Description of problem: Typically scripts include a line-ending character at the end of their output. Therefore, ProcessExecutionResults.getCapturedOutput() used by the plugin will include line end characters. However, the option 'availabilityOutputRegex' option makes no mention that: 1) The regex must match the entire output of the command, including line end characters 2) And constructs such as .* will not match the end of line characters. For example, if my command outputs 'foo\n' when run, the following will not work: <plugin-configuration> <c:simple-property name="executable" required="true" default="..." /> <c:simple-property name="availabilityOutputRegex" default="foo" /> </plugin-configuration> Nor will: <c:simple-property name="availabilityOutputRegex" default="foo.*" /> A work around is to include (?s) at the beginning <c:simple-property name="availabilityOutputRegex" default="(?s)foo.*" /> Suggested solutions: 1) Trim the whitespace by default from the output. This could possibly break some existing plugin users, but highly unlikely. 2) Include Pattern.DOTALL option by default. 3) Make some note that (?s) should be included if the output contains a line break. 4) Instead of using Pattern.matcher().matches() use Pattern.matcher().find() instead. 1 or 4 is probably the most intuitive solution. Whitespace is rarely important to consider. And just a guess, I doubt this plugin was well tested or this feature used in any case.
*** Bug 848261 has been marked as a duplicate of this bug. ***
This may be related to Bug 837731
trim whitespace/newlines from start/end of output string prior to applying regex filters. git commit to master 49fe82e
Setting back to ASSIGNED as the commit for this issue seems very bad as it changes the output of the script. This can result in failed matches if a valid or proper regular expression is used. Instead of trimming, we should use the correct matcher implementation as we are performing searches within output of a script that is most likely multi-line. This is the suggestion also mentioned in option 4) of comment 0.