kicad-developers team mailing list archive

[Wayne Stambaugh <stambaughw@xxxxxxxxx>]

Jon,

Would you please see if reverting commit ab74445df4 and loading the the
schematic with the 64 bit time stamps has been fixed by commit 175a2bc0b
when you get a chance?  If commit 175a2bc0b resolves the issue, then
your commit should be reverted.

Thanks,

Wayne

On 3/27/2019 9:44 AM, Jon Evans wrote:
> In my investigation of the issue in that forum thread, I found that
> eeschema was choking on the convert value being zero.
> 
> In that case, I have misunderstood the code and we should probably
> revert ab74445df4
> That commit fixed the loading/netlisting issues in the forum thread but
> perhaps it fixed it the wrong way.
> 
> -Jon
> 
> On Wed, Mar 27, 2019 at 8:35 AM Wayne Stambaugh <stambaughw@xxxxxxxxx
> <mailto:stambaughw@xxxxxxxxx>> wrote:
> 
>     Oops.
> 
>     On 3/27/2019 8:30 AM, Wayne Stambaugh wrote:
>     > Jon,
>     >
>     > I don't think setting m_Convert to 1 is the correct solution
>     unless the
>     > part actually has a convert.  According to the Doxygen comment:
> 
>     /**
>       * Shape identification for alternate body styles.  Set 0 if the item
>       * is common to all body styles.  This is commonly referred to as
>       * DeMorgan style and this is typically how it is used in KiCad.
>       */
> 
>     Unless someone changed this behavior this should still be valid.
> 
>     >
>     >
>     > On 3/23/2019 9:08 PM, Jon Evans wrote:
>     >> For context, I've been looking at the issues in this forum
>     >>
>     thread: https://forum.kicad.info/t/before-giving-up-on-kicad-one-last-attempt-erc-misery/15933/63
> 
>     This post is about the timestamps not the convert value.
> 
>     >>
>     >> I have already pushed a band-aid fix to "repair" these
>     situations, but
>     >> it looks like they are being caused by the EAGLE importer (or at
>     least
>     >> that is one path to this situation coming up)
>     >>
>     >> Specifically, it looks like m_convert is set to 0 in
>     >> SCH_COMPONENT::Init(), but this value is not valid (it should be at
>     >> least 1 for components).
>     >>
>     >> Are there any cases that I am missing, where it makes sense for a
>     >> component to have m_convert set to 0?  If not, it's an easy fix
>     in that
>     >> Init() to set it to 1.  If so, we can just update
>     the SCH_EAGLE_PLUGIN
>     >> to call SetConvert() wherever it creates new components.
>     >>
>     >> -Jon
>     >>
>     >> _______________________________________________
>     >> Mailing list: https://launchpad.net/~kicad-developers
>     >> Post to     : kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>
>     >> Unsubscribe : https://launchpad.net/~kicad-developers
>     >> More help   : https://help.launchpad.net/ListHelp
>     >>
> 
>     _______________________________________________
>     Mailing list: https://launchpad.net/~kicad-developers
>     Post to     : kicad-developers@xxxxxxxxxxxxxxxxxxx
>     <mailto:kicad-developers@xxxxxxxxxxxxxxxxxxx>
>     Unsubscribe : https://launchpad.net/~kicad-developers
>     More help   : https://help.launchpad.net/ListHelp
>