Re: Re: IT[X] fork, round 2
| From: | rocha at i-node dot com dot br | Date: | Wed, 12 Feb 2003 17:16:13 +0000 |
| Subject: | Re: Re: IT[X] fork, round 2 | ||
| References: | 1 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-13195@lists.php.net to get a copy of this message | ||
Bertrand Mansion <bmansion@mamasam.com> wrote:
> <paj@pearfr.org> wrote :
>
> > On Wed, 12 Feb 2003 16:23:48 +0100
> > Bertrand Mansion <bmansion@mamasam.com> wrote:
> >
> >> <paj@pearfr.org> wrote :
> >>
> >>> On Wed, 12 Feb 2003 17:02:52 +0300
> >>> Alexey Borzov <borz_off@cs.msu.su> wrote:
> >>>
> >>>> As for changing preg_replace() to str_replace() in variable
> >>>> substitution: it works faster and fixes an annoying bug, as
> >>> Bertrand> Mansion pointed out.
> >>>
> >>> He pointed something wrong here, that breaks BC, -1 here.
> >>
> >> Test it, then we will talk about it. If I "point" something wrong, I
> >> will admit it. Obviously, that's not the case.
> >
> > Will you ever stop being closed ? The bug you have reopened yesterday
> > points exactly on something that will be broken if you replace preg by a
> > simple (and faster only under some condition, but this is not a general
> > state).
> >
> > So, that ***BREAKS*** BC.
>
> Ah, ok, I get it ! Fixing bugs actually breaks BC.
> That's cool, thanks for your explanations...
>
> Forget it, I am tired. You don't need to reply to this one. TIA.
I have a Report class, using IT[X], and PHPLIB templates, also. Templates are
used through connectors so the system is almost independent of template brands.
But not too much. I am willing to send that to PEAR repository. First i was
waiting for developers to decide where to put the dirs, and the names, subdirs
etc. Now looks like i have to wait people discussing about, cache, pregs, et
all. So i thing is time form me to give my, very interested, opinion.
I use a lot preg's to apply functions on columns and do some math on SQL
results before work on the blocks at the template page. It looks like to me it
will be a lot more difficult to do the math without pregs. Sometimes i have
found that bug also, and have to work around it every time with some hack. Isnt
good to me (well, it is really bad ;) do not have pregs, but isn't good work
around that bug every time too.
To me, being practical, we have to go with both. I think using something like
setVarPReg/setVarStrReplace and/or flags could stop that discussion.
And i guess a flag defaulting to preg would not break BC.
S.Rocha
-------------------------------------------------
This mail sent through IMP: http://horde.org/imp/