Repository navigation
Conversation
|
There are (probably unrelated) build errors against the thrift header files: How should I go about those? |
|
Actually, PR #1528 is probably needed to get rid of the build errors. |
|
@pitrou Ignore all Parquet or Thrift related errors for now, at the moment all of them boil down that they wait for conda-forge/thrift-cpp-feedstock#17 |
|
Ok, then I think this PR is basically ready. |
|
Will review this today -- hope to get the PR backlog under control this week |
wesm
left a comment
There was a problem hiding this comment.
I made some minor stylistic comments but otherwise this looks good, thanks for refactoring / cleaning up this code path
There was a problem hiding this comment.
These should have braces -- I would have thought that clang-format would add them but maybe not
There was a problem hiding this comment.
I'm not sure how esoteric infinite iterators are in practice; at some point we might want some kind of "maximum size" to prevent hanging here
There was a problem hiding this comment.
NB. you can also indicate what std::min you want, e.g. std::min<int64_t>
There was a problem hiding this comment.
I think seqref.ref() is only writable on account of const PyObject* being unwieldy -- it might be less awkward to pass a mutable pointer to seqref
There was a problem hiding this comment.
This size argument should precede the "out" arguments (https://google.github.io/styleguide/cppguide.html#Function_Parameter_Ordering)
There was a problem hiding this comment.
If this is a "can't fail" check; use DCHECK_GE
|
I'm going to try to fix these issues right now so we can get this merged, and then your other PR rebased tomorrow |
|
@pitrou for future reference -- we prefer rebase to merge when updating from upstream, as it makes incremental PR maintenance a little bit easier |
c5314d1 to
a0b146b
Compare
Hmm, that's weird. IME rebasing produces lots of spurious conflicts when trying to reconcile other changes. For example trying to rebase or merge PR #1530 over the latest changes in this PR now produces lots of conflicts. |
Change-Id: I76f3c31831c5faf1081dfb6ee3adf7498aca0997
Change-Id: I03ec91b8a90533b217fc1a6832a87b7722dca9d0
ac8afdb to
0013889
Compare
|
Unrelated, but FTR, the Travis-CI connectivity issues have an issue open at travis-ci/travis-ci#9112 |
|
@pitrou Thank you for the reference to the Travis issue. I've retrigged all tests that were failed because of that, we're still waiting for AppVeyor. Once that is done, I'll merge. |
|
One Travis-CI job has failed with |
|
That was https://issues.apache.org/jira/browse/ARROW-1880 striking again. We haven't figured out yet what causes that, but it seems to be the Travis CI VMs having limited resources |
Iterator (not iterable) input was broken with pa.array() unless both type and size were explicitly passed. Author: Wes McKinney <wes.mckinney@twosigma.com> Author: Antoine Pitrou <pitrou@free.fr> Closes #1513 from pitrou/ARROW-2033-pa-array-iterator and squashes the following commits: 0013889 [Wes McKinney] Code review comments dc95be2 [Antoine Pitrou] Fix pyarrow.array with iterator input Change-Id: I930c3309e3fde12e65ede066b47985f67f7f4037
Accept passing a list of dicts to pa.array() if a struct type is given. Based on PR #1513. Author: Antoine Pitrou <antoine@python.org> Closes #1530 from pitrou/ARROW-1705-struct-array-from-dicts and squashes the following commits: 2b9133a [Antoine Pitrou] ARROW-1705: [Python] allow building array from dicts
Iterator (not iterable) input was broken with pa.array() unless both type and size were explicitly passed.