DX-61115 handle + sign during parsing +infinity double value #31
Conversation
|
Thanks for opening a pull request! If this is not a minor PR. Could you open an issue for this pull request on GitHub? https://github.com/apache/arrow/issues/new/choose Opening GitHub issues ahead of time contributes to the Openness of the Apache Arrow project. Then could you also rename the pull request title in the following format? or In the case of PARQUET issues on JIRA the title also supports: See also: |
| @@ -30,6 +30,9 @@ from_chars_result parse_infnan(const char *first, const char *last, T &value) n | |||
| minusSign = true; | |||
| ++first; | |||
| } | |||
| if (*first == '+') { | |||
There was a problem hiding this comment.
The comment at line 29 "C++17 20.19.3.(7.1) explicitly forbids '+' here" seems to be outdated
There was a problem hiding this comment.
not really, in fast_float version of 3.10.1( current is 3.8.1) https://github.com/fastfloat/fast_float/blob/main/include/fast_float/parse_number.h
was added
#ifdef FASTFLOAT_ALLOWS_LEADING_PLUS // disabled by default
if (*first == UC('+')) {
++first;
}
I created pr in apache arrow for updating fast_float version but i don't know when it will merged or will it be merged at all. So for now we could use this.
There was a problem hiding this comment.
and i don't want to remove any comments because it's a custom lib, and will be conflict on merging
Apache Arrow used [https://github.com/fastfloat/fast_float](fast_float lib). In latest version was added same code but in preprocessing block that undefined by default. Updating to latest version require al lot of code changes inside arrow, ci. So it will be easier to add this block directly to dremio until the version will be updated in apache arrow repo.