Kicad-developers] Debian and Ubuntu packaging.

On 02.05.2016 21:44, Chris Pavlina wrote: > Perhaps we could provide our own Debian and Ubuntu packages, like the Ubuntu > nightly PPA. That still wouldn't be legal, though -- the OpenSSL licence is incompatible with the GPL, and the result is undistributable, so any time the KiCad project ships

Re: [Kicad-developers] [RFC] On net ties, microwave tools & custom pad shapes, altogether.

Hi, On 03.05.2016 14:40, Tomasz Wlostowski wrote: > - net_tie: DRC treats the primitive as non-conducting and doesn't throw > a short circuit error (see drawing A) That requires the net tie to have a size that is at least larger than the minimum clearance of any of the netclasses involved. My