Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Awesome, thanks guys for the help in getting it there!
---
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Merged ð! Thanks @mattallenuk for your contribution.
---
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
I've pushed to the dev branch TINKERPOP-1977 with some minor fixes:
https://github.com/apache/tinkerpop/compare/tp32...TINKERPOP-1977
Travis started building it, if it succeeds I'll merge
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Its looking great!
I'll merge it to a dev branch into the project repo and fix the minor test
failure.
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@jorgebay I think I've got it now. Starting to see through the fog.
Hopefully that last push is the one :)
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Ah, thinking I've made a complete hash of this. My git rebase knowledge is
woeful lol.
I ran `git reset --hard ORIG_HEAD` on my local repo to get back to
pre-rebase.
I've tr
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
At this stage you have to use "rebase interactive" to manually select your
commits and skip the rest.
If you have a tool to do that, it would make your job a lot easier.
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
```
First, rewinding head to replay your work on top of it...
Generating patches: 100% (757/757), done.
Applying: valueMap should always return string keys
Using index info to re
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Ok, done that which way do I rebase? Tried
```
git checkout TINKERPOP-1977
git rebase origin/tp32
```
and getting some merges required with files I've not touched.
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
awesome!
First, lets restore `tp32` on your repo:
```
git fetch upstream tp32
git checkout tp32
git reset --hard upstream/tp32
git push origin tp32
```
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
origin https://github.com/mattallenuk/tinkerpop.git (fetch)
origin https://github.com/mattallenuk/tinkerpop.git (push)
upstreamhttps://github.com/apache/tinkerpop.git (fetch)
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
its not a problem :)
As long as you make sure you don't loose your commits on the process (make
sure you push a backup branch to your personal repository).
I think the issue is
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
If i look at https://github.com/mattallenuk/tinkerpop/commits/tp32, it
appears to be what you were expecting? Have I done it the wrong way around?
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
:( sorry
---
Github user spmallette commented on the issue:
https://github.com/apache/tinkerpop/pull/889
hmm - something still looks fishy:
https://github.com/mattallenuk/tinkerpop/commits/TINKERPOP-1977
we should have a string of your continuous commits on top of `tp32` but
they
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
I think the git branch is still based on`master` instead of `tp32`, that's
why github shows a lot of additional commits.
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@spmallette thanks I think I've got it right now. Not a git expert, still
learning :)
---
Github user spmallette commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Note that it's pulling in commits from master now. Not sure how you got to
that point exactly, but I think you need to just rebase your commits in your
local repository on your branch.
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@jorgebay I've rebased my TINKERPOP-1977 onto tp32 and made changes to the
documentation. Not sure I've done it right as I'm seeing lots of conflicts here
now lol!
Let me know if I'v
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Great contribution @mattallenuk!
We should merge it to `tp32`. @mattallenuk do you want to change the base
branch of this pull request and rebase? There shouldn't be any conflicts. We
c
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@spmallette, @jorgebay it has been a pleasure to contribute back, thank you
both for helping me get it done and for helping to improve my skills as the
pull has progressed. This is my first pr
Github user spmallette commented on the issue:
https://github.com/apache/tinkerpop/pull/889
I'm generally VOTE +1 but have some questions/points to make:
Is this ok pointed at master branch? any reason not to go to tp32/tp33, or
is this a feature we only want on 3.4.0 on maste
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Nice! Looks great!
VOTE: +1
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@jorgebay Yes that is a great idea have implemented and made the changes
requested. Have also updated the tests to use the PlainTextSaslAuthenticator
class.
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@jorgebay I've modified the Sasl Authenticator so that a mechanism can now
be passed in along with any mechanism options. I've changed the Sasl Plain
Authenticator to mimic the Java handler co
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@spmallette thank you I'll give that a go.
---
Github user spmallette commented on the issue:
https://github.com/apache/tinkerpop/pull/889
you need to send a `sasl` key with the `saslMechanism` otherwise the
message doesn't validate and you end up with an UNAUTHORIZED. Note the sending
of the `NULL_CHALLENGE`
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@spmallette you may be able to help me with this. I'm trying to understand
the SASL authentication process looking at how the Java code handles it
(https://github.com/apache/tinkerpop/blob/tp3
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Great, I'm working on making the SASL authenticator better so that it can
handle any mechanism and goes through the SASL client first authentication
flow. Hopefully have that done soon.
---
Github user spmallette commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@mattallenuk glad things are working on your end. jorge is on holiday for a
bit, but i'm sure he'll firm up his review as soon as he gets back.
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Ok, I have found the issue. Maven doesn't create a TLS secured gremlin
server for it's secure server. I changed my connection string to use a plain
web socket with no TLS and the tests have ru
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
I can run the tests outside of Maven with a local secure gremlin server and
the tests run fine. Whenever I try to run with Maven the connection to the
secure gremlin server spun-up by Maven do
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@spmallette, no worries.
Problem no 2 is a timeout on making connection to the secure gremlin
server, my plan is to try the tests on a local gremlin server that I control
and then tr
Github user spmallette commented on the issue:
https://github.com/apache/tinkerpop/pull/889
I'm not recognizing too much in the error you're seeing, but doesn't appear
to be related to maven configuration or anything like that, so I probably can't
help too much there.
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@spmallette, sorry, my apologies, haven't been able to look at this much
over the last few days. Have some time planned nearer end of the week to
progress this. I have Maven working but when I
Github user spmallette commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@mattallenuk if there are errors you can share from your build, i might be
able to direct you further.
---
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
We appreciate you taking the lead on this feature @mattallenuk . There is
no rush, we can leave the pull request open and wait for any progress.
---
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Still having issues with Maven build but think I'm getting closer to making
it work. Once working I'll get the integration tests working. Also going to
look at option B on the Sasl class.
--
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
`submit()` accepting `requestId` looks good!
About running the test locally, you should try (on root tinkerpop
directory):
- `mvn clean install -DskipTests` (run this once to generate
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@jorgebay Unfortunately I cannot get the integration tests to run locally
on my mac I am getting an error on maven around
gremlin-server:jar:3.4.0-snapshot not being found. I don't really know
Github user mallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@jorgebay. I think option b would be the better option, however, are there
any plans for Tinkerpop to support anything other than "PLAIN" SASL by default.
If other auth mechanisms can be configur
Github user jorgebay commented on the issue:
https://github.com/apache/tinkerpop/pull/889
Looking at Python and C# GLVs support for authentication, I think there are
2 possible routes:
a) Provide a simple plain text auth mechanism, similar to Python and C#
GLVs:
https://github.co
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@jorgebay, I've abstracted out the transport and made some changes to the
DriverRemoteConnection class to handle the authenticator response.
My only concern with the implementation is
Github user mattallenuk commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@jorgebay. I did actually do it this way originally but was trying to avoid
changing DriverRemoteConnection too much. Will upload these changes this week
for review.
@spmallette, you
Github user spmallette commented on the issue:
https://github.com/apache/tinkerpop/pull/889
@mattallenuk thanks for your contribution on this.
---
45 matches
Mail list logo
