[
https://issues.apache.org/jira/browse/ARROW-7661?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17025076#comment-17025076
]
Takuya Kato commented on ARROW-7661:
------------------------------------
I am now able to reproduce the problem by removing the last newline character
of the input file.
{code:bash}
$ vim /tmp/test.csv
$ cat /tmp/test.csv
"Name","Month","Change in %"
"Surrey Quays","Sep 18","1.01"
"Surrey Quays","Oct 18","0.38"
"Surrey Quays","Nov 18","0.97"
"Surrey Quays","Dec 18","1.28"
"Surrey Quays","Jan 19","2.43"
"Surrey Quays","Feb 19","2.49"
"Surrey Quays","Mar 19","0.81"
$ tail -c 1 /tmp/test.csv | xxd
00000000: 0a # ascii newline characeter
$ head -c 245 /tmp/test.csv > /tmp/test2.csv
$ tail -n 1 /tmp/test2.csv
"Surrey Quays","Mar 19","0.81"$
$ python3
>>> import os
>>> path = "/tmp/test2.csv"
>>> read_options = pc.ReadOptions(block_size=os.stat(path).st_size)
>>> df = pc.read_csv(path, read_options=read_options)
>>> df.to_batches()
[<pyarrow.lib.RecordBatch object at 0x7f1e99b956d8>, <pyarrow.lib.RecordBatch
object at 0x7f1e8d48fb38>]
{code}
Though, I'm not sure whether this should be fixed or not.
> [Python] Specify number of batches for read_csv
> -----------------------------------------------
>
> Key: ARROW-7661
> URL: https://issues.apache.org/jira/browse/ARROW-7661
> Project: Apache Arrow
> Issue Type: Bug
> Components: Python
> Affects Versions: 0.14.1, 0.15.0, 0.15.1
> Reporter: Sascha Hofmann
> Priority: Major
>
> We are reading a very simple csv (see below).
> The file is only 245 bytes so way below the default _block_size_ in the
> _ReadOptions_. Thus we would expect the resulting table to have only one
> batch. At least, if I understand correctly that a _block_ refers to the
> number of lines of certain byte size?
> The docs state: _This will determine multi-threading granularity as well as
> the size of individual chunks in the Table._ For me, that means also the size
> of individual batches?
> Previously, we thought by fixing the block_size to the total file size, we
> would ensure that even for files larger than 1MB we get a pa.Table with only
> one batch. This mini file seems to prove us wrong?
> Additionally, if I convert back and forth to pandas we get only one batch.
>
> To reproduce:
> {code:java}
> import os
> from pyarrow import csv as pc
> import pyarrow as pa
> path = "test.csv"
> read_options = pc.ReadOptions(block_size=os.stat(path).st_size)
> df = pc.read_csv(path, read_options=read_options)
> print(len(df.to_batches()))
> # returns 2
> print(pa.Table.from_batches([df.to_batches()[1]]).to_pandas())
> # returns the last line of the file
> pdf = df.to_pandas()
> ndf = pa.Table.from_pandas(pdf)
> print(len(ndf.to_batches()))
> # returns 1{code}
> test.csv:
> {code:java}
> "Name","Month","Change in %"
> "Surrey Quays","Sep 18","1.01"
> "Surrey Quays","Oct 18","0.38"
> "Surrey Quays","Nov 18","0.97"
> "Surrey Quays","Dec 18","1.28"
> "Surrey Quays","Jan 19","2.43"
> "Surrey Quays","Feb 19","2.49"
> "Surrey Quays","Mar 19","0.81"
> {code}
>
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
