At this moment BFA plugin has the risk of matching extreamly long lines (>100kb or even megabytes) by mistake. This is happening with tools that to join console output to a single line using separators. A notable example is the JSON output of Ansible which contains the console output as a JSON string.

This problem could be avoided if BFA would have either a global option line MAX_LINE_LENGTH or another option like MAX_MATCH_LENGTH. Either of them could be used to avoid accidents when the matching line is too big.

Please not that this bug has performance and stability implications because matching and processing very long lines requires high resources, both memory and CPU.

Another side effect of this bug is that a very long match line would break other features, like the ability to add comments to gerrit reviews. 

Some people could argue that it should be possible to limit the match length from the regular expression but this approach does not work in real life, where you can have like 100 patterns which are managed by a big number of jenkins users and nobody can police them to assure they all use matching limits.