I think this may be related to something merged in https://github.com/apache/arrow/pull/9041#issuecomment-752266447
We'll get a fix in ASAP -- sorry about that On Wed, Dec 30, 2020 at 4:29 AM Ying Zhou <yzhou7...@gmail.com> wrote: > Hi, > > When finalizing my Arrow2ORC C++ pull request I found a weird Rust-related > and IPC-related error in the Linter that didn’t happen just 2 days ago > despite my code having nothing to do with either Rust or IPC. Here is the > check: > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680 < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680> > Here is the part of the output I think is relevant: > > source_dir /arrow/cpp/src --quiet > 1413 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1413>[1/1] > cd /tmp/arrow-lint-_9i678pu/cpp-build && /usr/local/bin/python > /arrow/cpp/build-support/lint_cpp_cli.py /arrow/cpp/src > 1414 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1414>INFO:archery:Running > Python formatter (autopep8) > 1415 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1415>INFO:archery:Running > Python linter (flake8) > 1416 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1416>INFO:archery:Running > cmake-format linters > 1417 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1417>WARNING:archery:run-cmake-format > modifies files, regardless of --fix > 1418 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1418>INFO:archery:Running > apache-rat linter > 1419 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1419>INFO:archery:Running > R linter > 1420 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1420>INFO:archery:Running > Rust linter > 1421 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1421>Diff > in /arrow/rust/arrow/src/ipc/reader.rs at line 160: > 1422 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1422> > let null_count = struct_node.null_count() as usize; > 1423 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1423> > let struct_array = if null_count > 0 { > 1424 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1424> > // create struct array from fields, arrays and null data > 1425 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1425>- > StructArray::from(( > 1426 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1426>- > struct_arrays, > 1427 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1427>- > null_buffer, > 1428 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1428>- > )) > 1429 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1429>+ > StructArray::from((struct_arrays, null_buffer)) > 1430 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1430> > } else { > 1431 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1431> > StructArray::from(struct_arrays) > 1432 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1432> > }; > 1433 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1433>INFO:archery:Running > Docker linter > 1434 > < > https://github.com/apache/arrow/pull/8648/checks?check_run_id=1625423680#step:7:1434>Error: > `docker-compose --file /home/runner/work/arrow/arrow/docker-compose.yml run > --rm ubuntu-lint` exited with a non-zero exit code 1, see the process log > above. > > There is no C++ or reasonbly C++-related error. Does anyone know why the > error happen? > > I also would like to ask about table equality when some columns are > float/double. In this case do we have some built-in epsilon so that 0.6 == > 0.6 work? Right now I have separate tests for these types that looks like > the following and they are pretty clumsy: > > EXPECT_TRUE(outputTable->schema()->Equals(*(table->schema()))); > EXPECT_TRUE(outputTable->column(0) > ->chunk(0) > ->Slice(0, numRows / 2) > ->ApproxEquals(table->column(0)->chunk(1))); > EXPECT_TRUE(outputTable->column(0) > ->chunk(0) > ->Slice(numRows / 2, numRows / 2) > ->ApproxEquals(table->column(0)->chunk(3))); > EXPECT_TRUE(outputTable->column(1) > ->chunk(0) > ->Slice(0, numRows / 2) > ->ApproxEquals(table->column(1)->chunk(1))); > EXPECT_TRUE(outputTable->column(1) > ->chunk(0) > ->Slice(numRows / 2, numRows / 2) > ->ApproxEquals(table->column(1)->chunk(3))); > > Thanks, > Ying
