1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296
|
#!/usr/bin/env python3
#
# Copyright 2023 Ettus Research, a National Instruments Brand
#
# SPDX-License-Identifier: GPL-3.0-or-later
#
"""Changeset analyzer: Create list of required tests based on changes in a branch.
This script reads changes made on a particular branch compared to
another and computes a list of tests that need to be run to validate this branch.
"""
import argparse
import os
import pathlib
import re
import shutil
import subprocess
import sys
import requests
from ruamel import yaml
def parse_args():
"""Parse and return args."""
parser = argparse.ArgumentParser(
description=__doc__,
)
parser.add_argument(
"--source-branch", help="Source branch off diff. Defaults to the current branch."
)
parser.add_argument(
"--target-branch", default="master", help="Target branch off diff. Defaults to 'master'."
)
parser.add_argument(
"--repo-path",
default=".",
help="Path to the UHD git repository. Defaults to the current directory.",
)
parser.add_argument("--rule-file", help="Path to rules file.")
parser.add_argument("--set-azdo-var", help="Generate output to set an AzDO variable")
parser.add_argument("--list-tests", action="store_true", help="Show the generated test-list.")
parser.add_argument(
"--include-target",
action="store_true",
help="Include changes that originate from the target branch.",
)
parser.add_argument(
"--github-label-api-endpoint",
help="Endpoint to use for GitHub API requests read back labels.",
)
parser.add_argument("--github-token", help="GitHub token to use for API requests.")
parser.add_argument("-v", "--verbose", action="store_true", help="Verbose output")
return parser.parse_args()
def check_changeset_content(file, **kwargs):
"""Check changeset content for code and/or comments changes
For example, if a .cpp file only has a comment change, we can use this to
remove it from the "changed files" list. This is useful when we only want to
only include tests if the underlying code was actually changed, and not just
a comment (which cannot change the outcome of a hardware test).
The values of the include_content argument have the following meaning:
- 'code': at least one line with code changed
- 'code-only': all changed lines must be code changes
- 'comments': at least one line with comments changed
- 'comments-only': all changed lines must be comment changes
- 'code-and-comments': either code or comments changes (always True)
Arguments:
file: List of files that have been changed.
repo_path: Path to the UHD repository.
git_cmd: Path to the git command.
include_content: allowed values are: code, code-only, comments,
comments-only, code-and-comments
Returns: True if the file should be included, False otherwise.
"""
def identify_comment_line_cpp(line):
"""Identify if a line is a comment in a C++ file."""
line = line.strip()
return line.startswith("//") or line.startswith("/*") or line.startswith("*")
def identify_comment_line_py(line):
"""Identify if a line is a comment in a Python file."""
line = line.strip()
return line.startswith("#")
comment_identifer = {
".cpp": identify_comment_line_cpp,
".hpp": identify_comment_line_cpp,
".c": identify_comment_line_cpp,
".h": identify_comment_line_cpp,
".py": identify_comment_line_py,
}
include_content = kwargs.get("include_content", None)
if include_content == "code-and-comments":
# No need to check, result will always be true
return True
elif include_content in ["code", "code-only"]:
invert = True
elif include_content in ["comments", "comments-only"]:
invert = False
else:
raise ValueError(f"Unsupported argument: include_content={include_content}")
target_branch = kwargs["target_branch"]
if kwargs.get("include_target"):
target_branch += "..."
get_diff_args = [
shutil.which("git"),
"-C",
kwargs["repo_path"],
"diff",
"--no-color",
"--unified=0",
target_branch,
"--",
]
ext = os.path.splitext(file)[1]
if ext not in comment_identifer:
# If we have no rule to check, then always include the file.
return True
diff_lines = (
subprocess.check_output(get_diff_args + [file], encoding="utf-8").strip().split("\n")[4:]
)
line_matches = [comment_identifer[ext](line[1:])^invert for line in diff_lines if line[0] in ("-", "+")]
if include_content.endswith("-only"):
return all(line_matches)
else:
return any(line_matches)
def get_changed_files(repo_path, target_branch, source_branch, include_target):
"""Return a list of paths in the UHD repository that have are different between two branches.
Arguments:
repo_path: Path to the UHD repository.
target_branch: Branch to compare against (e.g., master).
source_branch: Branch to compare from. Defaults to the current branch.
include_target: Include changes that originate from the target branch.
"""
assert target_branch
# If include_target is false, then current (unstaged/uncommited) changes are
# not included. If we want to change this, then couple this with
# git diff --name-only (no further arguments)
if not include_target:
target_branch += "..."
git_cmd = shutil.which("git")
get_diff_args = [git_cmd, "diff", "--name-only", target_branch]
if source_branch:
get_diff_args.append(source_branch)
files = subprocess.check_output(get_diff_args, cwd=repo_path, encoding="utf-8")
return files.strip().split("\n")
def load_rules(rule_file):
"""Return the rules as a Python list."""
with open(rule_file, "r", encoding="utf-8") as rfd:
y = yaml.YAML(typ="safe", pure=True)
return y.load(rfd)
class RuleApplier:
"""Helper class to update an internal test list based on a set of rules."""
def __init__(self, rules, labels, **kwargs):
"""Initialize.
Arguments:
rules: List of rules to apply.
labels: List of labels relevant to the current changeset.
"""
self.rules = rules
self.labels = labels
self.args = kwargs
self.test_list = set()
def apply(self, filename, verbose=False):
"""Apply rules against a file."""
for rule in self.rules:
if self._apply_rule(rule, filename, verbose):
break
def apply_labels(self):
"""Apply rules based only on labels."""
for rule in self.rules:
if (
"label" in rule
and rule["label"] in self.labels
and ("re" not in rule and "name" not in rule)
):
self.test_list.update(rule.get("add", []))
self.test_list.difference_update(rule.get("remove", []))
for remove_re in rule.get("remove_re", []):
self.test_list.difference_update(
set([test for test in self.test_list if re.search(remove_re, test)])
)
def _apply_rule(self, rule, filename, verbose=False):
"""Apply a single rule.
Returns True if you can stop applying rules against this filename.
"""
# First: Check if this rule even applies to this file
if (
("re" in rule and not re.search(rule["re"], filename))
or ("name" in rule and rule["name"] != filename)
or ("name" not in rule and "re" not in rule)
):
return False
if verbose:
sys.stderr.write(f"Filename {filename} matches rule: {rule}\n")
if "label" in rule and not rule["label"] in self.labels:
if verbose:
sys.stderr.write(f"Skipping file based on missing label: {rule['label']}\n")
return False
if "label" in rule and verbose:
sys.stderr.write(f"Label {rule['label']} found\n")
include_content = rule.get("include_content", "code")
if not check_changeset_content(
filename, **self.args, include_content=include_content
):
if verbose:
sys.stderr.write(
f"Skipping {filename} based on content rule: include_content='{include_content}'\n"
)
return False
if "add" in rule:
self.test_list.update(rule["add"])
if "remove" in rule:
self.test_list.difference_update(rule["remove"])
# If stop is specified as False, then we can still apply more rules.
if "stop" in rule and not rule["stop"]:
return False
return True
def get_labels(api_endpoint, token):
"""Get a list of labels from the GitHub API."""
if not api_endpoint:
return []
if token:
headers = {"Authorization": f"Bearer {token}"}
else:
headers = {}
data = requests.get(api_endpoint, headers=headers).json()
labels = [label["name"] for label in data]
print("Labels:", ", ".join(labels))
return labels
def main():
"""Gogogo! Run script."""
args = parse_args()
rule_file = args.rule_file
if not rule_file:
rule_file = os.path.join(pathlib.Path(__file__).parent.resolve(), "changeset_testlist.yaml")
file_list = get_changed_files(
args.repo_path,
args.target_branch,
args.source_branch,
args.include_target,
)
labels = get_labels(args.github_label_api_endpoint, args.github_token)
rule_applier = RuleApplier(
load_rules(rule_file),
labels,
repo_path=args.repo_path,
target_branch=args.target_branch,
source_branch=args.source_branch,
include_target=args.include_target,
verbose=args.verbose,
)
for filename in file_list:
rule_applier.apply(filename, args.verbose)
rule_applier.apply_labels()
if args.set_azdo_var:
print(
f"##vso[task.setvariable variable={args.set_azdo_var};isoutput=true]"
+ ";".join(rule_applier.test_list)
)
if args.list_tests:
print("Required tests:")
print("---------------", end="")
print("\n* ".join(sorted([""] + list(rule_applier.test_list))))
if __name__ == "__main__":
sys.exit(main())
|