---

• installation: SMath Studio: Tools > Plugins... > Online Gallery;
  
• documentation: see attachment or Activebook;
  
• UI languages: ;
  
• Main features:
  
  • Selection/Navigation: mouse click, return key, mouse wheel, + and - keys, goto by leading characters (text and math);
    
  • Settings: double-click over the selected item (in the header or in the drop-down list) / context menu item;
    
  • Input: any vector/matrix/system of any size;
    
  • Output: place a variable in the region placeholder;
    
  • Size driven by the values (manual resize not allowed);
    
  • Adds the same controls in SMath Viewer;
  
  

Show Spoiler

---

Control Regions: CheckBoxList | ComboBoxList | ConditionallyFormattedLabel |  NumericUpDown | PathPicker | RadioButtonList | Slider

---

ComboBoxList - examples.sm (18 KiB) downloaded 1048 time(s).

last-minute fix for "goto by leading characters" uploaded (typing characters on collapsed drop-down list now expands the list).

Wrote

It seems that the control width is not adapting to the control content width.

Thank you Ioan, confirmed (even if needs more investigation, because I can't reproduce it every time...). I'll fix it ASAP.

A small workaround to adjust the size and to avoid the freeze of the canvas is to minimize and then restore the Smath window.


Best regards,

Davide

Fixed (collateral effects of API changes, nothing serious)

RC: SMath Studio 0.97.5301
- Code updated to fit recent API changes;
- Fixed sizes calculation logic to allow clicks even when the region becomes wider;
- [DEV] Removed OnKeyPress global hook;
- Password option is no more available, so from now the plugin is marked as Beta.


Best regards,

Davide

As stated in this post, I do not experience the problems reported SMath 0.97.5329 download topic. The attached files all run without problems and I can access all controls without crash.

---

Plugin Slider Region.sm (15 KiB) downloaded 267 time(s).

Thank you Martin

I don't know where is the difference (on my machine the error shown in the previous screenshot was made yesterday on the desktiop distribution and today with your portable distribution - both with the second-last version of the plugin)

Just to have a complete debug, you can see the dynamic preview when you change the input variable from the ComboBoxList region? I have issues both with the last and the second-last plugins.

Thanks in advance,

Davide

Wrote

Thank you Martin

I don't know where is the difference (on my machine the error shown in the previous screenshot was made yesterday on the desktiop distribution and today with your portable distribution - both with the second-last version of the plugin)

Just to have a complete debug, you can see the dynamic preview when you change the input variable from the ComboBoxList region? I have issues both with the last and the second-last plugins.

Thanks in advance,

Davide

The portable version seems to contain the version 0.97.5327 (release candidate), this is going to be corrected . But anyways, So far, I tried 0.97.5327 and 0.97.5329 with and without updated control regions and do not get a crash at all.

I am not sure what you mean by changing the input variable. Should I replace c by something else? The screenshots are with latest SMath and latest plugins.

---

I mean changing the value of the "Source" parameter in the form

---

I defined another matrix d, switched the input to it and don't see any problem.

---

Wrote

I defined another matrix d, switched the input to it and don't see any problem.

This is good, in one hand.

In the other hand, like Ioan, I have noticed issues on evaluations (trigggered twice, not triggered, triggered correctly...) both on the desktop and the portable distributions.

screencasts: evaluation triggered twice , evaluation not triggered

I have Win 7 x64 (home premium) and your same framework (2.0.50727.5477) but the behaviors is not the same (both on the desktop and the portable distributions).

Wrote

* after recent Davide's plugins rebuilt in the context of the SMath 0.97.5329, I have no more execution crashes;
* in the new version, after each event (click or change) on a ComboBoxList, the recalculation of the file is triggered; append is a test file; is this behavior general or machine dependent ?

Observed behaviour for Ioan's file (no difference between old and new plugin version)
- I had no crashes
- I get a re-calculation only and reproducible


This seems reasonable behaviour to me.

Well, maybe a day we will see a targeted recalculation (just on regions related with changed values)

Wrote

The only aspect that I do not agree in your evaluation is issued from the "axiom" that: the calculation of a page must start only:

* automatically, when this option is selected;
* by explicit user command from menu.

Other reasons (control click or control change events) to start of the calculations are not normals and are annoying , especially when working with big files.

Ioan,

I agree, that the control regions should obey the setting of the automatic re-calculation toggle (I normally have this "on" ). And in fact, at least the radiobutton list region does not care about the setting, even though it has a display flag which reflects the current re-calculate setting in the upper right corner.

If auto recalc is on, then I prefer getting immediate action upon changed values, e.g. if the values drive some graphics, as in my slider example. I'd vote for keeping it this way.

The bug is then that the recalc setting is not obeyed (which I can reproduce). Fixing that should help with your problem.

Here there are different things...

1) Recalculations (evaluations) does not obey "Auto calculation" settings

-> this can be fixed by me with a little bit different design of the plugins, or by Andrey, with a different handling of RequestForEvaluation. Actually I have no ideas about what is the best way (except that I think that if I want to trigger an evaluation, this should be possible), but is not a problem for me to make the fix from the plugin side.

2) Exceptions on opening settings' form.

-> Martin does not experience this issue at all, I have this issues since SMath 0.97.5327 (tested with portable version made from linux package), and this is experienced also by Ioan. This was fixed recompiling the plugins, but the different behavior is really strange.
I've made a clean installation of SMath, used portable versions, used the .NET framework fix tool but nothing changes; I'm not sure if the broken verions are those where the errors are not shown or those where the errors are experienced.

3) Multiple recalculations (evaluation triggered twice)

-> Here seems (correct me if I'm wrong) that here are different behaviors on the machines. Martin does not experience this issue at all, Ioan every times with several plugins (see attachments and comments of SS-192), and me just sometimes (not in my test file but in the Ioan file just with this plugin).

4) No recalculation ( using ( ... CreateTransaction() ) not triggered)

-> This can be seen changing the data source of the region; if the region is updated without closing the form it works, otherwise is broken. This is not experienced at all by Martin, I experience this since SMath 0.97.5327 (again tested on portable version)




P.S. the test on the 0.97.5327 was made just today, previously I have not made deep testings on that version

Wrote

Here there are different things...

1) Recalculations (evaluations) does not obey "Auto calculation" settings

-> this can be fixed by me with a little bit different design of the plugins, or by Andrey, with a different handling of RequestForEvaluation. Actually I have no ideas about what is the best way (except that I think that if I want to trigger an evaluation, this should be possible), but is not a problem for me to make the fix from the plugin side.

2) Exceptions on opening settings' form.

-> Martin does not experience this issue at all, I have this issues since SMath 0.97.5327 (tested with portable version made from linux package), and this is experienced also by Ioan. This was fixed recompiling the plugins, but the different behavior is really strange.
I've made a clean installation of SMath, used portable versions, used the .NET framework fix tool but nothing changes; I'm not sure if the broken verions are those where the errors are not shown or those where the errors are experienced.

3) Multiple recalculations (evaluation triggered twice)

-> Here seems (correct me if I'm wrong) that here are different behaviors on the machines. Martin does not experience this issue at all, Ioan every times with several plugins (see attachments and comments of SS-192), and me just sometimes (not in my test file but in the Ioan file just with this plugin).

4) No recalculation ( using ( ... CreateTransaction() ) not triggered)

-> This can be seen changing the data source of the region; if the region is updated without closing the form it works, otherwise is broken. This is not experienced at all by Martin, I experience this since SMath 0.97.5327 (again tested on portable version)




P.S. the test on the 0.97.5327 was made just today, previously I have not made deep testings on that version

Point 1 done by plugin design. No more evaluations if AutoRecalc is disabled.

I can do something also for the 4th point but requires more efforts (and will be just a workaround). Points 3 and behavior differences are in the fog, I don't understand where is the issue.

Thank you! I will take a look at this issue.

Thank you! I will take a look at this issue.

PLUGIN UPDATED SMath Studio 0.97.5346 and SMath Studio 0.97.5737

- Added Undo/Redo support;
- Plugins compiled for both the SS versions (no more crashes in SMath Studio 0.97.5346)

hello guys!
dear Davide Carpi,

as I've been using your very util plugin during
my tesis for the calculation of wooden joints,
I found out about a problem concerning the combined
use along with the Viewer.

It seems to me the once the Viewer loaded the file, all
the settings about the input variables for the comboboxList
are made. In the case, like you see in the example-pictures,
of multiple lists depending on their selection to decide
about the next-ones input, it comes to dificulties concerning
the update process.
The problem still consists with the newest version of SMath.

Is there a way to fix this problem?
I'd be glad aboud any advice of solution!
Thanks a lot!

Greetings,

Eckhard Bräuchler

Confirmed, I'll provide a fix ASAP.