From afa88c025e4b7c195637cf4d1d15fde931e2a6e5 Mon Sep 17 00:00:00 2001
From: Andrew Jackson <Andrew.Jackson@bl.uk>
Date: Mon, 11 Mar 2019 21:35:13 +0000
Subject: [PATCH] Ensure we start parsing full lines, for #239.

---
 .../main/java/org/archive/crawler/framework/CrawlJob.java   | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/engine/src/main/java/org/archive/crawler/framework/CrawlJob.java b/engine/src/main/java/org/archive/crawler/framework/CrawlJob.java
index 0718c292..c9aa8bb8 100644
--- a/engine/src/main/java/org/archive/crawler/framework/CrawlJob.java
+++ b/engine/src/main/java/org/archive/crawler/framework/CrawlJob.java
@@ -172,6 +172,12 @@ public class CrawlJob implements Comparable<CrawlJob>, ApplicationListener<Appli
             BufferedReader jobLogReader = new BufferedReader(
                     new InputStreamReader(jobLogIn));
             String line;
+            // If we sliced into the file, make sure we skip to the next line:
+            // (See https://github.com/internetarchive/heritrix3/issues/239)
+            if (startPosition != 0) {
+                line = jobLogReader.readLine();
+            }
+            // Parse lines looking for launch details:
             while ((line = jobLogReader.readLine()) != null) {
                 Matcher m = launchLine.matcher(line);
                 if (m.matches()) {