Mail Archives: geda-user/2017/02/14/11:51:02
This message is in MIME format. The first part should be readable text,
while the remaining parts are likely unreadable without MIME-aware tools.
--8323329-234458573-1487090947=:14019
Content-Type: TEXT/PLAIN; format=flowed; charset=UTF-8
Content-Transfer-Encoding: 8BIT
Hi,
as I pointed out in the the other thread[1], I just merged a few new
features into gEDA/gaf master. In order to use them, you need to build
the current development version of gEDA/gaf from the repository.[2]
1. Back annotation
You can now load back-annotation patches from Igor2's pcb-rnd into gschem
via Attributes -> Import patch… or the hotkey "t P" [3]. This works
similar to the text search: at the bottom of the window, a list of netlist
changes is shown, and when you click on a change, the window zooms to the
point in the schematic where the change should be done. After you fixed
something, you can re-import the patch (just click on "Find" again), and
the list will be updated.
2. Parametric subschematics
gnetlist now allows you to pass parameters to subschematics. In order to
do so, add one or more attributes of the form param=NAME=VALUE to the
subschematic symbol. Inside the subschematic, you can use $(NAME) in the
value part of an attribute; this will be replaced with VALUE.
While the current lightweight implementation works for any package
attributes retrieved by netlist backends, it doesn't influence the way the
netlister works, so you can't use parameters right now for example in
slot= or netname= attributes.
3. Power symbols
You can now define power symbols in a simple way: create a symbol with one
pin, some graphical representation, and a netname= attribute indicating
the net to which the pin should be connected. When using the symbol, you
can easily change the net by overriding the netname= attribute.
Power symbols defined in this way must not have a refdes= attribute
(because that indicates it's a regular symbol) or a pin= attribute
(because that would be a conflicting way to define a net) and must not be
graphical (because that would not stop the power symbol from working and
is most probably an error). Since the pin attributes of the single pin
don't have any effect, is it highly recommended to remove them.
4. I/O port symbols
Port symbols for use in a subschematic[4] can now be defined in an
analogous way: create a symbol with one pin, some graphical
representation, and a portname= attribute indicating the I/O port to which
the pin should be connected. Using the portname= attribute instead of the
refdes= attribute allows gnetlist to know which component is actually a
port, so it can warn you if there is no matching pin on the subschematic
symbol (when using the refdes= attribute, this silently generates a broken
netlist).
I have also prepared a patch which implements working buses, but since
there appear to be different concepts about what buses should be and how
they should work, I haven't merged it yet. If you want to have working
buses in gschem, please create a real example schematic which you want to
work once gEDA/gaf supports buses and send it to me.
Roland
[1] http://www.delorie.com/archives/browse.cgi?p=geda-user/2017/02/14/11:10:21
[2] http://wiki.geda-project.org/geda:gaf_building_git_version
[3] http://repo.hu/projects/pcb-rnd/devlog/20150830b_back_ann.html
[4] http://wiki.geda-project.org/geda:hierarchy
--8323329-234458573-1487090947=:14019--
- Raw text -