1011661 - [keyboard] requesting uppercase and lowercase for compositekey

Status: RESOLVED WONTFIX

(Firefox OS Graveyard :: Gaia::Keyboard, defect)

Description

[Raniere Silva]

Some times we need to have both lowercase and uppercase version of a compositekey in the virtual keyboard. An not great example is the key '.com'.

Comment 1

[Raniere Silva]

Attachment: patch

Comment 2

[Salvador de la Puente González [:salva]]

Comment on attachment 8424086 [details] [review]
patch

Please, provide a test case for this and check the comment on GitHub.

Comment 3

[Rudy Lu [:rudyl] (inactive after 2015/7/3)]

Comment on attachment 8424086 [details] [review]
patch

As you may already know, the code base in the original keyboard app would need an overhaul to improve its architecture and test coverage, so I am trying not to be too harsh for unit testing.

However, this patch would need some more modifications, according to my comments on the pull request.
Besides, do we have a specific use case for this change?

Thanks.

Comment 4

[Salvador de la Puente González [:salva]]

Yes, it is for the Google Summer of Code Raniere [1] is developing and I think it is a great user case. We are implementing LaTeX tags for mathematical functions and greek letters and, as you probably know, the lower alpha and upper alpha differs on the first letter: \alpha and \Alpha [2]

So here, when tapping the alpha-labeled key we need to send different composite values. Does it make sense for you?

What do you think about adding the new field? I saw you propose to keep the same but distinguishing between the different panels we have for lower and uppercase in [3], am I correct?

I'm pinging Raniere to urge him to address your comment.
Thank you very much, Rudy.

[1] http://www.google-melange.com/gsoc/proposal/public/google/gsoc2014/r_gaia_cs/5629499534213120
[2] http://wikieducator.org/Help:LaTeX_Symbol_Tables_-_Mathematics#Greek_Symbols
[3] https://github.com/mozilla-b2g/gaia/pull/19334#discussion-diff-13318211

Comment 5

[Raniere Silva]

Sorry for the delay.

Salvador, thanks for the words about use case.

Rudy, as Salvador said the use case is for my GSoC project. Information about the prototype is available at [1]. I believe that send small patches is better.

I don't understand the "distinguishing between the different panels".

I also update the patch.

[1] http://blog.rgaiacs.com/2014/06/03/flatfish_and_la_tex.html

Comment 6

[Rudy Lu [:rudyl] (inactive after 2015/7/3)]

Comment on attachment 8424086 [details] [review]
patch

Hi Salva, Raniere,

Yes, I know you're working on GSoC project, just want an example to be present here, so we can know why we need this change.
Thanks for the explanation, it is clear to me now.

Please refer to my comments here, do you think this can work?
https://github.com/mozilla-b2g/gaia/pull/19334/files#discussion_r13369213
Hope this is clear enough.

Comment 7

[Raniere Silva]

Tim, as we talk I rebased my patch. Could you take a look on it?

Comment 8

[Salvador de la Puente González [:salva]]

Cancellin ni for Tim. It's better to ask for feedback or review.

Comment 9

[Tim Guan-tin Chien [:timdream] (please needinfo)]

Comment on attachment 8424086 [details] [review]
patch

I think we can take this patch if

-- The part that wraps alt keys into multi-line are removed, since it is incomplete and should be dealt in another bug (we need to deal with menu area too and I really don't feel the syntax is correct)
-- We have unit tests in place to protect the feature added, since the layout that is using this feature is not checked-in yet.

==

So with this patch, we officially normalize alter key definitions into the form of [ { value: ... }, ... ], something equal to key rows. This is a good start, but I don't think we have figured out the relationship between upper case index and their keys yet. Since key objects is capable of specifying upper case value and lower case value independently, alt['l'] and alt['L'] can ideally referred to the same array and we can decide what to show or output on runtime (maybe we need to introduce |upperLockedCompositeKey| for key objects too?).

Anyway, I am just thinking aloud here. If you have better idea please share.

Comment 10

[BMO Automation]

Firefox OS is not being worked on