From b12c3155ef348392c36df1d157317443006486f7 Mon Sep 17 00:00:00 2001
From: Anush Loknath <anushloknath@gmail.com>
Date: Tue, 6 Oct 2026 11:47:08 +0530
Subject: [PATCH] Fix recon feature detection for quoted filenames

---
 openhack/deterministic_recon.py   | 15 +++++++------
 tests/test_deterministic_recon.py | 37 ++++++++++++++++++++++++++++---
 2 files changed, 42 insertions(+), 10 deletions(-)

diff --git a/openhack/deterministic_recon.py b/openhack/deterministic_recon.py
index f8b7c07..c0929a1 100644
--- a/openhack/deterministic_recon.py
+++ b/openhack/deterministic_recon.py
@@ -145,6 +145,8 @@ def _detect_features_fast(
             find_cmd.append("-o")
         find_cmd.extend(["-name", f"*{ext}"])
     find_cmd.append(")")
+    # Keep filenames literal across find, xargs, and grep.
+    find_cmd.append("-print0")
     for d in _GREP_EXCLUDE_DIRS:
         clean = d.rstrip("*").rstrip(".")
         find_cmd[2:2] = ["-not", "-path", f"*/{clean}/*"]
@@ -158,10 +160,10 @@ def _detect_features_fast(
         logger.warning(f"Find failed: {e}")
         return {}
 
-    if not file_list.strip():
+    if not file_list:
         return {}
 
-    file_count = file_list.count("\n")
+    file_count = file_list.count("\0")
 
     if file_count > 5000:
         # Large repo: fixed-string grep in parallel threads
@@ -179,7 +181,7 @@ def _detect_features_fast(
 
         def _check_feature(name_and_keywords):
             fname, keywords = name_and_keywords
-            cmd = ["xargs", "grep", "-Fl", "--max-count=1",
+            cmd = ["xargs", "-0", "grep", "-FlZ", "--max-count=1",
                    "--binary-files=without-match"]
             for kw in keywords:
                 cmd.extend(["-e", kw])
@@ -188,7 +190,7 @@ def _check_feature(name_and_keywords):
                     cmd, input=file_list, capture_output=True,
                     text=True, timeout=15,
                 )
-                matches = [l for l in proc.stdout.strip().split("\n") if l.strip()]
+                matches = [fp for fp in proc.stdout.split("\0") if fp]
                 if matches:
                     readable = fname.replace("_", " ").title()
                     return fname, [f"{readable} ({len(matches)} files)"]
@@ -210,7 +212,7 @@ def _check_feature(name_and_keywords):
     result: dict[str, list[str]] = {}
     for feature_name, patterns in feature_indicators.items():
         combined = "|".join(p for p, _ in patterns)
-        cmd_parts = ["xargs", "grep", "-El", "--max-count=1",
+        cmd_parts = ["xargs", "-0", "grep", "-ElZ", "--max-count=1",
                      "--binary-files=without-match", combined]
         try:
             proc = subprocess.run(
@@ -218,8 +220,7 @@ def _check_feature(name_and_keywords):
                 text=True, timeout=30,
             )
             files = []
-            for line in proc.stdout.strip().split("\n"):
-                fp = line.strip()
+            for fp in proc.stdout.split("\0"):
                 if not fp:
                     continue
                 try:
diff --git a/tests/test_deterministic_recon.py b/tests/test_deterministic_recon.py
index 3ed45cf..4c65a88 100644
--- a/tests/test_deterministic_recon.py
+++ b/tests/test_deterministic_recon.py
@@ -1,9 +1,15 @@
-import json
-from pathlib import Path
+import os
+
+import pytest
 from tests.conftest import write_file, write_json
 
 from openhack.tools.registry import ToolRegistry
-from openhack.deterministic_recon import run_deterministic_recon
+from openhack.deterministic_recon import (
+    _FEATURE_INDICATORS,
+    _detect_features_fast,
+    run_deterministic_recon,
+)
+from openhack.tools.filesystem import FileSystemTools
 
 
 class TestFeatureDetection:
@@ -53,3 +59,28 @@ def test_returns_expected_keys(self, tmp_path):
         assert "type" in result
         assert result["type"] == "recon_complete"
         assert "attack_surface" in result
+
+
+class TestFeatureDetectionPaths:
+    @pytest.mark.parametrize("file_count", [2, 5001], ids=["regex", "fixed-string"])
+    def test_quoted_paths_preserve_all_features(self, tmp_path, file_count):
+        write_file(tmp_path, "src/owner's uploads/don't_break.js", "multer();\n")
+        write_file(tmp_path, "auth.js", "bcrypt();\n")
+        for index in range(file_count - 2):
+            write_file(tmp_path, f"src/empty_{index}.js", "")
+
+        features = _detect_features_fast(FileSystemTools(tmp_path), _FEATURE_INDICATORS)
+
+        assert features == {
+            "file_uploads": ["File Uploads (1 files)"],
+            "auth_system": ["Auth System (1 files)"],
+        }
+
+    @pytest.mark.skipif(os.name == "nt", reason="Requires POSIX filename characters")
+    @pytest.mark.parametrize("filename", ['double"quote.js', "back\\slash.js", "new\nline.js", "tab\tname.js"])
+    def test_special_filename_is_counted_once(self, tmp_path, filename):
+        write_file(tmp_path, filename, "multer();\n")
+
+        features = _detect_features_fast(FileSystemTools(tmp_path), _FEATURE_INDICATORS)
+
+        assert features == {"file_uploads": ["File Uploads (1 files)"]}