Re: gEDA-user: Parts modified are not available until reloaded gschem
Hi Vinny, Once you have edited and saved a part, go back to the schematic, select the part, then select "Edit" and then "Update Component" or use the shortcut "ep." This should reload that component and update the schematic. If you have multiple instances of the component, highlight them all before doing the "ep" and it will update them all. If you have lots of them, it is probably easier to reload the schematic. Steve On Sat, 05 Apr 2008 15:58:25 -0700, Vincent Onelli <[EMAIL PROTECTED]> wrote: > Hi, > I am fairly new at gschem, I found that if I modify a part then save in > the same name the modification do not show when the part is reloaded > until gschem is reloaded. Same if the modified part is saved under > different name it is not available until gschem is reloaded. > Is there a way to get around? I will appreciate any help. Thank you in > advance > Vinny -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New user - Annoyances
Hi Peter, thanks for your reply. I am sure that a lot of users have learned the way it acts and have gotten used to it. Changing would be a bother. I just want to customize it to my liking. I don't like having to left click and then also right click to end my net. I just want to right click and be done with it. I also think that at that point a second right click should take you out of the add net mode and back to the select mode. Many times I have added a net and then try to go select something and the fool thing still thinks I am trying to add nets. Then I have to hit esc, then undo the wild net I just made, and then finally select what I originally wanted to select. Also, is there a way to highlight a whole net? I would like to highlight a net and have the whole length, wherever there is electrical connectivity, turn a different color so that I can easily trace a net visually. That capability may already be there, I just haven't found it yet. The biggest problem I see is the inconsistency in how the program acts to mouse clicks in the various modes, add net, add line, add component, etc. I think a standard scenario, such as left click repeats an action, right click repeats and ends the function, another left click restarts the function, whereas a second right click ends the mode and returns to select mode, that works consistently across most, if not all functions, should be the goal. I like that scenario best, but as long as it was consistent, one could learn and adapt more easily. In fact maybe it would be possible to have different sequence scenarios built in and be user selectable. That may be asking too much, I don't know how difficult that would be. > With regards the right-click to end behaviour, this is something you'll > have to edit in the source-code. Beware that there are some > configuration options which can set the right button to be a pop-up > menu, rather than being a place / cancel action. > > > If you're working on a Debian / Ubuntu box, I'd suggest grabbing the > Debian package source, adding your modifications and rebuilding. This > allows your package management to work properly. > > How did you install gEDA? (What distribution, and what gEDA version do > you have?) > I am running Kubuntu 6.10 on an Athlon 1700+ with 512M of memory. I installed gEDA with Adept from the Debian repositories. I have also now installed the source from the Debian repositories. I am not a Linux expert by any measure but I have compiled a few things when needed, usually following instructions on the web. > With that information we'll be able to help more about how to rebuild > the distribution package with modifications, or install directly from > source. > Thanks, Steve -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: New user - Annoyances
As a new gEDA used who has completed a couple schematics now, I have a few comments and questions. There are various inconsistencies in what the mouse buttons do between various functions. Some of these are annoying and could be modified to make it all easier to use. For example: When you are adding nets, the left button ends a net segment and you can continue the net with adding another segment with another left click, etc. However, when you are done and on the last segment, you have to left click it and then right click to stop adding to the net. It would be much simpler to just be able to right click on the last segment to complete the segment and stop the net. When you are adding lines, at the end of the first line segment, you left click as you do when adding nets, but that ends the line and you have to left click again to start a new segment. To be consistent, a left click should end the line segment, but moving the mouse should continue adding a new line segment until you right click which should add the last segment and stop the add line function. When adding components, a left click instantiates the component and each subsequent left click adds an other until the last one where you have to left click and then right click to stop adding. Again it would be nice to right click on the last instance, leaving it in place, and have that stop the add components function. I had a similar problem with the copy command stopping after just one copy. Now I have just found the multiple copy command which operates as I would have expected the copy command to work. The only thing I would change is to have it make the last copy and stop copying with a right click. I don't see any reason for having two copy commands at all. Now, my question is, is there the capability for a user to change this basic functionality through the key-mapping or config files? I have looked through the files and have not been able to find what I am looking for. Of course, I may not recognize it if I found it anyway. Or, are these operational functions things that would require modification the the program source files? If so, which files would I need to look into and where can I find info on recompiling the system? I am retired now but I have used several schematic capture systems over my career and I have not really liked most of them. I actually prefer gEDA in many ways over any of them. I just would like to make it even better. Thanks, Steve -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New user - Hierarchy problem
I think grefdes is what I remembered reading about. I have now tried it and in general, it does a nice job. Now I need to learn some Perl to understand it better. It does not seem to understand about schematic hierarchy however. It renames/renumbers even refdes names for the hierarchy inputs and outputs that don't need numbers. I had an input with a refdes of Vin and it changed it to Vin1. gnetlist -g drc2 then complained about the missing refdes because the page symbol was still looking for Vin not Vin1. Also, while this is a handy program, it does not solve my problem since it is operating at the schematic page level. I need something operating at the upper level of gnetlist so that when I am generating a bom or a netlist it will take into account the sub-schematic levels and modify the refdes accordingly. I have looked at the scheme files for generating the bom and pads netlist and so far, I can't figure out what they are doing and how to make them do what I want. I am not at all familiar with Scheme or GUILE so I guessing that's next on my list of things I need to learn. Steve Mon, 25 Feb 2008 17:11:51 -0800, John Griessen <[EMAIL PROTECTED]> wrote: > Kai-Martin Knaak wrote: >> On Mon, 25 Feb 2008 10:05:01 -0600, John Griessen wrote: >> >>> grefdes --pgskip 100 subschem1 subschem2 subschem3 subschem4 >> > he has multiple >> instances of a subsheet under a main sheet. So he needs a script that >> acts on the netlist rather than on the schematics. > > Oh... right.The subsheets are only one, placed four times. > He wants an instance referred to in the to sheet to trigger incremented > names > to replace subsheet names... > > grefdes is a model for a netlist text processing script though. > > JG -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: pads net list problem
Hi Dan, Yes, that seems to have fixed the problem. My layout guy was able to directly read in the netlist to PADS without any difficulty. Thanks, Steve On Mon, 25 Feb 2008 15:43:42 -0800, Dan McMahill <[EMAIL PROTECTED]> wrote: > could you verify that the attached patch works correctly? > > Thanks > -Dan > -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New user - Hierarchy problem
Yes, exactly. I thought I had seen something about someone else having done this before. There was something about changing the refdes from the composite sheet/part style into a flat numbering system. Maybe I'm just dreaming. Anyway, I'll take a look at grefdes tomorrow and see what I can make of it. Steve On Mon, 25 Feb 2008 17:11:51 -0800, John Griessen <[EMAIL PROTECTED]> wrote: > Kai-Martin Knaak wrote: >> On Mon, 25 Feb 2008 10:05:01 -0600, John Griessen wrote: >> >>> grefdes --pgskip 100 subschem1 subschem2 subschem3 subschem4 >> > he has multiple >> instances of a subsheet under a main sheet. So he needs a script that >> acts on the netlist rather than on the schematics. > > Oh... right.The subsheets are only one, placed four times. > He wants an instance referred to in the to sheet to trigger incremented > names > to replace subsheet names... > > grefdes is a model for a netlist text processing script though. > > JG -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: pads net list problem
I believe the answer to both these questions is yes. Steve On Mon, 25 Feb 2008 12:31:49 -0800, Stuart Brorson <[EMAIL PROTECTED]> wrote: >> If Pads expects CR-NL should the netlist generator output CR-NL? > > Interesting question. Is it safe to assume that PADS runs exclusively > on Windoze? > > Stuart > > > ___ > geda-user mailing list > geda-user@moria.seul.org > http://www.seul.org/cgi-bin/mailman/listinfo/geda-user -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: pads net list problem
Yes, I ran the file through unix2dos and that fixed the problem. Thanks to Dan McMahill and others who gave suggestions for this problem. Steve On Mon, 25 Feb 2008 11:27:01 -0800, Stuart Brorson <[EMAIL PROTECTED]> wrote: > You can always try running unix2dos on the file before shipping it to > your layout guy. That would fix any line end problems. > > Stuart > -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: pads net list problem
I have designed a circuit and created a pads netlist to send to the person who is doing my layout. He said that PADS will not accept the netlist file directly. He can get around the problem by first opening the file with Excel and then re-saving it. Then PADS will read it just fine. I will try to get more details about this, but for now, I was wondering if anybody else has run into this problem. Steve -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New user - Hierarchy problem
When creating a hierarchy you end up with refdes numbers with the top level refdes of the circuit symbol, a slash, and then the underlying refdes such as X101/R102. I would like to end up with refdes on the top level of R1, R2, etc. and parts from the underlying schematic be R101, R102, etc. for the first instance and R201, R202, etc. from the second instance, and so on. I have four copies of my underlying circuit instantiated on my top level schematic and all those X101/R102 refdes are too long and take up too much board space. I remember reading somewhere about a program or script to change all the refdes in a hierarchy design, but, now I can't find it. Does something exist to be able to do what I want? Steve -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New user - Hierarchy problem
I had put the "source=" attribute in the symbol for the underlying schematic page. That was enough for the proper hierarchy operations except for the drc and netlist generation. As per John's sample schematics, I added a "source=" attribute on each of the instantiated symbols at the top level and now everything is fine. Evidently the "source=" attribute doesn't get promoted, if that is the right context and terminology, and the drc and netlist operations need it to be at the top level. There probably is a way to ensure that the attribute is promoted, that I'll have to leave to another day. It is now working well enough for me to complete this project. Thanks to all who helped, Steve On Mon, 11 Feb 2008 14:55:51 -0800, John Griessen <[EMAIL PROTECTED]> wrote: > Steven Taylor wrote: >> My symbol, that I created, that represents my lower level schematic has >> pins with only pin numbers, pinseq numbers, and pin labels. The pin >> labels >> match the refdes on the IO connectors on the underlying schematic. > > OK, How about the refdes of the symbol that corresponds to the leaf cell > schematic > placed in the to schematic? > Did you give those different names? Mine have names like S1, S2, S3, > etc. > > Want me to create a schematic set with embedded symbols and send it to > you for another example to follow? > The schematic is for some free-published hardware... no problem to look > at it. > > John Griessen > -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New user - Hierarchy problem
Yes, John, that would be helpful. It would give me a second sample to try here and see where I may be going wrong. Thanks, Steve > > Want me to create a schematic set with embedded symbols and send it to > you for another example to follow? > The schematic is for some free-published hardware... no problem to look > at it. > > John Griessen > -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New user - Hierarchy problem
My symbol, that I created, that represents my lower level schematic has pins with only pin numbers, pinseq numbers, and pin labels. The pin labels match the refdes on the IO connectors on the underlying schematic. That is the way it was done on the gTAG example. I tried assigning pintypes to the IO connectors but that didn't seem to make any difference. Steve On Mon, 11 Feb 2008 13:51:33 -0800, John Griessen <[EMAIL PROTECTED]> wrote: > Steve Taylor wrote: > >> When I run the same command on the top level schematic, I get a long >> list >> of "Duplicated references for every symbol in the underlying schematic, >> I >> get another list for duplicate slot 1 of every symbol, and I get errors >> on >> three of my inputs, listing the components to which they should have >> been >> connected, saying the nets connect to only one pin. > > I run drc2 on a similar schematic and get just minor warnings > NOTE: Found pins without the 'pintype' attribute: U6:3 U6:4 U6:2 U6:5 > U6:6 U6:1 U6:8 U6:7 U5 > > WARNING: Pin(s) with pintype 'input/output': U2:6 > are connected by net 'GND' > to pin(s) with pintype 'power': U2:7 U2:8 > > These I know I can ignore in my case. > > The offpage symbols have attribs device=none pintype=pas > refdes=SENVDD1 pinnumber=1 > > Does your top schematic have symbols with pins that have the same name > as refdes=SENVDD1 with pinlabel=SENVDD1? > > This is going to be a symbol you create. > > John Griessen > -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New user - Hierarchy problem
Continuing my checking, I have found that *all* the nets on the lower level schematics are showing up in the net list file twice, first as separate nets and then again as included in the proper nets with all the pins from the lower levels combined with the top level connections. The reason those particular nets were being flaged is that they were the only nets that had only one pin connection on the lower level schematic. There must be a setting somewhere to stop this behavior. I just have no idea where it would be. Steve On Mon, 11 Feb 2008 13:47:54 -0800, Steven Taylor <[EMAIL PROTECTED]> wrote: > Actually, I have to take that back, the nets do show up correctly > connected for all of the IOs, but the pins on the lower level schematic > also show up in the net list as nets with only a single pin. So those > nets > in question are actually in the net list twice, once as single pin nets > and then again as correctly connected to the other pins that are also in > the net. Interesting. Now, why is this happening? Is it something I have > wrong or is it a bug? I am not doing my own board layout so I can just > edit out the offending nets in the netlist file before I send it to my > layout guy. It would be nice to know why it's happening though. > Steve > > > On Mon, 11 Feb 2008 13:09:40 -0800, Peter Clifton <[EMAIL PROTECTED]> > wrote: > >> >> On Mon, 2008-02-11 at 13:07 -0800, Steve Taylor wrote: >> >>> I am evidently doing something wrong and I need some help in pointing >>> me >>> in the right direction. >> >> It might be that the drc2 backend doesn't support this very well. Does >> the netlist look OK when you netlist for the circuit layout software >> you're using? >> > > > -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New user - Hierarchy problem
Actually, I have to take that back, the nets do show up correctly connected for all of the IOs, but the pins on the lower level schematic also show up in the net list as nets with only a single pin. So those nets in question are actually in the net list twice, once as single pin nets and then again as correctly connected to the other pins that are also in the net. Interesting. Now, why is this happening? Is it something I have wrong or is it a bug? I am not doing my own board layout so I can just edit out the offending nets in the netlist file before I send it to my layout guy. It would be nice to know why it's happening though. Steve On Mon, 11 Feb 2008 13:09:40 -0800, Peter Clifton <[EMAIL PROTECTED]> wrote: > > On Mon, 2008-02-11 at 13:07 -0800, Steve Taylor wrote: > >> I am evidently doing something wrong and I need some help in pointing me >> in the right direction. > > It might be that the drc2 backend doesn't support this very well. Does > the netlist look OK when you netlist for the circuit layout software > you're using? > -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: New user - Hierarchy problem
The net list comes out OK except for two of the nets which don't connect through between the upper level and the lower level schematics. Steve On Mon, 11 Feb 2008 13:09:40 -0800, Peter Clifton <[EMAIL PROTECTED]> wrote: > > On Mon, 2008-02-11 at 13:07 -0800, Steve Taylor wrote: > >> I am evidently doing something wrong and I need some help in pointing me >> in the right direction. > > It might be that the drc2 backend doesn't support this very well. Does > the netlist look OK when you netlist for the circuit layout software > you're using? > -- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user