Hello, On Thu, Dec 8, 2022 at 1:38 PM Eddie Epstein <eaepst...@gmail.com> wrote:
> It would be useful to understand what roles uimacpp is still needed. > Historically uimacpp code predated the existence of uimaj. > See > > https://web.archive.org/web/20060312040720id_/http://researchweb.watson.ibm.com/journal/sj/433/gotz.pdf This was very useful, thank you! I got the framework to compile and run in a laptop running Debian old stable (figure using old dependencies will help ramp up to newer ones). (Without ActiveMQ support though, I didn't try it as ActiveMQ-CPP is not packaged by Debian.) The autoconf scripts were quite outdated but with some fixing it compiled using system dependencies (i.e., the UCI, xercers and APR shipped by Debian). I need to use compiled dependencies to be able to build the SDK distribution. Also, I migrated the python wrapper from Python 2 to Python 3 and the python annotator seems to be working. I'd like to see whether I can run something from a virtualenv inside. You can follow the progress at my fork: https://github.com/DrDub/uima-uimacpp I still haven't found a killer bug so it is moving forward. The rest of the discussion in this thread is very enlightening regarding the shortcomings of UIMA-CPP. The lack of aggregate engines is... lacking of course. As my goal is Python, I have written a small concept code for discussion. I'll send it in another email. P
