Add paths to inputs of ScanSemgrep step (#952)

* add the input

* fix

* bump version and fix split

* add path key

patchwork/common/utils/input_parsing.py

@@ -0,0 +1,63 @@
+from __future__ import annotations
+from typing_extensions import Union, AnyStr
+from collections.abc import Iterable, Mapping
+
+__ITEM_TYPE = Union[AnyStr, Mapping]
+
+def __parse_to_list_handle_str(input_value: AnyStr, possible_delimiters: Iterable[AnyStr | None]) -> list[str]:
+ for possible_delimiter in possible_delimiters:
+ if possible_delimiter is None:
+ return input_value.split()
+
+ if possible_delimiter in input_value:
+ return input_value.split(possible_delimiter)
+
+ return []
+
+def __parse_to_list_handle_dict(input_value: Mapping, possible_keys: Iterable[AnyStr | None]) -> list[str]:
+ for possible_key in possible_keys:
+ if input_value.get(possible_key) is not None:
+ return input_value.get(possible_key)
+
+ return []
+
+def __parse_to_list_handle_iterable(input_value: Iterable[__ITEM_TYPE], possible_keys: Iterable[AnyStr | None]) -> list[str]:
+ rv = []
+ for item in input_value:
+ if isinstance(item, dict):
+ for possible_key in possible_keys:
+ if item.get(possible_key) is not None:
+ rv.append(item.get(possible_key))
+ else:
+ rv.append(item)
+
+ return rv
+
+def parse_to_list(
+ input_value: __ITEM_TYPE | Iterable[__ITEM_TYPE],
+ possible_delimiters: Iterable[AnyStr | None] | None = None ,
+ possible_keys: Iterable[AnyStr | None] | None = None
+) -> list[str]:
+ if len(input_value) < 1:
+ return []
+
+ if possible_delimiters is None:
+ possible_delimiters = []
+ if possible_keys is None:
+ possible_keys = []
+
+ value_to_parse = []
+ if isinstance(input_value, dict):
+ value_to_parse = __parse_to_list_handle_dict(input_value, possible_keys)
+ elif isinstance(input_value, str):
+ value_to_parse = __parse_to_list_handle_str(input_value, possible_delimiters)
+ elif isinstance(input_value, Iterable):
+ value_to_parse = __parse_to_list_handle_iterable(input_value, possible_keys)
+
+ rv = []
+ for value in value_to_parse:
+ stripped_value = value.strip()
+ if stripped_value == "":
+ continue
+ rv.append(stripped_value)
+ return rv

patchwork/steps/ScanSemgrep/ScanSemgrep.py

@@ -3,11 +3,13 @@
 from pathlib import Path
 
 from patchwork.common.utils.dependency import import_with_dependency_group
+from patchwork.common.utils.input_parsing import parse_to_list
 from patchwork.logger import logger
 from patchwork.step import Step, StepStatus
+from patchwork.steps.ScanSemgrep.typed import ScanSemgrepOutputs, ScanSemgrepInputs
 
 
-class ScanSemgrep(Step):
+class ScanSemgrep(Step, input_class=ScanSemgrepInputs, output_class=ScanSemgrepOutputs):
  def __init__(self, inputs: dict):
  super().__init__(inputs)
 
@@ -27,6 +29,9 @@ def __init__(self, inputs: dict):
  else:
  self.sarif_values = None
 
+ path_key = inputs.get("path_key", "path")
+ self.paths = parse_to_list(inputs.get("paths", ""), possible_delimiters=[",", None], possible_keys=[path_key])
+
  def run(self) -> dict:
  if self.sarif_values is not None:
  self.set_status(StepStatus.SKIPPED, "Using provided SARIF")
@@ -37,6 +42,8 @@ def run(self) -> dict:
 
  cmd = [
  "semgrep",
+ "scan",
+ *self.paths,
  *self.extra_args.split(),
  "--sarif",
  ]

patchwork/steps/ScanSemgrep/typed.py

@@ -7,6 +7,8 @@ class ScanSemgrepInputs(TypedDict, total=False):
  sarif_file_path: Annotated[str, StepTypeConfig(is_config=True, is_path=True)]
  sarif_values: str
  semgrep_extra_args: Annotated[str, StepTypeConfig(is_config=True)]
+ paths: str
+ path_key: str
 
 
 class ScanSemgrepOutputs(TypedDict):

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "patchwork-cli"
-version = "0.0.68"
+version = "0.0.69"
 description = ""
 authors = ["patched.codes"]
 license = "AGPL"