Fix segmentation fault in extract_from_url (#23)

Author: mkhazin-r7

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "pydomainextractor"
-version = "0.13.4"
+version = "0.13.5"
 authors = ["Gal Ben David <gal@intsights.com>"]
 edition = "2021"
 description = "A blazingly fast domain extraction library written in Rust"

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "pydomainextractor"
-version = "0.13.4"
+version = "0.13.5"
 description = "A blazingly fast domain extraction library written in Rust"
 authors = [
   {email = "gal@intsights.com"},

src/lib.rs

@@ -254,6 +254,9 @@ impl DomainExtractor {
             );
         }
 
+        if url_str.len() > 255 {
+            return Err(PyValueError::new_err("url is invalid: too long"));
+        }
         let mut domain_string = unsafe {
             DomainString::from_str_unchecked(url_str)
         };

tests/test_pydomainextractor.py

@@ -301,6 +301,14 @@ def test_syntactic_invalid_domains(
         ):
             self.domain_extractor.extract('com.')
 
+    def test_domain_too_long(
+        self,
+    ):
+        with self.assertRaises(
+            expected_exception=ValueError,
+        ):
+            self.domain_extractor.extract(f'{"very-long" * 255}.com')
+
     def test_extract_from_url(
         self,
     ):
@@ -349,6 +357,11 @@ def test_extract_from_url(
         ):
             self.domain_extractor.extract_from_url('co.uk')
 
+        with self.assertRaises(
+            ValueError,
+        ):
+            self.domain_extractor.extract_from_url(f'http://{"domain" * 255}co.uk:3030/some/path')
+
         self.assertEqual(
             first=self.domain_extractor.extract_from_url('http://www.google.com'),
             second={