These are chat archives for symengine/symengine

7th
Mar 2016
Isuru Fernando
@isuruf
Mar 07 2016 01:29
@myluszczak, are you here?
@chenchfort, rebase didn't go properly. Can you fix it?
Charles Chen
@chenchfort
Mar 07 2016 01:32
What went wrong?
Isuru Fernando
@isuruf
Mar 07 2016 01:33
@myluszczak's master branch was not updated and now there are 39 commits in the PR
Charles Chen
@chenchfort
Mar 07 2016 01:33
Oh it’s not passing the appveyor
I think it’s because he git rebase master instead of upstream/master?
Isuru Fernando
@isuruf
Mar 07 2016 01:36
It would have been the same if his master branch was updated
Can you do the following?
git checkout 399e428c725ceb77a1845f6dca4aa9a8c4544935
git fetch upstream master
git rebase upstream/master
Charles Chen
@chenchfort
Mar 07 2016 01:37
I see, yeah he didn’t pull the lastest commits from master.
It says fatal: reference is not a tree: 399e428c725ceb77a1845f6dca4aa9a8c4544935
Isuru Fernando
@isuruf
Mar 07 2016 01:42
Hmm, then @myluszczak should do it
Isuru Fernando
@isuruf
Mar 07 2016 02:34
@chenchfort, are you here?
Charles Chen
@chenchfort
Mar 07 2016 02:34
Yes
Isuru Fernando
@isuruf
Mar 07 2016 02:35
What you can do is first checkout polydicteq from origin
Charles Chen
@chenchfort
Mar 07 2016 02:35
Yup did that.
Oh wait, nvm
So git checkout origin/polydicteq?
Still gives me same fatal error
Isuru Fernando
@isuruf
Mar 07 2016 02:36
git fetch origin polydicteq
git reset --hard bb405ab
git cherry-pick --strategy=recursive -X theirs b44860c
git cherry-pick --strategy=recursive -X theirs 22caad6
git cherry-pick --strategy=recursive -X theirs 67542b0
git cherry-pick --strategy=recursive -X theirs 2114377
Then
git fetch upstream master
git rebase upstream/master
Charles Chen
@chenchfort
Mar 07 2016 02:40
Ok! Done.
Isuru Fernando
@isuruf
Mar 07 2016 02:41
Can you push to github now?
Charles Chen
@chenchfort
Mar 07 2016 02:42
Yes
Isuru Fernando
@isuruf
Mar 07 2016 02:42
Is it done now?
Charles Chen
@chenchfort
Mar 07 2016 02:43
Looks like there’s conflict with base.
Isuru Fernando
@isuruf
Mar 07 2016 02:43
Okay. Let me fix this for you
Charles Chen
@chenchfort
Mar 07 2016 02:44
Thanks.
Isuru Fernando
@isuruf
Mar 07 2016 02:46
@chenchfort, see #855
Next step for your team would be to send a new PR, starting from the master when #855 is merged and push in your new UnivariatePolynomial and UnivariateExprPolynomial
Charles Chen
@chenchfort
Mar 07 2016 02:55
Which is #810?
Isuru Fernando
@isuruf
Mar 07 2016 02:56
#810 has UnivariateSeries changes as well, which will take some time to finish and merge, but I think UnivariatePolynomial and UnivariateExprPolynomial is good to merge now
Charles Chen
@chenchfort
Mar 07 2016 02:57
OK! I’ll work on that.
Charles Chen
@chenchfort
Mar 07 2016 05:11
@isuruf #855 ready to merge
Isuru Fernando
@isuruf
Mar 07 2016 06:05
@chenchfort, can you rebase on top of master and clean up the history to few logical commits?
I mean about #856
Charles Chen
@chenchfort
Mar 07 2016 06:06
Ok. Resolving conflicts right now.
Isuru Fernando
@isuruf
Mar 07 2016 06:11
There'll be lots of conflicts due to recent changes. I'm not sure whether it's worth the time to resolve conflicts
Charles Chen
@chenchfort
Mar 07 2016 06:28
I’ve resolved the conflicts. Posted some questions on #856
Siddharth
@bollu
Mar 07 2016 06:56
@isuruf my history and your history are completely divergent :/ so I can't send a PR
@isuruf could you give me write access to the repository symengine/symengine.hs just for this one block of commits?
Isuru Fernando
@isuruf
Mar 07 2016 06:57
You can't send a PR, because your repo is not a fork of the symengine/symengine.hs repo
Siddharth
@bollu
Mar 07 2016 06:57
the problem is that our commit histories are completely different
Isuru Fernando
@isuruf
Mar 07 2016 06:58
Give me a sec
Siddharth
@bollu
Mar 07 2016 06:58
sure
Isuru Fernando
@isuruf
Mar 07 2016 07:04
@bollu, what you can do is
git remote add upstream git@github.com:symengine/symengine.hs
git fetch upstream master
git rebase upstream/master
Siddharth
@bollu
Mar 07 2016 07:05
will replayng my commits on top actually work?
I'm not so sure
replaying*
Isuru Fernando
@isuruf
Mar 07 2016 07:05
It did for me
Siddharth
@bollu
Mar 07 2016 07:05
hm, alright
Ralf Stephan
@rwst
Mar 07 2016 07:05
A simple alternative is to make a textual diff and apply it to a new branch
Siddharth
@bollu
Mar 07 2016 07:08
@isuruf symengine/symengine.hs#1
thanks for the help :)
Isuru Fernando
@isuruf
Mar 07 2016 07:09
Thanks. I'll review
Siddharth
@bollu
Mar 07 2016 07:09
I should cleanup the code, actually.
Isuru Fernando
@isuruf
Mar 07 2016 07:13
@rwst, I couldn't find the trac ticket for removal of directional infinity. Can you point me to it?
Ralf Stephan
@rwst
Mar 07 2016 07:15
Directional was not removed, but the Pynac imlementation limited to three directions. It happened on the fly while fixing infinity comparisons some time ago.
Isuru Fernando
@isuruf
Mar 07 2016 07:18
Ok. For infinities, what do you recommend that we do? One class with direction as argument, but with 3 directions initially so that it can be generalized later if needed?
Ralf Stephan
@rwst
Mar 07 2016 07:18
I think so, yes.
Isuru Fernando
@isuruf
Mar 07 2016 07:25
@bollu, I've enabled travis for the repo. Let me know if you need any more services enabled
Isuru Fernando
@isuruf
Mar 07 2016 08:36
@chenchfort are you here?
Charles Chen
@chenchfort
Mar 07 2016 08:36
Yes
Isuru Fernando
@isuruf
Mar 07 2016 08:37
Ah okay. Can you make a new branch in your git repo that's exactly the same as master, so that I can send a PR?
Charles Chen
@chenchfort
Mar 07 2016 08:38
Ok. What should I name it?
Isuru Fernando
@isuruf
Mar 07 2016 08:38
polynomial?
Charles Chen
@chenchfort
Mar 07 2016 08:39
Wait so, when I comment out lines 262 - 285, it’s complaining about base_dict2 not found which is declared there.
Sorry, it’s complaining about base_dict not base_dict2 on line 326
Isuru Fernando
@isuruf
Mar 07 2016 08:41
Let me send you a PR
Can you create a new branch?
Charles Chen
@chenchfort
Mar 07 2016 08:41
Ok, doing that now.
Done.
Isuru Fernando
@isuruf
Mar 07 2016 08:47
I sent you a PR
Charles Chen
@chenchfort
Mar 07 2016 08:47
Ok, merged.
And now?
Isuru Fernando
@isuruf
Mar 07 2016 08:48
You can send a PR with that branch and let's see what changes needs to be done
Can you send a PR?
Charles Chen
@chenchfort
Mar 07 2016 08:49
Yes.
So see if it passes the tests?
Isuru Fernando
@isuruf
Mar 07 2016 08:50
Yeah, there are some formatting issues there. Let's fix those
Also, you need to add methods like from_dict and from_vec like @myluszczak did
Charles Chen
@chenchfort
Mar 07 2016 09:04
To UnivariatePolynomial?
Isuru Fernando
@isuruf
Mar 07 2016 11:48
Yes
Isuru Fernando
@isuruf
Mar 07 2016 13:19
@chenchfort, there are lots of unrelated changes like formatting. You should remove those
Siddharth
@bollu
Mar 07 2016 15:59
@isuruf is it possible for you to enable CodeClimate as well?
Isuru Fernando
@isuruf
Mar 07 2016 16:01
Is it free?
Do I need to sign up for the trial?
Okay done. There was an issue previously and asked me to sign up for the trial. Now it's working fine
Siddharth
@bollu
Mar 07 2016 16:03
yes, it is free :) Thanks
Isuru Fernando
@isuruf
Mar 07 2016 16:05
@bollu, I left some comments on the PR
Charles Chen
@chenchfort
Mar 07 2016 16:11
@isuruf I won’t be able to talk to @myluszczak for now. So how should I implement UnivariatePolynomial::from_vec?
Isuru Fernando
@isuruf
Mar 07 2016 16:12
You can take a look at UnivaraiteIntPolynomial and do the same
Charles Chen
@chenchfort
Mar 07 2016 16:12
I think that is what I did.
myluszczak
@myluszczak
Mar 07 2016 17:24
@isuruf Thanks for fixing the errors I made rebasing. I'm sorry for being out of contact yesterday; in the future I'll try to be more conscientious about staying in contact and informing people when I am away from the keyboard
@chenchfort I'll have to go to class soon, so we can't talk now, but we can talk when we meet at noon if we need to
Charles Chen
@chenchfort
Mar 07 2016 17:27
@myluszczak It’s resolved now and UnivariatePolynomial::from_vec is working fine.
Siddharth
@bollu
Mar 07 2016 18:04
@isuruf I'll check them out and fix them