[
https://issues.apache.org/jira/browse/SOLR-6785?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18132090#comment-18132090
]
Nick Shanin commented on SOLR-6785:
-----------------------------------
>From code reading against apache/main b73f03976fa, the CSV writer escapes
>backslashes by default for polyfields
>(solr/core/src/java/org/apache/solr/response/CSVResponseWriter.java:303-306).
>A candidate is to document that default rather than change CSV output. Is that
>what you need?
🤖 AI-generated 🤖
> polyfields are encapsulated and escaped in csv
> ----------------------------------------------
>
> Key: SOLR-6785
> URL: https://issues.apache.org/jira/browse/SOLR-6785
> Project: Solr
> Issue Type: Bug
> Components: update
> Reporter: Michael Lawrence
> Priority: Major
>
> The wiki for UpdateCSV (which also presumably applies to the CSV response
> writer) states: "If an escape is specified, the encapsulator is not used
> unless also explicitly specified since most formats use either encapsulation
> or escaping, not both." That makes a lot of sense. However, the fix for
> SOLR-3959 makes it so that polyfields are always escaped, even when
> enscapsulation is enabled. Are we sure that SOLR-3959 was even a bug in the
> first place? Why should commas in the output of polyfields be treated any
> differently than commas elsewhere? One possible enhancement to the interface
> would be to use "\" as the separator by default when encapsulation is
> disabled, but I propose starting by reverting the "fix" for SOLR-3959.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
