kicad-developers team mailing list archive
Mailing list archive
Re: ok to svn up: first test and differences
> Where are the attached files ... ?
ack, I'm subscribed to this list through one email address, but posted
the original message from another address; when it bounced back, i
copied the message body, but not the attachments. try again!
> I tested the net version and netlist and erc seems work ok.
> But the current eeschema version does not read old annotation.
> Therefore the first annotation changes all components references (they
> are seen as not annotated).
> This will break all old boards made with previous eeschema version.
fixed. Made the default behavior to honor all old annotations and
apply them whenever a path-based alternate reference is not available.
> So i cannot test a new netlist with old existing large boards.
> Also eeschemas crashes after creating a new hierarchical sheet when
> entering this sheet (tested under XP)
> GDB said:
> Program received signal SIGSEGV, Segmentation fault.
> 0x004d51a5 in WinEDA_SchematicFrame::BestZoom (this=0xbd710c0)
> at f:/kicad-sourceforge/trunk/kicad/eeschema/schframe.cpp:356
> 356 dx = GetScreen()->m_CurrentSheetDesc->m_Size.x;
fixed. (that was a particularly dumb bug -- of course people might
want to make a sheet that does not have a schematic file yet!)
> Other (small) problem:
> the reference is not displayed after loading a new component from lib,
> until an annotation is made (or a part selection, for multi parts per
> package components).
> > I would also like to make it so component pins along with a segment
> > are connected to that wire segment. Having to place junction items
> > along a line of bypassing caps is kinda annoying and nonintuitive, but
> > I am afraid of changing the behavior for fear of breaking some
> > schematics. What do you think?
> Do not change the behavior.
> But a possible and welcomed enhancement (which does not break old
> schematic) could be automatically creates junctions after creating a
> wire, put or move a component.
> This is currently the case for new wires starting or ending on an
> existing wire.
> (and better: also remove useless junctions after moving or removing a
> wire or a component).
> This approach does not change the old behavior for netlist, erc and
> others features (dragging an item, deleting a connection for instance)
> and does not involves many changes in the existing code.
ahh, that's too much work. I need to work on other things ;)
It is not that hard to add the junction items, and i guess it makes
the schematic more clear.
> Again, thanks for your work.
well, thanks for yours!
> Jean-Pierre CHARRAS
> Maître de conférences
> Directeur d'études 2ieme année.
> Génie Electrique et Informatique Industrielle 2
> Institut Universitaire de Technologie 1 de Grenoble
> BP 67, 38402 St Martin d'Heres Cedex
> Recherche :
> GIPSA-LIS - INPG
> 46, Avenue Félix Viallet
> 38031 Grenoble cedex
------=_Part_5667_19893341.1203117856092 Content-Type: application/octet-stream; name=twocaps.sch
Content-Disposition: attachment; filename=twocaps.sch
[Attachment content not displayed.] ------=_Part_5667_19893341.1203117856092 Content-Type: application/octet-stream; name=twores.sch
Content-Disposition: attachment; filename=twores.sch
[Attachment content not displayed.] ------=_Part_5667_19893341.1203117856092--