Comment 1 by jteh on 2014-10-03 09:57
Changes:
Milestone changed from None to next

Comment 2 by jteh on 2014-10-03 10:45
Branch: t4521
Try build
Crash course:

• Move cursor to math, NVDA+alt+m to interact.
• Hierarchical navigation: right/left arrows to move next/previous, down/up arrows to move in/out.
• In a table, the usual control+alt+arrows to move between cells.

Comment 4 by peter on 2014-10-03 15:18
Hello,
testing on following page:
https://developer.mozilla.org/en-US/docs/Web/MathML/Examples/MathML_Pythagorean_Theorem
with this error:
ERROR - scriptHandler.executeScript (17:15:50):
error executing script: <bound method Gecko_ia2.script_moveByLine_forward of <virtualBuffers.gecko_ia2.Gecko_ia2 object at 0x056FCEB0>> with gesture u'down arrow'
Traceback (most recent call last):
File "scriptHandler.pyc", line 174, in executeScript
File "cursorManager.pyc", line 160, in script_moveByLine_forward
File "cursorManager.pyc", line 83, in caretMovementScriptHelper
File "speech.pyc", line 800, in speakTextInfo
File "speech.pyc", line 603, in speakTextInfo_addMath
File "mathPres__init.pyc", line 77, in ensureInit
File "mathPres\pymathspeak.pyc", line 9, in
ImportError: No module named mathspeak.en

is it necessary to install something to test it?
Peter

Comment 5 by jteh on 2014-10-03 23:16
Arrrg. I messed up the build. Stand by.

Comment 6 by jteh on 2014-10-04 12:35
Here's a fixed try build.

Comment 7 by driemer.riemer@... on 2014-10-04 15:16
Hi,
Nice work. I had great success with the initial try build. Here is one error on mathjacks website.
http://www.mathjax.org/demos/mathml-samples/

ERROR - scriptHandler.executeScript (09:14:37):
error executing script: <bound method Gecko_ia2.script_moveByLine_forward of <virtualBuffers.gecko_ia2.Gecko_ia2 object at 0x05A48770>> with gesture u'down arrow'
Traceback (most recent call last):
File "scriptHandler.pyc", line 174, in executeScript
File "cursorManager.pyc", line 160, in script_moveByLine_forward
File "cursorManager.pyc", line 83, in _caretMovementScriptHelper
File "speech.pyc", line 722, in speakTextInfo
File "speech.pyc", line 607, in _speakTextInfo_addMath
File "mathPres\pymathspeak.pyc", line 100, in getSpeechForMathMl
File "mathspeak_core.pyc", line 1929, in translate
File "mathspeak_core.pyc", line 1992, in _endElementHandler
File "mathspeak_core.pyc", line 1892, in _translate_mover
File "mathspeak_core.pyc", line 1843, in _stackText
File "mathspeak_core.pyc", line 1072, in _isOperator
TypeError: _isOperator() takes exactly 2 arguments (1 given)

Comment 8 by jteh (in reply to comment 7) on 2014-10-04 16:02
Replying to driemer.riemer@…:

File "mathspeak_core.pyc", line 1072, in _isOperator

TypeError: _isOperator() takes exactly 2 arguments (1 given)

Thanks. Fixed, but I didn't update the submodule or build just for this.

Comment 9 by jteh on 2014-10-10 03:48
Broadening the scope of this ticket to cover integrating liblouisutdml for braille, which I've now done.
Changes:
Changed title from "In-built math speech and interaction via pyMathSpeak" to "In-built math implementations"

Comment 10 by ajirving on 2014-10-27 18:21
I tried to test this, but after checking out the branch, when I do
git submodule update
I get an error
No submodule mapping found in .gitmodules for path 'include/liblouisutdml'

Am I doing something wrong or has liblouisutdml not been added propperly?

Comment 12 by jteh on 2014-10-28 00:09
It's a new submodule, so you need to do:

git submodule update --init

Comment 13 by ajirving on 2014-10-28 13:43
Unfortunately that still gives the same error. I've checked the .gitmodules file and there isn't anything about liblouisutdml (the only change in this branch is the addition of pymathspeak).

Comment 14 by jteh on 2014-10-28 21:28
Hmm. My bad. Thanks for spotting. Not sure how I managed that.

Fixed in 764556f.

Comment 15 by ajirving on 2014-10-31 15:27
Thank you, it works now. This looks like being a really useful feature. Two small things though:

1. The version of the pymathspeak module needs to be updated so that it includes your fix of _isOperator.
2. If the braille translation is longer than the display then the display always seems to go to the end rather than the beginning (at least in Firefox).

Comment 16 by jteh (in reply to comment 15) on 2014-11-01 01:19
Thanks for your feedback!

Replying to ajirving:

1. The version of the pymathspeak module needs to be updated so that it includes your fix of _isOperator.

Yeah. I figured I'd soon be making other changes to pyMathSpeak and didn't want too many submodule update commits, but I haven't worked on it in a while, so it's probably worth updating. I'm thinking of refactoring some of the changes I made, but that might be a while away.

1. If the braille translation is longer than the display then the display always seems to go to the end rather than the beginning (at least in Firefox).

Yeah, I've noticed that. I think I understand why, but haven't thought of a way to fix it yet. :(

Comment 17 by jteh on 2015-05-01 12:04
Since MathPlayer is free (though regrettably not open source) and it surpasses the functionality here, we've decided to include support for MathPlayer and are not going to pursue this pyMathSpeak/liblouisutdml implementation.
Changes:
Added labels: wontfix
State: closed

Comment 18 by driemer.riemer@... on 2015-05-01 13:14
Would it be possible to include this in an addon? The fact that mathplayer does that jumping out of expresssions thingy is really getting on my nerve.

Comment 19 by jteh on 2015-05-01 13:28
I don't have time to work on putting this into an add-on, but it shouldn't be too hard if someone wants to try. The branch is still there.

Regarding auto zoom out in MathPlayer, it's very possible that this will be optional in a future beta of MathPlayer.