xcircuit-dev — XCircuit news and discussion

[Xcircuit-dev] Re: Welcome to the "Xcircuit-dev" mailing list

[Martin A. <new...@te...>]

On Sun, 21 Nov 2004 12:57:46 -0800
xci...@li... wrote:

> Welcome to the Xci...@li... mailing list!
> 
> To post to this list, send your email to:
> 
>   xci...@li...
> 
> General information about the mailing list is at:
> 
>   https://lists.sourceforge.net/lists/listinfo/xcircuit-dev
> 
> If you ever want to unsubscribe or change your options (eg, switch to
> or from digest mode, change your password, etc.), visit your
> subscription page at:
> 
>   https://lists.sourceforge.net/lists/options/xcircuit-dev/news.malle%40telia.com
> 
> 
> You can also make such adjustments via email by sending a message to:
> 
>   Xci...@li...
> 
> with the word `help' in the subject or body (don't include the
> quotes), and you will get back a message with instructions.
> 
> You must know your password to change your options (including changing
> the password, itself) or to unsubscribe.  It is:
> 
>   flashback
> 
> If you forget your password, don't worry, you will receive a monthly
> reminder telling you what all your lists.sourceforge.net mailing list
> passwords are, and how to unsubscribe or change your options.  There
> is also a button on your options page that will email your current
> password to you.
> 
> You may also have your password mailed to you automatically off of the
> Web page noted above.

[Xcircuit-dev] different symbols for and,nand ...

[Holger P. <mai...@we...>]

Hi,

I am currently using Open Office to draw logic circuits etc. (for 
example half-adders. etc.)

This is not a good solution, so I discovered xcircuit. However, the 
logic symbols (such as and, nand, or... ) seem to be "old -style" and 
not conform to the DIN - rules. For example the "and" is a "&" in a box.

Are such symbols available in xcircuit?

thanks in advance,

Holger

[Xcircuit-dev] Re: XCircuit ASG segfault

[R. T. E. <tim...@mu...>]

Dear Xav,

> I'm very interested in the Automatic Schematic Generation feature
> of XCircuit 3.2. I just downloaded and compiled it under Linux
> (Fedora core 2), but it segfaults with all examples of netlists
> I feed it.

Please note that the latest revision 3.2 contains a preliminary bit
of ASG interface code that I hacked in, but it won't produce any
schematic even if it doesn't crash.  The main reason for posting it
was to pass it off to the author of the original SPAR code (the ASG
part) to figure out how to handle the switch from the coordinate
system of the original program to using xcircuit's internal coordinate
system.  For the first cut, I was looking at an example containing
only capacitors (it doesn't crash).  Due to some aspect of handling
xcircuit's finer grid spacing, it produces nothing.

As soon as we get the basic coordinate problems worked out, getting
rid of the segfaults is mostly a matter of tweaking the SPICE
parser so that 1) it recognizes many more types of components in
the SPICE deck, and 2) it cleans up memory after itself so that it
can be run twice in succession without crashing the system.  Progress
in implementing the proposed ASG functionality should be rapid after
that point.

Continue to consider the ASG part of XCircuit as pre-alpha, unstable,
and potentially broken until further notice.  Notice will be posted
on the XCircuit download page.

					Regards,
					Tim

[Xcircuit-dev] XCircuit ASG segfault

[Xavier B. <xav...@fr...>]

Hi,

I'm very interested in the Automatic Schematic Generation feature of
XCircuit 3.2. I just downloaded and compiled it under Linux (Fedora core
2), but it segfaults with all examples of netlists I feed it.
Here is a backtrace of a segfault when trying to parse
"coupling12.spice".

Regards,

	Xav

spiceparser: cell(0) is "TOP"
ERROR, line 0: AddNTermModule: No such object
        RESTR
ERROR, line 0: AddNTermModule: No such object
        RESTR
ERROR, line 0: AddNTermModule: No such object
        INDR
ERROR, line 0: AddNTermModule: No such object
        INDR
Placing internal modules...
Program received signal SIGSEGV, Segmentation fault.
0x080a0e79 in box_placement () at place.c:1497
1497                start_x = strings[i]->this->x_pos + strings[i]->this->x_size +
(gdb) bt
#0  0x080a0e79 in box_placement () at place.c:1497
#1  0x0808aa04 in Route (areastruct=0x8104420, bIsSparmode=0 '\0') at n2a.c:237
#2  0x0805ce8c in importspice () at files.c:1180
#3  0x080860ba in xcgettext (button=0x825ae60, callstruct=0x825b0a0, calldata=0x0) at xcircuit.c:953
#4  0x555884c9 in XtCallCallbacks () from /usr/X11R6/lib/libXt.so.6
#5  0x080b8375 in Select (w=0x825ae60, event=0xffffd3a0) at MenuBtn.c:1149
#6  0x555bc4f7 in _XtMatchAtom () from /usr/X11R6/lib/libXt.so.6
#7  0x555bca2b in _XtMatchAtom () from /usr/X11R6/lib/libXt.so.6
#8  0x555bd023 in _XtTranslateEvent () from /usr/X11R6/lib/libXt.so.6
#9  0x5559558b in XtDispatchEventToWidget () from /usr/X11R6/lib/libXt.so.6
#10 0x555960d4 in _XtOnGrabList () from /usr/X11R6/lib/libXt.so.6
#11 0x5559628f in XtDispatchEvent () from /usr/X11R6/lib/libXt.so.6
#12 0x555967e4 in XtAppMainLoop () from /usr/X11R6/lib/libXt.so.6
#13 0x0808a021 in local_xloop () at xcircuit.c:2981
#14 0x0808a0c8 in main (argc=1, argv=0xffffd4e4) at xcircuit.c:3055

[Xcircuit-dev] make error message

[<ar...@it...>]

When installing, I receive the following output from make:

Makefile:880: warning: overriding commands for target `all-recursive'
Makefile:405: warning: ignoring old commands for target `all-recursive'
Makefile:883: warning: overriding commands for target `install-recursive'
Makefile:405: warning: ignoring old commands for target `install-recursive'
make tcl
make[1]: Entering directory `/home/74/arae/xcircuit-3.2.26'
Makefile:880: warning: overriding commands for target `all-recursive'
Makefile:405: warning: ignoring old commands for target `all-recursive'
Makefile:883: warning: overriding commands for target `install-recursive'
Makefile:405: warning: ignoring old commands for target `install-recursive'
Making tcl library object
for subdir in ; do \
   test "$subdir" = . || (echo "Making tcl in $subdir"; \
   cd $subdir && make ) \
done
/bin/bash: -c: line 1: syntax error near unexpected token `;'
/bin/bash: -c: line 1: `for subdir in ; do  test "$subdir" = . || (echo "Making
tcl in $subdir";  cd $subdir && make )  done'
make[1]: *** [tcl] Error 2
make[1]: Leaving directory `/home/74/arae/xcircuit-3.2.26'
make: *** [all-recursive] Error 2

No unusual error messages from ./configure.

Any suggestions?

[Xcircuit-dev] RE: XCircuit startup error?

[R. T. E. <ti...@op...>]

Dear Romano,

>   Now, running /home/romano/tmp/bin/xcircuit simply opens a TkCon windows
>   with in it:
>
>   loading history file ... 0 events added
>   error in slave eval:
>   missing colon on line 1
>   Main console display active (Tcl8.4.5 / Tk8.4.5)
>   (tmp) 1 % 
>
>   and then it waits for a command. Probably I am a bit dense (it's just
>   coffee time here), but I cannot convince xcircuit to start up as ever... 

No, it's not the caffeine.  Once Tcl hits an error in the startup
sequence, it pretty much stops dead.  I suggest running xcircuit with
"xcircuit -noconsole" (or just "-noc") to bypass the console.  This may
circumvent the error.

The only thing I can think of is some problem with Tcl/Tk 8.4.5
(the most recent version I have is 8.4.3).  If you run
"/home/romano/tmp/lib/xcircuit-3.2/tkcon.tcl" from the Linux
command line, does it give the same error?  How about if you do
"home/romano/tmp/lib/xcircuit-3.2/tkcon.tcl", and when the
console comes up, you type in "package require Tk" followed by
"source /home/romano/tmp/lib/xcircuit-3.2/xcircuit.tcl"?

Somewhere in these steps you should (hopefully) get an error message
that is more meaningful than "missing colon on line 1".

					Regards,
					Tim

[Xcircuit-dev] XCircuit 3.2.24 --- no graphic interface?

[Romano G. <ro...@de...>]

Hi, 

    sorry to bother you, but... I compiled the new version of XCircuit in my
    Linux machine (I use, a lot, 3.1). To test it, I compiled with
    ./configure --prefix=/home/romano/tmp; make; make install and all was ok
    with no errors. 

    Now, running /home/romano/tmp/bin/xcircuits simply opens a TkCon windows
    with in it:

    loading history file ... 0 events added
    error in slave eval:
    missing colon on line 1
    Main console display active (Tcl8.4.5 / Tk8.4.5)
    (tmp) 1 % 

    and then it waits for a command. Probably I am a bit dense (it's just
    coffee time here), but I cannot convince xcircuit to start up as ever... 

    Do you have any advise for me? Thanks! 

                Romano                 


-- 
Romano Giannetti             -  Univ. Pontificia Comillas (Madrid, Spain)
Electronic Engineer - phone +34 915 422 800 ext 2416  fax +34 915 596 569

[Xcircuit-dev] RE: xcircuit white opaque fill?

[R. T. E. <tim...@mu...>]

Dear J)Bürnjakob,

> whenever i try to fill an object white and opaque, so it stands out
> on a grey background, it is drawn black on the screen.

The rectangle usually ends up drawn black while it is selected, due to
the "exclusive or" color function.  However, the rectangle is supposed
to return to white after it is no longer selected.  Selection is
persistant, so you can do multiple operations on the same element
(border, fill, color, etc.), and will even persist through a file save
or the creation of other elements, as long as you do not deselect with
the 3rd mouse button.  My suspicion is that this is what you are
experiencing.  Otherwise, I have never encountered the situation you
describe (although there were problems with early TCL-based versions
of xcircuit).  What system and, in particular, what video card do you
have?  Are you using the latest version of XCircuit, which is
3.2.23?

Note that there are several ways to get a white opaque fill;  you can
color the object white with 100% fill, or you can set the object to
be any color, set the "opaque" flag, and use 0% fill.  The difference
is the color of the border.

					Regards,
					Tim

[Xcircuit-dev] white opaque fill drawn black

[juedug <jue...@we...>]

hi there,
i only just started using xcircuit and so far it's been a pleasant program =
to=20
work with, but i've come across a fairly annoying bug (misconfiguration on =
my=20
side?) concerning screen output. whenever i try to fill an object white and=
=20
opaque, so it stands out on a grey background, it is drawn black on the=20
screen, yet in the actual eps file the fill is correct. any hints as to whe=
re=20
the problem lies? or is it in fact a bug?
thanks in advance
j=FCrnjakob

[Xcircuit-dev] Re: xcircuit make error on mac ox x (10.3.5)

[R. T. E. <tim...@mu...>]

Dear Dale,

> during the first make (run as "sudo make"), the link (?) step
> failed as follows:
>
> /usr/bin/libtool: unknown option character `b' in: -bind_at_load
> 
> From what I gather, -bind_at_load is a valid ld argument, but not
> for libtool.  Any ideas?

   The question is whether the compile works if you get rid of
"-bind_at_load" in the compile line.  Try deleting it from the
Makefile, and attempt a compile and install.

   I poked around on the web with Google for the error message
libtool spat back at you, and could only find one reference to
it on FinkBot.  Seems like a pretty rare error.

   Possibly some of the OS-X gurus listening in on xcircuit-dev
will have an idea?

					Regards,
					Tim

[Xcircuit-dev] make error on mac os x (10.3.5)

[Dale G. <dg...@re...>]

Since the most recent Fink version is 3.1.32, after playing with that 
I decided it would be good to get the most recent version (3.2.23) 
installed.  I'm running Mac OS X Panther (10.3.5), with a number of 
tools installed with Fink.  gcc is version 3.3.

./configure runs fine.  I didn't make any configuration changes.

However, during the first make (run as "sudo make"), the link (?) 
step failed as follows:

gcc -g -O2 -fPIC -o lib/tcl/xcircuit.dylib -dynamiclib -bind_at_load xcwrap.o \
         elements.o events.o filelist.o files.o fontfile.o formats.o 
functions.o help.o keybindings.o libraries.o menucalls.o netlist.o 
ngspice.o parameter.o python.o rcfile.o render.o schema.o selection.o 
text.o undo.o tclxcircuit.o tkSimple.o tkPixmap.o xcircuit.o 
-L/usr/X11R6/lib \
           -lc -lXt  -lSM -lICE -lm  -lX11 -lXpm
/usr/bin/libtool: unknown option character `b' in: -bind_at_load
Usage: /usr/bin/libtool -static [-] file [...] [-filelist 
listfile[,dirname]] [-arch_only arch] [-sacLT]
Usage: /usr/bin/libtool -dynamic [-] file [...] [-filelist 
listfile[,dirname]] [-arch_only arch] [-o output] [-install_name 
name] [-compatibility_version #] [-current_version #] [-seg1addr 0x#] 
[-segs_read_only_addr 0x#] [-segs_read_write_addr 0x#] 
[-seg_addr_table <filename>] [-seg_addr_table_filename 
<file_system_path>] [-all_load] [-noall_load]
make[2]: *** [lib/tcl/xcircuit.dylib] Error 1
make[1]: *** [tcl] Error 2
make: *** [all-recursive] Error 2

 From what I gather, -bind_at_load is a valid ld argument, but not for libtool.
Any ideas?

[Xcircuit-dev] Problems writing spice from latest xcircuit versions

[Graham P. <gra...@vl...>]

Hallo Tim,
I find that I cannot write spice netlists from the latest
versions of xcircuit: neither 3.1.38 nor 3.2.22. Version
3.1.32 is fine. You can test this with any of the schematics
on my web download at
http://www.vlsitechnology.org/library_5.0.tar.gz
in directory xcircuit/cells/vxlib
or use the PS file I have pasted below.
Can you fix this, or is something horribly wrong with my
schematics so that I am eternally stuck at 3.1.32!??
Best regards, Graham Petley
File an3_x2.ps follows.
%!PS-Adobe-3.0 EPSF-3.0
%%Title: an3_x2
%%Creator: Xcircuit v3.1
%%CreationDate: Sat May 22 14:43:29 2004
%%Pages: 1
%%BoundingBox: 68 68 484 424
%%DocumentNeededResources: font Helvetica-Bold font
Times-Roman 
%%EndComments
%%BeginProlog
%
%  PostScript prolog for output from xcircuit
%  Version: 2.4
%
%  Electrical circuit (and otherwise general) drawing program
%
%  Written by Tim Edwards 8/5/93--11/11/03 
(ti...@ba...)
%  The Johns Hopkins University
%
%%BeginResource: procset XCIRCproc 2.4 4
%
% supporting definitions --- these are the primary xcircuit
types.

/XCIRCsave save def
/topmat matrix currentmatrix def

/fontslant { /slant exch def [1 0 slant 1 0 0] 
    exch findfont exch makefont dup length dict /ndict exch
def
    { 1 index /FID ne { ndict 3 1 roll put } { pop pop }
ifelse } forall
    ndict definefont pop} def
/ul { dup type /stringtype eq showflag 1 eq and { gsave 
   currentpoint topmat setmatrix 0 0 moveto 2 index
stringwidth pop (_)
   false charpath flattenpath pathbbox grestore exch pop 1
index
   sub setlinewidth exch pop currentpoint 3 -1 roll add
moveto 0
   rlineto stroke moveto } if } def
/ol { dup type /stringtype eq showflag 1 eq and { gsave gsave
   currentpoint topmat setmatrix 2 index stringwidth pop 3
index
   true charpath flattenpath pathbbox grestore exch pop
   exch pop topmat setmatrix (_) true charpath pathbbox
grestore
   exch pop 1 index sub setlinewidth exch pop currentpoint
   exch 4 1 roll exch sub add moveto pop 0 rlineto stroke
   moveto } if } def
/stW { gsave currentpoint newpath moveto true charpath
flattenpath
	pathbbox pop exch pop sub grestore } def
/Ts {mark Tabs aload pop counttomark 1 add array astore /Tabs
exch def Tabs
	0 currentpoint pop put} def
/Tbn {mark Tabs aload pop counttomark dup 2 add 1 roll
cleartomark 1 sub} def
/Tb { 0 1 Tbn {Tabs exch get dup currentpoint pop lt
	{currentpoint exch pop moveto exit} {pop} ifelse } for } def
/Tf { Tbn -1 0 {Tabs exch get dup currentpoint pop gt
	{currentpoint exch pop moveto exit} {pop} ifelse } for } def
/qS { (aa) stW (a a) stW sub 4 div 0 Kn } def
/hS { qS qS } def
/pspc 0 def
/cf0 { scalefont setfont } bind def
/Kn { dup kY add /kY exch def rmoveto } bind def
/ss { /fscale fscale 0.67 mul def currentfont 0.67 cf0 0
fscale0 fscale mul
	0.33 mul neg Kn} def
/Ss { /fscale fscale 0.67 mul def currentfont 0.67 cf0 0
fscale0 fscale mul
	0.67 mul Kn } def
/ns { 0 kY neg Kn /kY 0 def /fscale 1.0 def xfont0 1.0 cf0 }
def
/CR { ns 0 /Bline Bline fscale0 neg add def Bline moveto }
def
/cf { dup type /realtype ne {1.0} if exch findfont exch kY 0
eq
	{ 40 mul dup /fscale0 exch def cf0 /xfont0 currentfont def}
	{fscale0 mul fscale mul cf0} ifelse } def
/ctmk { counttomark dup 2 add -1 roll pop } bind def
/label { gsave translate 0 0 moveto dup scale neg /rotval
exch def
	/just exch def just 384 and 0 gt {/mshow {pop} def} {/mshow
{show}
	def} ifelse just 16 and 0 gt {gsave rotval rotate 0 1
dtransform
	gsave pagemat setmatrix idtransform exch grestore 1 0
dtransform
	gsave pagemat setmatrix idtransform exch grestore dup abs
1e-9 lt
	{pop mul 0 gt} {3 1 roll pop pop 0 lt} ifelse grestore {-1
/rotval
	rotval neg def /just just dup 3 and 1 ne {3 xor} if def} {1}
ifelse
	exch -1e-9 lt {-1 /rotval rotval neg def /just just dup 12
and
	4 ne {12 xor} if def} {1} ifelse scale } if /showflag 0 def
	/fspc pspc def /Bline 0 def /Tabs 0 array def /fscale 1.0
def
	/kY 0 def gsave dup 1 add copy 0 exch 1 0 dtransform exch
atan rotate
	{exch dup type /stringtype eq {true charpath flattenpath}
{dup type
	/arraytype eq {exec} {12 string cvs true charpath
flattenpath} ifelse}
	ifelse} repeat pop pathbbox grestore 3 -1 roll pop 3 1 roll
just
	1 and 0 gt {just 2 and 0 gt {exch pop neg fspc sub} {exch
sub 0.5
	mul neg} ifelse} {pop neg fspc add} ifelse exch Bline exch
just 4
	and 0 gt {just 8 and 0 gt {exch pop neg fspc sub} {add 0.5
mul neg}
	ifelse} {pop neg fspc add} ifelse rotval rotate Kn
currentpoint
	translate /showflag 1 def /Bline 0 def /Tabs 0 array def
/fscale
	1.0 def /kY 0 def {dup type /stringtype eq {mshow} {dup type
	/arraytype eq {exec} {12 string cvs mshow} ifelse} ifelse}
repeat
	grestore } def
/pinlabel { 4 index 32 and 0 ne hlevel 0 eq or { /pspc 10 def
label
	/pspc 0 def } { pop pop pop pop pop {pop} repeat } ifelse }
def
/pinglobal { pinlabel } def
/infolabel { pinlabel } def

/scb { setrgbcolor } bind def  /sce { defColor aload pop scb
} bind def
/cRedef {/defColor currentcolor 3 array astore def} def
/begingate {dup type /dicttype ne {1 dict} if begin	% default
params
	dup type /dicttype ne {1 dict} if begin		% instanced params
	/hlevel hlevel 1 add def /defColor currentcolor sce 3 array
	astore def gsave sce translate 0 0 moveto neg rotate dup abs
scale
	} bind def
/endgate { /hlevel hlevel 1 sub def grestore defColor aload
pop cRedef
	scb end end} bind def

/hlevel 0 def
/tmpa [1 0 0 1 0 0] def
/gar {8 8 true tmpa {<c0 c0 00 00 0c 0c 00 00>} imagemask}
bind
{8 8 true tmpa {<30 70 60 02 03 07 06 20>} imagemask} bind
{8 8 true tmpa {<0c 1e 1e 0c c0 e1 e1 c0>} imagemask} bind
{8 8 true tmpa {<0f 0f 0f 0f f0 f0 f0 f0>} imagemask} bind
{8 8 true tmpa {<3f f3 e1 e1 f3 3f 1e 1e>} imagemask} bind
{8 8 true tmpa {<df cf 8f 9f fd fc f8 f9>} imagemask} bind
{8 8 true tmpa {<ff 3f 3f ff ff f3 f3 ff>} imagemask} bind 7
array astore def
/ppaint { gsave clip tmpa dup setmatrix pathbbox neg exch neg
4 2 roll
  neg 4 -1 roll 2 copy gt {exch} if 8 div ceiling 8 mul 4 2
roll neg 2 copy
  gt {exch} if 8 div ceiling 8 mul 3 -1 roll -8 5 -1 roll
  { 3 index exch 5 exch put dup -8 3 index { 3 index
  exch 4 exch put 3 index exec } for } for pop pop pop pop
grestore } bind def
/setstyles {
  currentlinewidth mul setlinewidth /style exch def
  style 1 and 0 gt not {closepath} if
  style 2 and 0 gt {currentlinewidth 4 mul dup 2 array astore
0 setdash} if
  style 4 and 0 gt {0.5 currentlinewidth 4 mul 2 array astore
0 setdash} if
  gsave style 16 and 0 gt { style 224 and -5 bitshift style
256 and 0 gt {
  7 exch sub 8 div dup 1 exch sub currentrgbcolor 3 array
astore
  {3 copy mul add 4 1 roll pop} forall pop pop setrgbcolor
eofill}
  {dup 7 lt {gar exch get ppaint} {pop eofill} ifelse}
ifelse}
  {style 256 and 0 gt {1 setgray eofill} if} ifelse grestore
style 8 and 0 gt 
  style 512 eq or {newpath} {stroke} ifelse grestore} def    


/polygon { gsave /num exch def moveto num 1 sub {lineto}
repeat setstyles } def
/xcarc   { gsave newpath arc setstyles } def
/elb { matrix currentmatrix 7 -1 roll 7 -1 roll translate 5 1
roll 4 -1 roll
	3 index div 1 scale } def
/ele { 0 4 1 roll 0 4 1 roll } bind def
/ellipse { gsave elb newpath ele arc setmatrix setstyles }
def
/pellip  { elb ele arc setmatrix } def
/nellip  { elb ele arcn setmatrix } def
/spline  { gsave moveto curveto setstyles } def
/polyc   { {lineto} repeat } bind def
/beginpath { gsave moveto } bind def
/endpath { setstyles } bind def
/bop	 { 1 setlinecap 0 setlinejoin 6 setmiterlimit 0 0 0 scb
cRedef } def
/psinsertion {/PSobj save def /showpage {} def bop rotate
translate dup scale} def
/end_insert {PSobj restore} def
/setpagemat {/pagemat matrix currentmatrix def} def
/inchscale  {setpagemat 0.375 mul dup scale} def
/cmscale    {setpagemat 0.35433071 mul dup scale} def

%%EndResource
%%EndProlog

% XCircuit output starts here.

%%BeginDocumentSetup

/an3_x2 {
% an3_x2 is_schematic
begingate
1 1.00 0 -48 -64 -48 -64 48 0 48 4 polygon
1 1.00 0 0 48 -90.00 90.00 xcarc
1.000 0.000 0.000 scb
(z) {/Helvetica-Bold cf} 2 4 0 1.00 96 0 pinlabel
(a) {/Helvetica-Bold cf} 2 7 0 1.00 -96 32 pinlabel
(c) {/Helvetica-Bold cf} 2 7 0 1.00 -96 -32 pinlabel
(b) {/Helvetica-Bold cf} 2 7 0 1.00 -96 0 pinlabel
sce
1 1.00 -96 32 -64 32 2 polygon
1 1.00 -96 0 -64 0 2 polygon
1 1.00 -96 -32 -64 -32 2 polygon
(an3_x2) {/Helvetica 0.400 cf} {/Helvetica cf} 3 20 0 1.00
-56 0 label
1 1.00 48 0 96 0 2 polygon
endgate
} def

/vddconn {
begingate
1 1.00 0 -16 0 16 2 polygon
1 1.00 -32 16 32 16 2 polygon
0.933 0.604 0.000 scb
(vdd) {/Times-Roman cf} 2 13 0 1.00 0 -16 pinglobal
endgate
} def

/vssconn {
begingate
0.933 0.604 0.000 scb
(vss) {/Times-Roman cf} 2 1 0 1.00 0 32 pinglobal
sce
1 1.00 0 32 0 0 2 polygon
1 1.00 -16 0 16 0 0 -32 -16 0 4 polygon
endgate
} def

/nmos {
<</width (16)  /length (2)  /model (n)  /idx (?)  /p_jst 21
/p_rot 0 >> begingate
1.000 0.000 0.000 scb
(g) {/Times-Roman cf} 2 7 0 1.00 -160 0 pinlabel
(s) {/Times-Roman cf} 2 13 0 1.00 0 -32 pinlabel
(d) {/Times-Roman cf} 2 9 0 1.00 0 32 pinlabel
0.180 0.545 0.341 scb
mark (u*2+14u') width ( ps=') (u*2+14u') width ( pd=')
(u*5u+12p') width 
( as=') (u*5u+12p') width (u ad=') length (u l=) width ( w=)
model 
( %pd %pg %ps vss ) idx (spice:m) {/Times-Roman cf} ctmk 4 0
1.00 -244 -139 
infolabel
mark ( %x %y) width ( ) length (sim:n %pg %pd %ps )
{/Times-Roman cf} ctmk 4 0 
1.00 -244 -187 infolabel
sce
mark width {/Times-Roman cf} ctmk p_jst p_rot 1.00 -28 0
label
1 1.00 0 32 -64 32 -64 -32 0 -32 4 polygon
1 1.00 -96 32 -96 -32 2 polygon
1 1.00 -96 0 -160 0 2 polygon
endgate
} def

/pmos {
<</width (16)  /length (2)  /model (p)  /idx (?)  /p_jst 21
/p_rot 0 >> begingate
1.000 0.000 0.000 scb
(s) {/Times-Roman cf} 2 9 0 1.00 0 32 pinlabel
(d) {/Times-Roman cf} 2 13 0 1.00 0 -32 pinlabel
(g) {/Times-Roman cf} 2 7 0 1.00 -160 0 pinlabel
0.180 0.545 0.341 scb
mark (u*2+14u') width ( ps=') (u*2+14u') width ( pd=')
(u*5u+12p') width 
( as=') (u*5u+12p') width (u ad=') length (u l=) width ( w=)
model 
( %pd %pg %ps vdd ) idx (spice:m) {/Times-Roman cf} ctmk 4 0
1.00 -196 -139 
infolabel
mark ( %x %y) width ( ) length (sim:p %pg %pd %ps )
{/Times-Roman cf} ctmk 4 0 
1.00 -196 -187 infolabel
sce
mark width {/Times-Roman cf} ctmk p_jst p_rot 1.00 -28 0
label
1 1.00 -96 32 -96 -32 2 polygon
1 1.00 0 32 -64 32 -64 -32 0 -32 4 polygon
1 1.00 -160 0 -128 0 2 polygon
1 1.00 -112 0 16 0.00 360.00 xcarc
endgate
} def

/point {
begingate
248 1.00 0 0 6 0.00 360.00 xcarc
1.000 0.000 0.000 scb
(x) {/Times-Roman cf} 2 29 0 1.00 0 0 pinlabel
endgate
} def


%%EndDocumentSetup

%%Page: 1 1
%%PageOrientation: Portrait
%%PageBoundingBox: 0 0 552 492
/pgsave save def bop
% an3_x2 is_symbol
1.0000 inchscale
2.6000 setlinewidth 800 704 translate

1.00 0 -192 368 vddconn
1.00 0 0 -448 vssconn
1.00 0 0 -352 <</width (24) >> nmos
1.00 0 -192 96 <</width (24) >> pmos
1.00 0 0 368 vddconn
1.00 0 192 368 vddconn
1 1.00 0 -416 0 -384 2 polygon
1 1.00 0 -320 0 -256 2 polygon
1 1.00 0 -192 0 -128 2 polygon
1 1.00 0 224 0 352 2 polygon
1 1.00 -192 320 -192 352 2 polygon
1 1.00 -192 320 -192 128 2 polygon
1 1.00 0 160 0 -64 2 polygon
1 1.00 192 352 192 320 2 polygon
1 1.00 192 256 192 0 -192 0 -192 64 4 polygon
1 1.00 -352 96 -352 -96 -160 -96 3 polygon
1 1.00 -160 192 -416 192 -416 -224 -160 -224 4 polygon
1 1.00 -160 -352 -480 -352 -480 288 32 288 4 polygon
1.00 0 0 0 point
1.00 0 192 0 point
1.00 0 -416 0 point
1.00 0 -480 64 point
1.00 0 -352 -64 point
1 1.00 -544 64 -480 64 2 polygon
1 1.00 -544 0 -416 0 2 polygon
1 1.00 -544 -64 -352 -64 2 polygon
1.000 0.000 0.000 scb
(a) {/Helvetica-Bold cf} 2 23 0 1.00 -544 64 pinlabel
(b) {/Helvetica-Bold cf} 2 23 0 1.00 -544 0 pinlabel
(c) {/Helvetica-Bold cf} 2 23 0 1.00 -544 -64 pinlabel
(zn) {/Helvetica-Bold cf} 2 20 0 1.00 192 -32 pinlabel
(n2) {/Helvetica-Bold cf} 2 20 0 1.00 0 -160 pinlabel
(n1) {/Helvetica-Bold cf} 2 20 0 1.00 0 -288 pinlabel
sce
1.00 0 0 -224 <</width (24) >> nmos
1.00 0 0 -96 <</width (24) >> nmos
1.00 0 0 192 <</width (24) >> pmos
1.00 0 192 288 <</width (24) >> pmos
1.00 0 352 96 <</width (38) >> pmos
1.00 0 352 -96 <</width (19) >> nmos
1 1.00 192 0 192 -96 2 polygon
1 1.00 352 64 352 -64 2 polygon
1.00 0 352 368 vddconn
1 1.00 352 352 352 128 2 polygon
1.00 0 352 -448 vssconn
1 1.00 352 -128 352 -416 2 polygon
1 1.00 352 0 416 0 2 polygon
1.00 0 192 96 point
1.00 0 352 0 point
1.000 0.000 0.000 scb
(z) {/Helvetica-Bold cf} 2 20 0 1.00 416 0 pinlabel
0.678 1.000 0.184 scb
512 1.00 -608 416 -608 -512 480 -512 480 416 4 polygon
pgsave restore showpage

%%Trailer
XCIRCsave restore
%%EOF

[Xcircuit-dev] XCIRCUIT.* domain names

[Christopher D. R. <cd...@en...>]

Hi folks,

A while back I was searching for a name for a product and jumped the gun and
registered XCIRCUIT.COM/NET/ORG before realizing that a naming conflict
existed with this open source project. The registrations are about to expire
and I'm not planning to renew them. Someone here may be interested in
snatching them up when they become available again. Maybe not too. Just
thought I would let you know.

- Chris

[Xcircuit-dev] Re: more on parameter selection in XCircuit

[Paul S. <su...@te...>]

I tried it out and it works like a charm.  :)

I was trying to draw a circuit today but was not successful.
Besides the frequent crashes (I have a "way" with software) there are a couple 
of things that XCircuit really needs.

1. A large "compliant" library of parts
2. Being able to move components around without having to rewire everything.

In the end I just gave up because it's such a pain having to make all the PCB 
compliant libraries before you I draw anything.
Maybe when I have a bit more determination I'll take another stab at it.

It seems that most people who use PCB use geda and not XCircuit for 
schematics. Am I using the wrong tool for the job?

Paul


On Tuesday, 22 June 2004 18:19, R. Timothy Edwards wrote:
> Dear Paul,
>
>    A few minutes ago, I updated the distribution of XCircuit on the
> website.  This version extends the parameter selection method and
> contains an example of a relatively sophisticated use of parameter
> selection in an updated "quadparts.lps" library.  This version allows
> several parameters to "track" a single variable.  This will probably
> not be especially useful in your case, for the PIC parts, unless
> there are specific modes in which clusters of pins change their
> meaning together.  If so, you might find the extended method
> particularly useful.
>
> 					Regards,
> 					Tim

[Xcircuit-dev] Re: Choice of parameters from list

[Paul S. <su...@te...>]

Hi Timothy

I played around a bit with what you mentioned in the previous e-mail and I 
sort of understand how it works. However I know nothing about TCL and my 
programming skills are not that hot so I had a couple of difficulties.

I created the parameter, inserted it into an label and saved it as a library.
However when I loaded the library I could not set the label using the "tcl set  
expression thing". It seems as if the expression was lost although all the 
syntax was in the library file.

However from what you've added below I think I should check out a new copy 
from CVS, compile it and see how things work.
When Soureforge is back up that is ... :)

Thanks
Paul


On Monday, 21 June 2004 19:55, R. Timothy Edwards wrote:
> Dear Paul,
>
>    I have implemented the method of choosing a parameter value from a
> list exactly as I described in the last email.  The command-line command
> "parameter make expression <name> {lindex {list} <index>}" will create
> the parameter to begin with.  If you select the object containing the
> parameter and type Control-P to get the list of parameters, you will see
> the list as before, and the value of the parameter will be the single
> item in the list at position <index> (i.e., the result of the expression
> as executed in the Tcl interpreter).  However, if you then click on the
> value, instead of the usual dialog, you will instead get a "listbox"
> containing all the items in the list.  Then you can click on the item
> to set the parameter, as you intended the method to work.
>
>    There is one problem I haven't solved yet, and that is that if you
> edit a string containing the parameter inside the object (for instance,
> the pin that you want to create using the list of parameters), and
> try to edit the parameter (which, being the result of a Tcl expression,
> cannot be changed), XCircuit will crash.  Hopefully I will fix that by
> sometime this afternoon.
>
> 						Regards,
> 						Tim

[Xcircuit-dev] Re: Choice of parameters from list

[R. T. E. <ti...@op...>]

Dear Paul,

   I have implemented the method of choosing a parameter value from a
list exactly as I described in the last email.  The command-line command
"parameter make expression <name> {lindex {list} <index>}" will create
the parameter to begin with.  If you select the object containing the
parameter and type Control-P to get the list of parameters, you will see
the list as before, and the value of the parameter will be the single
item in the list at position <index> (i.e., the result of the expression
as executed in the Tcl interpreter).  However, if you then click on the
value, instead of the usual dialog, you will instead get a "listbox"
containing all the items in the list.  Then you can click on the item
to set the parameter, as you intended the method to work.

   There is one problem I haven't solved yet, and that is that if you
edit a string containing the parameter inside the object (for instance,
the pin that you want to create using the list of parameters), and
try to edit the parameter (which, being the result of a Tcl expression,
cannot be changed), XCircuit will crash.  Hopefully I will fix that by
sometime this afternoon.

						Regards,
						Tim

[Xcircuit-dev] RE: PIC processors in XCircuit

[R. T. E. <ti...@op...>]

Dear Paul,

> Question 1 :
> What should I do with a pin label like : RA2/MCLR/CLKOUT/TX ?

The trick here is to be sure you're using the Tcl-based version of
Xcircuit, and use the "tcl expression" parameter type that I added
fairly recently to take care of this sort of thing (among its many
uses).  Not that this particular feature is immediately available
without some tweaking.  However, you can get an idea of how it
works by doing, on the XCircuit/Tcl command line:

   parameter make expression ptest {lindex {RA2 MCLR CLKOUT TX} 0}

This creates a new parameter named "ptest" in the current object
whose value is a Tcl expression "{lindex {list} 0}".  If you use
this parameter in, say, a text string (alt-p to insert the
parameter into a line of text), it will produce the first item
in the list.  Doing:

   parameter set ptest {lindex {RA2 MCLR CLKOUT TX} 1}
   parameter set ptest {lindex {RA2 MCLR CLKOUT TX} 2}
   parameter set ptest {lindex {RA2 MCLR CLKOUT TX} 3}

gets you the other items in the list.  In other words, the
parameter maintains a list of all available names for the pin,
but only shows one at a time.  This list is captured in the
output file (i.e., library).

To make this useful, of course, the above "parameter set" lines
need to be transparent to the end-user.  As-is, when the popup
window showing all the parameters is displayed, with "ptest" in
the "Key" column and "RA2" in the "Value" column, and you click
on "RA2", you get a dialog saying "Enter new value for parameter
ptest:".   Apart from the fact that this dialog doesn't work
(hmm, need to do something about that), it's not what you want.
The dialog (which is a Tcl script) should detect this form of
expression and instead of the dialog, pop up a listbox with the
available choices.  Unless you're an expert Tcl script writer
and want to tackle the problem, I'll take care of that one, and
you can go ahead and make your library with the above method
(unless I think of an even more clever way to do it).

> Question 2 :
> Would anyone appreciate a PIC library?
> Does anyone need one?

Several people have contributed PIC symbols in the past, so the
need is there, and they will be used.  Nobody has ever sent me a
library of PIC parts before, just individual parts that they
happened to be using.   I think that developing this method for
multiple-choice selection parameters will be extremely useful,
too.

					Regards,
					Tim

[Xcircuit-dev] Need some suggestions regarding IC library

[Paul S. <su...@te...>]

I'm making some PIC microcontroller parts for a XCircuit library and need some 
advice.

Microcontrollers have the tendency to have multiple functions per pin.
In my case some of the pins have up to four functions each.
Therefore the pin labels I have are too long to fit into a fair width IC and I 
don't want to make a 10 inch wide IC to fit the text in.

Question 1 :
What should I do with a pin label like : RA2/MCLR/CLKOUT/TX ?

At the moment I'm just throwing away all the text after the RA2 but what if 
someone wants to use the pin for one of the other functions?
It would be nice to have some sort of scrolling window on the pin descriptions 
so that the user can pick a description instead of having to retype all the 
pin descriptions just because I had to throw them away.
Any suggestions?

Question 2 :
Would anyone appreciate a PIC library?
Does anyone need one?

Thanks
Paul

[Xcircuit-dev] Re: Saving multiple pages in XCircuit

[R. T. E. <ti...@op...>]

Dear John,

> I create a multiple page file using the procedure in tutorial,
> saving each page. I exit XCIRCUIT.
> 
> I restart XCIRCUIT and reload the same file. I am able to view
> my previosuly created pages. I now create a new page and save
> it it with same original filename(the schematics checkbox in
> the WRITE PS dialog only shows 1 schematic present AFTER
> changing the filename- it doesn't seems to see the other
> pages). I exit XCIRCUIT.
> 
> When I restart and reload the file, I find that it only contains
> the mostly recently saved page and the previously created pages
> are lost.

   The error you mention was present in several XCircuit revisions
after making a change to the Tcl interface. . . the "imultiple"
variable was incorporated into the "XCOps" hash table variable,
and I missed a couple of (critical) instances.  This was fixed
some time ago (see the CHANGES file in the distribution top-level
directory).  Either your version of xcircuit is a bit out of date
(try 3.2.21), or the .tcl files were not overwritten in the install
directory (less likely).  I know that I made this change about the
same time I started implementing the bus notation, so for a while
the download page implied that those revisions that corrected the
problem were not to be trusted for netlist output.  However, since
revision 19, everything appears to be stable and working correctly.

					Regards,
					Tim

[Xcircuit-dev] Saving multiple page files

[John B. <joh...@nm...>]

Hi,

I create a multiple page file using the procedure in tutorial, saving
each page. I exit XCIRCUIT.

I restart XCIRCUIT and reload the same file. I am able to view
my previosuly created pages. I now create a new
page and save it it with same original filename(the schematics checkbox
in the WRITE PS dialog only shows 1 schematic present AFTER changing
the filename- it doesn't seems to see the other pages). I exit XCIRCUIT.

When I restart and reload the file, I find that it only contains
the mostly recently saved page and the previously created pages
are lost.

The only way to avoid this is to save each page in file one by one
before saving the new page and exiting XCIRCUIT. This can be tedious
when using XCIRCUIT with a number of different and large files.

Any advice/help appreciated
John D Barry


ps
when I click on the checkbox I get following error log

can't read "imultiple": no such variable
can't read "imultiple": no such variable
    while executing
"if {$imultiple == 1} {.output.textent.txt1  delete 0 end; 
.output.textent.but1 configure -text Apply; xcircuit::page  filename {}; 
focus .output.texte..."
    invoked from within
".output.title.imulti invoke"
    ("uplevel" body line 1)
    invoked from within
"uplevel #0 [list $w $cmd]"
    (procedure "tkCheckRadioInvoke" line 3)
    invoked from within
"tkCheckRadioInvoke .output.title.imulti"
    (command bound to event)

-- 
____________________________

John D Barry
NMRC
University College
Lee Maltings, Prospect Row
Cork, Ireland
tel: +353-21-4904240
fax: +353-21-4904058
joh...@nm...

[Xcircuit-dev] New contact info for me!

[R. T. E. <ti...@st...>]

Hello everybody,

   Tomorrow will be my last day at the Johns Hopkins University
Applied Physics Laboratory, where I've spent the last five years.
I was given a very generous offer by MultiGiG, Inc., which as
you may or may not know has been funding much of the development
work on magic versions 7.2 and 7.3;  for the past two years, I
have been working part-time for MultiGiG from my home in Maryland
on Thursdays and Fridays.  Starting next Monday, I will be
working full-time for MultiGiG, and commuting out to California
about once a month.  I will continue to work on EDA tool
development (magic, xcircuit, and the rest), and may be able to
accelerate much of that development in the coming year.

   From tomorrow, I will no longer be at the email address
ti...@st...;  you may continue to use the address
ti...@ba... that I post all over my website.
Meanwhile, the websites will continue to be at
http://bach.ece.jhu.edu/~tim/programs/magic and
http://xcircuit.ece.jhu.edu/; any changes will be posted
to this mailing list.

   Cheers!

				---Tim

[Xcircuit-dev] RE: XCircuit segmentation fault

[R. T. E. <ti...@st...>]

Dear Taiyo,

> Version 3.2.21 caused a segmentation fault when I clicked the "OK" button
> in the colour selection dialogue (as instructed in the tutorial)
> 
> Sorry if this is the ten millionth email you've recieved regarding this,
> but it's a habit mine to report bugs...

No, as an open source developer, I depend on the bug reports.  And I haven't
heard this one before, and it seems odd because it is apparently unrelated
to any code changes I've done recently.  However, I recently tracked down
a bug from a different report which could be related.  The error was that
routine "xc_alloccolor" (xcircuit.c, lines 2044 to 2056) supposedly returns
a value, but doesn't.  The strange thing about it is that it has worked;
apparently, the last value computed (pixval) ends up on the accumulator
and is picked up by the calling routine.  Still, it breaks on HP/UX and
may now be breaking on newer versions of gcc, and it needs to be fixed.
So I'll update the distribution, but for reference, the fix is to add

	return pixval;

at line 2056 of xcircuit.c (last line of routine xc_alloccolor).  If this
does *not* fix the problem, please let me know right away.

						Regards,
						Tim

[Xcircuit-dev] compile time error

[Sys A. <lin...@co...>]

Hello - i am having an issue with compiling on fedora core 2. sorry if 
this was already answered on the mailing list, i browsed through and 
couldn't find anything matching. (no search button for the archives?) "# 
./configure" completes okay, but when i run "# make", it fails with the 
following error:

---- cut ----
Making all in Xw
make[1]: Entering directory `/opt/software/xcircuit-3.2.21/Xw'
make[2]: Entering directory `/opt/software/xcircuit-3.2.21/Xw'
make[2]: Nothing to be done for `all-am'.
make[2]: Leaving directory `/opt/software/xcircuit-3.2.21/Xw'
make[1]: Leaving directory `/opt/software/xcircuit-3.2.21/Xw'
make[1]: Entering directory `/opt/software/xcircuit-3.2.21'
gcc -DPACKAGE_NAME=\"\" -DPACKAGE_TARNAME=\"\" -DPACKAGE_VERSION=\"\" 
-DPACKAGE_STRING=\"\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE=\"xcircuit\" 
-DVERSION=\"3.2\" -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 
-DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 
-DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -DPROTOTYPES=1 
-DSTDC_HEADERS=1 -DHAVE_STRING_H=1 -DHAVE_LIBM=1 -DSIZEOF_VOID_P=4 
-DSIZEOF_UNSIGNED_INT=4 -DSIZEOF_UNSIGNED_LONG=4 -DHAVE_LIBXT=1 
-DSTDC_HEADERS=1 -DHAVE_SETENV=1 -DHAVE_PUTENV=1 -DHAVE_DIRENT_H=1 
-DHAVE_VA_COPY=1 -DHAVE___VA_COPY=1 -DHAVE_U_CHAR=1 -DDOUBLEBUFFER=1 
-DHAVE_X11_XPM_H=1 -DHAVE_XPM=1 -DLGF=1 -DINPUT_FOCUS=1 -DGS_EXEC=\"gs\" 
  -DPROG_VERSION=3.2 -DCAD_DIR=\"/usr/local\" -DTEMP_DIR=\"/tmp\" 
-DRESOURCES_DIR=\"/usr/local/lib/xcircuit-3.2/app-defaults\" 
-DBUILTINS_DIR=\"/usr/local/lib/xcircuit-3.2\" -DBUILTINS_FILE=\"\" 
-DUSER_RC_FILE=\".xcircuitrc\" 
-DPROLOGUE_DIR=\"/usr/local/lib/xcircuit-3.2\" 
-DPROLOGUE_FILE=\"xcircps2.pro\" -DSTARTUP_FILE=\"startup.script\" 
-DLGF_LIB=\"lgf.lps\" -DASG_SPICE_LIB=\"asg_spice.lps\" -I. -I. -IXw 
-I/usr/X11R6/include  -g -O2 -c `test -f 'netlist.c' || echo './'`netlist.c
netlist.c: In function `new_tmp_pin':
netlist.c:1527: warning: passing arg 1 of `fprintf' from incompatible 
pointer type
netlist.c:1527: error: too few arguments to function `fprintf'
make[1]: *** [netlist.o] Error 1
make[1]: Leaving directory `/opt/software/xcircuit-3.2.21'
make: *** [all-recursive] Error 1
---- cut ----

[Xcircuit-dev] segfaults in version 3.2.21

[Taiyo R. <t...@iy...>]

Version 3.2.21 caused a segmentation fault when I clicked the "OK" button in 
the colour selection dialogue (as instructed in the tutorial)

I am using Mandrake Linux 9.1, and compiled from source. 

Sorry if this is the ten millionth email you've recieved regarding this, but 
it's a habit mine to report bugs...

Cheers!

-Taiyo

[Xcircuit-dev] Re: xcircuit feature request

[R. T. E. <ti...@st...>]

Dear Steve,

> 1. elastic wires, so you can drag boxes and have the wires remain
> connected

The XCircuit website has scattered comments about the current
development and eventual arrival of "ASG", or Automatic Schematic
Generation.  It is probably not obvious without knowing the details,
but if you can reliably generate a schematic automatically from a
netlist, then you can also rip up and regenerate parts of it on
demand.  This is what we're working on with the ASG implementation,
which also, as the name implies, should be able to generate a
schematic directly from a netlist such as a SPICE deck.

> 2. the selection drag as in autocad where if you drag left it selects
> only items entirely in the box and if you drag right it selects items
> that are at least partially in the box

Interesting method, and probably not difficult to implement.  I would,
however, like to clean up the basic selection mechanism first.  In
particular, some of the support routines for the "undo" function I
implemented recently should allow me to get rid of the somewhat
clumsy serial accept/reject mechanism for click-to-select (as opposed
to area box selections).  I have also had people suggest the
implementation of the "strokes" library.  But it's rather time
consuming to add support for everyone's favorite interface.  But of
course, I have to be told what the interfaces are before I can
decide if they're worth implementing or not, so thanks for the
description.

					Regards,
					Tim