ENH: improve performance of df.to_csv GH3054#3059

Merged

27 commits merged intomasterfrom

unknown repository

Mar 19, 2013

Commits on Mar 15, 2013

ENH: improve performance of df.to_csv GH3054
y-p
committed
ENH: to_csv using masking to simplify dtype processing

jreback
authored and
y-p
committed

Commits on Mar 16, 2013

ENH: more perf tweaks in df.to_csv
y-p
committed
PERF: cythonized parts of to_csv for increased perf

jreback
authored and
y-p
committed
PERF: more cython tweaks

jreback
authored and
y-p
committed
PERF: cythonize improved python version
y-p
committed
PERF: cythonized parts of to_csv for increased perf
jreback
committed
PERF: more speedups
jreback
committed
ENH: add chunksize parameter to DataFrame.to_csv to enable constant memory usage
jreback
committed
CLN: move repeated cast out of loop
y-p
committed
CLN: make guard more defensive
y-p
committed
REF: add com._ndarray_to_native_types

jreback
authored and
y-p
committed
REF: apply native type conv to ix, cols before write_csv
y-p
committed
PERF: added frame_to_csv2 vbench, revised frame_to_csv_mixed
jreback
committed

Commits on Mar 17, 2013

TST: test for to_csv on failing vbench

jreback
authored and
y-p
committed
ENH: refactor series from dict to list, eliminate one level of indirection

jreback
authored and
y-p
committed
ENH: replace variable lookup by constant .
y-p
committed
ENH: make chunks process constant element count

jreback
authored and
y-p
committed
PERF: avoid iteritems->iloc panelty for data conversion, use blocks
y-p
committed
TST: test for to_csv on failing vbench
jreback
committed
CLN: csv refactor

jreback
authored and
y-p
committed

Commits on Mar 18, 2013

ENH: add (undocumented) legacy kwd to df.to_csv, just in case
y-p
committed
TST: fail early on duplicate columns

jreback
authored and
y-p
committed

Commits on Mar 19, 2013