def test_aggregate():
test_sarif_files = find_test_data()
run_data_list = []
for sf in test_sarif_files:
with open(sf, mode="r") as report_file:
report_data = json.loads(report_file.read())
run_data_list += report_data["runs"]
with tempfile.NamedTemporaryFile(mode="w", encoding="utf-8", delete=False) as afile:
aggregate.jsonl_aggregate(run_data_list, afile.name)
afile.close()
with open(afile.name, "r") as outfile:
data = outfile.read()
assert data
os.unlink(afile.name)
def test_aggregate():
test_reports_dir = os.path.join(
os.path.dirname(os.path.realpath(__file__)), "data")
test_sarif_files = utils.find_files(test_reports_dir, ".sarif")
run_data_list = []
for sf in test_sarif_files:
with open(sf, mode="r") as report_file:
report_data = json.loads(report_file.read())
run_data_list += report_data["runs"]
with tempfile.NamedTemporaryFile(mode="w", encoding="utf-8",
delete=False) as afile:
aggregate.jsonl_aggregate(run_data_list, afile.name)
afile.close()
with open(afile.name, "r") as outfile:
data = outfile.read()
assert data
os.unlink(afile.name)
def summary(sarif_files, aggregate_file=None, override_rules={}):
"""Generate overall scan summary based on the generated
SARIF file
:param sarif_files: List of generated sarif report files
:param aggregate_file: Filename to store aggregate data
:param override_rules Build break rules to override for testing
:returns dict representing the summary
"""
report_summary = {}
build_status = "pass"
# This is the list of all runs which will get stored as an aggregate
run_data_list = []
for sf in sarif_files:
with open(sf, mode="r") as report_file:
report_data = json.loads(report_file.read())
# skip this file if the data is empty
if not report_data or not report_data.get("runs"):
LOG.warn("Report file {} is invalid. Skipping ...".format(sf))
continue
# Iterate through all the runs
for run in report_data["runs"]:
# Add it to the run data list for aggregation
run_data_list.append(run)
tool_desc = run["tool"]["driver"]["name"]
tool_name = find_tool_shortname(tool_desc)
# Initialise
report_summary[tool_name] = {
"tool": tool_name,
"description": tool_desc,
"critical": 0,
"high": 0,
"medium": 0,
"low": 0,
"status": "✅",
}
results = run.get("results", [])
metrics = run.get("properties", {}).get("metrics", None)
# If the result includes metrics use it. If not compute it
if metrics:
report_summary[tool_name].update(metrics)
else:
for aresult in results:
sev = aresult["properties"]["issue_severity"].lower()
report_summary[tool_name][sev] += 1
# Compare against the build break rule to determine status
default_rules = config.get("build_break_rules").get("default")
tool_rules = config.get("build_break_rules").get(tool_name, {})
build_break_rules = {
**default_rules,
**tool_rules,
**override_rules
}
for rsev in ["critical", "high", "medium", "low"]:
if build_break_rules.get("max_" + rsev) is not None:
if (report_summary.get(tool_name).get(rsev) >
build_break_rules["max_" + rsev]):
report_summary[tool_name]["status"] = "❌"
build_status = "fail"
# Should we store the aggregate data
if aggregate_file:
aggregate.jsonl_aggregate(run_data_list, aggregate_file)
LOG.info("Aggregate report written to {}\n".format(aggregate_file))
return report_summary, build_status
def summary(sarif_files,
depscan_files=None,
aggregate_file=None,
override_rules={}):
"""Generate overall scan summary based on the generated
SARIF file
:param sarif_files: List of generated sarif report files
:param aggregate_file: Filename to store aggregate data
:param override_rules Build break rules to override for testing
:returns dict representing the summary
"""
report_summary = {}
build_status = "pass"
# This is the list of all runs which will get stored as an aggregate
run_data_list = []
default_rules = config.get("build_break_rules").get("default")
depscan_default_rules = config.get("build_break_rules").get("depscan")
# Collect stats from depscan files if available
if depscan_files:
for df in depscan_files:
with open(df, mode="r") as drep_file:
dep_data = get_depscan_data(drep_file)
if not dep_data:
continue
# depscan-java or depscan-nodejs based on filename
dep_type = (os.path.basename(df).replace(".json", "").replace(
"-report", ""))
metrics, required_pkgs_found = calculate_depscan_metrics(
dep_data)
report_summary[dep_type] = {
"tool":
f"""Dependency Scan ({dep_type.replace("depscan-", "")})""",
"critical": metrics["critical"],
"high": metrics["high"],
"medium": metrics["medium"],
"low": metrics["low"],
"status": ":white_heavy_check_mark:",
}
report_summary[dep_type].pop("total", None)
# Compare against the build break rule to determine status
dep_tool_rules = config.get("build_break_rules").get(
dep_type, {})
build_break_rules = {**depscan_default_rules, **dep_tool_rules}
if override_rules and override_rules.get("depscan"):
build_break_rules = {
**build_break_rules,
**override_rules.get("depscan"),
}
# Default severity categories for build status
build_status_categories = (
"critical",
"required_critical",
"optional_critical",
"high",
"required_high",
"optional_high",
"medium",
"required_medium",
"optional_medium",
"low",
"required_low",
"optional_low",
)
# Issue 233 - Consider only required packages if available
if required_pkgs_found:
build_status_categories = (
"required_critical",
"required_high",
"required_medium",
"required_low",
)
for rsev in build_status_categories:
if build_break_rules.get("max_" + rsev) is not None:
if metrics.get(rsev) > build_break_rules["max_" +
rsev]:
report_summary[dep_type]["status"] = ":cross_mark:"
build_status = "fail"
for sf in sarif_files:
with open(sf, mode="r") as report_file:
report_data = json.load(report_file)
# skip this file if the data is empty
if not report_data or not report_data.get("runs"):
LOG.warn("Report file {} is invalid. Skipping ...".format(sf))
continue
# Iterate through all the runs
for run in report_data["runs"]:
# Add it to the run data list for aggregation
run_data_list.append(run)
tool_desc = run["tool"]["driver"]["name"]
tool_name = tool_desc
# Initialise
report_summary[tool_name] = {
"tool": tool_desc,
"critical": 0,
"high": 0,
"medium": 0,
"low": 0,
"status": ":white_heavy_check_mark:",
}
results = run.get("results", [])
metrics = run.get("properties", {}).get("metrics", None)
# If the result includes metrics use it. If not compute it
if metrics:
report_summary[tool_name].update(metrics)
report_summary[tool_name].pop("total", None)
else:
for aresult in results:
sev = aresult["properties"]["issue_severity"].lower()
report_summary[tool_name][sev] += 1
# Compare against the build break rule to determine status
tool_rules = config.get("build_break_rules").get(tool_name, {})
build_break_rules = {
**default_rules,
**tool_rules,
**override_rules
}
for rsev in ("critical", "high", "medium", "low"):
if build_break_rules.get("max_" + rsev) is not None:
if (report_summary.get(tool_name).get(rsev) >
build_break_rules["max_" + rsev]):
report_summary[tool_name][
"status"] = ":cross_mark:"
build_status = "fail"
# Should we store the aggregate data
if aggregate_file:
# agg_sarif_file = aggregate_file.replace(".json", ".sarif")
# aggregate.sarif_aggregate(run_data_list, agg_sarif_file)
aggregate.jsonl_aggregate(run_data_list, aggregate_file)
LOG.debug("Aggregate report written to {}\n".format(aggregate_file))
return report_summary, build_status