On Tue, Jun 25, 2013 at 10:29:26AM -0400, Louis-Dominique Dubeau scripsit:
I've written complex XSLT 1 and 2 stylesheets, mainly to process files
encoded using TEI. I would not dream of using anything else than XSLT
for these files. But I've also done conversion of the type Henry
mentioned dozens of times. In my experience it is simpler to avoid XSLT
for *this* kind of conversion.
In my experience, this is only true if the incoming CSV file is all of
clean, very simple, and highly infrequent.
If you're going to do a lot of whatever it is, XSLT 2 is nearly ideal;
you get your escaping for free, you can deal with the
never-to-be-sufficiently accursed quoted commas by tokenizing on "?,"?
after tokenizing on the newlines, the unparsed-text functions are very
handy things, matches, replace, and xsl:analyze-string really can do
about everything you'd want.
I realize it isn't a traditional way to think of XSLT, but, really, 2.0
is very nearly as good as perl for pure string handling tasks. _And_ it
won't let you commit some sin of omission with your character encoding.
-- Graydon
--~------------------------------------------------------------------
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
To unsubscribe, go to: http://lists.mulberrytech.com/xsl-list/
or e-mail: <mailto:xsl-list-unsubscribe(_at_)lists(_dot_)mulberrytech(_dot_)com>
--~--