utils/decompose_uri: Support URIs containing IPv6 addresses

An URI containing a bracketed or non-bracketed IPv6 address of the form
http://[::1]:42 was previously split on the first colon for port extraction,
causing a subsequent to_count() call to fail. Harden this to check for a
digits in the last :[0-9]+ component.

Fixes #4842

scripts/base/utils/urls.zeek

@@ -117,15 +117,14 @@ function decompose_uri(uri: string): URI
 			}
 		}
 
-	if ( /:/ in s )
+	if ( /:[0-9]*$/ in s )
 		{
-		# Parse location and port.
+		# Input ends with a numeric port or just colon: Strip it
-		parts = split_string1(s, /:/);
+		# for netlocation and convert any port digits into portnum.
-		u$netlocation = parts[0];
+		u$netlocation = gsub(s, /:[0-9]*$/, "");
-		if ( parts[1] != "" )
+		local portstr = s[|u$netlocation| + 1:];
-		    {
+		if ( portstr != "" )
-		    u$portnum = to_count(parts[1]);
+			u$portnum = to_count(portstr);
-		    }
 		}
 	else
 		{

testing/btest/Baseline/scripts.base.utils.decompose_uri/.stderr

@@ -0,0 +1 @@
+### BTest baseline data generated by btest-diff. Do not edit. Use "btest -U/-u" to update. Requires BTest >= 0.63.

testing/btest/Baseline/scripts.base.utils.decompose_uri/output

@@ -50,3 +50,30 @@ file:///documentation/faq.html?=v
 www.bro.org/?foo=bar
     -> [scheme=<uninitialized>, netlocation=www.bro.org, portnum=<uninitialized>, path=/, file_name=<uninitialized>, file_base=<uninitialized>, file_ext=<uninitialized>, params={\x0a\x09[foo] = bar\x0a}]
 
+http://[::1]:8080/?foo=bar&baz=qux
+    -> [scheme=http, netlocation=[::1], portnum=8080, path=/, file_name=<uninitialized>, file_base=<uninitialized>, file_ext=<uninitialized>, params={\x0a\x09[foo] = bar,\x0a\x09[baz] = qux\x0a}]
+
+http://[::1]/foo/bar
+    -> [scheme=http, netlocation=[::1], portnum=<uninitialized>, path=/foo/bar, file_name=bar, file_base=bar, file_ext=<uninitialized>, params=<uninitialized>]
+
+http://[::1]/foo/bar
+    -> [scheme=http, netlocation=[::1], portnum=<uninitialized>, path=/foo/bar, file_name=bar, file_base=bar, file_ext=<uninitialized>, params=<uninitialized>]
+
+[::1]:80/test/a/b.exe?a=b
+    -> [scheme=<uninitialized>, netlocation=[::1], portnum=80, path=/test/a/b.exe, file_name=b.exe, file_base=b, file_ext=exe, params={\x0a\x09[a] = b\x0a}]
+
+http://beeb:deed::1/test
+    -> [scheme=http, netlocation=beeb:deed:, portnum=1, path=/test, file_name=test, file_base=test, file_ext=<uninitialized>, params=<uninitialized>]
+
+http://beeb:deed::1:8080/test
+    -> [scheme=http, netlocation=beeb:deed::1, portnum=8080, path=/test, file_name=test, file_base=test, file_ext=<uninitialized>, params=<uninitialized>]
+
+https://en.wikipedia.org/wiki/Template:Welcome
+    -> [scheme=https, netlocation=en.wikipedia.org, portnum=<uninitialized>, path=/wiki/Template:Welcome, file_name=Template:Welcome, file_base=Template:Welcome, file_ext=<uninitialized>, params=<uninitialized>]
+
+https://[::1]:8080/wiki/Template:Welcome
+    -> [scheme=https, netlocation=[::1], portnum=8080, path=/wiki/Template:Welcome, file_name=Template:Welcome, file_base=Template:Welcome, file_ext=<uninitialized>, params=<uninitialized>]
+
+https://[::1]:8080/wiki/Template:Welcome?key=:&value=:
+    -> [scheme=https, netlocation=[::1], portnum=8080, path=/wiki/Template:Welcome, file_name=Template:Welcome, file_base=Template:Welcome, file_ext=<uninitialized>, params={\x0a\x09[key] = :,\x0a\x09[value] = :\x0a}]
+

testing/btest/scripts/base/utils/decompose_uri.zeek

@@ -1,5 +1,6 @@
 # @TEST-EXEC: zeek -b %INPUT > output
 # @TEST-EXEC: btest-diff output
+# @TEST-EXEC: btest-diff .stderr
 
 @load base/utils/urls
 
@@ -29,5 +30,21 @@ event zeek_init()
 	dc("https://www.bro.org/documentation/faq.html?=v");
 	dc("file:///documentation/faq.html?=v");
 	dc("www.bro.org/?foo=bar");
+
+	# Bracketed IPv6
+	dc("http://[::1]:8080/?foo=bar&baz=qux");
+	dc("http://[::1]/foo/bar");
+	dc("http://[::1]/foo/bar");
+	dc("[::1]:80/test/a/b.exe?a=b");
+
+	# Un-bracketed is ambiguous, but not causing errors.
+	dc("http://beeb:deed::1/test");
+	dc("http://beeb:deed::1:8080/test");
+
+	# Ensure colons in path or query parameters do not
+	# cause trouble.
+	dc("https://en.wikipedia.org/wiki/Template:Welcome");
+	dc("https://[::1]:8080/wiki/Template:Welcome");
+	dc("https://[::1]:8080/wiki/Template:Welcome?key=:&value=:");
 	}