Oops, I meant to mention Kirill Lisovsky as an early developer of neat
SXML tools, too.
Neil Van Dyke wrote on 07/28/2015 10:51 AM:
In short, it's a historical accident, but the confusion seems less
costly than a compromise would, IMHO.
Details...
15 years ago, the famous Oleg Kiselyov defined SXML:
http://okmij.org/ftp/Scheme/xml.html
Scheme people in general saw that SXML was good. Perhaps more a
motivation, we saw that Oleg's SSAX XML parser work was solid work
that no one wanted to redo (see papers links on above URL). Oleg,
Dmitry Lizorkin, Jim Bender, and others ended up making other XML
tools that used SXML.
My Scheme pragmatic HTML parser used an ad hoc sexp HTML format I'd
made up, but I converted my parser to use the de facto standard,
SXML. I then wrote other tools using SXML.
In parallel, IIRC, Racket (nee PLT Scheme) already had a help/Web
browser at the time, and a continuation-based Web server, and so were
already using their own XML and HTML stuff heavily.
A few/several years ago, I put considerable work into trying to unify
SXML and Racket xexprs. In the end, I threw out that work, when it
appeared that the differences I was trying to get rid of were actually
mostly wins. (One I recall offhand: arbitrary nesting means efficient
splicing in manipulations of large XML documents, as well as not
requiring a programmer to get all their ",@" right in programmatic
constructions of SXML like in Racket xexprs.)
As much as I wanted to resolve the accidental schism between SXML and
xexprs, the value of keeping SXML's advantages seemed more than the
value of eliminating the recurring SXML-xexpr confusion (of which you
are the latest victim). This discomfort will keep coming up, because
most of the neat HTML and XML tools for Racket actually use SXML, but
other Racket users and documents need xexpr.
I'm currently talking with Oleg about making a couple small tweaks to
the SXML standard, to make it more convenient for manually-written
HTML. I'm trying to bring my tools into strict compliance with
standard SXML. But I have no plans to change anything else about SXML
nor change my tools to use xexpr.
As an engineer, how dumb the situation looks at first glance is
annoying. But not so annoying that I'd push Racket to go to a lot of
work to rip up its carpet, and then take a floor-sander to Racket's
xexpr users.
BTW, even some of the most fervent Racket acolytes (you will know them
by their square brackets) will sneak some SXML, from time to time.
Neil V.
--
You received this message because you are subscribed to the Google Groups "Racket
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
