Re: Re: CVS Account Request: jordao
| From: | Xavier De Cock | Date: | Wed, 31 Oct 2007 16:01:35 +0000 |
| Subject: | Re: Re: CVS Account Request: jordao | ||
| References: | 1 2 3 4 5 6 7 | Groups: | php.pecl.dev |
| Request: | Send a blank email to pecl-dev+get-4914@lists.php.net to get a copy of this message | ||
Any information about this extension being added to pecl?
I'am personnaly going to use it so is there any plan to create the
package?
Thanks.
Le jeudi 26 juillet 2007 à 14:48 -0400, Olivier Hill a écrit :
> On 7/26/07, Carlos Jordão <carlosjordao@gmail.com> wrote:
> > Hi Olivier,
> > I have seen your page and your code. I "phpized" it and everything
> > compiled without problems, what led me to wander why it wasn't include in
> > PECL.
>
> (removed php-dev as it is not the right place for this)
>
> Because although I could commit it, I'm waiting for community consent.
> You also don't need to phpize it, a simple pecl install would have
> done it using the archive instead of a package/channel name.
>
> > Antony, as I said, I want to use Double Metaphone on my work. First, I
> > have thought propose the Olivier work to be included. Second, as the Double
> > Metaphone is my main goal, I will use it and test, to know how good it will
> > behave on my project - generate metaphones strings for brazilian names. My
> > first impression says it will be fine. If not, I would study Olivier's code
> > to make the appropriate modifications (with his supervision/permission) or
> > propose a new metaphone method, to be used. Anyway, I will read that PCEL
> > page you mentioned before submitting a new account request.
>
> There is not much to change regarding phonemes, as it is respecting a
> well known algorithm (Mr. Lawrence Philips' one). The only things that
> can be found are bugs in the code if it is not following the
> algorithm, or bugs in the way it is bridged with PHP. Please note that
> since it is a mathematical algorithm, it will produce phonemes that do
> not "sound" like the real word, but close names will have the same
> representations.
>
> If there is a problem with brazilian, you could try to do a Triple
> Metaphone ;) I have tested it for French, and the results were better
> than with Metaphone, which is normal since the second version of
> Metaphone has been done for languages other than English.
>
> Regards,
> Olivier
>
Attachment: [application/pgp-signature] Ceci est une partie de message numériquement signée signature.asc
Attachment: [application/pgp-signature] Ceci est une partie de message numériquement signée signature.asc