Hi Carlos, On Thursday 14 April 2005 17:26, Carlos Nieves Ónega wrote:
[snip] > > There are a few things. > > In your test file the nets and the device attributes are written > > across the symbol. They are hard to read. > > Fixed. > > > If you convert the AT90S8535_TQFP.src file, the bottom pins become > > unreachable. You have to translate the symbol away from the origin > > first. > > Something strange is happening: > - First I create the symbol > - Then I open it with gschem using: > gschem AT90S8535_TQFP.sym , > I can't see the bottom pins. If you look at the coords, the symbol > is correct. The coords at the bottom at the screen are just at > y=1400, so it's not displaying all the symbol. > - I open the SAME symbol in two steps: > - First running gschem > - And then File->Open and selecting the symbol > Everything is ok. > So I think this is a bug in gschem.... Sorry, I can not identify wheter this description was made before or after the last change. The first patch wrote the bottom pins to negativ y-Positions. gschem does not zoom/pan into negative coordinates (feature ;-)). The second patch is ok. everything works fine. > > The position of the graphical name and the refdes is just a matter > > of tast. It's ok for me even if I usually put the refdes to the > > upper right and the name to the upper left. > > Fixed. I'm curious: what do you do if there are pins on the top? The > name of the symbol will overlap with the pins... Well, if you have a really large symbol, you can place the name inside the symbol. > > > > P.S.: One of the tragesym options is "pinwidthvertikal". I think > > > it's mispelled: in english it would be "pinwidthvertical" (with a > > > "C" in "vertical")... > > > > Very funny, "vertikal" is the german translation of vertical. > > I included some code so it uses vertikal if it was changed (defined), > and vertical otherwise. > > On the other hand, I made tragesym a little more clever: If there are > pins on the top (or on the bottom) and the width specified is 0, then > it calculates the width based on the greater number of pins on the > top or on the bottom. > It could be extended, so it calculates automatically the symbol's > width based on the length of pin names and number of pins on the top > and on the bottom... just matter of figure out how to know the width > of the pin names. Don't make it too clever. Some graphical things are easier to do with a mouse than a text description. Great. Everything works fine. Regards Werner
