Miller assumes 64-bit integers, but in Go, the int type varies in size
depending on the architecture: 32-bit architectures have int
equivalent to int32. As a result, the supported range of integer
values is greatly reduced on 32-bit architectures compared to what is
suggested by the documentation.
This patch explicitly uses int64 wherever 64-bit integers are
assumed.
Test cases affected by the behaviour of the random generator are
updated to reflect the new values (the existing seed doesn't produce
the same behaviour since the way random values are generated has
changed).
Signed-off-by: Stephen Kitt <steve@sk2.org>
* To-do items for broader platform/go-version benchmarking
* neaten inferrer API
* extend type-inference unit-test cases
* Add benchmark scripts for comparing compiler versions
* mlr version in addition to mlr --version
* some go-benchmark files for Mac/Linux perf comparisons
* neaten perf-scripts
* merge
* type-scan optimization tests
* type-scan optimization infra
* test new inferrer
* mlr --time option
* include --cpuprofile and --traceprofile in on-line help
* sharpen inferred/deferred-type API distinction
* replace old inferrer with newer/faster
* update docs for new type-inferrer
* JIT mlrval type-interfence: mlrval package
* mlrmap refactor
* complete merge from #779
* iterating
* mlrval/format.go
* mlrval/copy.go
* bifs/arithmetic_test.go
* iterate on bifs/collections_test.go
* mlrval_cmp.go
* mlrval JSON iterate
* iterate applying mlrval refactors to dependent packages
* first clean compile in a long while on this branch
* results of first post-compile profiling
* testing
* bugfix in ofmt formatting
* bugfix in octal-supporess
* go fmt
* neaten
* regression tests all passing
* Rename inputChannel,outputChannel to readerChannel,writerChannel
* Rename inputChannel,outputChannel to readerChannel,writerChannel (#772)
* Start batched-reader API mods
* Singleton-list step for reader-batching at input
* CLI options for records-per-batch and hash-records
* Push channelized-reader logic into DKVP reader
* Push batching logic into chain-transformer, transformers, and channel-writer
* foo
* cmd/mprof and cmd/mprof2
* cmd/mprof3 and cmd/mprof4
* narrowed in on regexp-splitting on IFS/IPS as perf-hit
* neaten
* channelize nidx
* cmd/mprof5
* channelize CSV reader
* channelize NIDX reader
* Dedupe DKVP-reader and NIDX-reader source files
* channelize CSV-lite reader
* channelize XTAB reader
* batchify JSON reader
* channelize GEN pseudo-reader
* scripts for perf-testing on larger files
* merge with main for #776
* Fix record-batching for join and repl
* Fix comment-handling in channelized XTAB reader
* Fix bug found in positional-rename
