DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT <http://nagoya.apache.org/bugzilla/show_bug.cgi?id=25373>. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND INSERTED IN THE BUG DATABASE.
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=25373 Timestamp on generated .java and .class should match the one on .jsp ------- Additional Comments From [EMAIL PROTECTED] 2004-03-02 18:06 ------- The simple fact that workaround solutions are suggested is the best proof that this is a bug (without double quotes around it). If I have to precompile the JSP pages or selectively touch some of them then the server is broken, it is not doing what it is supposed to do: recompile pages that are out of sync. I am not so sure that the timestamp on .class files should indicate the time when they were generated. As this bug shows it may be better to use the same time as on the corresponding .jsp fil. Here is a parallel with a revision system like CVS. The timestamp of the files in my sandbox are not set to the time when they were checked out (created locally) but to the time when the corresponding file in the repository was last changed. This is the only way CVS can properly track what changed in my sandbox. This is very similar to our case. The sandbox file is the .class file and the repository file is the .jsp. The performance impact should be minimal and the benefit is fixing a bug which can create really obscure problems. This special timestamping should be done every time a page is compiled. Compared to the time needed to compile the page the time needed to set a timestamp is negligible (I doubt you can even detect it). Also this is one time deal, once compiled you don't need to timestamp anymore, so I would ignore the whole performance issue. The fact that recompilation of JSP pages should be done in development environments only is questionable. The fact that there is a bug remains. This bug can easily surface in development as well if periodically you deploy to an internal development or testing server. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
