CM-46872 - Fix Maven dependencies restore for SCA (#312)

Author: MarshalX

cycode/cli/files_collector/sca/base_restore_dependencies.py

@@ -59,14 +59,13 @@ def try_restore_dependencies(self, document: Document) -> Optional[Document]:
         manifest_file_path = self.get_manifest_file_path(document)
         restore_file_path = build_dep_tree_path(document.absolute_path, self.get_lock_file_name())
         relative_restore_file_path = build_dep_tree_path(document.path, self.get_lock_file_name())
-        working_directory_path = self.get_working_directory(document)
 
         if not self.verify_restore_file_already_exist(restore_file_path):
             output = execute_commands(
-                self.get_commands(manifest_file_path),
-                self.command_timeout,
+                commands=self.get_commands(manifest_file_path),
+                timeout=self.command_timeout,
                 output_file_path=restore_file_path if self.create_output_file_manually else None,
-                working_directory=working_directory_path,
+                working_directory=self.get_working_directory(document),
             )
             if output is None:  # one of the commands failed
                 return None
@@ -75,7 +74,7 @@ def try_restore_dependencies(self, document: Document) -> Optional[Document]:
         return Document(relative_restore_file_path, restore_file_content, self.is_git_diff)
 
     def get_working_directory(self, document: Document) -> Optional[str]:
-        return None
+        return os.path.dirname(document.absolute_path)
 
     @staticmethod
     def verify_restore_file_already_exist(restore_file_path: str) -> bool:

cycode/cli/files_collector/sca/go/restore_go_dependencies.py

@@ -43,6 +43,3 @@ def get_commands(self, manifest_file_path: str) -> list[list[str]]:
 
     def get_lock_file_name(self) -> str:
         return GO_RESTORE_FILE_NAME
-
-    def get_working_directory(self, document: Document) -> Optional[str]:
-        return os.path.dirname(document.absolute_path)

cycode/cli/files_collector/sca/maven/restore_maven_dependencies.py

@@ -30,34 +30,36 @@ def get_lock_file_name(self) -> str:
         return join_paths('target', MAVEN_CYCLONE_DEP_TREE_FILE_NAME)
 
     def try_restore_dependencies(self, document: Document) -> Optional[Document]:
-        restore_dependencies_document = super().try_restore_dependencies(document)
         manifest_file_path = self.get_manifest_file_path(document)
         if document.content is None:
-            restore_dependencies_document = self.restore_from_secondary_command(
-                document, manifest_file_path, restore_dependencies_document
-            )
-        else:
-            restore_dependencies_document.content = get_file_content(
-                join_paths(get_file_dir(manifest_file_path), self.get_lock_file_name())
-            )
+            return self.restore_from_secondary_command(document, manifest_file_path)
+
+        restore_dependencies_document = super().try_restore_dependencies(document)
+        if restore_dependencies_document is None:
+            return None
+
+        restore_dependencies_document.content = get_file_content(
+            join_paths(get_file_dir(manifest_file_path), self.get_lock_file_name())
+        )
 
         return restore_dependencies_document
 
-    def restore_from_secondary_command(
-        self, document: Document, manifest_file_path: str, restore_dependencies_document: Optional[Document]
-    ) -> Optional[Document]:
-        # TODO(MarshalX): does it even work? Ignored restore_dependencies_document arg
-        secondary_restore_command = create_secondary_restore_commands(manifest_file_path)
-        backup_restore_content = execute_commands(secondary_restore_command, self.command_timeout)
-        restore_dependencies_document = Document(
-            build_dep_tree_path(document.path, MAVEN_DEP_TREE_FILE_NAME), backup_restore_content, self.is_git_diff
+    def restore_from_secondary_command(self, document: Document, manifest_file_path: str) -> Optional[Document]:
+        restore_content = execute_commands(
+            commands=create_secondary_restore_commands(manifest_file_path),
+            timeout=self.command_timeout,
+            working_directory=self.get_working_directory(document),
         )
-        restore_dependencies = None
-        if restore_dependencies_document.content is not None:
-            restore_dependencies = restore_dependencies_document
-            restore_dependencies.content = get_file_content(MAVEN_DEP_TREE_FILE_NAME)
+        if restore_content is None:
+            return None
 
-        return restore_dependencies
+        restore_file_path = build_dep_tree_path(document.absolute_path, MAVEN_DEP_TREE_FILE_NAME)
+        return Document(
+            path=build_dep_tree_path(document.path, MAVEN_DEP_TREE_FILE_NAME),
+            content=get_file_content(restore_file_path),
+            is_git_diff_format=self.is_git_diff,
+            absolute_path=restore_file_path,
+        )
 
 
 def create_secondary_restore_commands(manifest_file_path: str) -> list[list[str]]:

cycode/cli/files_collector/sca/ruby/restore_ruby_dependencies.py

@@ -1,6 +1,3 @@
-import os
-from typing import Optional
-
 from cycode.cli.files_collector.sca.base_restore_dependencies import BaseRestoreDependencies
 from cycode.cli.models import Document
 
@@ -17,6 +14,3 @@ def get_commands(self, manifest_file_path: str) -> list[list[str]]:
 
     def get_lock_file_name(self) -> str:
         return RUBY_LOCK_FILE_NAME
-
-    def get_working_directory(self, document: Document) -> Optional[str]:
-        return os.path.dirname(document.absolute_path)

cycode/cli/files_collector/sca/sbt/restore_sbt_dependencies.py

@@ -1,6 +1,3 @@
-import os
-from typing import Optional
-
 from cycode.cli.files_collector.sca.base_restore_dependencies import BaseRestoreDependencies
 from cycode.cli.models import Document
 
@@ -17,6 +14,3 @@ def get_commands(self, manifest_file_path: str) -> list[list[str]]:
 
     def get_lock_file_name(self) -> str:
         return SBT_LOCK_FILE_NAME
-
-    def get_working_directory(self, document: Document) -> Optional[str]:
-        return os.path.dirname(document.absolute_path)

cycode/cli/files_collector/sca/sca_code_scanner.py

@@ -92,17 +92,16 @@ def get_project_file_ecosystem(document: Document) -> Optional[str]:
 
 def try_restore_dependencies(
     ctx: typer.Context,
-    documents_to_add: dict[str, Document],
     restore_dependencies: 'BaseRestoreDependencies',
     document: Document,
-) -> None:
+) -> Optional[Document]:
     if not restore_dependencies.is_project(document):
-        return
+        return None
 
     restore_dependencies_document = restore_dependencies.restore(document)
     if restore_dependencies_document is None:
         logger.warning('Error occurred while trying to generate dependencies tree, %s', {'filename': document.path})
-        return
+        return None
 
     if restore_dependencies_document.content is None:
         logger.warning('Error occurred while trying to generate dependencies tree, %s', {'filename': document.path})
@@ -114,10 +113,7 @@ def try_restore_dependencies(
         manifest_file_path = get_manifest_file_path(document, is_monitor_action, project_path)
         logger.debug('Succeeded to generate dependencies tree on path: %s', manifest_file_path)
 
-    if restore_dependencies_document.path in documents_to_add:
-        logger.debug('Duplicate document on restore for path: %s', restore_dependencies_document.path)
-    else:
-        documents_to_add[restore_dependencies_document.path] = restore_dependencies_document
+    return restore_dependencies_document
 
 
 def add_dependencies_tree_document(
@@ -128,7 +124,14 @@ def add_dependencies_tree_document(
 
     for restore_dependencies in restore_dependencies_list:
         for document in documents_to_scan:
-            try_restore_dependencies(ctx, documents_to_add, restore_dependencies, document)
+            restore_dependencies_document = try_restore_dependencies(ctx, restore_dependencies, document)
+            if restore_dependencies_document is None:
+                continue
+
+            if restore_dependencies_document.path in documents_to_add:
+                logger.debug('Duplicate document on restore for path: %s', restore_dependencies_document.path)
+            else:
+                documents_to_add[restore_dependencies_document.path] = restore_dependencies_document
 
     # mutate original list using slice assignment
     documents_to_scan[:] = list(documents_to_add.values())