Hi Chris, Could you share the output from when you installed the package? Snappy and the other compression libraries should be on in the binaries (see https://github.com/ursa-labs/arrow-r-nightly/runs/4052316735?check_suite_focus=true#step:4:625 for example), so I'm curious if there's anything in the install logs that help us understand what's up.
Neal On Sun, Oct 31, 2021 at 7:06 PM Chris Berthiaume <[email protected]> wrote: > Hello, > > After upgrading Arrow 5.0.0.2 to 6.0.0.2 in a Bioconductor 3.13 Docker > container, I started to see some new errors when reading Parquet files that > use snappy compression. I'm using the prebuilt Linux binary by setting > LIBARROW_BINARY=true during installation. Building arrow using the latest > nightly source fixes the issue. Is it possible the 6.0.0.2 prebuilt Linux > binary does not have snappy compression support enabled? The error is > copied below. > > Error: NotImplemented: Support for codec 'snappy' not built > In order to read this file, you will need to reinstall arrow with > additional features enabled. > Set one of these environment variables before installing: > > * LIBARROW_MINIMAL=false (for all optional features, including 'snappy') > * ARROW_WITH_SNAPPY=ON (for just 'snappy') > > See https://arrow.apache.org/docs/r/articles/install.html for details > Backtrace: > 1. popcycle::get.vct.by.file(db, vct_dir, > "2018_176/2018-06-25T20-03-48+00-00") test_files.R:210:2 > 4. arrow::read_parquet(...) > 5. base::tryCatch(reader$ReadTable(), error = read_compressed_error) > 6. base:::tryCatchList(expr, classes, parentenv, handlers) > 7. base:::tryCatchOne(expr, names, parentenv, handlers[[1L]]) > 8. value[[3L]](cond) > > Thanks, > Chris Berthiaume >
