Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen
- added provenance receive notifications
- added tests for attributes
- added upper limit of solr start parameter (1)
- enhanced documentation for upper limit
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
@JohannesDaniel Sounds good. Make sure it includes the request
documentation as well. That's going to be a big deal in making this easy for
new users to use. Thanks.
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
(means in 5-6 hous)
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
hi @MikeThomsen I will upload the changes this evening.
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
Have you had a chance to add the missing provenance tracking?
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
@JohannesDaniel I missed those tests for some reason. Basically I was just
looking for something that goes into deep pagination behavior. Disregard for
now. Thanks.
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen There are two tests that test Solr result paging:
```
testRetrievalOfFullResults()
testRetrievalOfFullResults2()
```
What do you mean exactly? A test that
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@ottobackwards you made my day!
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
@ottobackwards
> @JohannesDaniel If there are no results from a page request, does that
mean there is no output? I ask because I have a PR for that use case with
QueryElasticsearchHttp.
Github user ottobackwards commented on the issue:
https://github.com/apache/nifi/pull/2517
how are you building? Like what command line?
if you are using -Dmaven.test.skip=true you will have this problem
if you want to skip tests use -DskipTests
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@ottobackwards When I try this I always run into the same problem:
[ERROR] Failed to execute goal on project nifi-livy-processors: Could not
resolve dependencies for project
Github user ottobackwards commented on the issue:
https://github.com/apache/nifi/pull/2517
If you are building Nifi, and need to get things working after merging in
the new version poms, you can do a mvn clean install -DskipTests and that
should populate your local repo with the
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen I'm still impeded, because my Maven does not download the
NIFI-SNAPSHOTS. It downloads all dependencies except the SNAPSHOTS. When I
follow the links (e. g.
Github user ottobackwards commented on the issue:
https://github.com/apache/nifi/pull/2517
@JohannesDaniel If there are no results from a page request, does that mean
there is no output? I ask because I have a PR for that use case with
QueryElasticsearchHttp.
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
@JohannesDaniel I didn't see a test that tests deep pagination. If you
don't have one, please add one. You can just throw garbage data in there like
random strings. Also, you should add at least
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
Ok. Verified with a test run that it appears to run fine against a vanilla
Solr installation.
Docker: `docker run --name testsolr -p 8983:8983 solr:latest solr-demo`
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
> should be an expectation that there is some commonality between them.
Solr and ES are increasingly used for different use cases these days. Solr
tends to be more commonly used for pure
Github user ottobackwards commented on the issue:
https://github.com/apache/nifi/pull/2517
My comment is more about the way you work with the ES vs. the Solr
processors and if there is or should be an expectation that there is some
commonality between them. Like how this relates to
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@ottobackwards deep paging should be done with scrolling using cursor marks
(as it is done in GetSolr). simple paging can be done by sucessively increasing
the offset (start parameter).
Github user ottobackwards commented on the issue:
https://github.com/apache/nifi/pull/2517
Are there best practices for handling paging? I think the ES processors
handle this differently. Does it matter?
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen Thank you for your help with Git!!
@bbende
- Solr core is a test dependency again (I actually had a reason to add this
as a non-test dependency, but it was not a good
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
Builds just fine for me locally. If you're still having build problems, try
deleting everything under org/apache/nifi from .m2/repository in your home
folder.
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
Ok. I'll take a look and let you know sometime in a little while. Got some
other things on my plate at the moment.
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen done :)
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
@JohannesDaniel also, you don't need to wait for a new build to push the
changes. Overwriting everything will schedule the job to be killed and a new
one started.
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
@JohannesDaniel nope. If you did a rebase, those new commits would be
behind your commits, not merged ahead of them. What a rebase does is it sets
aside your commits, brings the base pointer for
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen or is this a normal thing that the commits of others are shown
here after the rebase?
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
@JohannesDaniel see my earlier comment about rebasing. You **cannot** just
do `git merge master` into another branch like that since master is the
foundation of that branch. Don't worry, though.
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen omg. now there are commits of others in the branch?? maybe I
should simply close this PR and insert the code into a new branch.
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
@JohannesDaniel go ahead and check in the change even if it's broken. I'll
help take a look.
---
Github user ottobackwards commented on the issue:
https://github.com/apache/nifi/pull/2517
If you are building and skipping the tests use -DskipTests=true. You can
also do to do at least one mvn install run as well so that your local repo has
everything.
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
Do this:
1. git checkout master
2. git pull upstream master (whatever you call github.com/apache/nifi
master)
3. git checkout NIFI-4516
4. git rebase master
5. git push
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen rebased everything, build for solr processors works without
any problems. however, when I try to build the whole application, I receive the
following error:
[ERROR] Failed to
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
@JohannesDaniel just do a rebase against master. You shouldn't use `git
merge master` for something like this. If you want to do a big squash event to
make the rebase easier, go ahead.
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen Yeah, this time I looked at this, but haven't had time to fix
this. I think I have to merge master into the branch due to updates. Local
build worked. But thanks for the ping :)
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
@JohannesDaniel Build failed because of missing imports:
> [ERROR] Failed to execute goal
org.apache.maven.plugins:maven-compiler-plugin:3.6.0:compile (default-compile)
on project
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
Hi @MikeThomsen
travic-ci test failed due to "The log length has exceeded the limit of 4 MB
(this usually means that the test suite is raising the same exception over and
over)." Is there
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
> Is that what you meant with integration tests? What do you mean with
FetchSolrIT? Do you have a link for an example?
You would create a JUnit test called `FetchSolrIT` and follow the
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen I will keep an eye on the CI tests in the future, thanks for
the advice. Actually, I did not take them into account as they frequently
appear to fail for no reason...
I tested
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
I have Solr Cloud in a docker session on the usual ports, so if you haven't
done them yet some integration tests (ex: `FetchSolrIT`) should be added as
well.
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
BTW, that's the reason why your Travis CI build failed last time. Keep an
eye on that because it'll slow down your code reviews.
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
Believe me, I understand. I just checked out your branch and ran into this
when doing a full build with `mvn clean install` from the root, can you
verify/fix?
> [ERROR] The project
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen ok, sorry, I am too impatient :D
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
Sorry, other stuff got in the way. I'll try to find some time soon.
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen any news?
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
Oh, sorry. Done.
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
Sorry, haven't had time. There's a merge conflict now. Can you fix that?
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@MikeThomsen Any news?
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
Refactored the treatment of flowfiles routed to relationship ORIGINAL. More
attributes are added to flowfiles for better descriptions of requests /
responses. Additionally, I adjusted some
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
Tested the processor in a local build where it worked as expected.
---
Github user MikeThomsen commented on the issue:
https://github.com/apache/nifi/pull/2517
You should consider making the output format configurable. The Solr
projects I've worked on in the past have used JSON instead of XML.
---
Github user JohannesDaniel commented on the issue:
https://github.com/apache/nifi/pull/2517
@ijokarumawak
To be aligned with processor GetSolr I implemented two options for the data
format of Solr results: Solr XML and record functions. However, facets and
stats are written to
52 matches
Mail list logo