[Scilab-users] New constant %chars to get sets of symbols

Dear Scilabers,

The need of easily accessing to some sets of symbols in Scilab -- like greek letters, etc -- has been reported for years, for instance in 2011 as the wish 9221, or more recently as the wish 16505 (July 2020).

While these reports have still no actual answer, I definitely share this opinion. In many usual situations it is possible and preferable to directly use and write symbols -- in the documentation, in graphics titles, legends or other annotations, etc -- that do not necessarily request LaTeX expressions or HTML entities, and that are then directly readable in the code.

For more than 3 years now, i have set in my startup file and i use on a regular way a constant built as a structure, whose fields are the name of some "classes" of characters.
Then, displaying the constant in the console, and copying/pasting selected characters from the display is straightforward, and always available.
I find it much more handy than if there were somewhere a menu to browse, a menu item to select, a popup showing a table of characters to expect, added to the selection/copy/pasting of the chosen character.

As well, i must confess that having a ALT+X or any other keys shortcut able to convert a series of 2 to 4 input unicodes to the corresponding character -- as proposed in wish #16505 -- would be of almost no help to me, because i do not remember unicodes of any non-ascii characters. Who does? Such a ALT-X shortcut is used for instance in the -- yet great -- Inskscape free drawing software. Then, each time that a greek letter or another symbol must be used, we need to find its unicodes in an extra document (most often on internet). It's definitely not handy.

So, i would like to propose to add this new constant to Scilab, for the benefit of all users.
In this respect, you will find there the proposed help page for such a new constant.
As for now, i have named it %chars:

• Most of existing Scilab constants have a name starting with %. We can keep this convention.
• A short name is desirable, to keep it handy.

As warned in the proposed page, and as experimented during 3 years on my side,
Character sets provided by %chars must not be considered as fixed and definitive. The number of sets, their names, hierarchy, and order can vary in the future, as well as the order of characters in the sets. In addition, a given character may belong to several sets.

Every comment is welcome about such a new constant.

Best regards
Samuel

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

As well, i must confess that having a ALT+X or any other keys shortcut able to convert a series of 2 to 4 input unicodes to the corresponding character -- as proposed in wish #16505 -- would be of almost no help to me, because i do not remember unicodes of any non-ascii characters. Who does? Such a ALT-X shortcut is used for instance in the -- yet great -- Inskscape free drawing software. Then, each time that a greek letter or another symbol must be used, we need to find its unicodes in an extra document (most often on internet). It's definitely not handy.

I agree with you here.
It is not a good idea to use such a shortcut that replicates what is already present at the OS level: on linux for example Crtl+Alt+Maj+u allows to type the unicode of a character (like 3BB for λ). Equivalent shortcuts exist under Windows and MacOS.

From my personal experience, my preferred implementation is the one used by Julia: type the LaTeX macro (like \lambda for λ) then Tab and you get the unicode character.
I assume implementation something like that in Scilab is quite a lot of work...

Your proposition might be a good compromise between ease of implementation and usefulness.

Cheers,

Antoine

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

On 29/12/2020 16:39, Samuel Gougeon wrote:

As well, i must confess that having a ALT+X or any other keys shortcut able to convert a series of 2 to 4 input unicodes to the corresponding character -- as proposed in wish #16505 -- would be of almost no help to me, because i do not remember unicodes of any non-ascii characters. Who does? Such a ALT-X shortcut is used for instance in the -- yet great -- Inskscape free drawing software. Then, each time that a greek letter or another symbol must be used, we need to find its unicodes in an extra document (most often on internet). It's definitely not handy.

I agree with you here.
It is not a good idea to use such a shortcut that replicates what is already present at the OS level: on linux for example Crtl+Alt+Maj+u allows to type the unicode of a character (like 3BB for λ). Equivalent shortcuts exist under Windows and MacOS.

From my personal experience, my preferred implementation is the one used by Julia: type the LaTeX macro (like \lambda for λ) then Tab and you get the unicode character.
I assume implementation something like that in Scilab is quite a lot of work...

Your proposition might be a good compromise between ease of implementation and usefulness.

Thank you for your input and for supporting the proposal, Antoine.

About any shortcut proposal, may i add that, yes it would be a more general solution than the selected characters in %chars. But while getting an external document referencing unicodes of some requested symbols, i don't see any reason to get the unicodes and use the shortcut, instead of directly copying the character from this external document and copying it wherever it is needed in Scilab or elsewhere, as in a documentation page edited with Notepadd++ for instance.
Therefore, the right external document to select is not a document listing unicodes, but more simply a document listing characters sets rendered without gif or other images. Before implementing this %chars, it was what i used to do.

I am not sure that implementing the Julia solution with LaTeX would improve a lot the situation:

• first, we still have to remember a code, the latex one ; this is simple for common characters, but get harder and harder for less frequent ones. While a selection among displayed/rendered characters do not need to remember any code. I have written 700 pages in LaTeX without wysiwym software like LyX, and hopefully i had always a hand on "A Guide to LaTeX-2e"  and its tables of illustrated codes to get the right one.
• Moreover, we can put in %chars some characters that have no LaTeX code.
• Finally, implementing a LaTeX shortcut could not be used when editing the documentation out of the console and Scinotes, except to render the character in order to then, anyway, copy/paste it wherever needed.

So, to me, the main purposes are

• to stop having to remember any code for the -- say 500 or 1000 -- most used characters, when no complex expression is required.
  
• to stop having to search in an external document when working with Scilab
  
• and possibly, to present classes of characters, what can help finding the required one.

Beyong this current topic and the trivial implementation of %chars, it could then even be useful to have an easy way to get the LaTeX code from a selected character, instead of the opposite!

Regards
Samuel

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

Le 31/12/2020 à 10:36, Antoine Monmayrant a écrit :

On 29/12/2020 16:39, Samuel Gougeon wrote:

As well, i must confess that having a ALT+X or any other keys shortcut able to convert a series of 2 to 4 input unicodes to the corresponding character -- as proposed in wish #16505 -- would be of almost no help to me, because i do not remember unicodes of any non-ascii characters. Who does? Such a ALT-X shortcut is used for instance in the -- yet great -- Inskscape free drawing software. Then, each time that a greek letter or another symbol must be used, we need to find its unicodes in an extra document (most often on internet). It's definitely not handy.

I agree with you here.
It is not a good idea to use such a shortcut that replicates what is already present at the OS level: on linux for example Crtl+Alt+Maj+u allows to type the unicode of a character (like 3BB for λ). Equivalent shortcuts exist under Windows and MacOS.

From my personal experience, my preferred implementation is the one used by Julia: type the LaTeX macro (like \lambda for λ) then Tab and you get the unicode character.
I assume implementation something like that in Scilab is quite a lot of work...

Your proposition might be a good compromise between ease of implementation and usefulness.

Thank you for your input and for supporting the proposal, Antoine.

About any shortcut proposal, may i add that, yes it would be a more general solution than the selected characters in %chars. But while getting an external document referencing unicodes of some requested symbols, i don't see any reason to get the unicodes and use the shortcut, instead of directly copying the character from this external document and copying it wherever it is needed in Scilab or elsewhere, as in a documentation page edited with Notepadd++ for instance.
Therefore, the right external document to select is not a document listing unicodes, but more simply a document listing characters sets rendered without gif or other images. Before implementing this %chars, it was what i used to do.

I am not sure that implementing the Julia solution with LaTeX would improve a lot the situation:

• first, we still have to remember a code, the latex one ; this is simple for common characters, but get harder and harder for less frequent ones. While a selection among displayed/rendered characters do not need to remember any code. I have written 700 pages in LaTeX without wysiwym software like LyX, and hopefully i had always a hand on "A Guide to LaTeX-2e"  and its tables of illustrated codes to get the right one.
• Moreover, we can put in %chars some characters that have no LaTeX code.
• Finally, implementing a LaTeX shortcut could not be used when editing the documentation out of the console and Scinotes, except to render the character in order to then, anyway, copy/paste it wherever needed.

So, to me, the main purposes are

• to stop having to remember any code for the -- say 500 or 1000 -- most used characters, when no complex expression is required.
  

Well, I don't see how this should work then.
How do I select λ in your proposed solution? Should I have to visually scan a 500-symbol long list?
I think I missed something in your proposal.

• 
  
• to stop having to search in an external document when working with Scilab
  
• and possibly, to present classes of characters, what can help finding the required one.

Beyong this current topic and the trivial implementation of %chars, it could then even be useful to have an easy way to get the LaTeX code from a selected character, instead of the opposite!

Well, here is my assumption (that might be wrong): most of the people trying to use λ or ∆ might be aware that they are called lambda and Delta and from there, the LaTeX naming convention is usually quite sensible: \lambda, \Delta.

I think there is quite a difference between remembering Ctrl+Maj+Alt+u+03BB and remembering \lambda+Tab to get λ!
For me, the second solution is way more user friendly... :-)

Antoine

Regards
Samuel

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

.../...

I am not sure that implementing the Julia solution with LaTeX would improve a lot the situation:

• first, we still have to remember a code, the latex one ; this is simple for common characters, but get harder and harder for less frequent ones. While a selection among displayed/rendered characters do not need to remember any code. I have written 700 pages in LaTeX without wysiwym software like LyX, and hopefully i had always a hand on "A Guide to LaTeX-2e"  and its tables of illustrated codes to get the right one.
• Moreover, we can put in %chars some characters that have no LaTeX code.
• Finally, implementing a LaTeX shortcut could not be used when editing the documentation out of the console and Scinotes, except to render the character in order to then, anyway, copy/paste it wherever needed.

So, to me, the main purposes are

• to stop having to remember any code for the -- say 500 or 1000 -- most used characters, when no complex expression is required.
  

Well, I don't see how this should work then.
How do I select λ in your proposed solution? Should I have to visually scan a 500-symbol long list?
I think I missed something in your proposal.

We enter and display
--> %chars   // (OK not here. See the proposed documentation for the full display)

or for a chosen class

--> %chars.greek
 ans  =
  lower = "αβδεϵζηθικλμνξοπρστυφϕχψωάϐέήϑίϊϰόϱςύϋΰϖώ"
  upper = "ΑΒΓΔΕΖΗΘϴΙΪΚΛΜΝΞΟΠΡΣΤΥΫΦΧΨΩ KΩ℧"


Then we select λ, and copy/paste it where needed.

• to stop having to search in an external document when working with Scilab
  
• and possibly, to present classes of characters, what can help finding the required one.

Beyong this current topic and the trivial implementation of %chars, it could then even be useful to have an easy way to get the LaTeX code from a selected character, instead of the opposite!

Well, here is my assumption (that might be wrong): most of the people trying to use λ or ∆ might be aware that they are called lambda and Delta and from there, the LaTeX naming convention is usually quite sensible: \lambda, \Delta.

Yes, these are the frequent easy characters to remember.

I think there is quite a difference between remembering Ctrl+Maj+Alt+u+03BB and remembering \lambda+Tab to get λ!
For me, the second solution is way more user friendly... :-)

For sure, but, still, i won't remember \Diamond (why with a capital?), \diamondsuit (all in lowercases), \vdash, etc etc codes without first sawing them rendered.

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users



On 31/12/2020 13:44, Samuel Gougeon wrote:

Le 31/12/2020 à 10:36, Antoine Monmayrant a écrit :

On 29/12/2020 16:39, Samuel Gougeon wrote:

As well, i must confess that having a ALT+X or any other keys shortcut able to convert a series of 2 to 4 input unicodes to the corresponding character -- as proposed in wish #16505 -- would be of almost no help to me, because i do not remember unicodes of any non-ascii characters. Who does? Such a ALT-X shortcut is used for instance in the -- yet great -- Inskscape free drawing software. Then, each time that a greek letter or another symbol must be used, we need to find its unicodes in an extra document (most often on internet). It's definitely not handy.

I agree with you here.
It is not a good idea to use such a shortcut that replicates what is already present at the OS level: on linux for example Crtl+Alt+Maj+u allows to type the unicode of a character (like 3BB for λ). Equivalent shortcuts exist under Windows and MacOS.

From my personal experience, my preferred implementation is the one used by Julia: type the LaTeX macro (like \lambda for λ) then Tab and you get the unicode character.
I assume implementation something like that in Scilab is quite a lot of work...

Your proposition might be a good compromise between ease of implementation and usefulness.

Thank you for your input and for supporting the proposal, Antoine.

About any shortcut proposal, may i add that, yes it would be a more general solution than the selected characters in %chars. But while getting an external document referencing unicodes of some requested symbols, i don't see any reason to get the unicodes and use the shortcut, instead of directly copying the character from this external document and copying it wherever it is needed in Scilab or elsewhere, as in a documentation page edited with Notepadd++ for instance.
Therefore, the right external document to select is not a document listing unicodes, but more simply a document listing characters sets rendered without gif or other images. Before implementing this %chars, it was what i used to do.

I am not sure that implementing the Julia solution with LaTeX would improve a lot the situation:

• first, we still have to remember a code, the latex one ; this is simple for common characters, but get harder and harder for less frequent ones. While a selection among displayed/rendered characters do not need to remember any code. I have written 700 pages in LaTeX without wysiwym software like LyX, and hopefully i had always a hand on "A Guide to LaTeX-2e"  and its tables of illustrated codes to get the right one.
• Moreover, we can put in %chars some characters that have no LaTeX code.
• Finally, implementing a LaTeX shortcut could not be used when editing the documentation out of the console and Scinotes, except to render the character in order to then, anyway, copy/paste it wherever needed.

So, to me, the main purposes are

• to stop having to remember any code for the -- say 500 or 1000 -- most used characters, when no complex expression is required.
  

Well, I don't see how this should work then.
How do I select λ in your proposed solution? Should I have to visually scan a 500-symbol long list?
I think I missed something in your proposal.

• 
  
• to stop having to search in an external document when working with Scilab
  
• and possibly, to present classes of characters, what can help finding the required one.

Beyong this current topic and the trivial implementation of %chars, it could then even be useful to have an easy way to get the LaTeX code from a selected character, instead of the opposite!

Well, here is my assumption (that might be wrong): most of the people trying to use λ or ∆ might be aware that they are called lambda and Delta and from there, the LaTeX naming convention is usually quite sensible: \lambda, \Delta.

I think there is quite a difference between remembering Ctrl+Maj+Alt+u+03BB and remembering \lambda+Tab to get λ!
For me, the second solution is way more user friendly... :-)

Antoine

Regards
Samuel

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

_______________________________________________
users mailing list
[hidden email]
https://antispam.utc.fr/proxy/1/c3RlcGhhbmUubW90dGVsZXRAdXRjLmZy/lists.scilab.org/mailman/listinfo/users

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

Please read

Le 31/12/2020 à 15:43, Samuel Gougeon a écrit :

For sure, but, still, i won't remember \Diamond (why with a capital?), \diamondsuit (all in lowercases), \vdash, etc etc codes without first seeing them rendered.

;-)

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

Le 31/12/2020 à 15:27, Antoine Monmayrant a écrit :

.../...

I am not sure that implementing the Julia solution with LaTeX would improve a lot the situation:

• first, we still have to remember a code, the latex one ; this is simple for common characters, but get harder and harder for less frequent ones. While a selection among displayed/rendered characters do not need to remember any code. I have written 700 pages in LaTeX without wysiwym software like LyX, and hopefully i had always a hand on "A Guide to LaTeX-2e"  and its tables of illustrated codes to get the right one.
• Moreover, we can put in %chars some characters that have no LaTeX code.
• Finally, implementing a LaTeX shortcut could not be used when editing the documentation out of the console and Scinotes, except to render the character in order to then, anyway, copy/paste it wherever needed.

So, to me, the main purposes are

• to stop having to remember any code for the -- say 500 or 1000 -- most used characters, when no complex expression is required.
  

Well, I don't see how this should work then.
How do I select λ in your proposed solution? Should I have to visually scan a 500-symbol long list?
I think I missed something in your proposal.

We enter and display
--> %chars   // (OK not here. See the proposed documentation for the full display)

or for a chosen class

--> %chars.greek
 ans  =
  lower = "αβδεϵζηθικλμνξοπρστυφϕχψωάϐέήϑίϊϰόϱςύϋΰϖώ"
  upper = "ΑΒΓΔΕΖΗΘϴΙΪΚΛΜΝΞΟΠΡΣΤΥΫΦΧΨΩ KΩ℧"

OK, I see better what you propose.
But you are trading remembering a code (ie \lambda for λ) for remembering which class the symbol you are looking for belongs to...
Again, for some of them, it might be obvious (ie \lambda is easy, so is %chars.greek for a Greek symbol) but for some others it's far from obvious.
Like where would you put your \Diamond or \vdash?
I've used my share of LaTeX IDEs and all the symbols assistants failed me in the same way: they give you easy and obvious access to symbols you already know by heart (ie \alpha is in Greek, top first element) but are a useless mess when looking for more obscure symbols (why is \bigstar in Misc-Math, between \blacklozenge and \spadsuit ?)

Then we select λ, and copy/paste it where needed.

• to stop having to search in an external document when working with Scilab
  
• and possibly, to present classes of characters, what can help finding the required one.

Beyong this current topic and the trivial implementation of %chars, it could then even be useful to have an easy way to get the LaTeX code from a selected character, instead of the opposite!

Well, here is my assumption (that might be wrong): most of the people trying to use λ or ∆ might be aware that they are called lambda and Delta and from there, the LaTeX naming convention is usually quite sensible: \lambda, \Delta.

Yes, these are the frequent easy characters to remember.

I think there is quite a difference between remembering Ctrl+Maj+Alt+u+03BB and remembering \lambda+Tab to get λ!
For me, the second solution is way more user friendly... :-)

For sure, but, still, i won't remember \Diamond (why with a capital?), \diamondsuit (all in lowercases), \vdash, etc etc codes without first sawing them rendered.

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

On 31/12/2020 15:43, Samuel Gougeon wrote:

.../...
We enter and display
--> %chars   // (OK not here. See the proposed documentation for the full display)

or for a chosen class

--> %chars.greek
 ans  =
  lower = "αβδεϵζηθικλμνξοπρστυφϕχψωάϐέήϑίϊϰόϱςύϋΰϖώ"
  upper = "ΑΒΓΔΕΖΗΘϴΙΪΚΛΜΝΞΟΠΡΣΤΥΫΦΧΨΩ KΩ℧"

OK, I see better what you propose.
But you are trading remembering a code (ie \lambda for λ) for remembering which class the symbol you are looking for belongs to...

%chars displays all of them, on less than a screen (50 characters per line x 20 lines make 1000 characters ;-). It is illustrated in the provided help page.
And it is hierarchical. So remembering 2 to 10 trivial fields names is enough (instead of 1000 codes), if you wish to display subsets.

Again, for some of them, it might be obvious (ie \lambda is easy, so is %chars.greek for a Greek symbol) but for some others it's far from obvious.
Like where would you put your \Diamond or \vdash?

I may not understand the question. Please see the documentation. Both are already included in my current %chars illustrated in the doc.

I've used my share of LaTeX IDEs and all the symbols assistants failed me in the same way: they give you easy and obvious access to symbols you already know by heart (ie \alpha is in Greek, top first element) but are a useless mess when looking for more obscure symbols (why is \bigstar in Misc-Math, between \blacklozenge and \spadsuit ?)

+1. That's the point. This is why most often i display the whole %chars. Just %chars. Without any codes.

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

Dear,

if the content of e.g. %chars.greek.lower is a string:

Why not even subdivide this further, such as:

%chars.greek.lower.alpha = "α"

%chars.greek.lower.beta = "β"

Because then this would reintroduce some codes (and even worse, paths to codes), and break the overall visibility over sets, which is definitely against the main purposes of the proposal.

Cheers
Samuel

_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

Hello to all

(and best wishes for the new year)

> De : Samuel Gougeon
> Envoyé : mardi 29 décembre 2020 16:40
>
> So, i would like to propose to add this new constant to Scilab, for the benefit of all users.
> In this respect, you will find
> http://bugzilla.scilab.org/attachment.cgi?id=5206

That suits me.
For the time being, I'm using my personal Wikipedia page in exactly the same way https://fr.wikipedia.org/wiki/Utilisateur:Cdang/bo%C3%AEte_%C3%A0_outils#Caract%C3%A8res_sp%C3%A9ciaux
so that's perfect for me.

Best regards

--
Christophe Dang Ngoc Chan
Mechanical calculation engineer

General
This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error), please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden.
_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users

Hello all,

For your information, this proposal has been merged and will be available on the next release. Thanks Samuel for this idea and for the code !

Regards,

Clément

-----Original Message-----
From: users <[hidden email]> On Behalf Of Dang Ngoc Chan, Christophe
Sent: Monday, January 4, 2021 9:30 AM
To: [hidden email]; Users mailing list for Scilab <[hidden email]>
Subject: Re: [Scilab-users] {EXT} New constant %chars to get sets of symbols

Hello to all

(and best wishes for the new year)

> De : Samuel Gougeon
> Envoyé : mardi 29 décembre 2020 16:40
>
> So, i would like to propose to add this new constant to Scilab, for the benefit of all users.
> In this respect, you will find
> http://bugzilla.scilab.org/attachment.cgi?id=5206

That suits me.
For the time being, I'm using my personal Wikipedia page in exactly the same way https://fr.wikipedia.org/wiki/Utilisateur:Cdang/bo%C3%AEte_%C3%A0_outils#Caract%C3%A8res_sp%C3%A9ciaux
so that's perfect for me.

Best regards

--
Christophe Dang Ngoc Chan
Mechanical calculation engineer

General
This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error), please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden.
_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users
_______________________________________________
users mailing list
[hidden email]
http://lists.scilab.org/mailman/listinfo/users