From 741d95f428514448e4dc25e1c0dec724e47c6fb2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Adri=C3=A1n=20Chaves?= <adrian@chaves.io>
Date: Tue, 8 Aug 2023 11:51:36 +0200
Subject: [PATCH] Only prefix absolute URLs with / (#34)

---
 src/protego.py        | 14 +++++---------
 tests/test_protego.py | 16 +++++++++++++++-
 2 files changed, 20 insertions(+), 10 deletions(-)

diff --git a/src/protego.py b/src/protego.py
index 3ee76a6..3b2c0fe 100644
--- a/src/protego.py
+++ b/src/protego.py
@@ -49,13 +49,6 @@ def _is_valid_directive_field(field):
     )
 
 
-def _enforce_path(pattern):
-    if pattern.startswith("/"):
-        return pattern
-
-    return "/" + pattern
-
-
 class _URLPattern(object):
     """Internal class which represents a URL pattern."""
 
@@ -179,6 +172,9 @@ def _quote_path(self, path):
         return path or "/"
 
     def _quote_pattern(self, pattern):
+        if pattern.startswith("https://") or pattern.startswith("http://"):
+            pattern = "/" + pattern
+
         # Corner case for query only (e.g. '/abc?') and param only (e.g. '/abc;') URLs.
         # Save the last character otherwise, urlparse will kill it.
         last_char = ""
@@ -444,11 +440,11 @@ def _parse_robotstxt(self, content):
 
             elif field in _ALLOW_DIRECTIVE:
                 for rule_set in current_rule_sets:
-                    rule_set.allow(_enforce_path(value))
+                    rule_set.allow(value)
 
             elif field in _DISALLOW_DIRECTIVE:
                 for rule_set in current_rule_sets:
-                    rule_set.disallow(_enforce_path(value))
+                    rule_set.disallow(value)
 
             elif field in _SITEMAP_DIRECTIVE:
                 self._sitemap_list.append(value)
diff --git a/tests/test_protego.py b/tests/test_protego.py
index df12384..3571d11 100644
--- a/tests/test_protego.py
+++ b/tests/test_protego.py
@@ -1,7 +1,8 @@
-# encoding=utf-8
 from datetime import time
 from unittest import TestCase
 
+import pytest
+
 from protego import Protego, _RuleSet
 
 
@@ -1139,3 +1140,16 @@ def test_parse_time_period(self):
         start_time, end_time = rs._parse_time_period("0500 0600", separator=" ")
         self.assertEqual(start_time, time(5, 0))
         self.assertEqual(end_time, time(6, 0))
+
+
+@pytest.mark.parametrize(
+    "allow,disallow,url,allowed",
+    [
+        ("*/p", "/", "http://example.com/page", True),
+        ("/page", "*/*.htm", "https://example.com/page.htm", False),
+    ],
+)
+def test_leading_asterisk(allow, disallow, url, allowed):
+    content = f"User-Agent: *\n" f"allow: {allow}\n" f"disallow: {disallow}\n"
+    rp = Protego.parse(content)
+    assert rp.can_fetch(url, "*") == allowed