From 8d09cc4e5005b0a1c8bf41918058d2c21c2f773e Mon Sep 17 00:00:00 2001
From: Adam Miller <adam@archive.org>
Date: Tue, 7 Oct 2025 11:04:12 -0700
Subject: [PATCH] feat: Pass along script tag attributes to JS extractor for
 configurable rejection.

---
 .../extractor/ConfigurableExtractorJS.java    | 37 ++++++++++++++++++-
 .../modules/extractor/ExtractorHTML.java      | 21 ++++++++++-
 .../modules/extractor/ExtractorJS.java        |  4 ++
 3 files changed, 59 insertions(+), 3 deletions(-)

diff --git a/modules/src/main/java/org/archive/modules/extractor/ConfigurableExtractorJS.java b/modules/src/main/java/org/archive/modules/extractor/ConfigurableExtractorJS.java
index 9cdc7fc4..b506f2fc 100644
--- a/modules/src/main/java/org/archive/modules/extractor/ConfigurableExtractorJS.java
+++ b/modules/src/main/java/org/archive/modules/extractor/ConfigurableExtractorJS.java
@@ -126,7 +126,31 @@ public class ConfigurableExtractorJS extends ExtractorJS {
     public void addRejectRelativeIgnoreList(String ignoreString) {
 		rejectRelativeIgnoreSet.add(ignoreString);
     }
-    
+
+
+    /**
+     * List of regular expressions that will block extraction if matched against a script tag attribute.
+     * When script blocks are extracted from HTML, the tag attributes are passed as the attributeContext parameter
+     * to considerStrings. If any of the patterns on this list match the attributeContext, then the script block is not
+     * processed.
+     * This can be used to block processing of script blocks
+     */
+    private List<String> rejectScriptAttributeRegexList = new ArrayList<>();
+    private List<Pattern> rejectScriptAttributeRegexListPatterns = new ArrayList<>();
+
+    public List<String> getRejectScriptAttributeRegexList() {
+        return rejectScriptAttributeRegexList;
+    }
+    public void setRejectScriptAttributeRegexList(List<String> patterns) {
+        rejectScriptAttributeRegexList = patterns;
+        rejectScriptAttributeRegexListPatterns = new ArrayList<>();
+        for (String p : patterns) {
+            rejectScriptAttributeRegexListPatterns.add(Pattern.compile(p, Pattern.CASE_INSENSITIVE));
+        }
+    }
+    public void addRejectScriptAttributeRegex(String pattern) {
+        rejectScriptAttributeRegexListPatterns.add(Pattern.compile(pattern, Pattern.CASE_INSENSITIVE));
+    }
     
     @Override
     protected boolean shouldAddUri(CrawlURI curi, String candidate) {
@@ -181,5 +205,14 @@ public class ConfigurableExtractorJS extends ExtractorJS {
 
 		return false;
 	}
-    
+    public long considerStrings(Extractor ext,
+                                CrawlURI curi, CharSequence cs, String attributeContext) {
+        if(!attributeContext.isEmpty())
+            for(Pattern p : rejectScriptAttributeRegexListPatterns) {
+                if(p.matcher(attributeContext).matches()) {
+                    return 0;
+                }
+            }
+        return super.considerStrings(ext, curi, cs, attributeContext);
+    }
 }
diff --git a/modules/src/main/java/org/archive/modules/extractor/ExtractorHTML.java b/modules/src/main/java/org/archive/modules/extractor/ExtractorHTML.java
index 2be77f06..3562611d 100644
--- a/modules/src/main/java/org/archive/modules/extractor/ExtractorHTML.java
+++ b/modules/src/main/java/org/archive/modules/extractor/ExtractorHTML.java
@@ -721,7 +721,6 @@ public class ExtractorHTML extends ContentExtractor implements InitializingBean
         }
     }
 
-
     /**
      * Extract the (java)script source in the given CharSequence.
      *
@@ -735,6 +734,21 @@ public class ExtractorHTML extends ContentExtractor implements InitializingBean
         }
     }
 
+    /**
+     * Extract the (java)script source in the given CharSequence.
+     *
+     * @param curi  source CrawlURI
+     * @param cs    CharSequence of javascript code
+     * @param attributeContext  attributes of the tag from which the script
+     *                          was extracted (e.g. "onload", "onclick", "data-", etc.)
+     */
+    protected void processScriptCode(CrawlURI curi, CharSequence cs, String attributeContext) {
+        if (getExtractorJS() != null && getExtractJavascript()) {
+            numberOfLinksExtracted.addAndGet(
+                    getExtractorJS().considerStrings(this, curi, cs, attributeContext));
+        }
+    }
+
     static final String JAVASCRIPT = "(?i)^javascript:.*";
 
     /**
@@ -1041,6 +1055,11 @@ public class ExtractorHTML extends ContentExtractor implements InitializingBean
 
         // then, apply best-effort string-analysis heuristics
         // against any code present (false positives are OK)
+        if(endOfOpenTag > 6) {
+            String attributes = sequence.subSequence(6, endOfOpenTag).toString().strip();
+            processScriptCode(
+                    curi, sequence.subSequence(endOfOpenTag, sequence.length()), attributes);
+        }
         processScriptCode(
             curi, sequence.subSequence(endOfOpenTag, sequence.length()));
     }
diff --git a/modules/src/main/java/org/archive/modules/extractor/ExtractorJS.java b/modules/src/main/java/org/archive/modules/extractor/ExtractorJS.java
index 530e4c05..703f313b 100644
--- a/modules/src/main/java/org/archive/modules/extractor/ExtractorJS.java
+++ b/modules/src/main/java/org/archive/modules/extractor/ExtractorJS.java
@@ -137,6 +137,10 @@ public class ExtractorJS extends ContentExtractor {
             CrawlURI curi, CharSequence cs) {
         return considerStrings(ext, curi, cs, false);
     }
+    public long considerStrings(Extractor ext,
+            CrawlURI curi, CharSequence cs, String attributeContext) {
+        return considerStrings(ext, curi, cs, false);
+    }
     
     public long considerStrings(Extractor ext, 
             CrawlURI curi, CharSequence cs, boolean handlingJSFile) {