Thanks Andrew, I'll definitely subscribe to mulberrytech since what I'm trying to do is more than a one person task:-). Currently I'm filtering out the CR's and need to read the output because it is yet imperfect and requires some post-transform fixing up.
cheers, Roger ----- Original Message ----- From: "Andrew Welch" <[EMAIL PROTECTED]> To: "Roger I Martin PhD" <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]> Sent: Wednesday, August 20, 2003 4:49 AM Subject: RE: text output encoding UTF-16 and controlling LF and CR > Older versions of javac with Unicode source code did not have a problem > with > LF CR. Hi Roger, I can't really comment on why java might not like CR LF line endings, but I have some suggestions on how to remove them. If you are performing the transform and compiling the result as part of say a batch process, you could add an intermediary step where a java class accepts a stream of characters and writes everything back out unless it's a CR. Alternatively, if you don't intend on reading the java produced from the transform, you could add the template: <xsl:template match="text()"> <xsl:value-of select="normalize-space()"/> </xsl:template> ...which will give you one long line with no line breaks at all. If you aren't already, you may want to subscribe to the xsl-list http://www.mulberrytech.com/xsl/xsl-list/ you will find more like minded people there, especially a few mathml'ers. I would suggest you post what you are doing, not only to get non-xalan specific xslt help, but just to air your ideas. cheers andrew
