Package io.jenkins.plugins.explain_error

Class PipelineLogExtractor

java.lang.Object
io.jenkins.plugins.explain_error.PipelineLogExtractor
public class PipelineLogExtractor extends Object
Utility for extracting log lines related to a failing build or pipeline step and computing a URL that points back to the error source.

For WorkflowRun (Pipeline) builds, this class walks the flow graph to locate the node that originally threw the error, reads a limited number of log lines from that step, and records a node-specific URL that can be used to navigate to the failure location. When no failing step log can be found, or when the build is not a pipeline, it falls back to the standard build console log.

If the optional pipeline-graph-view plugin is installed, the generated URL is compatible with its overview page so that consumers can deep-link directly into the failing node from error explanations.

  • Field Details

  • Constructor Details

    • PipelineLogExtractor

      public PipelineLogExtractor(Run<?,?> run, int maxLines)

  • Method Details

    • getFailedStepLog

      public List<String> getFailedStepLog() throws IOException
      Extracts the log output of the step(s) that caused the pipeline failure, combining results from multiple strategies so that parallel failures (e.g. both a Rspec test failure and a RuboCop offense) are all captured.
      1. Strategy 1 — ErrorAction multi-collect: walks the FlowGraph and collects logs from all nodes with ErrorAction and an associated LogAction (explicit uncaught exceptions). Unlike the original single-return approach, this accumulates logs from every failing step up to maxLines total, covering parallel failures such as multiple Rspec pod crashes.
      2. Strategy 3 — Error pattern scan (always runs as supplement): reads the full build console log and appends lines matching common error patterns (with surrounding context) that were not already captured by Strategy 1. This fills the gap left by catchError + sh(returnStatus:true) + error() pipelines (where no LogAction exists on the error() step) and catches errors that appear early in large build logs.
      Falls back to run.getLog(maxLines) (last N lines of console) only if all strategies produce no results.
      Returns:
      A non-null list of log lines combining all relevant failure output, capped at maxLines.
      Throws:
      IOException - if there is an error reading the build logs.

    • getUrl

      public String getUrl()
      Returns the URL associated with the extracted log.

      When getFailedStepLog() finds a failed pipeline step with an attached log and the pipeline-graph-view plugin is available, this will point to the pipeline overview page with the failing node preselected. Otherwise, it falls back to the build's console output URL.

      Returns:
      the Jenkins URL for either the pipeline overview of the failing step or the build console output, or null if getFailedStepLog() has not been invoked successfully.