Re: GSoC 2011 - Draft Proposal

thank you for the feedback

looked at the code sections you suggested
and realized adding new kind of text style is probably too much scope.

here are the alternatives I can think of
1) reducing the scope of the project to header repeating
        may be I am underestimating the work load of this task,
        but I felt like this is not enough as the whole project.
        will there be enough work to be done just in this task?

2) when ever a template is select all the styles for cells and text in
the selected table,
    will be changed.
    pros: will be using existing styles.it will be as same as if
someone went cell by cell and set the styles manually
    we just do it programmatically .
    cons: will not solve the table header problem.

3) modify TableLayout stuff without touching text styles.
    unlike character and paragraph style, TableLayout seems to less
coupled with rest of the code.
    so by modifying TableLayout it will be possible to archive most of
the styling options with relatively less changes.
=====================================================

please comment on what would be a better choice.

=====================================================

sorry if "I'm confident this will be finished on time" sounds little
'smug' or uninformative.
so here is some more info,
     I have been a ACM ICPC participant representing my university.
     our team placed world rank 26 and 39 in IEEE Xtreme last two years.
     I enjoy programming very much.

my current internship will continue until 12-Aug, but I'll be able to
work required
hours during this period.

=====================================================
http://i.imgur.com/QFiot.jpg

-- 
by,
            Manu
(R Chathura Manuranga Perera)
> Hi Manu,
>
> Regarding table styles have you looked at the Styles code in AbiWord?
>
> The core of it is in:
>
> src/text/ptbl/pt_PT_Styles.(h,cpp) but there are assumptions about
> styles throughout src/text/fmt/xp/*.(h,cpp).
>
> We actually only two types of styles character and paragraph.
> Extending those to include a new type of style will involve all sorts
> of changes through the code. This is quite a big project because we
> have a lot of hardwired assumptions that just these types of styles
> exist. Fixing this would certainly be a good thing but it would the
> first thing that needs doing before you could look at implementing a
> new type of style.
>
> Rather than simply saying "I'm confident this will be finished on
> time" you should give examples of verifiable code you have developed.
> In particular if you can show examples of how you personally solved
> problems with little outside assistance, particularly if you did this
> just because you were interested and it was a fun thing to do, that
> would carry a lot of weight with us.
>
> That sort thing demonstrates to us that you have the potential to
> stick with the AbiWord project after Google stops paying you and
> become a long term contributor to the project.
>
> Finally regarding your internship. How long does it go for and what
> hours are you expected to work for it? We expect our GSoC students to
> work roughly 40 hours per week on their project.
>
>
> Best Regards,
>
> Martin
>
> On Tue, Apr 5, 2011 at 1:30 PM, manu [ranga] <rcmperera@gmail.com> wrote:
>>
>> Draft Proposal - GSoC 2011 - AbiWord
>>
>> Name:
>>    R.C.Manuranga Perera
>>
>> Email:
>>    rcmperera@gmail.com
>>
>> Project Title:
>>    Table improvements
>>
>> Synopsis:
>>    this is based on Martin Sevior's idea displayed on idea page[1].
>>    but rather than implementing it as "header repeat" functionality,
>>    this project proposes it to be implemented as table styling template.
>>    this proposed "table template" will include color, font, align,
>>    ect. for both body and header. users will be able to easily apply a
>>    template to a selected table.also they can make and save new template
>>    styles of their own.
>>
>> Benefits to the AbiWord:
>>   not only this will solve the "big tables header repeat" problem,
>>   but also add a feature that will enable making nice looking tables quickly.
>>
>> Deliverables:
>>   will add new tab to "Format Table" dialog box,  which will allow
>>   users to select template for tables and customize templates
>>
>> Project Details:
>>  introducing "table header repeat" as well as other table style as a
>>  "table template" will have many benefits.
>>  this will make the backend keep separate data models for table data
>>  and meta/style data, which is a good design.
>>  style data will be able be switched without change raw table date.
>>  this will avoid any data repetition in the actual table data, since
>>  even many rows of the same data is visible, actual data model is
>>  untouched.
>>
>>  default templates can be provided for many common patterns eg:-
>>  alternating dark blue and light blue rows with bold headers repeated
>>  for each page
>>
>> Project Schedule:
>>   I am confident that I can manage to stick to google's scheduled
>>   time-line[2] or better even though I am currently interning at
>>   WSO2[3].
>>
>> Bio:
>>  I am Manu (R.C.M. Perera) from Sri Lanka.I am an undergraduate
>>  studying at of University of Moratuwa.
>>  have a good experience both in C++ and C.
>>
>> [1] http://www.abisource.com/wiki/Google_Summer_of_Code_2011#Table_improvements
>> [2] http://www.google-melange.com/gsoc/events/google/gsoc2011
>> [3] http://wso2.com/
>>
>> ========================================================================
>> please let me know any feedback/suggestions/comments on this proposel
>> --
>> by,
>>            Manu
>> (R Chathura Manuranga Perera)
>>