[antlr-interest] inconsistent line endings in generated parser and lexer

[Doucette, Charles]

I have a grammar file which I have been editing in Antlrworks on Windows.
Apparently, every one of those lines ends with CR-LF.
Every other line automatically generated by ANTLR ends with LF.
When you combine those two streams into the automatically generated parser and lexer, the line endings are inconsistent.
When I try to add those files to my source control tool (and perhaps I shouldn't try since it is generated), Subversion,
I get an error that the line endings are inconsistent.
It can handle all CRs (Mac?), all LFs (Unix) or all CR-LFs (Windows) - but it apparently doesn't like a mixture.
Is it possible to configure ANTLR to generate consistent line endings?
If not, I'd like to request that as a feature.
Today, perhaps I must just remove CRs from my grammar input file before processing it.

Thanks,
Chuck

________________________________

Charles E. Doucette *  617-399-1122 *  Vaultus Mobile Technologies, 263 Summer St., Boston, MA 02210 *  http://www.vaultus.com<http://www.vaultus.com/>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.antlr.org/pipermail/antlr-interest/attachments/20080422/471f7277/attachment-0001.html