amis-devl — AMIS development list

[Amis-devl] Bangalore weekly status report: Nov 8

[Enable I. <ena...@ya...>]

Hi
 
Tasks done this week:

TTS
- Fine tuning for TTS functions
- getSelectedVoice function working
- getVoices working
- setVoice working
- setPunctuation and getPunctuation working
Audio
- Testing and debugging for audio plugin (for 9 out of 11 functions)
- Analysis on boost library and design of the smil time code handler
- Test cases: new scenarios for audio added.
NCX parse
- Design study of proposed design in progress
 
Plan for the week 08 Nov - 14 Nov 2004
TTS
- Solving issues for TTS functions such as setVoice and getVoice
- Testing TTS  plugin (with 19 out of 21 functions)
Audio
- Testing to continue
- Work to setup mffm timescale library 
NCX parse
- Design to continue - flow of adding elements during SAX for NCX with existing design
 
Issues
- Require code input from Mark ASAP (pending from a week)
(Problem is that mffm timescale library is not provided as a single project, it needs to be build from four existing open source projects which have some missing files. Projects are: mffm timescale mffm timecode, fftw c++ wrapper for fftw and fftw. Dinesh has tried to build projects, but there are some missing files. Mark has complete project. If this is provided, the work will get done faster)
 
FYI :Friday (Nov 12) and Monday (Nov 15) are holidays due to Diwali and Ramzan
 
Regards
 
Shanti



Ms. Shanti Raghavan
Founder, Managing Trustee
EnAble India 
 
Phone: (080) 2552 9657
Cell: 98453 13919
Email: ena...@ya...
Addr: #694 6A Cross, Koramangala 3rd Blk, Bangalore 34 
 






Yahoo! India Matrimony: Find your life partneronline.

Re: [Amis-devl] banglore weekly reports

[Pranav L. <sli...@ya...>]

Hi,

I am concerned about operability in a network environment. What if some one 
wants to build a thin client interface? 

Re: [Amis-devl] banglore weekly reports

[Dinesh K. <din...@ho...>]

hey pranav,

I can think of the following advantages, marisa can add more.

1. Avoid duplication of data: in the old approach, you maintain information 
about a file in memory, and when user needs to have all the sections and 
subsections available to them at once, in a UI treeview kind of thing, in 
order to provide that option the entire list of heading needs to be sent as 
and xml text, which will be a huge memory hit.
2. Simple structure: unified approach for both the components will make the 
system simple.

best regards
Dinesh


another thing is that xml adds ten times the
----- Original Message ----- 
From: "Pranav Lal" <sli...@ya...>
To: <ami...@li...>
Sent: Thursday, November 04, 2004 10:11 PM
Subject: Re: [Amis-devl] banglore weekly reports


> Hi,
>
> What would be the advantage of this approach?
>
> Pranav
> At 09:47 PM 11/2/2004, Dinesh Kaushal wrote:
> in the new model, we have access to the datastructure directly, and to
> access the information we simplly use functions and object pointers, i.e.
> say you have got access to a node and the address of the node is in the
> pointer currentNode, then to know the audio content of that node, you will
> simplly call a function say 
> currentNode->getAudio(&src,&clipBegin,&clipEnd).
> if you want to goto the next node you can call currentNode->nextNode() or
> some such means. Marisa is developing the model, and I hope to finish the
> model for ncx.
>
>
> ----- Original Message -----
> From: "Pranav Lal" <sli...@ya...>
> To: <ami...@li...>
> Sent: Tuesday, November 02, 2004 8:14 PM
> Subject: Re: [Amis-devl] banglore weekly reports
>
>
> > Hi,
> > <snip The new model likely be on pointers rather than xml based output.
> > PL] Hmm; elaborate please.
> >
> > Pranav
> >
> >
> > -------------------------------------------------------
> > This SF.Net email is sponsored by:
> > Sybase ASE Linux Express Edition - download now for FREE
> > LinuxWorld Reader's Choice Award Winner for best database on Linux.
> > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
> > _______________________________________________
> > Amis-devl mailing list
> > Ami...@li...
> > https://lists.sourceforge.net/lists/listinfo/amis-devl
> >
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by:
> Sybase ASE Linux Express Edition - download now for FREE
> LinuxWorld Reader's Choice Award Winner for best database on Linux.
> http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by:
> Sybase ASE Linux Express Edition - download now for FREE
> LinuxWorld Reader's Choice Award Winner for best database on Linux.
> http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
> 

Re: [Amis-devl] banglore weekly reports

[Pranav L. <sli...@ya...>]

Hi,

What would be the advantage of this approach?

Pranav
At 09:47 PM 11/2/2004, Dinesh Kaushal wrote:
in the new model, we have access to the datastructure directly, and to
access the information we simplly use functions and object pointers, i.e.
say you have got access to a node and the address of the node is in the
pointer currentNode, then to know the audio content of that node, you will
simplly call a function say currentNode->getAudio(&src,&clipBegin,&clipEnd).
if you want to goto the next node you can call currentNode->nextNode() or
some such means. Marisa is developing the model, and I hope to finish the
model for ncx.


----- Original Message -----
From: "Pranav Lal" <sli...@ya...>
To: <ami...@li...>
Sent: Tuesday, November 02, 2004 8:14 PM
Subject: Re: [Amis-devl] banglore weekly reports


 > Hi,
 > <snip The new model likely be on pointers rather than xml based output.
 > PL] Hmm; elaborate please.
 >
 > Pranav
 >
 >
 > -------------------------------------------------------
 > This SF.Net email is sponsored by:
 > Sybase ASE Linux Express Edition - download now for FREE
 > LinuxWorld Reader's Choice Award Winner for best database on Linux.
 > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
 > _______________________________________________
 > Amis-devl mailing list
 > Ami...@li...
 > https://lists.sourceforge.net/lists/listinfo/amis-devl
 >


-------------------------------------------------------
This SF.Net email is sponsored by:
Sybase ASE Linux Express Edition - download now for FREE
LinuxWorld Reader's Choice Award Winner for best database on Linux.
http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
_______________________________________________
Amis-devl mailing list
Ami...@li...
https://lists.sourceforge.net/lists/listinfo/amis-devl

RE: [Amis-devl] goToSection clarification: what should be section 1

[Pranav L. <sli...@ya...>]

Marisa,
<snip This still makes sense to me.
PL] Will implementt accordingly.

Pranav 

RE: [Amis-devl] goToSection clarification: what should be section 1

[Marisa D. <ma...@di...>]

Pranav,

Here is what I wrote to you on October 23:

--quote--
3.5.1 =
third top level heading, fifth child of that, and
first child of that.
Tree structure implied by heading nesting.

start counting top level headings after the heading
one class="title" element.

--end quote--

This still makes sense to me.

I'm in Amsterdam and have limited time for email this
week but will try to answer any more questions asap.

/marisa

----you wrote:
Hi Markku,
<snip for example, information about what is on the
cover,
producer information, etc.
PL] The producer information will be given by another
component. NCCParse looks only at the body of the NCC
file. The first h1 is usually the title here. Marisa
suggests that I check for the class attribute having a
value of title and treat that as section 1. The
problem with this approach is that the user could get
confused. If she goes to section 1 then she would land
on the title and not chapter 1.

Am I just being paranoid?

Pranav


At 02:26 AM 11/3/2004, Markku Hakkinen wrote:
For DAISY 2.02 Books, you should take the first
heading as the first
section.  Even though it is a title, it may be
followed by content the user
wishes to hear, for example, information about what is
on the cover,
producer information, etc.


> -----Original Message-----
> From: ami...@li...
> [mailto:ami...@li...] On
Behalf Of Pranav Lal
> Sent: Tuesday, November 02, 2004 9:31 AM
> To: amis-devl-lists.sourceforge.net
> Subject: [Amis-devl] goToSection clarification: what
should
> be section 1
>
> Hi all,
>
> In the goToSection routine, what should I consider
as section
> 1? Should it be the first h1 in the document or the
second h1
> since the first h1 is usually the title?
>
> Pranav
>
--------------------------------------------------------------
>
--------------------------------------------------------------
> -----------
> Pranav 

Re: [Amis-devl] banglore weekly reports

[Dinesh K. <din...@ho...>]

in the new model, we have access to the datastructure directly, and to 
access the information we simplly use functions and object pointers, i.e. 
say you have got access to a node and the address of the node is in the 
pointer currentNode, then to know the audio content of that node, you will 
simplly call a function say currentNode->getAudio(&src,&clipBegin,&clipEnd). 
if you want to goto the next node you can call currentNode->nextNode() or 
some such means. Marisa is developing the model, and I hope to finish the 
model for ncx.


----- Original Message ----- 
From: "Pranav Lal" <sli...@ya...>
To: <ami...@li...>
Sent: Tuesday, November 02, 2004 8:14 PM
Subject: Re: [Amis-devl] banglore weekly reports


> Hi,
> <snip The new model likely be on pointers rather than xml based output.
> PL] Hmm; elaborate please.
>
> Pranav
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by:
> Sybase ASE Linux Express Edition - download now for FREE
> LinuxWorld Reader's Choice Award Winner for best database on Linux.
> http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
> 

RE: [Amis-devl] goToSection clarification: what should be section 1

[Pranav L. <sli...@ya...>]

Hi Markku,
<snip for example, information about what is on the cover,
producer information, etc.
PL] The producer information will be given by another component. NCCParse 
looks only at the body of the NCC file. The first h1 is usually the title 
here. Marisa suggests that I check for the class attribute having a value 
of title and treat that as section 1. The problem with this approach is 
that the user could get confused. If she goes to section 1 then she would 
land on the title and not chapter 1.

Am I just being paranoid?

Pranav


At 02:26 AM 11/3/2004, Markku Hakkinen wrote:
For DAISY 2.02 Books, you should take the first heading as the first
section.  Even though it is a title, it may be followed by content the user
wishes to hear, for example, information about what is on the cover,
producer information, etc.


 > -----Original Message-----
 > From: ami...@li...
 > [mailto:ami...@li...] On Behalf Of Pranav Lal
 > Sent: Tuesday, November 02, 2004 9:31 AM
 > To: amis-devl-lists.sourceforge.net
 > Subject: [Amis-devl] goToSection clarification: what should
 > be section 1
 >
 > Hi all,
 >
 > In the goToSection routine, what should I consider as section
 > 1? Should it be the first h1 in the document or the second h1
 > since the first h1 is usually the title?
 >
 > Pranav
 > --------------------------------------------------------------
 > --------------------------------------------------------------
 > -----------
 > Pranav
 >
 >
 >
 > -------------------------------------------------------
 > This SF.Net email is sponsored by:
 > Sybase ASE Linux Express Edition - download now for FREE
 > LinuxWorld Reader's Choice Award Winner for best database on Linux.
 > http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
 > _______________________________________________
 > Amis-devl mailing list
 > Ami...@li...
 > https://lists.sourceforge.net/lists/listinfo/amis-devl
 >




-------------------------------------------------------
This SF.Net email is sponsored by:
Sybase ASE Linux Express Edition - download now for FREE
LinuxWorld Reader's Choice Award Winner for best database on Linux.
http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
_______________________________________________
Amis-devl mailing list
Ami...@li...
https://lists.sourceforge.net/lists/listinfo/amis-devl

RE: [Amis-devl] goToSection clarification: what should be section 1

[Markku H. <hak...@da...>]

For DAISY 2.02 Books, you should take the first heading as the first
section.  Even though it is a title, it may be followed by content the user
wishes to hear, for example, information about what is on the cover,
producer information, etc.


> -----Original Message-----
> From: ami...@li... 
> [mailto:ami...@li...] On Behalf Of Pranav Lal
> Sent: Tuesday, November 02, 2004 9:31 AM
> To: amis-devl-lists.sourceforge.net
> Subject: [Amis-devl] goToSection clarification: what should 
> be section 1
> 
> Hi all,
> 
> In the goToSection routine, what should I consider as section 
> 1? Should it be the first h1 in the document or the second h1 
> since the first h1 is usually the title?
> 
> Pranav
> --------------------------------------------------------------
> --------------------------------------------------------------
> -----------
> Pranav
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by:
> Sybase ASE Linux Express Edition - download now for FREE 
> LinuxWorld Reader's Choice Award Winner for best database on Linux.
> http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
> 

Re: [Amis-devl] banglore weekly reports

[Pranav L. <sli...@ya...>]

Hi,
<snip The new model likely be on pointers rather than xml based output.
PL] Hmm; elaborate please.

Pranav 

[Amis-devl] goToSection clarification: what should be section 1

[Pranav L. <no...@vs...>]

Hi all,

In the goToSection routine, what should I consider as section 1? Should it 
be the first h1 in the document or the second h1 since the first h1 is 
usually the title?

Pranav
---------------------------------------------------------------------------------------------------------------------------------------
Pranav

[Amis-devl] converting SMIL Time Clock values in milliseconds

[Dinesh K. <din...@ho...>]

hi all,

I have experimented with boost C++ library for handling time strings, and it 
seems that we will have to develop our own algorithm to do that, logic for 
which is given below. There are two problems with boost, I believe all the 
available libraries might have the same problem. one problem is that it can 
not understand Metric and another problem with boost library that it 
considers missing values from the right side. for example, if we have 29:30 
which in smil means 29 minutes and 30 seconds, but boost date time library 
considers it as 29 hours and 30 minutes.

If there is any implementation which can handle smil time values, it will 
help.

Below is the official SMIL definition of clock value:

<quote>

Clock-value         ::= ( Full-clock-value | Partial-clock-value |
Timecount-value )
Full-clock-value    ::= Hours ":" Minutes ":" Seconds ("." Fraction)?
Partial-clock-value ::= Minutes ":" Seconds ("." Fraction)?
Timecount-value     ::= Timecount ("." Fraction)? (Metric)?
Metric              ::= "h" | "min" | "s" | "ms"
Hours               ::= DIGIT+; any positive number
Minutes             ::= 2DIGIT; range from 00 to 59
Seconds             ::= 2DIGIT; range from 00 to 59
Fraction                ::= DIGIT+
Timecount           ::= DIGIT+
2DIGIT                    ::= DIGIT DIGIT
DIGIT                        ::= [0-9]

For Timecount values, the default metric suffix is "s" (for seconds). No
embedded white space is allowed in clock values, although leading and
trailing white space characters will be ignored.

</quote>

Following design flow is our starting point.

note: there is one confusion, what if the metric is h and value is in 
fraction? i.e. 2.3h, what is the real value? is it 2 hours 18 minutes or 2 
hours 30 minutes. My guess is that it is 2 hours and 18 minutes.

Variable listing:

wholeTimeString
fractionTimeString
mMetric
mHour
mMinute
mSecond
mMillisecond
mTotalTime:


Flow description

1. look for the metric and in case it is not found, consider it as second. 
if metric found, remove it from the string and stor appropriate value in the 
mMetric variable which is a Metric enum with millisecond, second, minute, 
and hour as it's members.

to find metric, from the right of the string we move to left, looking for 
the characters (leaving the white space), we check the first character:

There are four possibilities:

A. s: it might be s or ms, so look for the left character of s, which might 
be m or a numeric. if the character is m then mMetric is equal to 
millisecond else it is second.
B. n: it can only be min, so look for the two characters i and m in movement 
towards the left of the n, if found, mMetric equal to minute. if not found, 
consider it as an error and ignoring it  assume default value of mMetric to 
be second.
C. h: it is the hour, and mMetric is equal to hour.
d. numeric or any other alphabet consider mMetric equal to second and remove 
characters till a numeric is found.

2. look for the period "."
if period found, then store left part of the string before the dot in 
wholeTimeString variable, and right part in fractionTimestring variable.
if period not found, then store the entire string in wholeTimeString 
variable.

3. Look for the colons in wholeTimeString variable, there are three 
possibilities.
In the following text, conversion stands for converting from string to the 
integer type and empty string is considered as zero value.

A. . There are no colons in the wholeTimeString. calculate the value as per 
the metric.
note: confusion here, what if the metric is h and value is in fraction? i.e. 
2.3h, what is the real value? is it 2 hours 18 minutes or 2 hours 30 
minutes. My guess is that it is 2 hours and 18 minutes.

determining the position for each fragment.
(i) . metric is h: consider wholeTimeString as hour, store converted value 
in mHour  and consider the fractionTimeString  as minute, store converted 
value in mMinute.
(ii). metric is min: consider wholeTimeString as minute, store converted 
value in mMinute and consider fractionTimeString as seconds, store converted 
value in mSecond.
(iii) metric is s: consider wholeTimeString as seconds, store converted 
value in mSecond and consider fractionTimeString as milliseconds, store 
converted value in mMillisecond.
(iv) metric is ms: consider wholeTimeString as milliseconds, store converted 
value in mMillisecond and consider fractionTimeString as zero.
B. There is one colon in the wholeTimeString. it is partial time value of 
the form (minute:second.fraction), consider left of colon as minute, store 
converted value in mMinute and consider right of colon as second, store 
converted value in mSecond and consider fractionTimeString as milliseconds, 
store converted value in mMillisecond.
C. There are 2 colons in the wholeTimeString. it is full time value of the 
form (hour: minute: seconds.fraction), consider left of first colon as hour, 
store converted value in mHour; consider right of first colon and left of 
second colon as minute, store converted value in mMinute; and consider right 
of second colon as second, store converted value in mSecond; and consider 
fractionTimeString as milliseconds, store converted value in mMillisecond.

4. Final calculation:
convert all of them to milliseconds, add them all and store them in 
mTotalTime.

mHour to be multiplied by 3600000
mMinute to be multiplied by 60000
mSecond to be multiplied by 1000

Notes:
1. one second has 1000 milliseconds.
2. No optimization for multiplication is to be done as compiler can take 
care of that.

----- Original Message ----- 
From: "Markku Hakkinen" <hak...@di...>
To: "'Dinesh Kaushal'" <din...@ho...>
Sent: Tuesday, October 26, 2004 6:56 AM
Subject: Calculating SMIL Time Clock values in milliseconds


> Hi Dinesh,
>
> To answer your question from earlier today, here is the VB starting point
> for the conversion of likely SMIL clock values to milliseconds.
> Below is the official SMIL definition of clock value:
>
> <quote>
>
> Clock-value         ::= ( Full-clock-value | Partial-clock-value |
> Timecount-value )
> Full-clock-value    ::= Hours ":" Minutes ":" Seconds ("." Fraction)?
> Partial-clock-value ::= Minutes ":" Seconds ("." Fraction)?
> Timecount-value     ::= Timecount ("." Fraction)? (Metric)?
> Metric              ::= "h" | "min" | "s" | "ms"
> Hours               ::= DIGIT+; any positive number
> Minutes             ::= 2DIGIT; range from 00 to 59
> Seconds             ::= 2DIGIT; range from 00 to 59
> Fraction                ::= DIGIT+
> Timecount           ::= DIGIT+
> 2DIGIT                    ::= DIGIT DIGIT
> DIGIT                        ::= [0-9]
>
> For Timecount values, the default metric suffix is "s" (for seconds). No
> embedded white space is allowed in clock values, although leading and
> trailing white space characters will be ignored.
> </quote>
>
> So, in my VB code solution I did two things, assuming all DAISY books used
> the following simplification of timecodes:
> 1. if the time value contained only:
>   a number with or without a decimal point, and with or without the suffix
> s, such
>   as nnn[.nnn][s]
>   then multiply the value by 1000 to get the milliseconds
>
> 2. if the value contained any colons, then I assumed it as a time code 
> based
> on hours, minutes, and seconds, and cheated by using VB's built function 
> to
> isolate the time components from a timeclock string, resulting in the 
> simple
> function:
>
> Public Function time2msec(ti As String) As Long
>    h = Hour(ti)
>    m = Minute(ti)
>    s = Second(ti)
>    l& = h * 360000 + m * 60000 + s * 1000
>    time2msec = l&
> End Function
>
> Now, I would recommend writing a more robust parser of the time SMIL time
> code, which really is quite straight forward, given the definition above.
> I hope this helps, but if you need more, let me know.
>
> Regards,
> mark
>
>
> 

Re: [Amis-devl] banglore weekly reports

[Dinesh K. <din...@ho...>]

Hey pranav,
During my descussions with marisa, she has suggested a new model for 
handling the navigation files, and one unified model might be used for ncc 
and ncx. she has suggested to help on the ncc part.

The new model likely be on pointers rather than xml based output.

regards

----- Original Message ----- 
From: "Pranav Lal" <sli...@ya...>
To: <ami...@li...>
Sent: Saturday, October 30, 2004 4:58 AM
Subject: Re: [Amis-devl] banglore weekly reports


> Dinesh,
> <snip Flow needs to be written of adding ncx elements via sax
> colaborating with marisa for ncc and ncx combined in one model.
> PL] Could you clarify?
>
> Pranav
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by:
> Sybase ASE Linux Express Edition - download now for FREE
> LinuxWorld Reader's Choice Award Winner for best database on Linux.
> http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
> 

[Amis-devl] DeMeglio weekly reports

[Marisa D. <ma...@di...>]

Done 25-31 October
---
* MOUs sent to Dipendra
* Wiki online with some docs
* basic integration of GUI + Smil Engine complete
* Amis Common library finished

Plan 01-07 November
---
* Away in Amsterdam until Nov 9th

-- 
Marisa DeMeglio
Software Development Coordinator
DFA Project
www.daisy-for-all.org

[Amis-devl] Pranav Lal weekly report

[Pranav L. <no...@vs...>]

Done 25 to 30 October 2004
got feedback from Marisa on NCCParse
altered goToSection to conform to feedback
added memory cleanup code
changed documentation  to reflect changes
fixed a bug in moveNext and movePrevious

to do
01 November to 05 November
fix goToSection code and get it verified
get memory cleanup verified and make any changes if required


Pranav
---------------------------------------------------------------------------------------------------------------------------------------
Pranav

Re: [Amis-devl] banglore weekly reports

[Marisa D. <ma...@di...>]

This is a (fairly new) modification of the original design, which allows 
us to more efficiently synchronize the view and data model.  More info 
should appear on the wiki during the coming weeks.  NCC Parse 
requirements have not changed so there is nothing for you to worry about 
there.

Pranav Lal wrote:

> Dinesh,
> <snip Flow needs to be written of adding ncx elements via sax
> colaborating with marisa for ncc and ncx combined in one model.
> PL] Could you clarify?
>
> Pranav
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by:
> Sybase ASE Linux Express Edition - download now for FREE
> LinuxWorld Reader's Choice Award Winner for best database on Linux.
> http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
>
>
>

Re: [Amis-devl] banglore weekly reports

[Pranav L. <sli...@ya...>]

Dinesh,
<snip Flow needs to be written of adding ncx elements via sax
colaborating with marisa for ncc and ncx combined in one model.
PL] Could you clarify?

Pranav 

[Amis-devl] wiki

[Marisa D. <ma...@di...>]

1. we have a wiki on the website for general document collaboration and 
architecture sketches.  eventually we can use it for FAQ's, user guides, 
etc etc.

2.  I put a few initial docs in the wiki, I figure what these can do is 
serve as a whiteboard for our thoughts and ideas.  Actual requirements 
will not change (they are fixed elsewhere as txt files) but the steps to 
meeting the requirements can be something we use the wiki for.  Right 
now the wiki docs contain a brief module description and some API 
functions (both taken from the task details docs).  Just a starting point.

3. As editing is allowed by anyone, please use the sandbox 
(http://amis.sourceforge.net/tiki/tiki-editpage.php?page=SandBox) for 
testing out your wiki skills.  And don't edit a module page unless you 
are a) me or b) writing that module.  If you want a login so your name 
is associated with your edits, let me know.

all for now
marisa

Marisa DeMeglio
Software Development Coordinator
DFA Project
www.daisy-for-all.org



Dinesh Kaushal wrote:

> 25 oct 2004 to 31 oct 2004
> audio:
>
> 9 functions released for testing
> testing going on with test cases prepared earlier
> discussion with mark regarding tsm with mffmtimescale library,
> searching for converter to convert smil time format to a usable time 
> format
> to be used such as millisecond, mark has provide partial code in vb, a
> prommissing library is boost.
>
> tts:
> partial implementation of 14 functions
> problems to be solved are:
> in url format, potential problems with characters which are part of the
> format, may be solved when we move new recommended format by marisa.
> since testing program are available for url format, new API format is 
> still
> not done.
>
> things to be done next week
>
> getting information from sapi5 regarding the language of the selected
> synthesiser,
> getting all the synthesiser voices from sapi5,
> setting the voice
> implementation of punctuation function,
>
> NAV engine:
>
> design modification in progress based on marisa's design
>
> things to be done
>
> Flow needs to be written of adding ncx elements via sax
> colaborating with marisa for ncc and ncx combined in one model.
>
>
>
> -------------------------------------------------------
> This Newsletter Sponsored by: Macrovision For reliable Linux 
> application installations, use the industry's leading
> setup authoring tool, InstallShield X. Learn more and evaluate today. 
> http://clk.atdmt.com/MSI/go/ins0030000001msi/direct/01/
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
>
>
>

[Amis-devl] banglore weekly reports

[Dinesh K. <din...@ho...>]

25 oct 2004 to 31 oct 2004
audio:

9 functions released for testing
testing going on with test cases prepared earlier
discussion with mark regarding tsm with mffmtimescale library,
searching for converter to convert smil time format to a usable time format
to be used such as millisecond, mark has provide partial code in vb, a
prommissing library is boost.

tts:
partial implementation of 14 functions
problems to be solved are:
in url format, potential problems with characters which are part of the
format, may be solved when we move new recommended format by marisa.
since testing program are available for url format, new API format is still
not done.

things to be done next week

getting information from sapi5 regarding the language of the selected
synthesiser,
getting all the synthesiser voices from sapi5,
setting the voice
implementation of punctuation function,

NAV engine:

design modification in progress based on marisa's design

things to be done

Flow needs to be written of adding ncx elements via sax
colaborating with marisa for ncc and ncx combined in one model.

[Amis-devl] DeMeglio weekly reports

[Marisa D. <ma...@di...>]

These reports will show up every Monday; if you are working on some 
task, please follow the same procedure.  Short lists are sufficient.

Done 18-24 October
---
* Installed wiki software on the website, this will be our document 
collaboration model
* Smil Engine: extracted error codes & messages, put in AmisCommon
* reviewed NCC Parse and sent feedback to Pranav
* coded MFC menus, toolbars, basic dialogs.  placeholders now there for 
these items.
* attached (empty) gui logic class to MFC controls
* initial plugin architecture discussion with mark
* Amis Common Library:
        - tested metadata class
        - added booklist class to add/remove filepaths from a text file
        - started writing TitleAuthorParse class + associated utilities
* talked with dinesh and shanti sunday night about nav parse, audio, tts 
modules
 
Plan 25-31 October
---
* Get MOUs out (administrative)
* Set up Wiki with documents + index (website)
* Finish AmisCommon library
* Hook up SMIL Engine to Amis GUI.

Notes:
I'll be gone from Nov 1-9 but available occasionally by email.

-- 
Marisa DeMeglio
Software Development Coordinator
DFA Project
www.daisy-for-all.org

Re: [Amis-devl] code formatter

[Marisa D. <ma...@di...>]

the one advantage i see is that you could set up a batch process for all 
your project files.  i think the formatting itself is about the same.

Pranav Lal wrote:

> any advantages over the vc++ v6 formatting that one does with ctrl+a 
> and alt+f8?
>
> Pranav
> At 11:01 PM 10/7/2004, Marisa DeMeglio wrote:
> Indents code well:
> http://prdownloads.sourceforge.net/astyle/astyle_1.15.3_win32.zip?download 
>
>
> run like this:
> prompt>astyle code.file
>
>
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
> Use IT products in your business? Tell us what you think of them. Give us
> Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out 
> more
> http://productguide.itmanagersjournal.com/guidepromo.tmpl
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
>
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
> Use IT products in your business? Tell us what you think of them. Give us
> Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out 
> more
> http://productguide.itmanagersjournal.com/guidepromo.tmpl
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
>
>
>

Re: [Amis-devl] code formatter

[Pranav L. <sli...@ya...>]

any advantages over the vc++ v6 formatting that one does with ctrl+a and 
alt+f8?

Pranav
At 11:01 PM 10/7/2004, Marisa DeMeglio wrote:
Indents code well:
http://prdownloads.sourceforge.net/astyle/astyle_1.15.3_win32.zip?download

run like this:
prompt>astyle code.file





-------------------------------------------------------
This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
Use IT products in your business? Tell us what you think of them. Give us
Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more
http://productguide.itmanagersjournal.com/guidepromo.tmpl
_______________________________________________
Amis-devl mailing list
Ami...@li...
https://lists.sourceforge.net/lists/listinfo/amis-devl

Re: [Amis-devl] NCC Parse problems

[Marisa D. <ma...@di...>]

I think we talked about the first item (level sensitive) before, that 
feedback had been a mistake on my part so just disregard it.

Pranav Lal wrote:

> Hi Marisa,
>
> Just going through my archives reviewing NCC feedback. You mention 
> that goToNextHeading and goToPreviousHeading should not be level 
> sensitive. I wonder if this still applies? I can set the granularity 
> level but then there will be no other command that is level sensitive 
> so there will be no point in setting any granularity level.
>
>
> Apologies if we have discussed this before.
>
> Pranav
> At 11:51 PM 7/21/2004, Marisa DeMeglio wrote:
> Noticed today:
>
> A. heading navigation
> Both commands:
> goToPreviousHeading
> goToNextHeading
>
> Give: <amisData source="NCCParse"
> type="message"><errmsg>no_heading_at_or_above_t
> his_level</errmsg></amisData>
>
> First of all this feedback is incorrect.  Second of all, these commands
> should not be level-sensitive.
>
> B. page navigation
> goToNextPage
> goToPreviousPage
> Give: <amisData source="NCCParse"
> type="message"><errmsg>on_last_page</errmsg></a
> misData>
> When I was near the beginning of the document, so far navigating using
> headings.
>
> C. Output
> Output format is contradictory.  return data type says "message" but
> actual contents are in an <errmsg> tag.  Remove errmsg tag and give
> return data as messages not errors.  Note this strategy cannot apply to
> all return messages/errors; actual categorizing depends.  Post output
> docs as stated in AMIS Project conventions and we'll check for
> conformity.  Easier to debug in documentation than in practice!
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1 today.
> http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
>
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: IT Product Guide on ITManagersJournal
> Use IT products in your business? Tell us what you think of them. Give us
> Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out 
> more
> http://productguide.itmanagersjournal.com/guidepromo.tmpl
> _______________________________________________
> Amis-devl mailing list
> Ami...@li...
> https://lists.sourceforge.net/lists/listinfo/amis-devl
>
>
>

Re: [Amis-devl] NCC Parse problems

[Pranav L. <sli...@ya...>]

Hi Marisa,

Just going through my archives reviewing NCC feedback. You mention that 
goToNextHeading and goToPreviousHeading should not be level sensitive. I 
wonder if this still applies? I can set the granularity level but then 
there will be no other command that is level sensitive so there will be no 
point in setting any granularity level.


Apologies if we have discussed this before.

Pranav
At 11:51 PM 7/21/2004, Marisa DeMeglio wrote:
Noticed today:

A. heading navigation
Both commands:
goToPreviousHeading
goToNextHeading

Give: <amisData source="NCCParse"
type="message"><errmsg>no_heading_at_or_above_t
his_level</errmsg></amisData>

First of all this feedback is incorrect.  Second of all, these commands
should not be level-sensitive.

B. page navigation
goToNextPage
goToPreviousPage
Give: <amisData source="NCCParse"
type="message"><errmsg>on_last_page</errmsg></a
misData>
When I was near the beginning of the document, so far navigating using
headings.

C. Output
Output format is contradictory.  return data type says "message" but
actual contents are in an <errmsg> tag.  Remove errmsg tag and give
return data as messages not errors.  Note this strategy cannot apply to
all return messages/errors; actual categorizing depends.  Post output
docs as stated in AMIS Project conventions and we'll check for
conformity.  Easier to debug in documentation than in practice!



-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
_______________________________________________
Amis-devl mailing list
Ami...@li...
https://lists.sourceforge.net/lists/listinfo/amis-devl

[Amis-devl] code formatter

[Marisa D. <ma...@di...>]

Indents code well:
http://prdownloads.sourceforge.net/astyle/astyle_1.15.3_win32.zip?download

run like this:
prompt>astyle code.file