[Seaside-dev] Seaside2.8a1-jmt.480

[Philippe Marschall]

2007/9/14, John Thornborrow <john at pinesoft.co.uk>:
> The only intended change was #encode:on:, so on that basis, please do
> reject this, I obviously messed up a merge :).
>
> However, I agree that the standards do not require the extra characters
> to be escaped, yet they are most useful to have.

I guess you could come up with a use for everything. I'm sure there
are cases where the fact that Seaside writes all the html on one line
is a problem as well. We try to optimized for the general case and not
the special case.

> An example of this is
> when using a WYSIWYG text editor such as TinyMCE (this is what provoked
> me to add those characters.)

Can you elaborate a bit what the problem there is?

> I seem to remember that there was a separate package for "useful" items,
> should this be considered for that package?

Keith runs Seaside Jetsam, you'll have to check with him.

Cheers
Philippe

> Regards,
> John.
>
> Philippe Marschall wrote:
> > Hi
> >
> > First, thanks for your contribution. As far as I can see
> > Seaside2.8a1-jmt.480 consists mainly of two changes: WAHtmlEncoder
> > class >> #encode:on: and the merging of Seaside2.8a1-mb.420.
> >
> > WAHtmlEncoder class >> #encode:on:
> > This is not needed because in xhtml only <, &, and " need escaping.
> > Other characters never need any entity escaping if you use the correct
> > character set and encoding. Using named entity references makes a
> > character unnecessarily use up more bytes. All the comparisons make
> > the escaping much slower and escaping is a hotspot. Seaside spends a
> > considerable amount of time there. This is one of the rare cases where
> > there is a real world measurable speed difference between #do: and
> > #to:do:.
> >
> > Seaside2.8a1-mb.420 has been rejected for the following reasons:
> > - It makes changes that are only useful for SeasideTesting
> > - It introduces several changes made by Keith that were reject and are
> > now part of Seaside Jetsam
> >
> > SeasideTesting has a history of breaking for every SeasideRelease and
> > a long time until it was fixed. If we had full SeasideTesting test
> > cases for all the functional tests things might look different. In
> > general we would prefer a testing framework like Albatros that does
> > not require any changes to client code.
> >
> > So I fear we have to reject Seaside2.8a1-jmt.480 as a whole but don't
> > let this stop you from future contributions.
> >
> > Cheers
> > Philippe
> > _______________________________________________
> > seaside-dev mailing list
> > seaside-dev at lists.squeakfoundation.org
> > http://lists.squeakfoundation.org/mailman/listinfo/seaside-dev
> >
> >
> >
> >
>
>
> Pinesoft Computers are registered in England, Registered number: 2914825. Registered office: 266-268 High Street, Waltham Cross, Herts, EN8 7EA
>
>
>
> This message has been scanned for viruses by BlackSpider MailControl - www.blackspider.com
>
> _______________________________________________
> seaside-dev mailing list
> seaside-dev at lists.squeakfoundation.org
> http://lists.squeakfoundation.org/mailman/listinfo/seaside-dev
>