[nvda] Re: localizing NVDA and making custom scripts
- From: "Michael Curran" <mick@xxxxxxxxxx>
- To: <nvda@xxxxxxxxxxxxx>
- Date: Sun, 28 Jan 2007 00:10:26 +1100
Hi Peter,
Cleverson (on this list) currently is maintaining one set of files for one
particular language.
However, we havn't yet successfully set up the msgmurge gettext utility to
enable a maintainer to be able to murge the new messages with the old so
that they only have to translate any new ones, rather than retranslating all
of them each time.
The framework is certainly all there for internationalization, you just need
to read up on gettext in Python.
There is a good section on it in the official python manual under
internationalization.
NVDA only now will run with Python 2.5 (due to it now using a few of the
useful aditions to the sintax that Python 2.5 has).
Creating customized support for controls is not too hard, its all managed in
the NVDAObjects directlry.
Every object (control) in the operating system is represented by NVDA as an
NVDAObject.
NVDAObjects/baseType.py contains the base NVDAObject class which contains
all the main properties and methods an NVDAObject should have.
There are then NVDAObjects written specifically for different accessiblity
APIs and application APIs.
IAccessible.py contains most of the NVDAObjects that are usually used with
NVDA.
The bottom of IAccessible.py has a windowClass,objectRole to NVDAObject map
where you can see what NVDAObjects are used for what.
Its simply just a matter of writing an NVDAObject that is based on a more
general one, and changing some of the methods to suit your needs.
Then of course you will have to add it to the classMap at the bottom, if in
deed it is an IAccessible (MSAA) object you are talking about.
Sapi5 should handle unicode characters fine (everything in NVDA is handled
as unicode internally, as long as it was received as unicode from the
Windows API). When ever NVDA uses a windows API call, it makes sure it uses
the unicode version.
I have successfully read websites in Mandarin Chinese with a sapi5 synth for
that language with no problems.
Mick
----- Original Message -----
From: "Peter Vágner" <peter.v@xxxxxxxxxxx>
To: <nvda@xxxxxxxxxxxxx>
Sent: Saturday, January 27, 2007 7:49 AM
Subject: [nvda] localizing NVDA and making custom scripts
Hello listers,
I was impressed with this project when rev 163 was published on the site.
I looked a bit to the python language so at least I can understand some
code now hopefully. Then I was about to write a synth driver for NVDA to
be able to use a mbrola synthesis. I failed and discarded the idea when
realized that mbrola has to be installed prior to runing. Then I installed
python 2.4 and all the required dependencies to run nvda from source. I
leaved when rev 209 was available in the SVN.
My first question is: will actual code run with python 2.4 or do I need to
upgrade to play with it (and perhaps come up with any custom script
eventually)?
Another thing I am interested in is localizing NVDA to my native language.
I know before moving to sourceforge there was a brazilian portuguese set
of files in the SVN. When I looked to the SVN trunk today I can see no
such stuff just a blank directory named locale. Is NVDA ready to be
adapted to pass national characters to the synthesizer if it supports
them? or are there any hints on localizing?
The last thing I can think of at the moment is how to access custom
controls. It is quite common amonk programmers to adapt some of the
standard windows controls to suit their needs but unluckily these do not
provide all the accessibility features as their originated objects do. Can
you recommend any technique to use when trying to discover how a specific
object can be made accessible? In particular I would be happy if NVDA
would be able to read list of files in total commander file manager one
day. I do understand that for edit boxes and static texts NVDA is just
sending a gettext command to these controls. Can something be done if this
fails? Or can highlighted text be somehow detected on the screen?
Sorry for too many questions at once If something has already been
discussed just let me know please and I will try to locate it in the list
archives.
thanks for such a nice project...
Peter
To post messages to the list send email to
nvda@xxxxxxxxxxxxx
To unsubscribe from the list send a blank message to:
nvda-request@xxxxxxxxxxxxx
with 'unsubscribe' in the Subject field.
Thank you for your continued support of Nonvisual Desktop Access, an open
source free screen reader for Microsoft Windows:
http://www.kulgan.net/nvda/
To post messages to the list send email to
nvda@xxxxxxxxxxxxx
To unsubscribe from the list send a blank message to:
nvda-request@xxxxxxxxxxxxx
with 'unsubscribe' in the Subject field.
Thank you for your continued support of Nonvisual Desktop Access, an open
source free screen reader for Microsoft Windows:
http://www.kulgan.net/nvda/
- Follow-Ups:
- [nvda] Re: localizing NVDA and making custom scripts
- From: Peter Vágner
- References:
- [nvda] localizing NVDA and making custom scripts
- From: Peter Vágner
Other related posts:
- » [nvda] localizing NVDA and making custom scripts
- » [nvda] Re: localizing NVDA and making custom scripts
- » [nvda] Re: localizing NVDA and making custom scripts
Hello listers,I was impressed with this project when rev 163 was published on the site. I looked a bit to the python language so at least I can understand some code now hopefully. Then I was about to write a synth driver for NVDA to be able to use a mbrola synthesis. I failed and discarded the idea when realized that mbrola has to be installed prior to runing. Then I installed python 2.4 and all the required dependencies to run nvda from source. I leaved when rev 209 was available in the SVN. My first question is: will actual code run with python 2.4 or do I need to upgrade to play with it (and perhaps come up with any custom script eventually)? Another thing I am interested in is localizing NVDA to my native language. I know before moving to sourceforge there was a brazilian portuguese set of files in the SVN. When I looked to the SVN trunk today I can see no such stuff just a blank directory named locale. Is NVDA ready to be adapted to pass national characters to the synthesizer if it supports them? or are there any hints on localizing? The last thing I can think of at the moment is how to access custom controls. It is quite common amonk programmers to adapt some of the standard windows controls to suit their needs but unluckily these do not provide all the accessibility features as their originated objects do. Can you recommend any technique to use when trying to discover how a specific object can be made accessible? In particular I would be happy if NVDA would be able to read list of files in total commander file manager one day. I do understand that for edit boxes and static texts NVDA is just sending a gettext command to these controls. Can something be done if this fails? Or can highlighted text be somehow detected on the screen?
Sorry for too many questions at once If something has already been discussed just let me know please and I will try to locate it in the list archives.
thanks for such a nice project... Peter To post messages to the list send email to nvda@xxxxxxxxxxxxx To unsubscribe from the list send a blank message to: nvda-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field.Thank you for your continued support of Nonvisual Desktop Access, an open source free screen reader for Microsoft Windows:
http://www.kulgan.net/nvda/
- [nvda] Re: localizing NVDA and making custom scripts
- From: Peter Vágner
- [nvda] localizing NVDA and making custom scripts
- From: Peter Vágner