delorie.com/archives/browse.cgi   search  
Mail Archives: geda-user/2014/03/05/14:44:17

X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Message-ID: <1394048274.2135.34.camel@AMD64X2.fritz.box>
Subject: Re: [geda-user] gschem attributes only for symbols pins and nets?
From: Stefan Salewski <mail AT ssalewski DOT de>
To: geda-user AT delorie DOT com
Date: Wed, 05 Mar 2014 20:37:54 +0100
In-Reply-To: <20140305115559.5F2ED807ABF4@turkos.aspodata.se>
References: <1393977186 DOT 2845 DOT 14 DOT camel AT AMD64X2 DOT fritz DOT box>
<20140305115559 DOT 5F2ED807ABF4 AT turkos DOT aspodata DOT se>
X-Mailer: Evolution 3.10.4
Mime-Version: 1.0
Reply-To: geda-user AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: geda-user AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com

On Wed, 2014-03-05 at 12:55 +0100, karl AT aspodata DOT se wrote:
> Stefan:
> > http://ssalewski.de/tmp/prop.png
> 
> I like your millimeter paper like grid.

Yes, I also like that better than the background graphics of most other
tools. (But I do offer other backgrounds also, something like . . . . or
+ + + + for example.)  

> Your transistors, do C and E have their pins inside the circle ?
> 

Currently I have no symbols drawn by myself, for testing I am using an
schematics shipped with an older gEDA version. My clipping of line-width
makes the line-width 0 of the gEDA symbols look a little bit
different... 

> About the "sym" thing; nice tought a little confusing at first.
> 

Yes. there is not much room for labels. Basically my whole design is for
a plain full HD display, so we have 1920 pixel vertically, and my goal
was to use only max. 20 % or maybe 320 pixel for the tool palette at the
left. Meaning of the widgets should become clear fast for beginners --
not obviously widgets have a tooltip text. (Of course now ultra HD
displays become affordable, but on these displays users generally have
windows of different programs at the same time visible, so we should not
waste too much areas of the display.)

> If you show x and y, perhaps you should make a cross or something for 
> the symbols attachment point.

Yes, symbols have a x mark where the absolute origin is, and each
attribute has such a mark also. Seems that I have turned that off
currently...

> 
> Orientaion, is that that the same as the symbols angle of rotation ?
> Why do you have a checkmark for orientation, shouldn't 0 (= no rotation 
> suffice) ?

Orientation is angle and mirror, checkmark indicates active mirror. Yes,
there was no space for a describing label, but checkmark has a tooltip
text.

> 
> The checkmarks for the attributes, is that for visibility ?
> 

Yes. Name and Value can be visible or invisible each. gschem uses a
really strange notation for that. 

> I liked your nine text "Origin" circles.

Plain GTK CheckButton widgets, with borders set to zero to by CSS to
allow dense position.

> 
> The "+ - <> 100 - + 300 - +" near the bottom, what's that for ?

"+" for add a new attribute, "-" for delete currently selected
attribute, "<>" is for spreading all attributes: We may insert in the
numeric fields 0 and 300 and click '<>' and all attributes are aligned
with 300 units vertical distance.

All that is not really finished now, changes can occur.

For the attribute coordinates, I still ask myself if I should display
absolute or relative coordinates (relative to symbol origin).

> 
> > Basically I am satisfied, it is compact and not too ugly.
> 
> Hmm...
> 

Changes can still occur. My first goals was to check if I can fit all
important information into a not to large area, and I think that is OK.
Of course we may later introduce some more margin/borders to the
widgets. Maybe fully remove some, i.e. the bulky ComboBox widget for the
color index of the attributes. (I think attributes should have the
default color, and later there may be other tools to change color.) 

> > Now I wonder if gschem supports attached attributes for all elements,
> > even text elements, or only for symbols, pins and nets.
> 
> T 200 200 5 8 1 1 0
> test=true
> {
> T 100 100 5 8 1 1 0
> hello=3
> }
> 
> seems to work all right, move the "test" and the "3" will move along.
> docs/wiki/geda-file_format_spec.html says
> 
>  "Attributes are attached to the previous object."
> 
> and doesn't state any restrictions, but:
> 
> T 200 200 5 8 1 1 0
> test=true
> {
> T 100 100 5 8 1 1 0
> hello=3
> {
> T 300 300 5 8 1 1 0
> deep=yes
> }
> }
> 
> didn't work.
> 
> ...
> > Indeed I can not really imagine what sense attributes for boxes or text
> > elements can make -- but on the other hand when the attributes widget
> > already exist, why not show it always.
> 
> What about (try it):
> 
> T 200 200 5 8 1 1 0
> Value=2  = 8
> {
> T 280 280 5 5 1 1 0
> power=3
> }
> 

Thanks for testing. I think I will provide attributes for all elements,
later I may hide the widgets if no one will need them

> > A related question: For boxes we have mesh and hatch, with angle and
> > size parameters. Has someone ever used that for schematics? May cairo
> > drawing routine supports that from the early days, but I wonder if I
> > should offer all the related parameters for the user interface -- it may
> > confuse beginners, and when no one use it...
> 
> Don't know.

I think I will offer all too (for a box we have many properties) later I
can hide some.
> 

Best regards

Stefan Salewski


- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019