Well, I found how to fix it, JDOM has an API
Format fmt = Format.getRawFormat();
fmt.setLineSeparator("\n");
xmloutputter.setFormat(fmt);
I guess we must have upgraded from an earlier version of JDOM at some point
and they changed the default behavior of how it formats line breaks.
On Tue, Mar 30, 2010 at 8:37 PM, Henry Minsky <[email protected]>wrote:
> I'm looking at some failures in the test lztest/lztest-textheight.lzx, and
> one of them is due to CRLF sequences ( '\r\n' ) ending up in text field
> content, when the text content comes from an inline dataset, as in the
> example below.
>
> Does anyone recall any change over the last while which would be adding
> \r\n line termination to dataset content as it is
> parsed by the compiler? Since I'm on a Mac, there are not any \r chars are
> in my lzx source file, just
> single \n newlines.
>
> When I compile to script, I see that the compiler has emitted \r\n
> sequences from the dataset XML.
>
> testds = canvas.lzAddLocalData("testds", "<data>\r\n
> <text>ABC\r\nDEF
> --
> Henry Minsky
> Software Architect
> [email protected]
>
>
>
--
Henry Minsky
Software Architect
[email protected]
