Mit dem Diagramm-Editor fällt mir direkt keine Lösung ein.
Aber es hindert dich niemand daran, ein Element mit gekreuzten Linien zu zeichnen...

Vielleicht hilft dir das weiter:
In der ersten Zeile der QET-Datei steht die Version, mit der die Datei gespeichert wurde.
Zum Beispiel version="0.80" für QET 0.8, etc.
Öffne die Datei mit einem Text-Editor, dann kannst Du es lesen.

Maybe this solves your problem:
https://github.com/qelectrotech/qelectr … e_24v.elmt

You know that these devices are discontinued for three years now?
Even service has been stopped last year (according to Schneider-Website)!

Hello Nathan,

what's wrong with the available element?
The name says, it's the 115/230V-Version of XPS-BCE.
If you need the 24V-Version of the element with separated terminals S13/S22 and S12/S23: Feel free to create that element and post it here as an attachment. We will publish it in official collection, when available!

Best regards
  plc-user

405

(9 replies, posted in Elements)

QET_ElementScaler now checks polygon values for validity and does not transfer that polygon to the output element (a message is displayed at stderr). Additionally the definition-line of the output-element is correct in that case so that the element can directly be used in QET.
Available here: https://github.com/plc-user/QET_ElementScaler

406

(9 replies, posted in Elements)

handy817257 wrote:

Dewe43_full.elmt file attached.

Your element contains three polygons with invalid values for the respective points.

<polygon x1="nan" x2="nan" antialias="false" y2="nan" y5="nan" x5="nan" x4="nan" style="line-style:normal;line-weight:thin;filling:none;color:black" x6="nan" y6="nan" x3="nan" y1="nan" y3="nan" y4="nan"/>
 
<polygon x1="nan" x2="nan" antialias="false" y2="nan" y5="nan" x5="nan" x4="nan" style="line-style:normal;line-weight:thin;filling:none;color:black" x6="nan" y6="nan" x3="nan" y1="nan" y3="nan" y4="nan"/>
 
<polygon x4="nan" x10="nan" x29="nan" style="line-style:normal;line-weight:thin;filling:lightgray;color:black" y10="nan" x30="nan" x1="nan" x12="nan" y9="nan" x23="nan" y7="nan" x11="nan" x28="nan" x17="nan" y2="nan" x5="nan" y31="nan" x14="nan" x7="nan" y5="nan" y19="nan" y20="nan" y17="nan" x6="nan" x20="nan" x18="nan" y21="nan" y30="nan" y8="nan" y15="nan" y18="nan" y11="nan" y4="nan" y13="nan" y6="nan" y16="nan" y14="nan" x15="nan" y29="nan" antialias="false" x19="nan" x26="nan" y24="nan" y26="nan" x22="nan" x25="nan" y25="nan" y27="nan" y23="nan" x9="nan" x3="nan" x2="nan" x31="nan" y3="nan" x8="nan" x21="nan" x24="nan" y1="nan" x13="nan" x16="nan" y12="nan" x27="nan" y22="nan" y28="nan"/>

I cannot say from here where the polygons or the invalid values come from. However, it is reasonable to assume that the polygons were not created using the QET element editor. But you can only answer this question yourself.
With these invalid polygon points, the element editor also gets confused when calculating the component dimensions for the definition line and then the diagram editor cannot display the element correctly.
So everything can be explained.
But this might be a good example of why the element editor should perhaps carry out a plausibility check for the values read in when opening an elmt file?

[Edit]:
QET_ElementScaler doesn't get confused by these invalid values any more. See below.

407

(9 replies, posted in Elements)

If your component is so large or complex that it cannot be displayed, you should rethink your concept! In my opinion, it makes sense for a circuit diagram to divide large and complex devices into sensibly subdivided subgroups. When I look at some of the elements in the collection, I think to myself: No electrical engineer did this! Even the creator won't be able to cope with it in six months' time!

That's why I definitely go down the route of strictly separating the circuit diagram and the layout diagrams!

I'm curious: Which component is it?

408

(6 replies, posted in Code)

Hello Christoph,

christoph wrote:

-- Could NOT find ECM (missing: ECM_DIR)
--
* ECM (required version >= 5.77.0), Extra CMake Modules., <HERE WAS A URL>

on Debian-Systems there is a Package

apt install extra-cmake-modules

Maybe there is something similar on your apple...

409

(13 replies, posted in Elements)

Hello jhonathan disner,

Please make sure that the elements you provide are an improvement on the existing elements and that they have meaningful names.
Some elements have the name "Name of the element" and some have some kind of directory name...

Best regards
  plc-user

Hallo Fabian,

die Einstellmöglichkeiten hast Du alle durchgetestet?

Salut stephane,

(at the moment) there is no separate help page for the elements, hence the confusing title. However, the actual help text should be the same as for the conductors.

Best regards
  plc-user

Vielleicht helfen dir meine Einstellungen bei der Fehlersuche?
Ich fand das jedenfalls alles ziemlich hakelig, bis das lief...

413

(2 replies, posted in EN : Help, suggestions, discussions, ...)

Use element-editor to create elements.
With dxf-importer you can import dxf into element-editor.
No other importers are implemented!

Hallo Stephan,

ich habe noch diese beiden installiert:

apt install libkf5coreaddons-dev
apt install libkf5widgetsaddons-dev

What I haven't really questioned yet:
Is there a reason why the effort with SingleApplication is made to allow only one instance of QET at a time?

Of course, as a user you then have to make sure that the component library is only changed in one and the same instance in order to avoid inconsistencies in the sqlite file. But QET only touches it for writing when components change, doesn't it?

That seems to be a "problem" with your packing-application:
Some applications show the directories in a straight list but when you right-click on the 7z-file you should have the option to unpack the 7z following the directory-structure.
In the attachment you see "my" unpacker: looks and works perfect!

Can you explain what you mean?
Maybe with a screenshot of your file-manager?

For me the folder-structure looks right  ...  I just use the latest ReadyToUse-Version!  nomicons/smile

418

(18 replies, posted in Code)

Ooops ... No, no, no ... don't get me wrong!
It wasn't meant that way!!! 

I have a good job and earn my money for living and "a bit" more!  nomicons/smile

Please spend the donated money for QET-Project!

419

(18 replies, posted in Code)

Health is the most important thing, besides family and friends!
(a little money might help a bit, too...nomicons/wink)

420

(18 replies, posted in Code)

@Laurent: +2

scorpio810 wrote:

(...) yes schemes of cars was very different than industry...

You are completely right, Laurent!
Electrical "installations" in cars are very different!  nomicons/cwy

422

(18 replies, posted in Code)

o.k.  –  Merci, Laurent !

In this case, the learning curve is not quite so steep for me: I "only" have to dive deeper into QT and QET programming!
I've hardly ever had anything to do with sqlite as a programmer.

As I said: For now I just want to understand how QET works "under the hood"... and maybe contribute some things.

"Still learning!" (even at my age nomicons/wink)

423

(18 replies, posted in Code)

Let me ask more specific:

As a user: If I make a change in the circuit diagram, where is this change saved internally? In the XML structure?
It is clear that this will (at some point) be transferred to a sqlite database in memory.

Or is it so that the sqlite DB in memory is used during editing the diagram and the XML structure is only adapted for saving?

To ask technically: (sqlite XOR XML)?   nomicons/wink

Na ja ... "unnötige Drückerei" ... kommt ja immer darauf an, was Du machen willst!

Wenn Du einmal ein Potenzial und die zugehörigen Leitereigenschaften definiert hast und an das Potenzial einen weiteren Draht anschließen willst, macht es schon Sinn, die Eigenschaften überall gleich zu haben:
Potenzial definiert -> neuen Draht dazu -> Eigenschaften vom neuen Draht passen! nomicons/smile

Wieso sollte sich zwischendurch Leiterfarbe oder Querschnitt ändern, oder das Potenzial einen anderen Namen bekommen?
An ein Potenzial gehört eine Farbe, ein Querschnitt, etc. pp.
Wenn zwischendrin z.B. eine Steckverbindung ist, kann ich mir das noch vorstellen...
Aber vor und hinter einem Schalter oder einer Sicherung ist definitiv nicht dasselbe Potenzial, also müssen die Eigenschaften neu eingestellt werden!

Gib' doch mal deinen Anwendungsfall preis, bei dem unterschiedliche Leitereigenschaften für ein Potenzial Verwendung finden sollen (musst ja keine Geheimnisse verraten!).

425

(18 replies, posted in Code)

Salut Laurent,

scorpio810 wrote:

What could be very constructive would be to add the conductors wire section, color, cable, element label, terminal name connected to each end in the sqlite database in a new table, a bit like nomeclature aka BOM.

I would like to understand your comment better, therefore my questions.

How or in what is the internal structure of a folio stored at runtime of QET?
Is it the XML structure as it is in the resulting qet file, or is an sqlite database used internally?
Is there a place where this is documented?
Please don't say: "The source code is the documentation" nomicons/wink

Of course, I would also like the connection information to become an internal QET generation for a cable plan or wiring instructions for the electrician at some point: I don't really like such things with python plugins! This can be quite a hassle on some systems... 
How this can/should be designed for the user is, in my opinion, of secondary importance for the time being: The internal structures need to be understood and extended.