Implementation of crawling results normalization

stamparm · stamparm · commit a660828cec15 · 2019-10-31T22:07:16.000+01:00
diff --git a/lib/core/option.py b/lib/core/option.py
@@ -1953,6 +1953,7 @@ def _setKnowledgeBaseAttributes(flushAll=True):
     kb.mergeCookies = None
     kb.multipleCtrlC = False
     kb.negativeLogic = False
+    kb.normalizeCrawlingChoice = None
     kb.nullConnection = None
     kb.oldMsf = None
     kb.orderByColumns = None
diff --git a/lib/core/settings.py b/lib/core/settings.py
@@ -18,7 +18,7 @@
 from thirdparty.six import unichr as _unichr
 
 # sqlmap version (<major>.<minor>.<month>.<monthly commit>)
-VERSION = "1.3.10.38"
+VERSION = "1.3.10.39"
 TYPE = "dev" if VERSION.count('.') > 2 and VERSION.split('.')[-1] != '0' else "stable"
 TYPE_COLORS = {"dev": 33, "stable": 90, "pip": 34}
 VERSION_STRING = "sqlmap/%s#%s" % ('.'.join(VERSION.split('.')[:-1]) if VERSION.count('.') > 2 and VERSION.split('.')[-1] == '0' else VERSION, TYPE)
diff --git a/lib/utils/crawler.py b/lib/utils/crawler.py
@@ -195,6 +195,26 @@ def crawlThread():
             for url in threadData.shared.value:
                 kb.targets.add((urldecode(url, kb.pageEncoding), None, None, None, None))
 
+        if kb.normalizeCrawlingChoice is None:
+            message = "do you want to normalize "
+            message += "crawling results [Y/n] "
+
+            kb.normalizeCrawlingChoice = readInput(message, default='Y', boolean=True)
+
+        if kb.normalizeCrawlingChoice:
+            seen = set()
+            results = OrderedSet()
+
+            for target in kb.targets:
+                match = re.search(r"/[^/?]*\?.*\Z", target[0])
+                if match:
+                    key = re.sub(r"=[^=&]*", "=", match.group(0))
+                    if key not in seen:
+                        results.add(target)
+                        seen.add(key)
+
+            kb.targets = results
+
         storeResultsToFile(kb.targets)
 
 def storeResultsToFile(results):
