Device tree plugin fixes

src/plugins/analysis/device_tree/code/device_tree.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import re
 from typing import TYPE_CHECKING, Dict
 
 from semver import Version
@@ -13,13 +14,16 @@
 if TYPE_CHECKING:
     import io
 
+DT_BINARY_DATA_REGEX_1 = re.compile(r'\[[0-9a-f ]{32,}]')
+DT_BINARY_DATA_REGEX_2 = re.compile(r'<(0x[0-9a-f]+ ?){10,}>')
+
 
 class AnalysisPlugin(AnalysisPluginV0):
     def __init__(self):
         metadata = self.MetaData(
             name='device_tree',
             description='get the device tree in text from the device tree blob',
-            version=Version(2, 0, 1),
+            version=Version(3, 0, 0),
             system_version=None,
             mime_blacklist=[*MIME_BLACKLIST_COMPRESSED, 'audio', 'image', 'video'],
             timeout=10,
@@ -28,6 +32,9 @@ def __init__(self):
         super().__init__(metadata=metadata)
 
     def summarize(self, result: Schema) -> list[str]:
+        if not result.device_trees:
+            return []
+
         models = [device_tree.model for device_tree in result.device_trees if device_tree.model]
 
         if not models:
@@ -50,6 +57,7 @@ def analyze(
         while (offset := binary.find(DeviceTree.Header.MAGIC, offset)) >= 0:
             try:
                 device_tree = DeviceTree.from_binary(binary, offset=offset)
+                device_tree.string = self.replace_binary_data(device_tree.string)
                 # We found a valid device tree.
                 # Skip only the header because device trees may contain device trees themselves.
                 offset += DeviceTree.Header.SIZE
@@ -68,3 +76,12 @@ def get_tags(self, result: Schema, summary: list[str]) -> list[Tag]:
                 color=TagColor.ORANGE,
             ),
         ]
+
+    @staticmethod
+    def replace_binary_data(device_tree: str) -> str:
+        # textual device tree data can contain huge chunks of binary data
+        # -> remove them from the result if they are too large
+        return DT_BINARY_DATA_REGEX_2.sub(
+            '(BINARY DATA ...)',
+            DT_BINARY_DATA_REGEX_1.sub('(BINARY DATA ...)', device_tree),
+        )

src/plugins/analysis/device_tree/test/test_device_tree.py

@@ -38,4 +38,21 @@ def test_multiple_device_trees(file, analysis_plugin):
 @pytest.mark.parametrize('file', [TEST_FP, TEST_BROKEN])
 def test_no_device_trees(file, analysis_plugin):
     result = analysis_plugin.analyze(io.FileIO(file), {}, {})
+    summary = analysis_plugin.summarize(result)
     assert result.device_trees == []
+    assert summary == []
+
+
+@pytest.mark.AnalysisPluginTestConfig(plugin_class=AnalysisPlugin)
+@pytest.mark.parametrize(
+    ('input_dts', 'expected_result'),
+    [
+        ('', ''),
+        ('data = [01 23 45 67 89 ab cd ef 01 23 45 67 89 ab cd ef];', 'data = (BINARY DATA ...);'),
+        ('data = <0x01 0x2345 0x67 0x89 0xabcdef 0x1234 0x56 0x78 0x90 0xab 0xcd>;', 'data = (BINARY DATA ...);'),
+        ('data = [01 23 45 67];', 'data = [01 23 45 67];'),  # short entries should not be replaced
+        ('data = <0x01 0x2345 0x67>;', 'data = <0x01 0x2345 0x67>;'),  # short entries should not be replaced
+    ],
+)
+def test_replace_binary_data(analysis_plugin, input_dts, expected_result):
+    assert analysis_plugin.replace_binary_data(input_dts) == expected_result

src/plugins/analysis/device_tree/view/device_tree.html

@@ -3,7 +3,7 @@
 
 {% macro device_tree_table_cell(dt_data, colspan="1") %}
     <td colspan="{{ colspan }}">
-        <pre class="dt-pre">{{ dt_data | hide_dts_binary_data | nice_generic | safe }}</pre>
+        <pre class="dt-pre">{{ dt_data | nice_generic | safe }}</pre>
     </td>
 {% endmacro %}
 
@@ -22,15 +22,15 @@
             max-height: 768px;
          }
     </style>
-    <script src="{{ url_for('static', filename='highlight.js/languages/dts.min.js') }}"></script>
+    <script src="{{ url_for('static', filename='node_modules/@highlightjs/cdn-assets/languages/dts.min.js') }}"></script>
     <script>
         async function async_highlight(element) {
             element.innerHTML = hljs.highlight(
                 element.innerHTML.replaceAll("&lt;", "<").replaceAll("&gt;", ">"),
                 {language: 'dts'}
             ).value;
-        };
-        document.addEventListener('DOMContentLoaded', (event) => {
+        }
+        document.addEventListener('DOMContentLoaded', (_) => {
             document.querySelectorAll('.dt-pre').forEach((element) => {
                 async_highlight(element);
             });
@@ -53,8 +53,8 @@
                 <table class="table table-bordered table-sm m-0" style="width: 100%">
                     {% for key, value in device_tree.header.items() -%}
                         <tr>
-                            <td width="200px">{{ key | replace_underscore }}</td>
-                            <td width="125px">{{ value | nice_number }}</td>
+                            <td style="width: 200px;">{{ key | replace_underscore }}</td>
+                            <td style="width: 125px;">{{ value | nice_number }}</td>
                             <td>{{ value | hex }}</td>
                         </tr>
                     {%- endfor %}

src/test/unit/web_interface/test_filter.py

@@ -457,20 +457,6 @@ def test_sort_cve_result(input_list, expected_result):
     assert id_list == expected_result
 
 
-@pytest.mark.parametrize(
-    ('input_dts', 'expected_result'),
-    [
-        ('', ''),
-        ('data = [01 23 45 67 89 ab cd ef 01 23 45 67 89 ab cd ef];', 'data = (BINARY DATA ...);'),
-        ('data = <0x01 0x2345 0x67 0x89 0xabcdef 0x1234 0x56 0x78 0x90 0xab 0xcd>;', 'data = (BINARY DATA ...);'),
-        ('data = [01 23 45 67];', 'data = [01 23 45 67];'),  # short entries should not be replaced
-        ('data = <0x01 0x2345 0x67>;', 'data = <0x01 0x2345 0x67>;'),  # short entries should not be replaced
-    ],
-)
-def test_hide_dts_data(input_dts, expected_result):
-    assert flt.hide_dts_binary_data(input_dts) == expected_result
-
-
 @pytest.mark.parametrize(
     ('input_', 'expected_result'),
     [

src/web_interface/components/jinja_filter.py

@@ -167,7 +167,6 @@ def _setup_filters(self):
                 'group_dict_list_by_key': flt.group_dict_list_by_key,
                 'group_changed_text_files': flt.group_path_dict_by_dirs,
                 'hex': hex,
-                'hide_dts_binary_data': flt.hide_dts_binary_data,
                 'infection_color': flt.infection_color,
                 'is_list': lambda item: isinstance(item, list),
                 'is_text_file_or_image': flt.is_text_file_or_image,

src/web_interface/filter.py

@@ -540,12 +540,6 @@ def linter_reformat_issues(issues) -> dict[str, list[dict[str, str]]]:
     return reformatted
 
 
-def hide_dts_binary_data(device_tree: str) -> str:
-    # textual device tree data can contain huge chunks of binary data -> hide them from view if they are too large
-    device_tree = re.sub(r'\[[0-9a-f ]{32,}]', '(BINARY DATA ...)', device_tree)
-    return re.sub(r'<(0x[0-9a-f]+ ?){10,}>', '(BINARY DATA ...)', device_tree)
-
-
 def get_searchable_crypto_block(crypto_material: str) -> str:
     """crypto material plugin results contain spaces and line breaks -> get a contiguous block without those"""
     blocks = crypto_material.replace(' ', '').split('\n')