Separate the reporting of superscripts and subscripts from the report font attributes setting

[codeofdusk]

Link to issue number:

Related to #3727.

Summary of the issue:

The reporting of superscript and subscript is controlled by the "report font attributes" setting (which is far too verbose for reading of mathematical/scientific documents using the <sup>/<sub> HTML tags).

Description of how this pull request fixes the issue:

Adds a new setting, "report superscripts and subscripts", which controls whether superscripted/subscripted text is read.

Testing performed:

With all possible combinations of "report superscripts and subscripts" and "report font attributes", tested that HTML documents in Chrome using the <sup> tag read correctly and that no non-selected formatting information is reported.

Known issues with pull request:

I'm not sure how to make this work in Braille – does NVDA even present superscripts/subscripts to Braille displays?

Change log entry:

== Changes ==

• The reporting of superscripts and subscripts is now controlled separately to the reporting of font attributes. (Separate the reporting of superscripts and subscripts from the report font attributes setting #10919)

[LeonarddeR]

I think text position is ambiguous, as alignment also says something about the position of the text.
May be report script changes or report subscript and superscript?

[codeofdusk]

I think text position is ambiguous, as alignment also says something about the position of the text.
May be report script changes or report subscript and superscript?

Fixed.

[AppVeyorBot]

• PR introduces Flake8 errors
• Build (for testing PR)

See test results for failed build of commit 45f7a89490

[LeonarddeR]

SOrry for me not being explicit enough

I think that it is ok to call the option reportScript, as that's way shorter than reportSubScriptsAndSUperScripts. At first glance, you could leave the label as is, may be @feerrenrut has suggestions as what wording is used to cover both sub and superscript

Could you also add a config spec update that:

• If report attributes is off (default), report script stays off
• If report attributes is on, turn on script as well.

[codeofdusk]

I prefer "report superscripts and subscripts" to "report script".

[feerrenrut]

Thanks @codeofdusk

[josephsl]

Hi,

I'm afraid this introduced a regression: a KeyError exception is raised when NVDA+F is pressed to obtain formatting information.

Thanks.

[josephsl]

Traceback:

ERROR - scriptHandler.executeScript (20:38:42.761) - MainThread (3928):
error executing script: <bound method GlobalCommands.script_reportFormatting of <globalCommands.GlobalCommands object at 0x07620F10>> with gesture 'NVDA+f'
Traceback (most recent call last):
  File "scriptHandler.pyc", line 205, in executeScript
  File "globalCommands.pyc", line 1435, in script_reportFormatting
  File "globalCommands.pyc", line 1401, in _reportFormattingHelper
  File "textInfos\__init__.pyc", line 564, in getFormatFieldSpeech
  File "speech\__init__.pyc", line 2181, in getFormatFieldSpeech
KeyError: 'reportSuperscriptsAndSubscripts'

This is seen when pressing NVDA+F from browse mode documents. Thanks.

[Adriani90]

Is it possible to exclude the reporting of baseline when pressing quick navigation key in browse mode? When I enabe superscripts and subscripts, NVDA always reports "base line" when I press quick navigation keys (i.e. b for buttons in the commenting area). This is quite annoying on some websites. This should be only reported when NVDA detects a mathematical content or if this is part of a text block while navigating with reading commands.

[codeofdusk]

Most likely, open a separate issue and hopefully someone will take it.

[ABuffEr]

Hi,
sorry, @codeofdusk , but I'm confused by this feature. If I enable or disable it, what should it happen reading
A<sup><em>i,j</em></sup>
rendered in a web page? I notice no differences in Chrome or Firefox, excluding what reported by @Adriani90.