Re: [PHP-DOC] Errata cleanup: From: Rasmus Lerdorf (rasmus <email protected>)
Date: 07/31/00

> I've been manually cleaning these out, but if anybody already _has_
> a decent script to batch through all of the posts, and replace the
> tagging (say, replace <B>and</B> with *'s, so <b>BOLD</b> becomes
> *BOLD*, and convert <PRE>code</PRE> to a <p>code<p>),that would
> go a long ways towards errata/notes legibility.... Some of the
> <pre> tags seem to be ignored entirely , but some are not (huh?),
> depending on how you're looking at it (in manual, in edit mode,
> in errata listing mode)

That's just because the tools don't treat them the same way.

> http://www.php.net/manual/function.addslashes.php has an errata
> from bruce cadiz that looks like a mess on screen on in edit
> mode, for this reason (not to mention the A tags)... this is just
> one of _many_.
>
> Hence, a suggestion for removing them entirely from the db, rather
> than trying to re-parse on the fly, or spend lots of human time on
> the edits.

What we really need is a simple markup system that we define and tell
people to use and then have the scripts parse and verify these.

We can't just turn on arbitrary html tags because people can easily make
them manual pages unreadable by putting very long lines inside <pre>
</pre> for example, or adding a <table> tag and then not closing it,
etc. And we certainly don't want a full HTML parsing engine to have to
chew through each note.

I don't have time to dive into this stuff right now, but all the tools are
in CVS (co phpweb), so any of you guys with CVS accounts can modify grab
these scripts and have a whack at fixing them. I can provide mysqldumps
of whatever tables you need to recreate the entire environment on your own
machines.

-Rasmus