[
https://issues.apache.org/jira/browse/IMPALA-10350?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17316999#comment-17316999
]
Zoltán Borók-Nagy commented on IMPALA-10350:
--------------------------------------------
[~amargoor] I think strtod is fine, we just hit the limitations of double
precision with the value -0.43149576573887374.
[https://onlinegdb.com/Bk90zB2rd] (C++17)
[https://onlinegdb.com/ByecxQHhBO] (Java)
Lemire's algorithm has a fast path that can be used in most cases:
[https://github.com/lemire/fast_double_parser/blob/e4f6319bfa9cbc829f7f99ae88c1d2fb205c15e8/include/fast_double_parser.h#L893]
It uses a similar representation that Impala is using, i.e. an integer + scale
(power).
It also has a secondary fast path:
[https://github.com/lemire/fast_double_parser/blob/e4f6319bfa9cbc829f7f99ae88c1d2fb205c15e8/include/fast_double_parser.h#L921]
And if compute_float_64() fails it falls back to strtod:
[https://github.com/lemire/fast_double_parser/blob/e4f6319bfa9cbc829f7f99ae88c1d2fb205c15e8/include/fast_double_parser.h#L1254-L1257]
Probably we could try to use compute_float_64() and when it fails we could just
fall back similarly.
Based on my previous comment google/wuffs uses a different represantation, i.e.
we'd need to generate the string representation of the decimal value first.
> Impala loses double precision because of DECIMAL->DOUBLE cast
> -------------------------------------------------------------
>
> Key: IMPALA-10350
> URL: https://issues.apache.org/jira/browse/IMPALA-10350
> Project: IMPALA
> Issue Type: Bug
> Components: Backend
> Reporter: Zoltán Borók-Nagy
> Assignee: Amogh Margoor
> Priority: Major
> Labels: correctness, ramp-up
> Attachments: test.c
>
>
> Impala might loses presision of double values. Reproduction:
> {noformat}
> create table double_tbl (d double) stored as textfile;
> insert into double_tbl values (-0.43149576573887316);
> {noformat}
> Then inspect the data file:
> {noformat}
> $ hdfs dfs -cat
> /test-warehouse/double_tbl/424097c644088674-c55b910100000000_175064830_data.0.txt
> -0.4314957657388731{noformat}
> The same happens if we store our data in Parquet.
> Hive writes don't lose precision. If the data was written by Hive then Impala
> can read the values correctly:
> {noformat}
> $ bin/run-jdbc-client.sh -t NOSASL -q "select * from double_tbl;"
> Using JDBC Driver Name: org.apache.hive.jdbc.HiveDriver
> Connecting to: jdbc:hive2://localhost:21050/;auth=noSasl
> Executing: select * from double_tbl
> ----[START]----
> -0.43149576573887316
> ----[END]----{noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
