From 34624da8abc16dbfb40cd4685d9025bb776048f4 Mon Sep 17 00:00:00 2001
From: Ryan Brandenburg <rybrande@microsoft.com>
Date: Fri, 19 Dec 2025 11:13:54 -0800
Subject: [PATCH 1/2] Handle large json failures

---
 providers/process/scancode.js | 36 ++++++++++++++++++++++-------------
 1 file changed, 23 insertions(+), 13 deletions(-)

diff --git a/providers/process/scancode.js b/providers/process/scancode.js
index c32e0440..76c1fdbb 100644
--- a/providers/process/scancode.js
+++ b/providers/process/scancode.js
@@ -59,6 +59,10 @@ class ScanCodeProcessor extends AbstractProcessor {
         request.markDead('Error', error ? error.message : 'ScanCode run failed')
         throw error
       }
+      else {
+        this.logger.error(error, request.meta)
+        request.markRequeue('ScanCode Error', error.message)
+      }
     }
   }
 
@@ -91,19 +95,25 @@ class ScanCodeProcessor extends AbstractProcessor {
   // Scan the results file for any errors that are not just timeouts or other known errors
   // TODO do we need to do this anymore
   _hasRealErrors(resultFile) {
-    const results = JSON.parse(fs.readFileSync(resultFile))
-    return results.files.some(
-      file =>
-        file.scan_errors &&
-        file.scan_errors.some(error => {
-          return !(
-            error.includes('ERROR: Processing interrupted: timeout after') ||
-            error.includes('ValueError:') ||
-            error.includes('package.json') ||
-            error.includes('UnicodeDecodeError')
-          )
-        })
-    )
+    try{
+      const results = JSON.parse(fs.readFileSync(resultFile))
+      return results.files.some(
+        file =>
+          file.scan_errors &&
+          file.scan_errors.some(error => {
+            return !(
+              error.includes('ERROR: Processing interrupted: timeout after') ||
+              error.includes('ValueError:') ||
+              error.includes('package.json') ||
+              error.includes('UnicodeDecodeError')
+            )
+          })
+      )
+    } catch (e) {
+      // This might happen if the results file is >512mb, but regardless our error handling should not fail.
+      this.logger.error(`Could not parse ScanCode results file ${resultFile}: ${e.message}`)
+      return true
+    }
   }
 
   _detectVersion() {

From 8125198840277655645ef41c9b7b2f6b69f27140 Mon Sep 17 00:00:00 2001
From: Ryan Brandenburg <rybrande@microsoft.com>
Date: Fri, 19 Dec 2025 11:24:28 -0800
Subject: [PATCH 2/2] Cleanup

---
 providers/process/scancode.js | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/providers/process/scancode.js b/providers/process/scancode.js
index 76c1fdbb..7bc84ba2 100644
--- a/providers/process/scancode.js
+++ b/providers/process/scancode.js
@@ -54,15 +54,12 @@ class ScanCodeProcessor extends AbstractProcessor {
         maxBuffer: 5 * 1024 * 1024
       })
     } catch (error) {
+      this.logger.error(error, request.meta)
       // TODO see if the new version of ScanCode has a better way of differentiating errors
       if (this._isRealError(error) || this._hasRealErrors(file.name)) {
         request.markDead('Error', error ? error.message : 'ScanCode run failed')
         throw error
       }
-      else {
-        this.logger.error(error, request.meta)
-        request.markRequeue('ScanCode Error', error.message)
-      }
     }
   }
 
@@ -112,6 +109,7 @@ class ScanCodeProcessor extends AbstractProcessor {
     } catch (e) {
       // This might happen if the results file is >512mb, but regardless our error handling should not fail.
       this.logger.error(`Could not parse ScanCode results file ${resultFile}: ${e.message}`)
+      // Assume it's broken if we could not parse it
       return true
     }
   }