[
https://issues.apache.org/jira/browse/CAY-2961?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Andrus Adamchik updated CAY-2961:
---------------------------------
Description:
This was either a legacy behavior that we tried to handle or a self-inflicted
injury, but Cayenne explicitly reverse-engineers PostgreSQL "text" column as
CLOB (even though the driver categorizes it as a VARCHAR).
Per PostgreSQL docs, there's really no difference between the two in
performance, and their philosophy is that setting an arbitrary length limit is
meaningless, and hence `text` is slightly preferred as a general String column
and is very common.
CLOBs are both heavy and incorrect here (you can't stream PG text). Let's map
is as VARCHAR instead of CLOB
was:
This was either a legacy behavior that we tried to handle or a self-inflicted
injury, but Cayenne explicitly reverse-engineers PostgreSQL "text" column as
CLOB (even though the driver categorizes it as a VARCHAR).
Per PostgreSQL docs, there's really no difference between the two in
performance, and their philosophy is that setting an arbitrary length limit is
meaningless, and hence `text` is slightly preferred as a general String column
and is very common.
CLOBs are both heavy and incorrect here (you can't stream PG text). Let's map
is as LONGVARCHAR instead of CLOB (just VARCHAR will trigger missing length
warnings on the Cayenne side)
> PostgreSQL "text" column is reverse-engineered as CLOB
> ------------------------------------------------------
>
> Key: CAY-2961
> URL: https://issues.apache.org/jira/browse/CAY-2961
> Project: Cayenne
> Issue Type: Bug
> Reporter: Andrus Adamchik
> Assignee: Andrus Adamchik
> Priority: Major
> Fix For: 5.0-M3
>
>
> This was either a legacy behavior that we tried to handle or a self-inflicted
> injury, but Cayenne explicitly reverse-engineers PostgreSQL "text" column as
> CLOB (even though the driver categorizes it as a VARCHAR).
> Per PostgreSQL docs, there's really no difference between the two in
> performance, and their philosophy is that setting an arbitrary length limit
> is meaningless, and hence `text` is slightly preferred as a general String
> column and is very common.
> CLOBs are both heavy and incorrect here (you can't stream PG text). Let's map
> is as VARCHAR instead of CLOB
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
