On Mon, 10 Sep 2018 12:35:22 -0400 Wes McKinney <[email protected]> wrote: > Yes, that's the error. Any patch that has a base prior to the > parquet-cpp merge has to be rebased. I'm not sure what I did wrong (if > anything) since I effectively cherry-picked 318 commits into master, > but for some reason it's fouled up the part of the merge script that > adds the squashed commit messages
Apparently it's the "git merge" invocation that fails to find the right ancestor. Regards Antoine. > On Mon, Sep 10, 2018 at 12:14 PM Antoine Pitrou <[email protected]> wrote: > > > > > > Hi Wes, > > > > I've just got the following error trying to merge a PR after rebasing, > > is that what you meant? > > > > > > $ ./dev/merge_arrow_pr.py > > ARROW_HOME = /home/antoine/arrow > > PROJECT_NAME = arrow > > Which pull request would you like to merge? (e.g. 34): 2492 > > > > === Pull Request #2492 === > > title ARROW-3170: [C++] Experimental readahead spooler > > source pitrou/ARROW-501-readahead > > target master > > url https://api.github.com/repos/apache/arrow/pulls/2492 > > > > Proceed with merging pull request #2492? (y/n): y > > Depuis https://github.com/apache/arrow > > * [nouvelle référence] refs/pull/2492/head -> PR_TOOL_MERGE_PR_2492 > > Depuis github.com:apache/arrow > > * [nouvelle branche] master -> PR_TOOL_MERGE_PR_2492_MASTER > > 498215fb..a42d4bf1 master -> apache/master > > Basculement sur la branche 'PR_TOOL_MERGE_PR_2492_MASTER' > > La fusion automatique a réussi ; stoppée avant la validation comme demandé > > Traceback (most recent call last): > > File "./dev/merge_arrow_pr.py", line 375, in <module> > > merge_hash = merge_pr(pr_num, target_ref) > > File "./dev/merge_arrow_pr.py", line 199, in merge_pr > > merge_message_flags) > > File "./dev/merge_arrow_pr.py", line 101, in run_cmd > > output = subprocess.check_output(cmd) > > File "/home/antoine/miniconda3/envs/pyarrow/lib/python3.7/subprocess.py", > > line 376, in check_output > > **kwargs).stdout > > File "/home/antoine/miniconda3/envs/pyarrow/lib/python3.7/subprocess.py", > > line 453, in run > > with Popen(*popenargs, **kwargs) as process: > > File "/home/antoine/miniconda3/envs/pyarrow/lib/python3.7/subprocess.py", > > line 756, in __init__ > > restore_signals, start_new_session) > > File "/home/antoine/miniconda3/envs/pyarrow/lib/python3.7/subprocess.py", > > line 1499, in _execute_child > > raise child_exception_type(errno_num, err_msg, err_filename) > > OSError: [Errno 7] Argument list too long: 'git' > > > > > > Regards > > > > Antoine. > > > > > > > > On Sat, 8 Sep 2018 13:08:56 -0400 > > Wes McKinney <[email protected]> wrote: > > > I'm on plane wifi right now so it's hard for me to investigate too > > > much, but for the time being any outstanding patches should be rebased > > > before running the merge script. > > > > > > Note that any committer can rebase a contributor's patch if the > > > contributor has not disallowed it. Please post here if you have any > > > questions or issues > > > On Sat, Sep 8, 2018 at 12:50 PM Wes McKinney <[email protected]> wrote: > > > > > > > > > > > There's some strangeness with our merge script after the parquet-cpp > > > > codebase graft -- I just reverted the most recent commit and am taking > > > > a look > > > > > > > > > > > On Sat, 8 Sep 2018 13:08:56 -0400 > > Wes McKinney <[email protected]> wrote: > > > > > I'm on plane wifi right now so it's hard for me to investigate too > > > much, but for the time being any outstanding patches should be rebased > > > before running the merge script. > > > > > > Note that any committer can rebase a contributor's patch if the > > > contributor has not disallowed it. Please post here if you have any > > > questions or issues > > > On Sat, Sep 8, 2018 at 12:50 PM Wes McKinney <[email protected]> wrote: > > > > > > > > > > > There's some strangeness with our merge script after the parquet-cpp > > > > codebase graft -- I just reverted the most recent commit and am taking > > > > a look > > > > > >
