[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

gEDA-user: IEC417 + U? errors



John,

The IEC417 got me into a lot of trouble -- several times.  Blech.

"U?" caused a good deal of pain.

Maybe the IEC417 library should be scrapped if it is the cause of these
problems?

Or wait, should it be properly attributed and kept?  

Is there a replacement anyone is working on?

Phil



John Doty <jpd@xxxxxxxxxxxxx> wrote:

> Which symbol are you using for ground? DO NOT use ground.sym from the  
> IEC417 library: it doesn't have proper attributes.
> 
> In fact, *never* use anything from the IEC417 library in a circuit.  
> It's only good for abridged diagrams in documents. The "pins"  
> generally represent ports, not node connections. Dangerous to newbies  
> to have it in the standard distribution...
> 
> On Jan 7, 2006, at 11:42 AM, Jose, Marshall wrote:
> 
> > I've grepped "U?" with no luck. The netlist reports node "U?-?" is  
> > connect to the ground node, to which a large number of other pins  
> > are connected.
> >
> > I'm pretty sure it's related to a special object like a ground or  
> > io symbol, but I'm kinda fuzzy on how those should be created and  
> > treated.
> >
> > There is the following code in s_net.c, which is one of the source  
> > files for gnetlist; I looked here for a clue, but I'm unfamiliar  
> > with the method gnetlist uses:
> > ====================================================================== 
> > ========
> >     /* apply the hierarchy name to the uref */
> >     uref = s_hierarchy_create_uref(pr_current, temp_uref,  
> > hierarchy_tag);
> >
> >     if (uref && pinnum) {
> >         string = (char *) malloc(sizeof(char) *
> >                                  strlen(uref) + strlen(pinnum) +
> >                                  strlen("  ") + 1);
> >
> >         sprintf(string, "%s %s", uref, pinnum);
> >     } else {
> >         if (pinnum) {
> >             string = (char *) malloc(sizeof(char) *
> >                                      strlen(pinnum) +
> >                                      strlen("POWER") + strlen("  ")  
> > + 1);
> >             sprintf(string, "POWER %s", pinnum);
> >         } else {
> >             string = (char *) malloc(sizeof(char) *
> >                                      strlen("U?") + strlen("?") +
> >                                      strlen("  ") + 1 +  
> > hierarchy_tag_len);
> >
> >             if (hierarchy_tag) {
> >                 misc =
> >                     s_hierarchy_create_uref(pr_current, "U?",
> >                                             hierarchy_tag);
> >                 sprintf(string, "%s ?", misc);
> >                 free(misc);
> >             } else {
> >                 sprintf(string, "U? ?");
> >             }
> >
> >             fprintf(stderr, "Missing Attributes (refdes and pin  
> > number)\n");
> >         }
> >     }
> > ====================================================================== 
> > ========
> >
> > Marshall
> >
> > -----Original Message-----
> > From: owner-geda-user@xxxxxxxx on behalf of John Doty
> > Sent: Sat 1/7/2006 1:33 PM
> > To: geda-user@xxxxxxxx
> > Subject: Re: gEDA-user: gnetlist finding "U?"
> >
> > Try looking for U? with Attributes->Find Specific Text in gschem. Or
> > look for it in the file with a text editor.
> >
> > The Missing Attributes message suggests you may find it useful to run
> > gsymcheck -vv over your symbol files.
> >
> > On Jan 7, 2006, at 11:20 AM, Jose, Marshall wrote:
> >
> >> I'm getting the complaint "Missing Attributes (refdes and pin
> >> number)" when I make a BOM with gnetlist, which is annoying enough.
> >> But I'm also finding the node "U?-?" in my netlist when I run
> >> gsch2pcb on the design. Curiously, "U?" is not found in the BOM list.
> >>
> >> I'm looking for suggestions as to its cause at this point because
> >> I've run out of clues. Any takers?
> >>
> >> Marshall
> >>
> >
> > John Doty              Noqsi Aerospace, Ltd.
> > jpd@xxxxxxxxxxxxx
> >
> >
> >
> > <winmail.dat>
> 
> John Doty              Noqsi Aerospace, Ltd.
> jpd@xxxxxxxxxxxxx
> 
> 
>