v0.5.23

v0.5.23 (2020-01-31)

Diff since v0.5.22

Closed issues:

• Error when writing dataframe with a BigInt field (#568)
• footerskip counts \r\n termination as two lines (#566)

Merged pull requests:

• Fix #568 by not relying on split_sign from Base which doesn't work fo… (#570) (quinnj)
• Fix #566. The issue here was our footerskip logic reversed the input … (#569) (quinnj)
• update description of CI specification (#563) (bkamins)

v0.5.22

v0.5.22 (2020-01-09)

Diff since v0.5.21

Closed issues:

• How to use nonunique on CSV read's dataframe? (#559)

Merged pull requests:

• Fix an issue reading columns inferred as Dates.Time that also contain… (#561) (quinnj)

v0.5.21

v0.5.21 (2019-12-22)

Diff since v0.5.20

Closed issues:

• Header parsing: Delimiters at end of line makes parsing continue in next line (#555)
• Unexpected whitespace handling in column names (#554)
• dev vs. add CSV (#553)
• Threaded CSV fails with segfault (#552)
• Writing files with large # of columns polls forever? (#538)
• Tab-completion in REPL hangs (#537)

Merged pull requests:

• When ignoring repeated delimiters, our column name detection logic wa… (#556) (quinnj)

v0.5.20

v0.5.20 (2019-12-16)

Diff since v0.5.19

Closed issues:

• Some crashes in csv-comparison (#550)
• hcat fails if one column is only 'missing' values (#548)

Merged pull requests:

• Fix copy for pooled columns. Fixes #550 (#551) (quinnj)

v0.5.19

v0.5.19 (2019-12-14)

Diff since v0.5.18

Closed issues:

• fix categorical for views (#547)
• should be possible to write to IO that don't support seek (#544)
• no method matching + error (#543)
• wrong CSV writing if DataFrame's names contain a delimiter (#540)
• CSV breaks filter! (#539)
• Problems with threads and LazyArrays (#536)
• CSV.write structures (#505)

Merged pull requests:

• Drop LazyArrays dependency in favor of our own CSV.Column2 which we c… (#546) (quinnj)
• Don't fail if writing IO doesn't support seek. Fixes #544 (#545) (quinnj)
• Make sure we treat all column names as strings so they get proper quo… (#541) (quinnj)

v0.5.18

v0.5.18 (2019-11-20)

Diff since v0.5.17

Closed issues:

• Dependency on Profile (#534)
• Non-deterministic parsing of missingstring (#533)
• Data types: Float16 not possible? (#532)
• Setting missingstrings=\["NA"\] makes some columns not have names (#530)
• Unable to append! DataFrames created with CSV.read (#529)

Merged pull requests:

• If a parsed column name is empty (i.e. zero length), then generate a … (#535) (quinnj)

v0.5.17

v0.5.17 (2019-11-18)

Diff since v0.5.16

Merged pull requests:

• Default to copycols=true in DataFrames constructor (#531) (nalimilan)

v0.5.16

v0.5.16 (2019-11-12)

Diff since v0.5.15

Closed issues:

• CSV.File could have more flexible interface (#348)

Merged pull requests:

• Make CSV.File subtype AbstractVector and cleanup iteration (#525) (quinnj)

v0.5.15

v0.5.15 (2019-11-11)

Diff since v0.5.14

Closed issues:

• what's blocking LazyArrays from being updated? (#528)
• CSV.read cannot detect "Time" type string (#526)
• CSV.read errors on Ubuntu 18.04 CSV 0.5.14 (#524)
• [real data test] Small dataset warnings etc (#523)

Merged pull requests:

• Fix #526 by making sure we can auto-detect Time types (#527) (quinnj)
• Fixed up the read from zip files examples (#520) (xiaodaigh)

v0.5.14

v0.5.14 (2019-10-19)

Diff since v0.5.13

Closed issues:

• CSV.write() fails on DataFrame with cell with empty string and --check-bounds=yes (#515)
• "allowmissing is a deprecated keyword argument" - how should I change my code? (#514)
• jq/mem3 getting errors; due to using [email protected] (#512)
• Skip unparseable rows (#509)
• bug with comment flag in read (#508)
• Speed up getproperty(r::Row2, nm::Symbol)? (#507)
• CSV-row filtering when reading (#503)
• Update compatibility with CategoricalArrays.jl (#502)
• propertynames does not return column names on a CSV.Row2 (#501)
• CSV.Row2 supports getindex, but CSV.Row does not (#500)
• threaded is 10x slower sometimes (#499)
• ERROR: ArgumentError: Reference array points beyond the end of the pool (#495)
• memory mapping failed: Cannot allocate memory (#488)
• Getting TaskFailed Exception error on 1.3 rc1 for Fannie Mae 2003Q3.txt performance (#484)
• Scientific notation write issue in v0.5 (#480)
• Error: could not create file mapping: The operation comleted successfully (#424)

Merged pull requests:

• Add ability to detect a CSV.Row2 value directly (avoiding intermediat… (#518) (quinnj)
• Fix issue in CSV.write for empty strings; fixes #515 (#516) (quinnj)
• Ensure FilePaths work (#511) (morris25)
• Refactor internals to allow better memory efficiency (#510) (quinnj)