I don't know much about Tika, but I think it only detects type of file and extracts meta-information from the documents, so many parts of POI are not triggered by it. Also it usually works off of beta-versions of POI, so it usually takes a few weeks until it uses changed code.
Dominik. On Thu, Dec 25, 2014 at 6:42 PM, Andreas Beeker <[email protected]> wrote: > Hi Dominik, > > On 25.12.2014 10:28, Dominik Stadler wrote: >> Let me know what you think or if it is causing unexpected problems. It >> may fail for newly added test-documents, either because it triggers >> some bug or it needs to be excluded for some reason (broken file, >> password protected file, ...). >> > > I think that's a good approach - I'm always a bit concerned when doing > rudimentary > changes, that the existing tests use the heap of test files only in a very > specific manner. > Btw. (I haven't checked it yet) but isn't TIKA doing something similar? > > As we are in the beginning of a new beta phase, I don't mind, if the > jenkins/gump builds fail for a > bit longer than usual *) > > Thanks and happy holidays to you too, > Andi. > > *) ... as long as my commits are not the reason for it :) > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
