Closed bgyori closed 5 years ago
On my machine, the two files
[error] /home/bmg16/.ivy2/cache/org.maltparser/maltparser/jars/maltparser-1.9.0.jar:appdata/features/liblinear/conllu/2PlanarEager.xml [error] /home/bmg16/.ivy2/cache/org.clulab/processors-modelsmain_2.11/jars/processors-modelsmain_2.11-7.5.1.jar:appdata/features/liblinear/conllu/2PlanarEager.xml
differ in line endings. The clulab version has CRLF and the malt version just has LF. I'm trying to track down how that might have come about. It may be that some git settings are messed up, possibly mine.
Yes, it looks like git "helpfully" converted everything in processors to CRLF, which normally doesn't matter unless I do the release and they get into something other than github. I'm working on it. There will probably have to be a 7.5.2 release of processors to fix it.
I reproduced the error, then re-cloned processors with the correct settings, did a publishLocal, then updated reach to use the updated version, and the problem goes away. So, @MihaiSurdeanu, I need to do a PR to processors to update the version to 7.5.2, then republish to maven (which you want to know about), then do a PR to reach to use the updated version. Sorry about the mistake, everyone.
That's not quite right. The publishing process will automatically bump the version to 7.5.2. I'll add a comment to CHANGES.md, though.
Please do!
On Wed, May 8, 2019 at 2:39 PM Keith Alcock notifications@github.com wrote:
I reproduced the error, then re-cloned processors with the correct settings, did a publishLocal, then updated reach to use the updated version, and the problem goes away. So, @MihaiSurdeanu https://github.com/MihaiSurdeanu, I need to do a PR to processors to update the version to 7.5.2, then republish to maven (which you want to know about), then do a PR to reach to use the updated version. Sorry about the mistake, everyone.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/clulab/reach/issues/627#issuecomment-490660838, or mute the thread https://github.com/notifications/unsubscribe-auth/AAI75TRQX77HSO4M2P7LPVTPUNCA3ANCNFSM4HLUQZZQ .
Is it OK to include @BeckySharp 's "Enabling finding triggers through named mentions" as well?
@BeckySharp ?
It was merged i think, there's a unit test, i mean, as far as I'm concerned, yes.... asking bc it's a PR from me?
On Wed, May 8, 2019 at 3:54 PM Mihai Surdeanu notifications@github.com wrote:
@BeckySharp https://github.com/BeckySharp ?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/clulab/reach/issues/627#issuecomment-490680900, or mute the thread https://github.com/notifications/unsubscribe-auth/ABJCPCJRQ2UVKDM66I2STILPUNKZNANCNFSM4HLUQZZQ .
Maybe it's bigger than a 0.0.1 change, perhaps it can only be used with an updated version of reach, ... I was insufficiently paranoid the last time, so just checking. Thanks!
sure? :) hope it was helpful, I doubt the change will affect reach (bc it adds prev unavailable ways to write rule, rather than changing existing functionality), but hopefully if I'm wrong unit tests in reach (which i haven't run) will catch it?
It should be fixed (famous last words). I'll add a unit test to help make sure it doesn't happen again. reach master should use processors 7.5.2.
Just tested the build and it works. Thanks!
With the latest master, when I try to do
sbt assembly
to build a fat jar, I get some dependency conflict errors. I don't have any local changes to my config files and I deleted my ivy cache before the build. Does anyone know what the problem could be?