Open andrewkroh opened 3 years ago
I was curious so I started looking upstream and found https://github.com/FasterXML/jackson-dataformats-text/issues/233 which appears to have been fixed in 2.12.
Pinging @elastic/es-core-infra (Team:Core/Infra)
Thanks for reporting and the investigation upstream @andrewkroh. It looks like 2.12 was just released a few weeks ago. I'm going to mark this as help wanted
so anyone can pick it up if they feel so empowered. In theory, simply bumping our Jackson version should enable support.
Bumped version of jackson and tested this. Please have a look @elasticmachine
Elasticsearch version (
bin/elasticsearch --version
):Plugins installed: []
JVM version (
java -version
):OS version (
uname -a
if on a Unix-like system): Our official Docker container based on CentOS Linux 8Description of the problem including expected versus actual behavior:
When Elasticsearch unmarshals YAML scalar values in hexadecimal format it interprets the value as a string, but it should be a number. For example
0x1
becomes"0x01"
when it should a1
. This causes problems when you expected the data type to be a number (like when reading the param value within painless).Reference: https://yaml.org/spec/1.2/spec.html#id2766934
Relates: https://github.com/elastic/kibana/issues/85486
Steps to reproduce:
Content-Type: application/yaml
. Include a hexidecimal number.Note that
number_base16
is"0x1"
. It should a 1.