main: Add option to ignore symlinks

When analyzing a complete rootfs (which might not be the rootfs of the
analyzing system) symlink within that rootfs might be broken. In
particular absolute symlinks. However, if by chance such a symlink
currently points to a valid binary in your system, this binary pointed
to is analyzed. This commit adds the possibility to ignore symlinks to
files (symlinks to dirs are already ignored by default). This allows to
solve the issue described above, and if the whole rootfs is analyzed
there shouldn't be a loss of information (because all the binaries will
be analyzed anyway). Additionally, this also saves some time when
performing the analysis.

This commit also involves some refactoring of the walk_filepath_list()
function.

Author: Maximilian Blenk

checksec/__main__.py

@@ -8,6 +8,7 @@
     -w WORKERS --workers=WORKERS    Specify the number of process pool workers [default: 4]
     -j --json                       Display results as JSON
     -s LIBC --set-libc=LIBC         Specify LIBC library to use to check for fortify scores (ELF)
+    -i --ignore-symlinks            Ignore symlinks to files
     -d --debug                      Enable debug output
     -h --help                       Display this message
 """
@@ -27,15 +28,20 @@
 from .utils import lief_set_logging
 
 
-def walk_filepath_list(filepath_list: List[Path], recursive: bool = False) -> Iterator[Path]:
+def walk_filepath_list(filepath_list: List[Path], recursive: bool = False, ignore_symlinks: bool = False) \
+            -> Iterator[Path]:
+    def error_handler(exception):
+        # Fail loudly
+        raise exception
+
     for path in filepath_list:
         if path.is_dir() and not path.is_symlink():
-            if recursive:
-                for f in os.scandir(path):
-                    yield from walk_filepath_list([Path(f)], recursive)
-            else:
-                yield from (Path(f) for f in os.scandir(path))
-        elif path.is_file():
+            for root, _, filenames in os.walk(path, onerror=error_handler):
+                for filename in filenames:
+                    yield from walk_filepath_list([Path(root).joinpath(filename)], recursive, ignore_symlinks)
+                if not recursive:
+                    break
+        elif path.is_file() and (not ignore_symlinks or not path.is_symlink()):
             yield path
 
 
@@ -72,6 +78,7 @@ def main(args):
     json = args["--json"]
     recursive = args["--recursive"]
     libc_path = args["--set-libc"]
+    ignore_symlinks = args["--ignore-symlinks"]
 
     # logging
     formatter = "%(asctime)s %(levelname)s:%(name)s:%(message)s"
@@ -107,7 +114,7 @@ def main(args):
             # we need to consume the iterator once to get the total
             # for the progress bar
             check_output.enumerating_tasks_start()
-            count = sum(1 for i in walk_filepath_list(filepath_list, recursive))
+            count = sum(1 for i in walk_filepath_list(filepath_list, recursive, ignore_symlinks))
             check_output.enumerating_tasks_stop(count)
             with ProcessPoolExecutor(
                 max_workers=workers, initializer=worker_initializer, initargs=(libc_path,)
@@ -116,7 +123,7 @@ def main(args):
                     check_output.processing_tasks_start()
                     future_to_checksec = {
                         pool.submit(checksec_file, filepath): filepath
-                        for filepath in walk_filepath_list(filepath_list, recursive)
+                        for filepath in walk_filepath_list(filepath_list, recursive, ignore_symlinks)
                     }
                     for future in as_completed(future_to_checksec):
                         filepath = future_to_checksec[future]