998 replies to this topic

[Taapat]

I have not used a phone with buttons for a long time. So get a Nokia phone from a colleague today to see how de facto letters are inputted with the number keys in Latvian.

And I see that I am right.

The input works in alphabetical order, as I had created, not according to EU 202 130.

Little google and I found a document that confirms it and proposes to make changes in EU 202 130: http://www.likit.lt/...i_comments.html

 

Therefore, I can repeat that in the Latvian layout the letters needs to be returned in the same way as before.

Special symbols on 1 or 0 you can create as you feel better.

Edited by Taapat, 6 September 2018 - 14:02.

[ims]

@Ian,

 

yes, it is ok, but added chars with "umlaut" (äëïöüÄËÏÖÜ) is unnecessary. I know, that it was in this example, but  this chars are not in CZ alphabet and no one word exist in CZ with umlaut. It was just adding lat1 chars for more universal sms keyboards...

 

Our full alphabet is here (if it will be displayable for you). No umlaut.

 

note: ch is standalone char in alphabet, but in ascii tables not exist (on keyboards not exist too), is writed as c and h

 

Btw - here is freq. characteristic of our chars

[IanSav]

Hi Taapat,

 

That document appears to want to change the order of the glyphs so that all lowercase letters come before the number.  The English Latin characters come first followed by the locale characters then the digit.  That is what I was offered but you objected!

 

The draft is a request to change the standard but I don't think it was accepted.  Still I think the proposed standard is better and what I originally offered.

 

As per your previous agreement I have updated Latvian to the format specified in the standards document and adjusted to the proposed amendment.  This matches your requests from the previous posts.

 

Regards,

Ian.

 

[IanSav]

Hi All,

 

This is the latest test version of NumericalTextInput.py that has the characters for all languages adjusted into lowercase, digit, uppercase order.  The sequence of the lowercase and uppercase characters is as listed in the ETSI ES 202 130 V2.1.2 document.

 

I hope I have corrected Czech as directed by Ims and Dutch as directed by Rob.  I have also added Persian.  I hope that Persian Prince can tell me if my efforts are acceptable.

 

Hopefully these changes are acceptable and the changes can be submitted into the repository.

 

Regards,

Ian.

 

Attached Files

•  NumericalTextInput.py   13.77KB   4 downloads

[IanSav]

Hi All,

 

Please do NOT use the version of NumericalTextInput.py in my last post.  There is a debugging line that locked that version to Dutch.  (That was the last language I tested.)

 

This version will correctly switch to the appropriate locale.

 

Regards,

Ian.

 

Attached Files

•  NumericalTextInput.py   13.73KB   2 downloads

[ims]

Hi Ian,

 

pls, try next version :-)

Allmost all umlauts are under Czech removed - it is ok, but still are there:öÖ under key 9

But in this version you moved numbers to end of lowercases, but it must be after "basic" (english) chars => example:

 

abc2áčABCÁČ

def3ďéěDEFĎÉĚ

ghi4íGHIÍ

...

[IanSav]

Hi Ims,

 

pls, try next version :-)

Allmost all umlauts are under Czech removed - it is ok, but still are there:öÖ under key 9

But in this version you moved numbers to end of lowercases, but it must be after "basic" (english) chars => example:

 

abc2áčABCÁČ

def3ďéěDEFĎÉĚ

ghi4íGHIÍ

...

 

The extra characters have now been properly removed.

 

The lowercase, digit, uppercase is was what Taapat was demanding.  In the document he linked (http://www.likit.lt/...i_comments.html) there is a draft amendment to ETSI ES 202 130 that says that the numbers *should* be moved between the lowercase and uppercase letters.

 

Are you willing to accept this proposed revision of the standard for the Czech layout?

 

Regards,

Ian.

[ims]

I prefer what is used many years in our country ... abc2áčABCÁČ

[IanSav]

Hi Ims,

 

I will revert Czech but I really think you should please consider moving to the newer format.

 

Regards,

Ian.

 

Attached Files

•  NumericalTextInput.py   13.72KB   4 downloads

[ims]

Thx ...

 

 

 

[Dimitrij]

IanSav

Can you make the adaptation of the Russian language in a new virtual keyboard?
I will be very grateful to you.

[Taapat]

@IanSav I want to repeat again. I am talking only about the Latvian layout.
You can see that the document in my link refers to the Lithuanian layout, the same is with the Latvian layout. But I do not want to say that in the same way changes need to be made in all languages, such as Czech.
 
Document indicates that there are differences between what all the time is used de facto and between what is specified in the standard.
In Latvian de facto is used the alphabetical layout "aābcč2AĀBCČ".
According to ETSI, EU 202 130 should be "abc2āčABCĀČ".

[IanSav]

Hi Dimitrij,

 

Russian (locale ru_RU) has already been added and updated in both the VirtualKeyBoard.py and NumericalTextInput.py modules. 

 

If you are having any difficulties please let me know.

 

Regards,

Ian.

[IanSav]

Hi Taapat,

 

@IanSav I want to repeat again. I am talking only about the Latvian layout.
You can see that the document in my link refers to the Lithuanian layout, the same is with the Latvian layout. But I do not want to say that in the same way changes need to be made in all languages, such as Czech.
 
Document indicates that there are differences between what all the time is used de facto and between what is specified in the standard.
In Latvian de facto is used the alphabetical layout "aābcč2AĀBCČ".
According to ETSI, EU 202 130 should be "abc2āčABCĀČ".

 

As per your request and agreement I used ETSI, EU 202 130 with the amended format as per the proposed update document you posted.  I don't know what is wrong now.  The mixed English and local characters is not standard in EITHER document.  The difference is simply where the digit should be placed.

 

Regards,

Ian.

Edited by IanSav, 6 September 2018 - 18:58.

[zeros]

Hey,
I hope you don't change these things at Estonian, or should I worry and check? For now Estonian is OK and I'd like it to remain so. Since that is the case, I assume I don't have to check your new files?
Another thing, it would be nice if you could add to documentation an examples to a developer, how he/she can call the keyboard in his plugins. It's just a thought, because the documentation you've written is good.

[IanSav]

Hi Zeros,

 

I am so turned around trying to make everyone happy that I think you should check that I haven't broken Estonian.

 

The way that NumericalTextInput is called can be very different.  Compare how the NumericalTextInput is used from config.py and how it is called from MovieSelection.py.  The former uses the UI while the latter does not!    There is no one way to call it or use it.  The way it is called or used depends on what you are trying to achieve.

 

If you would like any assistance in using NumericalTextInput in your code please ask and I will try to help.

 

Regards,

Ian.

[Taapat]

@IanSav Yesterday, when I got a point from ims to the documentation for his Czech language, I found that it coming from ETSI ES 202 130. I found this standard and checked that mixed English and local characters were not preserved for the Latvian language.

Then I wrote ok, if there is a standard then follow it.

 

However, today, I'm checking a real phone and see that it does not work according to this standard.

After that, I found a document to which i point before, that there is a proposal to change the standard ETSI EU 202 130, since many languages, including Latvian, do not use the layout as in standard.

Therefore, I wrote today that it is necessary to restore the layout of Latvian as before, in alphabetical order. Because it is used de facto in Latvia.

Sorry for deception yesterday.

 

Of course, we can say if there is a standard, then use it. But it seems to me that this standart for the Latvian language has not been developed by a person who knows the Latvian language, and therefore it does not correspond to what is actually used in Latvia.

[IanSav]

Hi Taapat,

 

Here is the latest version of NumericalTextInput with Latvian re-sequenced in contradiction to the standards but as requested.  If there are any complaints about the Latvian sequence I will revert to the documented standard.

 

Regards,

Ian.

Attached Files

•  NumericalTextInput.py   13.72KB   2 downloads

Edited by IanSav, 7 September 2018 - 06:17.

[Taapat]

Thx.

Here is corrected Latvian layot:

(None, None, None),
(None, None, None),
(u"a\u0101bc\u010D2A\u0100BC\u010C", u"A\u0100BC\u010C2", u"a\u0101bc\u010D2"),
(u"de\u0113f3DE\u0112F", u"DE\u0112F3", u"de\u0113f3"),
(u"g\u0123hi\u012B4G\u0122HI\u012A", u"G\u0122HI\u012A4", u"g\u0123hi\u012B4"),
(u"jk\u0137l\u013C5JK\u0136L\u013B", u"JK\u0136L\u013B5", u"jk\u0137l\u013C5"),
(u"mn\u0146o6MN\u0145O", u"MN\u0145O6", u"mn\u0146o6"),
(u"prs\u01617qPRS\u0160Q", u"PRS\u01607Q", u"prs\u01617q"),
(u"tu\u016Bv8TU\u016AV", u"TU\u016AV8", u"tu\u016Bv8"),
(u"z\u017E9wxyWXYZ\u017D", u"Z\u017D9WXY", u"z\u017E9wxy")

I remove the letters ō and ŗ that are not in the Latvian alphabet (they were not there before).

 

In addition, I moved the English letters q,w,x,y that not in the Latvian alphabet behind the digits as evidenced in the recommendation for the Lithuanian language (it was not there before).

[IanSav]

Hi Taapat,

 

Your changes significantly diverge from ALL the documented standards.  I suspect that this layout is going to annoy some people.  Given that this change is likely to only affect your fellow countrymen, and I have wasted enough time on this one issue, I am going to merge your changes and move on.  If there are ANY complaints or issues then I will revert to the documented standards.

 

Regards,

Ian.