That is why we have that hardcoded dep ATM but in tomee it creates as much issues for apps not using that featire than it solves problem for others :(
Le 4 avr. 2018 17:24, "Matthew Broadhead" <matthew.broadh...@nbmlaw.co.uk> a écrit : > i posted these links to the issue. did anyone see them? they seem > important to the problem > https://stackoverflow.com/questions/6340802/java-xpath-apach > e-jaxp-implementation-performance > https://blogs.sap.com/2009/12/04/performance-improvements-in > -nw-java-applications-with-xml-processing/ > > On 08/12/2017 17:08, Matthew Broadhead wrote: > >> ok i created an issue https://bz.apache.org/bugzilla >> /show_bug.cgi?id=61875 >> >> On 08/12/2017 16:29, Jeremy Boynes wrote: >> >>> On Dec 8, 2017, at 3:58 AM, Matthew Broadhead < >>>> matthew.broadh...@nbmlaw.co.uk> wrote: >>>> >>>> here is a patch which removes the xalan dependency. but it breaks the >>>> ForEachTagTest. >>>> i notice that every constructor generates a JSTLXPathCompiler. could it >>>> not be a singleton? >>>> >>> Thanks for the patch. I tried applying it but it looks like the support >>> classes from o.a.t.standard.xpath e.g. JSTLXPathCompiler were not included >>> - please could you update. It would also be easier to read the diff if you >>> removed the old code rather than commenting it out. >>> >>> How about creating an new issue in Bugzilla and attaching patches to >>> that. >>> >>> Cheers >>> Jeremy >>> >>> >>> --------------------------------------------------------------------- >>> To unsubscribe, e-mail: taglibs-user-unsubscr...@tomcat.apache.org >>> For additional commands, e-mail: taglibs-user-h...@tomcat.apache.org >>> >>> >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: taglibs-user-unsubscr...@tomcat.apache.org >> For additional commands, e-mail: taglibs-user-h...@tomcat.apache.org >> >> >
