Repository navigation
Release 2.0.6 - #87
Merged
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Release 2.0.6: argument validation, temp-file and reader cleanup, and join column order. Bumps
project.cljand the README install snippets; the changes below were merged in #83–#86.Bugs fixed
computeno longer deletes an existing output file: all argument validation now runs before the output file is touched (Validate compute and aggregate arguments before side effects #83).:selectnames are rejected with aTypeExceptionlisting them, on every dataframe type; joined frames used to throw a bareIndexOutOfBoundsException(Validate compute and aggregate arguments before side effects #83).num-workeris checked (integer, 1–8) for every compute path; it was only checked for group-by/aggregate, and 0 or negative values were accepted everywhere (Validate compute and aggregate arguments before side effects #83).aggregatechecks that the number of new column names matches the number of aggregated columns (Validate compute and aggregate arguments before side effects #83).gb/modereturns tied modes sorted instead of in hash order (Validate compute and aggregate arguments before side effects #83)..clojask/groupedand.clojask/joinare deleted when the compute that wrote them finishes, including on failure; they used to linger until the next compute (Delete temp group files when the compute that wrote them finishes #84).inner-join/outer-joinwithout:selectnow keep the caller's column order (1_*first) in header, data and preview, even when the smaller-side heuristic swaps the inputs internally (Keep argument column order when a join swaps its sides #86).Resource handling
:closefunction for early stops (Close abandoned CSV readers; bump clojask-io to 1.0.7 #85).dataframeconstruction opens its input once instead of three to four times, and closes it (Close abandoned CSV readers; bump clojask-io to 1.0.7 #85).preview,print-dfand the pipeline error pre-check close their reader in afinally, also when the previewed pipeline throws (Close abandoned CSV readers; bump clojask-io to 1.0.7 #85).Known, unchanged
bindextensions still open stat-only readers they never close.Local:
lein test52 tests / 198 assertions, 0 failures (JDK 25) on the code tree of this release.