[
https://issues.jboss.org/browse/TEIID-5709?page=com.atlassian.jira.plugin...
]
Steven Hawkins commented on TEIID-5709:
---------------------------------------
Most of our logic is UCS2 based, but this is only noted with regards to Teiid string
sorting behavior. Some routines are surrogate aware, such as Clob upper/lower.
Issues identified so far:
- the Teiid length function reports the UCS2 length (same as H2), but different than
Oracle (which has explicit LENGTH_ functions) and Postgresql
-- similar issues exist with ascii, initCap, translate, trim
- the char type can only hold characters in the bmp.
- XMLFunctions name escaping does not properly handle surrogate pairs
- there may be issues with object names, but that requires more review (SQLStringVisitor,
resolving logic, etc.).
Consistent character handling beyond bmp
----------------------------------------
Key: TEIID-5709
URL:
https://issues.jboss.org/browse/TEIID-5709
Project: Teiid
Issue Type: Bug
Components: Server
Reporter: Steven Hawkins
Assignee: Steven Hawkins
Priority: Major
Fix For: 12.2
There are many places in the code that only consider each 16 bit character when we should
consult the full code point.
--
This message was sent by Atlassian Jira
(v7.12.1#712002)