This repository was archived by the owner on May 7, 2023. It is now read-only.
Releases: synw/df
Releases · synw/df
0.4.0-nullsafety.0
- Updates df to support null safety
DataFrame.meanColandDataMatrix.mean_now require you to specify how to treat nullsDataFrame.sortandDataFrame.sort_now require you to either specify how to treat nulls or to provide a custom sort function- All methods that previously accepted or returned
dynamicnow returnObjectorObject?
All DataFrame columns have nullable data types. A future update may support non-nullable columns-if you'd like this feature comment and/or thumbs up it's corresponding github issue
0.3.0
Updates to support the quote escaping sections of the CSV standard (see: https://tools.ietf.org/html/rfc4180 sections 2.5-2.7):
supports using escape quotes to escape commas
supports using escape quotes to escape double quotes
supports using escape quotes to escape newlines
BREAKING CHANGE TO THE DF PACKAGE. Here are the breaking changes, with examples:
Double quotes around a field are now consumed
example csv input row: a,"b",c\n
current df output values: {'a', '"b"', 'c'}
df output values after PR: {'a', 'b', 'c'}
An un-escaped double quote or an escaping double quote without a closing escape quote will throw an exception
example csv input row: a,b"c,d\n
current df output: {'a', 'b"c', 'd'}
df output values after PR: FormatException('A field contained an escape...') <- explanatory error message is included
Numeric records accompanied by whitespace will be inferred as Strings with whitespace
example csv input row: a, 1 \n
current df output values: {'a', 1} <--- parsed as int
df output values after PR: {'a', ' 1 '} <---- remains a string