maria-developers team mailing list archive

[Alexander Barkov <bar@xxxxxxxxxxx>]

Hi Tamás,

On 04/17/2013 09:53 PM, Tamás Kövesdan wrote:
> Hello,
>
> I'm interested in this project. I have read the other thread about
> this. I've checked the available regex implementations and I've
> written a draft application. Is there any potential mentor who would
> review my application and help to improve it?

Thanks for your interest in this project!

Please feel free to send your application to me.

I'll try to write a detailed list of requirements for this
task soon. Here is a brief list:

1. Ideally, after replacing the regex library,
the regular expression functions should be able to:


a. Work with all MariaDB character sets:
- 8bit
- Unicode: utf8, utf16, utf16le, utf32
- Asian multi-byte: sjis, cp932, ujis, eucjpms, gbk, gb2312, euckr)


b. Follow the comparison rules defined in MariaDB collation,
i.e. take into account things like case and accent sensitivity
for the current collation:

SELECT 'o' RLIKE '<o with diaeresis>' COLLATE utf8_unicode_ci   -> TRUE
SELECT 'o' RLIKE '<o with diaeresis>' COLLATE utf8_hungarian_ci -> FALSE


c. Support modern regex features like look-aheads/look-behinds,
non-greedy modifiers, may be even recursion, etc.



2. The library should be distributed under some permissive library
(e.g. LGPL, BSD, MIT, etc). GPL is not desirable.


Greetings.

> Thanks in advance.
>
> Best whishes,
> Tamás Kövesdán
>
> _______________________________________________
> Mailing list: https://launchpad.net/~maria-developers
> Post to     : maria-developers@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~maria-developers
> More help   : https://help.launchpad.net/ListHelp