refactor: Drop fact_helper_file dependency

It is replaced by the much simpler magic.py

docsrc/conf.py

@@ -69,7 +69,6 @@
     'docker',
     'docker-compose',
     'email-validator',
-    'fact_helper_file',
     'flaky',
     'flask',
     'flask_login',

src/install/requirements_common.txt

@@ -27,8 +27,6 @@ pydantic==2.1.1
 # Config parsing
 toml==0.10.2
 
-git+https://github.com/fkie-cad/fact_helper_file.git
-
 # Common code modules
 git+https://github.com/fkie-cad/common_helper_files.git
 git+https://github.com/fkie-cad/common_helper_filter.git

src/plugins/analysis/file_type/code/file_type.py

@@ -4,11 +4,11 @@
 from typing import List
 
 import pydantic
-from fact_helper_file import get_file_type_from_path
 from pydantic import Field
 
 from analysis.plugin import AnalysisPluginV0
 from analysis.plugin.compat import AnalysisBasePluginAdapterMixin
+from helperFunctions import magic
 
 if typing.TYPE_CHECKING:
     import io
@@ -39,9 +39,7 @@ def summarize(self, result: Schema) -> List[str]:
     def analyze(self, file_handle: io.FileIO, virtual_file_path: str, analyses: dict) -> Schema:
         del virtual_file_path, analyses
 
-        file_dict = get_file_type_from_path(file_handle.name)
-
         return AnalysisPlugin.Schema(
-            mime=file_dict['mime'],
-            full=file_dict['full'],
+            mime=magic.from_file(file_handle.name, mime=True),
+            full=magic.from_file(file_handle.name, mime=False),
         )

src/plugins/analysis/qemu_exec/code/qemu_exec.py

@@ -16,11 +16,11 @@
 from common_helper_files import get_binary_from_file, safe_rglob
 from docker.errors import DockerException
 from docker.types import Mount
-from fact_helper_file import get_file_type_from_path
 from requests.exceptions import ReadTimeout
 
 import config
 from analysis.PluginBase import AnalysisBasePlugin
+from helperFunctions import magic
 from helperFunctions.docker import run_docker_container
 from helperFunctions.tag import TagColor
 from helperFunctions.uid import create_uid
@@ -125,7 +125,10 @@ def _find_relevant_files(self, extracted_files_dir: Path):
         result = []
         for path in safe_rglob(extracted_files_dir):
             if path.is_file() and not path.is_symlink():
-                file_type = get_file_type_from_path(path.absolute())
+                file_type = {
+                    'full': magic.from_file(path.absolute(), mime=False),
+                    'mime': magic.from_file(path.absolute(), mime=True),
+                }
                 if self._has_relevant_type(file_type):
                     result.append((f'/{path.relative_to(Path(self.root_path))}', file_type['full']))
         return result

src/test/acceptance/test_io_routes.py

@@ -1,8 +1,8 @@
 import pytest
-from fact_helper_file import get_file_type_from_binary
 
+from helperFunctions import magic
 from storage.db_interface_comparison import ComparisonDbInterface
-from test.common_helper import create_test_firmware
+from test.common_helper import create_test_firmware  # pylint: disable=wrong-import-order
 
 COMPARE_RESULT = {
     'general': {'a': {'id1': '<empty>', 'id2': '<empty>'}, 'b': {'id1': '<empty>', 'id2': '<empty>'}},
@@ -68,4 +68,4 @@ def test_pdf_download(self, test_client, backend_db):
         assert response.status_code == 200, 'pdf download failed'  # noqa: PLR2004
         device = self.test_fw.device_name.replace(' ', '_')
         assert response.headers['Content-Disposition'] == f'attachment; filename={device}_analysis_report.pdf'
-        assert get_file_type_from_binary(response.data)['mime'] == 'application/pdf'
+        assert magic.from_buffer(response.data) == 'application/pdf'

src/test/integration/helperFunctions/test_pdf.py

@@ -2,8 +2,7 @@
 import os
 from pathlib import Path
 
-from fact_helper_file import get_file_type_from_binary
-
+from helperFunctions import magic
 from helperFunctions.pdf import build_pdf_report
 from test.common_helper import TEST_FW
 
@@ -21,5 +20,5 @@ def test_build_pdf_report():
 
     pdf_path = build_pdf_report(TEST_FW, docker_mount_base_dir)
 
-    assert get_file_type_from_binary(pdf_path.read_bytes())['mime'] == 'application/pdf'
+    assert magic.from_buffer(pdf_path.read_bytes(), mime=True) == 'application/pdf'
     assert pdf_path.name == f"{TEST_FW.device_name.replace(' ', '_')}_analysis_report.pdf"

src/unpacker/unpack.py

@@ -6,10 +6,9 @@
 from time import time
 from typing import TYPE_CHECKING, Optional
 
-from fact_helper_file import get_file_type_from_path
-
 import config
 from analysis.PluginBase import sanitize_processed_analysis
+from helperFunctions import magic
 from helperFunctions.fileSystem import file_is_empty, get_relative_object_path
 from helperFunctions.tag import TagColor
 from objects.file import FileObject
@@ -94,7 +93,8 @@ def generate_objects_and_store_files(
                 continue
             current_file = FileObject(file_path=str(path))
             current_virtual_path = get_relative_object_path(path, extraction_dir)
-            current_file.temporary_data['parent_fo_type'] = get_file_type_from_path(parent.file_path)['mime']
+            current_file.temporary_data['parent_fo_type'] = magic.from_file(parent.file_path)
+
             if current_file.uid not in extracted_files:
                 # the same file can be contained multiple times in one archive -> only the VFP needs an update
                 self.unpacking_locks.set_unpacking_lock(current_file.uid)

src/web_interface/components/io_routes.py

@@ -6,10 +6,10 @@
 from time import sleep
 
 import requests
-from fact_helper_file import get_file_type_from_binary
 from flask import Response, make_response, redirect, render_template, request
 
 import config
+from helperFunctions import magic
 from helperFunctions.database import get_shared_session
 from helperFunctions.pdf import build_pdf_report
 from helperFunctions.task_conversion import check_for_errors, convert_analysis_task_to_fw_obj, create_analysis_task
@@ -82,7 +82,7 @@ def _prepare_file_download(self, uid: str, packed: bool = False) -> str | Respon
     def _get_file_download_mime(self, binary: bytes, uid: str) -> str:
         type_analysis = self.db.frontend.get_analysis(uid, 'file_type')
         mime = type_analysis.get('mime') if type_analysis is not None else None
-        return mime or get_file_type_from_binary(binary)['mime']
+        return mime or magic.from_buffer(binary)
 
     @roles_accepted(*PRIVILEGES['download'])
     @AppRoute('/ida-download/<compare_id>', GET)