[
https://issues.apache.org/jira/browse/SUREFIRE-2212?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17792851#comment-17792851
]
ASF GitHub Bot commented on SUREFIRE-2212:
------------------------------------------
michael-o closed pull request #687: [SUREFIRE-2212] surefire-report-parser:
parse element content only if needed to prevent OOME
URL: https://github.com/apache/maven-surefire/pull/687
> OutOfMemoryError raised when parsing files with huge stderr/stdout output in
> surefire-report-parser
> ---------------------------------------------------------------------------------------------------
>
> Key: SUREFIRE-2212
> URL: https://issues.apache.org/jira/browse/SUREFIRE-2212
> Project: Maven Surefire
> Issue Type: Bug
> Components: surefire-shared-utils
> Affects Versions: 3.2.2
> Reporter: Ramon Bisswanger
> Assignee: Michael Osipov
> Priority: Major
> Fix For: 3.2.3
>
>
> While parsing large reports, we identified {{OutOfMemoryError}} being raised
> in the {{surefire-report-parser}} in case the test file contains large CDATA
> sections below {{system-err}} / {{system-out}} in the report.
> While this is using a SAX parser, it parses all element contents into a
> string and that can cause issues for large test reports.
> Proposed fix: only keep element contents in memory in case the value is used
> somewhere for the result.
> i.e. {{system-err}} / {{system-out}} is just discarded.
> Sample project which is demonstrating the issue by creating a huge dummy test
> report and then uses the parser:
> [https://github.com/bisswanger/surefire-parser-sample] (see README for
> details)
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
