[
https://issues.apache.org/jira/browse/GROOVY-11200?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17778302#comment-17778302
]
ASF GitHub Bot commented on GROOVY-11200:
-----------------------------------------
yuhengfdada opened a new pull request, #1972:
URL: https://github.com/apache/groovy/pull/1972
# Analysis
As mentioned by Paul's comment in the jira, this is a bug in
`JsonParserCharArray`. The problem does not exist in
`JsonParserUsingCharacterSource`.
After debugging through both classes, both parser methods look exactly the
same, except that `CharacterSource` would automatically add a `10`(line feed)
char at the end, which triggers an exception. But for `JsonParserCharArray` it
ends right at the last curly brace so the exception is overlooked.
# Fix
To keep the behavior consistent across `JsonParserCharArray` and
`JsonParserUsingCharacterSource`, a line feed character is added to the end of
the char array on initialization of `JsonParserCharArray`.
# Concern
Performance
> JsonSlurper parses badly format JSON without throwing exception
> ---------------------------------------------------------------
>
> Key: GROOVY-11200
> URL: https://issues.apache.org/jira/browse/GROOVY-11200
> Project: Groovy
> Issue Type: Bug
> Components: JSON
> Affects Versions: 4.0.14
> Reporter: John Yin
> Priority: Major
>
> JsonSlurper should throw a JsonException when parsing the following input:
> {code:json}
> """
> {
> "a":1,
> "b": {
> "c":2
> }"""
> {code}
>
> But it parses the string successfully. On the other hand, it does throw a
> JsonException when parsing
>
> {code:json}
> """
> {
> "a":1,
> "b": {
> "c":2
> }
> """
> {code}
> where the ending """ is on a separate line.
>
> I think it should fail for both.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
