safa-developer Mailing List for screen access for all
Brought to you by:
dineshkaushal
Menu
▾
▴
safa-developer — This list is used to discuss the development issues for the safa project.
You can subscribe to this list here.
| 2003 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(5) |
Jul
(36) |
Aug
(24) |
Sep
|
Oct
(10) |
Nov
(4) |
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2004 |
Jan
|
Feb
(4) |
Mar
|
Apr
(6) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: <n.r...@ex...> - 2004-04-27 08:34:17
|
All, Ever realized that Microsoft has released something to the open source community??? Here is the first one (that i know of!!!) Building our SAFA installation using this WiX toolkit could be a good idea. Follow: http://sourceforge.net/projects/wix/ <http://sourceforge.net/projects/wix/> raghav. -----Original Message----- From: Dinesh Kaushal [mailto:din...@ho...] Sent: Monday, April 12, 2004 12:55 PM To: saf...@li... Subject: [Safa-developer] FAQ Some Frequently asked questions about Unicode on windows. how does window NT based system provide unicode support and still compatible with windo 9X programs Windows API (Application programming Interface) consists of thousands of functions. functions which are dealing with strings are of two types, once with A appended to their name meaning they support ansi strings. once with W appended on their name means that they support wide char or another way of saying that they support Unicode characters using UTF-16 encoding. all the windows NT based systems have full unicode support meaning that they have W function functions for all the string operations. whare as in case of window 9X, only some limited functions have W appended to them. to support ansi based applications on window NT based operating systems, all the functions with W also have there counter part A or ansi functions. these ansi functions try to convert wide char values to ansi, based on the code page information and if they are not successful, they return Question mark "?" or ascii value 63. When ansi functions are called with ansi text, they simplly append each character value with zeros to make unicode or wide char values before calling wide char function on windows NT based operating systems. -------- Can window 9X systems display unicode text? yes, but with some limitations. The applications which directly work with ExtTextOut, which is implemented as ANSI and unicode function on all platforms after windows 95, can display Unicode text. For complex scripts like indian languages, they need to use uniscribe which is available for windows 9x operating systems as an additional layer. That is the reason internet explorer can display indian language text on windows 9x systems. window 9x based systems do not allow senging text messages in unicode, neither they allow creating windows which can support unicode. that is the reason the controls available on window 9x based system do not support display of unicode text. ----- can windows 9X based systems input unicode characters? no, as the input mechanism on 9X based system does not support unicode functions. |
|
From: Ravikumar M S <rav...@az...> - 2004-04-12 10:04:49
|
Hi,
Thank you very much dinesh for the info about the UNICODE. Actually it
was my task, you made my work simple.
Thanks once again.
With regards,
Ravikumar M. S.
-----Original Message-----
From: Dinesh Kaushal [mailto:din...@ho...]
Sent: Monday, April 12, 2004 12:55 PM
To: saf...@li...
Subject: [Safa-developer] FAQ
Some Frequently asked questions about Unicode on windows.
how does window NT based system provide unicode support and still compatible
with windo 9X programs
Windows API (Application programming Interface) consists of thousands of
functions. functions which are dealing with strings are of two types, once
with A appended to their name meaning they support ansi strings. once with W
appended on their name means that they support wide char or another way of
saying that they support Unicode characters using UTF-16 encoding. all the
windows NT based systems have full unicode support meaning that they have W
function functions for all the string operations. whare as in case of window
9X, only some limited functions have W appended to them. to support ansi
based applications on window NT based operating systems, all the functions
with W also have there counter part A or ansi functions. these ansi
functions try to convert wide char values to ansi, based on the code page
information and if they are not successful, they return Question mark "?" or
ascii value 63.
When ansi functions are called with ansi text, they simplly append each
character value with zeros to make unicode or wide char values before
calling wide char function on windows NT based operating systems.
--------
Can window 9X systems display unicode text?
yes, but with some limitations. The applications which directly work with
ExtTextOut, which is implemented as ANSI and unicode function on all
platforms after windows 95, can display Unicode text. For complex scripts
like indian languages, they need to use uniscribe which is available for
windows 9x operating systems as an additional layer. That is the reason
internet explorer can display indian language text on windows 9x systems.
window 9x based systems do not allow senging text messages in unicode,
neither they allow creating windows which can support unicode. that is the
reason the controls available on window 9x based system do not support
display of unicode text.
-----
can windows 9X based systems input unicode characters?
no, as the input mechanism on 9X based system does not support unicode
functions.
|
|
From: Dinesh K. <din...@ho...> - 2004-04-12 07:24:51
|
Some Frequently asked questions about Unicode on windows. how does window NT based system provide unicode support and still = compatible with windo 9X programs Windows API (Application programming Interface) consists of thousands of = functions. functions which are dealing with strings are of two types, = once with A appended to their name meaning they support ansi strings. = once with W appended on their name means that they support wide char or = another way of saying that they support Unicode characters using UTF-16 = encoding. all the windows NT based systems have full unicode support = meaning that they have W function functions for all the string = operations. whare as in case of window 9X, only some limited functions = have W appended to them. to support ansi based applications on window NT = based operating systems, all the functions with W also have there = counter part A or ansi functions. these ansi functions try to convert = wide char values to ansi, based on the code page information and if they = are not successful, they return Question mark "?" or ascii value 63. When ansi functions are called with ansi text, they simplly append each = character value with zeros to make unicode or wide char values before = calling wide char function on windows NT based operating systems. -------- Can window 9X systems display unicode text? yes, but with some limitations. The applications which directly work = with ExtTextOut, which is implemented as ANSI and unicode function on = all platforms after windows 95, can display Unicode text. For complex = scripts like indian languages, they need to use uniscribe which is = available for windows 9x operating systems as an additional layer. That = is the reason internet explorer can display indian language text on = windows 9x systems. window 9x based systems do not allow senging text = messages in unicode, neither they allow creating windows which can = support unicode. that is the reason the controls available on window 9x = based system do not support display of unicode text.=20 ----- can windows 9X based systems input unicode characters? no, as the input mechanism on 9X based system does not support unicode = functions.=20 |
|
From: Dinesh K. <din...@ho...> - 2004-04-12 07:22:42
|
Understanding Unicode computers manipulate bits, which is considered to be the basic unit of = data while working with=20 computers of today. a bit can represent two possible states, i.e. 0 or = 1. from the beginning of=20 computing, these bits were combined to form numbers or characters. in = the beginning byte was=20 the commonly used unit for these collected bits. each byte was composed = of 8 bits and could=20 represent a number or a character depending on the kind of operations = performed on them. a byte=20 could hold a number with a possible alternative values of 256. how this = works, is rooted in the=20 mathematics. a number represented with base k and n positions has value = in decimal (10 base=20 system) as k^(n-1)+ k^(n-2) + ... + k^2 + k^1 + k^0. Whare ^ is used to = denote the power. So taking k =3D 2 and n =3D 8 we get: 2^7 + 2^6+ 2^5 + 2^4 + 2^3 + 2^2 + = 2^1 + 2^0 =3D 128 + 64 + 32 +=20 16 + 8 + 4 + 2 + 1 =3D 255. since 0 is also one possibility, we have 255 = + 1 =3D 256.=20 in later days of computing, it was realized, that this range was farely = limited, so basic unit of calculation was extended to 16 bits and then = to 32 bits and now to 64 bits per unit of calculation.=20 but still the size of character remained to be at 8 bit unit for a long = time as most of the computing was done in the western countries, for = them it was sufficient. another issue was to=20 save the space, so, no extention.=20 initially the character was only based on 7 bits, so we only had 128 = possible character=20 representations. as english has 52 characters, for 26 upper and lower, + = 10 for numbers and some=20 others, it was mostly sufficient. This 7 bit character representation = was known as American=20 Standard Code for Information Interchange (ASCII). then when the need to = extend this code arose, due to some special characters in western = European scripts, it was extended to be 8 bit character set and named as = American National Standards Institute=20 (ANSI) code. this provided scope to have 256 alternative character = representations, and this=20 extra code space was used by various people acording to there needs. = europeans used it to add=20 special characters that were part of german or french language, and = Chinese, japanese and koreans=20 used it for there script.=20 Similarly in india efforts were on from 70s to create a standard that=20 could cater to all the indian languages. As all indian language scripts = are based on bhrammi=20 an ancient script, hence they have similar structure. taking benefit of = the similar characters=20 across the scripts, a common script for all Indian languages was created = which was named Indian=20 standard code for information interchange (ISCII). This was sufficient as long as the document did not have non indian = scripts. Since ISCII and=20 other scripts were using the same code space, having indian language and = other languages such as japanese and other european languages was not = possible simultaneously. Also japanese and chinese languages required larger code space, So = another mechanism was proposed=20 in which the code space was to be extended and the character unit was = change from 8 to 16 bits.=20 various schemes were proposed, and finally it was named unicode. since = unicode used 16 bits, it=20 could represent 65536 possible alternative character values. it was = sufficient for most of the=20 scripts used today, but not for all the scripts and special characters. =20 so finally the unicode now has 20 bits, and various schemes depending on = the requirements of a=20 specific script. using these 20 bits, unicode defines different = character space called plains. Basic plain uses only 16 bits, and extra = 4 bits are used to select other 16 plains, making total of 17 plains. Unicode defines a scaler value for each character, but to store and = transmit these characters, we need some encoding schemes, and following = are three alternatives used with unicode characters. UTF-32 in UTF-32, a character consist of 32 bits, and can theoretically = represent 4 billion values. but=20 unicode has only defined upto 20 bits. for around 1 million character = values. this is the most=20 simple form, but with a big problem, that it consumes alot of space. for = a person who is only=20 using english or other western european scripts, will only need 8 bits, = and hence will be=20 waisting 3 times the space actually needed. UTF-16 To solve the problem of space, UTF-16 was created. in UTF-16, each unit = is of 16 bits and when=20 ever some value which is greater than 16 bits, we use surrogate pairs. = each component of a pair=20 is again of 16 bits, with special codes to identify that they are part = of surrogate pairs. for=20 more information about it refer to RFC2781.=20 UTF-8 Even using UTF-16 waists alot of space for the scripts which only use = ASCII as they only use 7=20 bits for information storage. for example languages like english, whare = no value above 127 is=20 used for character representation, alot of space is waisted. so a clever = scheme is devised. In=20 this scheme, to represent character values upto 127, (ascii characters), = only one byte is used.=20 for character values above 127 more than one byte is used. now how does = computer know while=20 reading a byte stream which one is a single byte or a double byte? this = can be understood by=20 having a look at the bit representation of the characters. A character = is stored as a sequence of=20 bits from left to right, the left most bit being the most significant = bit.=20 if the left most bit is zero, it is a single byte representation. since = out of 8 bits left most=20 is used to determine the if single or multiple bytes are used, all the = other bits will count upto=20 7. so it will only represent ascii values. when we need more than 7 = bits, we need to extend the=20 representation to more than one byte. in that case left most bit of all = the bytes will be 1. To=20 know how many bytes form part of one character value, the left most byte = of the multi byte=20 sequence will have it's left most bits set to 1 based on the number of = bytes in the sequence. if=20 the character requires three bytes, the left most byte will have three = bits set to one followed=20 by a 0. Since 4 bits are already used, we have only 4 bits available for = actual data. for each of=20 the other bytes on the right, we have left most bit set to 1 followed by = 0. so we have 6 bits for=20 actual data. now in a three byte data we have 4 + 6 + 6 =3D 16 bits, = which can represent all the=20 unicode characters for the basic unicode range. for indian characters = which use minimum of 12=20 bits as the range is between hex 0x900 to 0x97f or binary 0000 1001 0000 = 0000 to 0000 1001 0111=20 1111, so we need at least 12 data bits. since right most byte can only = accomodate 6 bits in multi=20 byte character and in two byte representation left most byte can only = hold 5 bits, we need more=20 than 2 bytes and the number is 3 bytes.=20 Unicode and windows All window 9X system use ANSI internally and also support limited = Unicode. For example ExtTextOut function in window 9X does have = Unicode and ANSI versions.=20 All window NT based Operating systems i.e windows NT/2000/XP/2003 = natively support unicode. |
|
From: Dinesh K. <din...@ho...> - 2004-04-12 06:57:36
|
Following may be a handy information for anyone working with indian text = on windows. ------ How to work with indian text in notepad and wordpat. for notepad=20 If you want to save a file in plain text in notepad, then enable indian = languages in windows <put the link for the enabling indian languages in = windows>. once the indian language input is enabled, open notepad and = select the desired language from the language bar or use left alt + left = Shift to switch to the desired input language and type your text in the = desired language and then save the file, choose save as. in the save as = dialog box select encoding as unicode or unicode big-indian or UTF-8. = either of the options will work. When ever you open the file in notepad = it will show the text as it is. The Encoding option is not available by = pressing tab, you need to select it with mouse. For wordpad=20 do typing as mentioned above, and then select the file type as unicode = text document, or rich text document during save as. Both the options = will allow the saving the indian language text. Note: for notepad it is the encoding setting and for wordpad it is the = file type. |
|
From: Dr. A. D. <ana...@sa...> - 2004-02-24 00:19:50
|
Dear Dinesh, Persisting with the continue button did install the software. I was very happy to see it work. Many congratulations to you for giving us safa. I could not however find a readme file on the cd. Can you please mail me one by attachment? -------------------- With Regards, Anand Deep, |
|
From: Dr. A. D. <ana...@sa...> - 2004-02-23 01:17:30
|
Dear dinesh, Pressing the continue button does not make any difference. It brings us back to the same exit install dialogue box. -------------------- With Regards, Anand Deep, ----- Original Message ----- From: "Dinesh Kaushal" <din...@ho...> To: <saf...@li...> Sent: Sunday, February 22, 2004 1:58 PM Subject: Re: [Safa-developer] safa failed to install > Hello Dr. Anand Deep > I am still investigating the cause, but in the meanwhile you can select > continue in the installer and software will be installed and work. > > thanks > > ----- Original Message ----- > From: "Dr. Anand Deep" <ana...@sa...> > To: <saf...@li...> > Sent: Saturday, February 21, 2004 3:17 PM > Subject: [Safa-developer] safa failed to install > > > > Dear Dinesh, > > > > I tried to install safa 1.0 on my system. The installer gave the following > > error message: > > error 1904 the module failed to register. > > What can be the cause? > > -------------------- > > With Best wishes, > > Anand Deep, > > > > > > > > > > ------------------------------------------------------- > > SF.Net is sponsored by: Speed Start Your Linux Apps Now. > > Build and deploy apps & Web services for Linux with > > a free DVD software kit from IBM. Click Now! > > http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click > > _______________________________________________ > > Safa-developer mailing list > > Saf...@li... > > https://lists.sourceforge.net/lists/listinfo/safa-developer > > > > > ------------------------------------------------------- > SF.Net is sponsored by: Speed Start Your Linux Apps Now. > Build and deploy apps & Web services for Linux with > a free DVD software kit from IBM. Click Now! > http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click > _______________________________________________ > Safa-developer mailing list > Saf...@li... > https://lists.sourceforge.net/lists/listinfo/safa-developer > |
|
From: Dinesh K. <din...@ho...> - 2004-02-22 08:48:30
|
Hello Dr. Anand Deep I am still investigating the cause, but in the meanwhile you can select continue in the installer and software will be installed and work. thanks ----- Original Message ----- From: "Dr. Anand Deep" <ana...@sa...> To: <saf...@li...> Sent: Saturday, February 21, 2004 3:17 PM Subject: [Safa-developer] safa failed to install > Dear Dinesh, > > I tried to install safa 1.0 on my system. The installer gave the following > error message: > error 1904 the module failed to register. > What can be the cause? > -------------------- > With Best wishes, > Anand Deep, > > > > > ------------------------------------------------------- > SF.Net is sponsored by: Speed Start Your Linux Apps Now. > Build and deploy apps & Web services for Linux with > a free DVD software kit from IBM. Click Now! > http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click > _______________________________________________ > Safa-developer mailing list > Saf...@li... > https://lists.sourceforge.net/lists/listinfo/safa-developer > |
|
From: Dr. A. D. <ana...@sa...> - 2004-02-21 10:02:10
|
Dear Dinesh, I tried to install safa 1.0 on my system. The installer gave the following error message: error 1904 the module failed to register. What can be the cause? -------------------- With Best wishes, Anand Deep, |
|
From: Dinesh K. <din...@ho...> - 2003-11-03 04:15:11
|
the code that I had sent yesterday can also be used with braille = embossers to print unicode hindi with little modification. regards Dinesh |
|
From: Dinesh K. <din...@ho...> - 2003-11-02 14:25:22
|
no, I don't think so. ----- Original Message ----- From: "Pranav Lal" <no...@vs...> To: <saf...@li...> Sent: Sunday, November 02, 2003 6:15 PM Subject: Re: [Safa-developer] braille display code > Can one test this with out a braille display? > > Pranav > At 05:37 PM 11/2/2003 Dinesh Kaushal wrote: > attached is the vb6 code for displaying the hindi unicode text on an alva > braille display. > > it is a single class which can be instantiated and displayBraille method > may be called. > > with little modifications more languages can be added. > > regards > > > > > > ------------------------------------------------------- > This SF.net email is sponsored by: SF.net Giveback Program. > Does SourceForge.net help you be more productive? Does it > help you create better code? SHARE THE LOVE, and help us help > YOU! Click Here: http://sourceforge.net/donate/ > _______________________________________________ > Safa-developer mailing list > Saf...@li... > https://lists.sourceforge.net/lists/listinfo/safa-developer > |
|
From: Pranav L. <no...@vs...> - 2003-11-02 12:58:42
|
Can one test this with out a braille display? Pranav At 05:37 PM 11/2/2003 Dinesh Kaushal wrote: attached is the vb6 code for displaying the hindi unicode text on an alva braille display. it is a single class which can be instantiated and displayBraille method may be called. with little modifications more languages can be added. regards |
|
From: Anupam B. <an...@cs...> - 2003-10-10 12:22:59
|
We are on the verge of testing our TTS with a screen reader developed by = us.=20 It will take a little more time. We hope to provide you the Hindi and Bengali TTS with suitable SAPI = enabled API. In the meanwhile, it will be nice if Dinesh can provide me with the = detailed specs of the screen reader that he desires. Anupam Basu ----- Original Message -----=20 From: Dinesh Kaushal=20 To: saf...@li...=20 Sent: Wednesday, October 08, 2003 3:33 PM Subject: [Safa-developer] need open source tts Hi All, We have been using viavoice tts till now for developing and testing = the software. but if we want to release the software then we need a good = quality tts for english which is open source or free. =20 another issue is regarding hindi tts. =20 Hindi tts that we have been using till date, is from IIIT hyderbad. = this tts does not allow the API to be open, so we need to remove that = tts also before we make the software open source. =20 I think we need the code to be open so that we can participate more = directly. =20 regards Dinesh =20 =20 =20 |
|
From: Dinesh K. <din...@ho...> - 2003-10-08 10:56:50
|
MessageI guess, we can use freeware/shareware tts, but I am not aware of =
some good quality once.
in the new design it is a matter of removing one file and couple of =
lines of code from safa.
but in that case testing might not be consistent.
regards
Dinesh
----- Original Message -----=20
From: Manish Agrawal=20
To: saf...@li...=20
Sent: Wednesday, October 08, 2003 4:03 PM
Subject: RE: [Safa-developer] need open source tts
Maybe the code that is common to both english and hindi (and I'm =
convinced there's a lot of that) can be made open source to work with =
english screen readers only and you can do the hindi specific thing at =
your end.
I understand this is not as simple as that and we will need to =
disentangle the code calling the hindi tts from the rest but I would see =
that as a worthwhile excercise.
Do we really need an open source english TTS? can we not use a =
freeware/shareware TTS engine with an open source project?
~Manish
-----Original Message-----
From: saf...@li... =
[mailto:saf...@li...] On Behalf Of Dinesh =
Kaushal
Sent: Wednesday, October 08, 2003 3:34 PM
To: saf...@li...
Subject: [Safa-developer] need open source tts
Hi All,
We have been using viavoice tts till now for developing and testing =
the software. but if we want to release the software then we need a good =
quality tts for english which is open source or free.
another issue is regarding hindi tts.
Hindi tts that we have been using till date, is from IIIT hyderbad. =
this tts does not allow the API to be open, so we need to remove that =
tts also before we make the software open source.
I think we need the code to be open so that we can participate more =
directly.
regards
Dinesh
|
|
From: Manish A. <mag...@sa...> - 2003-10-08 10:34:12
|
Maybe the code that is common to both english and hindi (and I'm convinced there's a lot of that) can be made open source to work with english screen readers only and you can do the hindi specific thing at your end. I understand this is not as simple as that and we will need to disentangle the code calling the hindi tts from the rest but I would see that as a worthwhile excercise. Do we really need an open source english TTS? can we not use a freeware/shareware TTS engine with an open source project? ~Manish -----Original Message----- From: saf...@li... [mailto:saf...@li...] On Behalf Of Dinesh Kaushal Sent: Wednesday, October 08, 2003 3:34 PM To: saf...@li... Subject: [Safa-developer] need open source tts Hi All, We have been using viavoice tts till now for developing and testing the software. but if we want to release the software then we need a good quality tts for english which is open source or free. =20 another issue is regarding hindi tts. =20 Hindi tts that we have been using till date, is from IIIT hyderbad. this tts does not allow the API to be open, so we need to remove that tts also before we make the software open source. =20 I think we need the code to be open so that we can participate more directly. =20 regards Dinesh =20 =20 =20 |
|
From: HP K. <hpk...@rb...> - 2003-10-08 10:13:31
|
Hi Dinesh The screen reader project at Bangalore also got stuck at this point. I = guess you will have to get in touch with MSAA folks. Firstly because other screen readers are able to do that so, it is = possible. Alternatively, one can customise other browser which is more = accessable. How about the doccumentation of the DOM model put out by W3C? Harish ----- Original Message -----=20 From: Dinesh Kaushal=20 To: saf...@li...=20 Sent: Wednesday, October 08, 2003 3:03 PM Subject: Re: [Safa-developer] html parser hi Harish, I am looking for an html parser so that I can get the text from MSIE = and get the structure for reading in safa. I am currently able to get the plain text from MSIE but, when I get = text from the page along with the type of text such as whether it is = link or paragraph, MSHTML does not provide that information to me = correctly. in one example I used document.body.all collection whare I got elements twice or thrice and the other was document.FirstChild and IHTMLDOMNode::nextSibling combination which = does not give all the element on the page. I need something that can read the = page like the screen reader does. thanks and regards Dinesh |
|
From: Dinesh K. <din...@ho...> - 2003-10-08 10:03:54
|
Hi All, We have been using viavoice tts till now for developing and testing the = software. but if we want to release the software then we need a good = quality tts for english which is open source or free. another issue is regarding hindi tts. Hindi tts that we have been using till date, is from IIIT hyderbad. this = tts does not allow the API to be open, so we need to remove that tts = also before we make the software open source. I think we need the code to be open so that we can participate more = directly. regards Dinesh |
|
From: Dinesh K. <din...@ho...> - 2003-10-08 09:38:39
|
Hi All, I have uploaded amis hindi version on safa site. the software is made up = of amis core and the hindi english text reading plugin. You don't need to download amis separately, but follow the instructions = in the AmisReadme the next link on the site. happy testing Dinesh |
|
From: Dinesh K. <din...@ho...> - 2003-10-08 09:33:33
|
hi Harish, I am looking for an html parser so that I can get the text from MSIE and = get the structure for reading in safa. I am currently able to get the plain text from MSIE but, when I get text = from the page along with the type of text such as whether it is link or = paragraph, MSHTML does not provide that information to me correctly. in one example I used document.body.all collection whare I got elements twice or thrice and the other was document.FirstChild and IHTMLDOMNode::nextSibling combination which does = not give all the element on the page. I need something that can read the = page like the screen reader does. thanks and regards Dinesh |
|
From: HP K. <hpk...@rb...> - 2003-10-08 08:20:54
|
Hi Dinesh
In case you are looking for a tool to build a browser then it can be =
very easily done using I think it is called Webster activex control.
Harish
----- Original Message -----=20
From: Dinesh Kaushal=20
To: saf...@li...=20
Sent: Tuesday, October 07, 2003 6:10 PM
Subject: [Safa-developer] html parser
Hi all,
if any one knows about a good html parser, then please let me know. we =
need it for processing the text on internet pages properly.
regards
Dinesh
----- Original Message -----=20
From: Dinesh Kaushal=20
To: saf...@li...=20
Sent: Tuesday, October 07, 2003 5:42 PM
Subject: [Safa-developer] new version for testing
Hi all,
I apologize for the long silence, now we have two new softwares.
1. the new version of the safa software which has alot of new =
features and=20
2. version of amis which will read daisy text books in english and =
Hindi.
although the hindi daisy book is still not available, we can test =
the features with the english text book.
daisy player with new features is ready and I will post it by =
tomorrow, but safa screen reader is posted and you can download.
this is again a test version, not a final release.
for more details readme page is also available after the download =
link for the software.
one needs viavoice installed in the computer in order to use the =
software.
safa now reads:=20
selection and table information in ms word
email and internet pages without links and other information
although links are read when tabbing through.
most of the basic windows controls such as push button, radio =
button, checkbox, listbox and list view, treeview, menues, and combo =
boxes are now read.=20
safa also speaks total number of items and current item index except =
in treeview.
its still an experimental version as we are just understanding how =
to access various parts of windows.
we need more people who can work on this project.
regards
Dinesh
|
|
From: Dinesh K. <din...@ho...> - 2003-10-07 12:40:42
|
Hi all, if any one knows about a good html parser, then please let me know. we = need it for processing the text on internet pages properly. regards Dinesh ----- Original Message -----=20 From: Dinesh Kaushal=20 To: saf...@li...=20 Sent: Tuesday, October 07, 2003 5:42 PM Subject: [Safa-developer] new version for testing Hi all, I apologize for the long silence, now we have two new softwares. 1. the new version of the safa software which has alot of new features = and=20 2. version of amis which will read daisy text books in english and = Hindi. although the hindi daisy book is still not available, we can test the = features with the english text book. daisy player with new features is ready and I will post it by = tomorrow, but safa screen reader is posted and you can download. this is again a test version, not a final release. for more details readme page is also available after the download link = for the software. one needs viavoice installed in the computer in order to use the = software. safa now reads:=20 selection and table information in ms word email and internet pages without links and other information although links are read when tabbing through. most of the basic windows controls such as push button, radio button, = checkbox, listbox and list view, treeview, menues, and combo boxes are = now read.=20 safa also speaks total number of items and current item index except = in treeview. its still an experimental version as we are just understanding how to = access various parts of windows. we need more people who can work on this project. regards Dinesh |
|
From: Dinesh K. <din...@ho...> - 2003-10-07 12:12:31
|
Hi all, I apologize for the long silence, now we have two new softwares. 1. the new version of the safa software which has alot of new features = and=20 2. version of amis which will read daisy text books in english and = Hindi. although the hindi daisy book is still not available, we can test the = features with the english text book. daisy player with new features is ready and I will post it by tomorrow, = but safa screen reader is posted and you can download. this is again a test version, not a final release. for more details readme page is also available after the download link = for the software. one needs viavoice installed in the computer in order to use the = software. safa now reads:=20 selection and table information in ms word email and internet pages without links and other information although links are read when tabbing through. most of the basic windows controls such as push button, radio button, = checkbox, listbox and list view, treeview, menues, and combo boxes are = now read.=20 safa also speaks total number of items and current item index except in = treeview. its still an experimental version as we are just understanding how to = access various parts of windows. we need more people who can work on this project. regards Dinesh |
|
From: Dinesh K. <din...@ho...> - 2003-08-26 04:22:32
|
MessageWhat about user convenience?
en-us is the locale identifier, used internally by programs.
will the creation of directory with name like en-us or hi-in will be =
clear to everyone?
enu also seems to be waig.
dinesh
----- Original Message -----=20
From: Manish Agrawal=20
To: saf...@li...=20
Sent: Monday, August 25, 2003 9:00 AM
Subject: RE: [Safa-developer] configuration file storage
we should use the two or four letter locale identifiers en-us, hi etc.
neither enu, nor english(united states) is a well known convention as =
are the four letter culture/locale IDs. you can find a full listing of =
these in msdn.
Regards
Manish
-----Original Message-----
From: Dinesh Kaushal [mailto:din...@ho...]=20
Sent: Sunday, August 24, 2003 11:55 AM
To: saf...@li...
Subject: [Safa-developer] configuration file storage
Hi all,
I have developed the component for fileManager as described in the =
design architectural document.=20
current component allows us to create configuration files as the =
following example.
config\enu\winword.10\keymap.xml
with language id for US english,
application MS word XP
and keymap for mapping keys to functions.
I am contempleting between the example above or
config\english (united states)\winword.10\keymap.xml
for hindi options are:
config\hin\winword.10\keymap.xml
or
config\hindi\winword.10\keymap.xml
what do you think, which one is more natural or which one should we =
use?
regards
Dinesh
|
|
From: Dinesh K. <din...@ho...> - 2003-08-26 04:22:06
|
Hi all, most of our prototype that we have already build is basically in vb6 and = I am considering options between various platforms. options are:=20 vb6, dotnet platform, vc6 or unmanaged c++. 1. vb6 already the prototype is in vb6 and the benifit in vb6 is of = exceptable runtime speed and faster development. limitation is that of when we need multiple threads=20 2. vc6: benefit is of more control and virtually no limit to what we can = do with it, but the development time grows many folds. more development = time means greater expense. 3. dotnet Platform: alot of control and already built in facilities for = application development. limitation is that of win32 API whare = performance is impacted by 12 to 20 extra assembly instructions for each = call between dotnet code and win32 API. another issue is that of extra = runtime installation for the first time user of the software due to the = size of the dotnet runtime which is around 27MB. 4. vc7 or unmanaged / managed c++: it allows more control on win32 API = and virtually no extra runtime instructions for win32 API calls, but = that too adds extra 27MB for dotnet runtime. what do you suggest, which one can we use? any extra information or suggestion will help. regards Dinesh |
|
From: Manish A. <mag...@sa...> - 2003-08-25 03:30:50
|
we should use the two or four letter locale identifiers en-us, hi etc. neither enu, nor english(united states) is a well known convention as are the four letter culture/locale IDs. you can find a full listing of these in msdn. Regards Manish -----Original Message----- From: Dinesh Kaushal [mailto:din...@ho...]=20 Sent: Sunday, August 24, 2003 11:55 AM To: saf...@li... Subject: [Safa-developer] configuration file storage Hi all, I have developed the component for fileManager as described in the design architectural document.=20 current component allows us to create configuration files as the following example. =20 config\enu\winword.10\keymap.xml =20 with language id for US english, application MS word XP and keymap for mapping keys to functions. =20 I am contempleting between the example above or =20 config\english (united states)\winword.10\keymap.xml =20 for hindi options are: config\hin\winword.10\keymap.xml or config\hindi\winword.10\keymap.xml =20 what do you think, which one is more natural or which one should we use? =20 regards Dinesh =20 |
2 messages has been excluded from this view by a project administrator.
