Python
/
ruff
mirror of https://github.com/astral-sh/ruff
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Clean up Python implementation; enable all projects

This commit is contained in:
Zanie 2023-10-26 10:27:23 -05:00
parent 1f4e87f043
commit 1c820b0200
8 changed files with 293 additions and 346 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

225
python/ruff-ecosystem/ruff_ecosystem/check.py
View File

@ -1,3 +1,6 @@
"""
Execution, comparison, and summary of `ruff check` ecosystem checks.
"""
from __future__ import annotations from __future__ import annotations
import asyncio import asyncio
@ -11,7 +14,7 @@ from subprocess import PIPE
from typing import TYPE_CHECKING, Iterator, Self, Sequence from typing import TYPE_CHECKING, Iterator, Self, Sequence
from ruff_ecosystem import logger from ruff_ecosystem import logger
from ruff_ecosystem.markdown import markdown_project_section, markdown_details from ruff_ecosystem.markdown import markdown_details, markdown_project_section
from ruff_ecosystem.types import ( from ruff_ecosystem.types import (
Comparison, Comparison,
Diff, Diff,
@ -33,40 +36,7 @@ CHECK_DIFF_LINE_RE = re.compile(
) )
async def compare_check( def markdown_check_result(result: Result) -> str:
ruff_baseline_executable: Path,
ruff_comparison_executable: Path,
options: CheckOptions,
cloned_repo: ClonedRepository,
) -> Comparison:
async with asyncio.TaskGroup() as tg:
baseline_task = tg.create_task(
ruff_check(
executable=ruff_baseline_executable.resolve(),
path=cloned_repo.path,
name=cloned_repo.fullname,
options=options,
),
)
comparison_task = tg.create_task(
ruff_check(
executable=ruff_comparison_executable.resolve(),
path=cloned_repo.path,
name=cloned_repo.fullname,
options=options,
),
)
baseline_output, comparison_output = (
baseline_task.result(),
comparison_task.result(),
)
diff = Diff.new(baseline_output, comparison_output)
return Comparison(diff=diff, repo=cloned_repo)
def summarize_check_result(result: Result) -> str:
# Calculate the total number of rule changes # Calculate the total number of rule changes
all_rule_changes = RuleChanges() all_rule_changes = RuleChanges()
for _, comparison in result.completed: for _, comparison in result.completed:
@ -154,82 +124,6 @@ def summarize_check_result(result: Result) -> str:
return "\n".join(lines) return "\n".join(lines)
def add_permalink_to_diagnostic_line(repo: ClonedRepository, line: str) -> str:
match = CHECK_DIFF_LINE_RE.match(line)
if match is None:
return line
pre, inner, path, lnum, post = match.groups()
url = repo.url_for(path, int(lnum))
return f"{pre} <a href='{url}'>{inner}</a> {post}"
async def ruff_check(
*, executable: Path, path: Path, name: str, options: CheckOptions
) -> Sequence[str]:
"""Run the given ruff binary against the specified path."""
logger.debug(f"Checking {name} with {executable}")
ruff_args = options.to_cli_args()
start = time.time()
proc = await create_subprocess_exec(
executable.absolute(),
*ruff_args,
".",
stdout=PIPE,
stderr=PIPE,
cwd=path,
)
result, err = await proc.communicate()
end = time.time()
logger.debug(f"Finished checking {name} with {executable} in {end - start:.2f}s")
if proc.returncode != 0:
raise RuffError(err.decode("utf8"))
# Strip summary lines so the diff is only diagnostic lines
lines = [
line
for line in result.decode("utf8").splitlines()
if not CHECK_SUMMARY_LINE_RE.match(line)
]
return lines
@dataclass(frozen=True)
class CheckOptions(Serializable):
"""
Ruff check options
"""
select: str = ""
ignore: str = ""
exclude: str = ""
# Generating fixes is slow and verbose
show_fixes: bool = False
# Limit the number of reported lines per rule
max_lines_per_rule: int | None = 50
def markdown(self) -> str:
return f"select {self.select} ignore {self.ignore} exclude {self.exclude}"
def to_cli_args(self) -> list[str]:
args = ["check", "--no-cache", "--exit-zero"]
if self.select:
args.extend(["--select", self.select])
if self.ignore:
args.extend(["--ignore", self.ignore])
if self.exclude:
args.extend(["--exclude", self.exclude])
if self.show_fixes:
args.extend(["--show-fixes", "--ecosystem-ci"])
return args
@dataclass(frozen=True) @dataclass(frozen=True)
class RuleChanges: class RuleChanges:
""" """
@ -357,3 +251,112 @@ def limit_rule_lines(diff: Diff, max_per_rule: int | None = 100) -> list[str]:
reduced.append(f"... {hidden_count} changes omitted for rule {code}") reduced.append(f"... {hidden_count} changes omitted for rule {code}")
return reduced return reduced
def add_permalink_to_diagnostic_line(repo: ClonedRepository, line: str) -> str:
match = CHECK_DIFF_LINE_RE.match(line)
if match is None:
return line
pre, inner, path, lnum, post = match.groups()
url = repo.url_for(path, int(lnum))
return f"{pre} <a href='{url}'>{inner}</a> {post}"
async def compare_check(
ruff_baseline_executable: Path,
ruff_comparison_executable: Path,
options: CheckOptions,
cloned_repo: ClonedRepository,
) -> Comparison:
async with asyncio.TaskGroup() as tg:
baseline_task = tg.create_task(
ruff_check(
executable=ruff_baseline_executable.resolve(),
path=cloned_repo.path,
name=cloned_repo.fullname,
options=options,
),
)
comparison_task = tg.create_task(
ruff_check(
executable=ruff_comparison_executable.resolve(),
path=cloned_repo.path,
name=cloned_repo.fullname,
options=options,
),
)
baseline_output, comparison_output = (
baseline_task.result(),
comparison_task.result(),
)
diff = Diff.from_pair(baseline_output, comparison_output)
return Comparison(diff=diff, repo=cloned_repo)
async def ruff_check(
*, executable: Path, path: Path, name: str, options: CheckOptions
) -> Sequence[str]:
"""Run the given ruff binary against the specified path."""
logger.debug(f"Checking {name} with {executable}")
ruff_args = options.to_cli_args()
start = time.time()
proc = await create_subprocess_exec(
executable.absolute(),
*ruff_args,
".",
stdout=PIPE,
stderr=PIPE,
cwd=path,
)
result, err = await proc.communicate()
end = time.time()
logger.debug(f"Finished checking {name} with {executable} in {end - start:.2f}s")
if proc.returncode != 0:
raise RuffError(err.decode("utf8"))
# Strip summary lines so the diff is only diagnostic lines
lines = [
line
for line in result.decode("utf8").splitlines()
if not CHECK_SUMMARY_LINE_RE.match(line)
]
return lines
@dataclass(frozen=True)
class CheckOptions(Serializable):
"""
Ruff check options
"""
select: str = ""
ignore: str = ""
exclude: str = ""
# Generating fixes is slow and verbose
show_fixes: bool = False
# Limit the number of reported lines per rule
max_lines_per_rule: int | None = 50
def markdown(self) -> str:
return f"select {self.select} ignore {self.ignore} exclude {self.exclude}"
def to_cli_args(self) -> list[str]:
args = ["check", "--no-cache", "--exit-zero"]
if self.select:
args.extend(["--select", self.select])
if self.ignore:
args.extend(["--ignore", self.ignore])
if self.exclude:
args.extend(["--exclude", self.exclude])
if self.show_fixes:
args.extend(["--show-fixes", "--ecosystem-ci"])
return args

18
python/ruff-ecosystem/ruff_ecosystem/cli.py
View File

@ -117,21 +117,3 @@ def parse_args() -> argparse.Namespace:
) )
return parser.parse_args() return parser.parse_args()
def get_executable_path(name: str) -> str | None:
# Add suffix for Windows executables
name += ".exe" if sys.platform == "win32" else ""
path = os.path.join(sysconfig.get_path("scripts"), name)
# The executable in the current interpreter's scripts directory.
if os.path.exists(path):
return path
# The executable in the global environment.
environment_path = shutil.which("ruff")
if environment_path:
return environment_path
return None

115
python/ruff-ecosystem/ruff_ecosystem/defaults.py
View File

@ -1,64 +1,67 @@
"""
Default projects for ecosystem checks
"""
from ruff_ecosystem.projects import CheckOptions, Project, Repository from ruff_ecosystem.projects import CheckOptions, Project, Repository
# TODO: Consider exporting this as JSON # TODO(zanieb): Consider exporting this as JSON and loading from there instead
DEFAULT_TARGETS = [ DEFAULT_TARGETS = [
# Project(repo=Repository(owner="DisnakeDev", name="disnake", ref="master")), Project(repo=Repository(owner="DisnakeDev", name="disnake", ref="master")),
# Project(repo=Repository(owner="PostHog", name="HouseWatch", ref="main")), Project(repo=Repository(owner="PostHog", name="HouseWatch", ref="main")),
# Project(repo=Repository(owner="RasaHQ", name="rasa", ref="main")), Project(repo=Repository(owner="RasaHQ", name="rasa", ref="main")),
# Project(repo=Repository(owner="Snowflake-Labs", name="snowcli", ref="main")), Project(repo=Repository(owner="Snowflake-Labs", name="snowcli", ref="main")),
# Project(repo=Repository(owner="aiven", name="aiven-client", ref="main")), Project(repo=Repository(owner="aiven", name="aiven-client", ref="main")),
# Project(repo=Repository(owner="alteryx", name="featuretools", ref="main")), Project(repo=Repository(owner="alteryx", name="featuretools", ref="main")),
# Project( Project(
# repo=Repository(owner="apache", name="airflow", ref="main"), repo=Repository(owner="apache", name="airflow", ref="main"),
# check_options=CheckOptions(select="ALL"), check_options=CheckOptions(select="ALL"),
# ), ),
# Project(repo=Repository(owner="aws", name="aws-sam-cli", ref="develop")), Project(repo=Repository(owner="aws", name="aws-sam-cli", ref="develop")),
# Project(repo=Repository(owner="bloomberg", name="pytest-memray", ref="main")), Project(repo=Repository(owner="bloomberg", name="pytest-memray", ref="main")),
Project( Project(
repo=Repository(owner="bokeh", name="bokeh", ref="branch-3.3"), repo=Repository(owner="bokeh", name="bokeh", ref="branch-3.3"),
check_options=CheckOptions(select="ALL"), check_options=CheckOptions(select="ALL"),
), ),
# Project(repo=Repository(owner="commaai", name="openpilot", ref="master")), Project(repo=Repository(owner="commaai", name="openpilot", ref="master")),
# Project(repo=Repository(owner="demisto", name="content", ref="master")), Project(repo=Repository(owner="demisto", name="content", ref="master")),
# Project(repo=Repository(owner="docker", name="docker-py", ref="main")), Project(repo=Repository(owner="docker", name="docker-py", ref="main")),
# Project(repo=Repository(owner="freedomofpress", name="securedrop", ref="develop")), Project(repo=Repository(owner="freedomofpress", name="securedrop", ref="develop")),
# Project(repo=Repository(owner="fronzbot", name="blinkpy", ref="dev")), Project(repo=Repository(owner="fronzbot", name="blinkpy", ref="dev")),
# Project(repo=Repository(owner="ibis-project", name="ibis", ref="master")), Project(repo=Repository(owner="ibis-project", name="ibis", ref="master")),
# Project(repo=Repository(owner="ing-bank", name="probatus", ref="main")), Project(repo=Repository(owner="ing-bank", name="probatus", ref="main")),
# Project(repo=Repository(owner="jrnl-org", name="jrnl", ref="develop")), Project(repo=Repository(owner="jrnl-org", name="jrnl", ref="develop")),
# Project(repo=Repository(owner="latchbio", name="latch", ref="main")), Project(repo=Repository(owner="latchbio", name="latch", ref="main")),
# Project(repo=Repository(owner="lnbits", name="lnbits", ref="main")), Project(repo=Repository(owner="lnbits", name="lnbits", ref="main")),
# Project(repo=Repository(owner="milvus-io", name="pymilvus", ref="master")), Project(repo=Repository(owner="milvus-io", name="pymilvus", ref="master")),
# Project(repo=Repository(owner="mlflow", name="mlflow", ref="master")), Project(repo=Repository(owner="mlflow", name="mlflow", ref="master")),
# Project(repo=Repository(owner="model-bakers", name="model_bakery", ref="main")), Project(repo=Repository(owner="model-bakers", name="model_bakery", ref="main")),
# Project(repo=Repository(owner="pandas-dev", name="pandas", ref="main")), Project(repo=Repository(owner="pandas-dev", name="pandas", ref="main")),
# Project(repo=Repository(owner="prefecthq", name="prefect", ref="main")), Project(repo=Repository(owner="prefecthq", name="prefect", ref="main")),
# Project(repo=Repository(owner="pypa", name="build", ref="main")), Project(repo=Repository(owner="pypa", name="build", ref="main")),
# Project(repo=Repository(owner="pypa", name="cibuildwheel", ref="main")), Project(repo=Repository(owner="pypa", name="cibuildwheel", ref="main")),
# Project(repo=Repository(owner="pypa", name="pip", ref="main")), Project(repo=Repository(owner="pypa", name="pip", ref="main")),
# Project(repo=Repository(owner="pypa", name="setuptools", ref="main")), Project(repo=Repository(owner="pypa", name="setuptools", ref="main")),
# Project(repo=Repository(owner="python", name="mypy", ref="master")), Project(repo=Repository(owner="python", name="mypy", ref="master")),
# Project( Project(
# repo=Repository( repo=Repository(
# owner="python", owner="python",
# name="typeshed", name="typeshed",
# ref="main", ref="main",
# ), ),
# check_options=CheckOptions(select="PYI"), check_options=CheckOptions(select="PYI"),
# ), ),
# Project(repo=Repository(owner="python-poetry", name="poetry", ref="master")), Project(repo=Repository(owner="python-poetry", name="poetry", ref="master")),
# Project(repo=Repository(owner="reflex-dev", name="reflex", ref="main")), Project(repo=Repository(owner="reflex-dev", name="reflex", ref="main")),
# Project(repo=Repository(owner="rotki", name="rotki", ref="develop")), Project(repo=Repository(owner="rotki", name="rotki", ref="develop")),
# Project(repo=Repository(owner="scikit-build", name="scikit-build", ref="main")), Project(repo=Repository(owner="scikit-build", name="scikit-build", ref="main")),
# Project( Project(
# repo=Repository(owner="scikit-build", name="scikit-build-core", ref="main") repo=Repository(owner="scikit-build", name="scikit-build-core", ref="main")
# ), ),
# Project(repo=Repository(owner="sphinx-doc", name="sphinx", ref="master")), Project(repo=Repository(owner="sphinx-doc", name="sphinx", ref="master")),
# Project(repo=Repository(owner="spruceid", name="siwe-py", ref="main")), Project(repo=Repository(owner="spruceid", name="siwe-py", ref="main")),
# Project(repo=Repository(owner="tiangolo", name="fastapi", ref="master")), Project(repo=Repository(owner="tiangolo", name="fastapi", ref="master")),
# Project(repo=Repository(owner="yandex", name="ch-backup", ref="main")), Project(repo=Repository(owner="yandex", name="ch-backup", ref="main")),
# Project( Project(
# repo=Repository(owner="zulip", name="zulip", ref="main"), repo=Repository(owner="zulip", name="zulip", ref="main"),
# check_options=CheckOptions(select="ALL"), check_options=CheckOptions(select="ALL"),
# ), ),
] ]

175
python/ruff-ecosystem/ruff_ecosystem/format.py
View File

@ -1,3 +1,7 @@
"""
Execution, comparison, and summary of `ruff format` ecosystem checks.
"""
from __future__ import annotations from __future__ import annotations
import re import re
@ -6,7 +10,7 @@ from asyncio import create_subprocess_exec
from dataclasses import dataclass from dataclasses import dataclass
from pathlib import Path from pathlib import Path
from subprocess import PIPE from subprocess import PIPE
from typing import TYPE_CHECKING, Self, Sequence from typing import TYPE_CHECKING, Sequence
from unidiff import PatchSet from unidiff import PatchSet
@ -15,13 +19,13 @@ from ruff_ecosystem.markdown import markdown_project_section
from ruff_ecosystem.types import Comparison, Diff, Result, RuffError from ruff_ecosystem.types import Comparison, Diff, Result, RuffError
if TYPE_CHECKING: if TYPE_CHECKING:
from ruff_ecosystem.projects import ClonedRepository, Project from ruff_ecosystem.projects import ClonedRepository
FORMAT_IGNORE_LINES = re.compile("^warning: `ruff format` is a work-in-progress.*") FORMAT_IGNORE_LINES = re.compile("^warning: `ruff format` is a work-in-progress.*")
def summarize_format_result(result: Result) -> str: def markdown_format_result(result: Result) -> str:
lines = [] lines = []
total_lines_removed = total_lines_added = 0 total_lines_removed = total_lines_added = 0
total_files_modified = 0 total_files_modified = 0
@ -58,7 +62,7 @@ def summarize_format_result(result: Result) -> str:
lines.extend( lines.extend(
markdown_project_section( markdown_project_section(
title=title, title=title,
content=patch_set_with_permalinks(patch_set, comparison.repo), content=format_patchset(patch_set, comparison.repo),
options=project.format_options, options=project.format_options,
project=project, project=project,
) )
@ -77,6 +81,57 @@ def summarize_format_result(result: Result) -> str:
return "\n".join(lines) return "\n".join(lines)
def format_patchset(patch_set: PatchSet, repo: ClonedRepository) -> str:
"""
Convert a patchset to markdown, adding permalinks to the start of each hunk.
"""
lines = []
for file_patch in patch_set:
for hunk in file_patch:
# Note: When used for `format` checks, the line number is not exact because
# we formatted the repository for a baseline; we can't know the exact
# line number in the original
# source file.
hunk_link = repo.url_for(file_patch.path, hunk.source_start)
hunk_lines = str(hunk).splitlines()
# Add a link before the hunk
link_title = file_patch.path + "~L" + str(hunk.source_start)
lines.append(f"<a href='{hunk_link}'>{link_title}</a>")
# Wrap the contents of the hunk in a diff code block
lines.append("```diff")
lines.extend(hunk_lines[1:])
lines.append("```")
return "\n".join(lines)
async def compare_format(
ruff_baseline_executable: Path,
ruff_comparison_executable: Path,
options: FormatOptions,
cloned_repo: ClonedRepository,
):
# Run format without diff to get the baseline
await ruff_format(
executable=ruff_baseline_executable.resolve(),
path=cloned_repo.path,
name=cloned_repo.fullname,
options=options,
)
# Then get the diff from stdout
diff = await ruff_format(
executable=ruff_comparison_executable.resolve(),
path=cloned_repo.path,
name=cloned_repo.fullname,
options=options,
diff=True,
)
return Comparison(diff=Diff(diff), repo=cloned_repo)
async def ruff_format( async def ruff_format(
*, *,
executable: Path, executable: Path,
@ -113,66 +168,6 @@ async def ruff_format(
return lines return lines
async def black_format(
*,
executable: Path,
path: Path,
name: str,
) -> Sequence[str]:
"""Run the given black binary against the specified path."""
logger.debug(f"Formatting {name} with {executable}")
black_args = []
start = time.time()
proc = await create_subprocess_exec(
executable.absolute(),
*black_args,
".",
stdout=PIPE,
stderr=PIPE,
cwd=path,
)
result, err = await proc.communicate()
end = time.time()
logger.debug(f"Finished formatting {name} with {executable} in {end - start:.2f}s")
if proc.returncode != 0:
raise RuffError(err.decode("utf8"))
lines = result.decode("utf8").splitlines()
return [line for line in lines if not FORMAT_IGNORE_LINES.match(line)]
async def compare_format(
ruff_baseline_executable: Path,
ruff_comparison_executable: Path,
options: FormatOptions,
cloned_repo: ClonedRepository,
):
# Run format without diff to get the baseline
await ruff_format(
executable=ruff_baseline_executable.resolve(),
path=cloned_repo.path,
name=cloned_repo.fullname,
options=options,
)
# Then get the diff from stdout
diff = await ruff_format(
executable=ruff_comparison_executable.resolve(),
path=cloned_repo.path,
name=cloned_repo.fullname,
options=options,
diff=True,
)
return create_format_comparison(cloned_repo, FormatDiff(lines=diff))
def create_format_comparison(repo: ClonedRepository, diff: str) -> FormatComparison:
return FormatComparison(diff=diff, repo=repo)
@dataclass(frozen=True) @dataclass(frozen=True)
class FormatOptions: class FormatOptions:
""" """
@ -182,55 +177,3 @@ class FormatOptions:
def to_cli_args(self) -> list[str]: def to_cli_args(self) -> list[str]:
args = ["format", "--diff"] args = ["format", "--diff"]
return args return args
@dataclass(frozen=True)
class FormatDiff(Diff):
"""A diff from ruff format."""
lines: list[str]
def __bool__(self: Self) -> bool:
"""Return true if this diff is non-empty."""
return bool(self.lines)
@property
def added(self) -> set[str]:
return set(line for line in self.lines if line.startswith("+"))
@property
def removed(self) -> set[str]:
return set(line for line in self.lines if line.startswith("-"))
@dataclass(frozen=True)
class FormatComparison(Comparison):
diff: FormatDiff
repo: ClonedRepository
@dataclass(frozen=True)
class FormatResult(Result):
comparisons: tuple[Project, FormatComparison]
def patch_set_with_permalinks(patch_set: PatchSet, repo: ClonedRepository) -> str:
lines = []
for file_patch in patch_set:
for hunk in file_patch:
# Note: The line number is not exact because we formatted the repository for
# a baseline; we can't know the exact line number in the original
# source file.
hunk_link = repo.url_for(file_patch.path, hunk.source_start)
hunk_lines = str(hunk).splitlines()
# Add a link before the hunk
link_title = file_patch.path + "~L" + str(hunk.source_start)
lines.append(f"<a href='{hunk_link}'>{link_title}</a>")
# Wrap the contents of the hunk in a diff code block
lines.append("```diff")
lines.extend(hunk_lines[1:])
lines.append("```")
return "\n".join(lines)

21
python/ruff-ecosystem/ruff_ecosystem/main.py
View File

@ -6,8 +6,8 @@ from pathlib import Path
from typing import TypeVar from typing import TypeVar
from ruff_ecosystem import logger from ruff_ecosystem import logger
from ruff_ecosystem.check import compare_check, summarize_check_result from ruff_ecosystem.check import compare_check, markdown_check_result
from ruff_ecosystem.format import compare_format, summarize_format_result from ruff_ecosystem.format import compare_format, markdown_format_result
from ruff_ecosystem.projects import ( from ruff_ecosystem.projects import (
Project, Project,
RuffCommand, RuffCommand,
@ -38,6 +38,7 @@ async def main(
logger.debug("Using cache directory %s", cache) logger.debug("Using cache directory %s", cache)
logger.debug("Checking %s targets", len(targets)) logger.debug("Checking %s targets", len(targets))
# Limit parallelism to avoid high memory consumption
semaphore = asyncio.Semaphore(max_parallelism) semaphore = asyncio.Semaphore(max_parallelism)
async def limited_parallelism(coroutine: T) -> T: async def limited_parallelism(coroutine: T) -> T:
@ -61,15 +62,15 @@ async def main(
) )
comparisons_by_target = dict(zip(targets, comparisons, strict=True)) comparisons_by_target = dict(zip(targets, comparisons, strict=True))
errors, successes = [], [] # Split comparisons into errored / completed
errored, completed = [], []
for target, comparison in comparisons_by_target.items(): for target, comparison in comparisons_by_target.items():
if isinstance(comparison, Exception): if isinstance(comparison, Exception):
errors.append((target, comparison)) errored.append((target, comparison))
continue else:
completed.append((target, comparison))
successes.append((target, comparison)) result = Result(completed=completed, errored=errored)
result = Result(completed=successes, errored=errors)
match format: match format:
case OutputFormat.json: case OutputFormat.json:
@ -77,9 +78,9 @@ async def main(
case OutputFormat.markdown: case OutputFormat.markdown:
match command: match command:
case RuffCommand.check: case RuffCommand.check:
print(summarize_check_result(result)) print(markdown_check_result(result))
case RuffCommand.format: case RuffCommand.format:
print(summarize_format_result(result)) print(markdown_format_result(result))
case _: case _:
raise ValueError(f"Unknown target Ruff command {command}") raise ValueError(f"Unknown target Ruff command {command}")
case _: case _:

4
python/ruff-ecosystem/ruff_ecosystem/markdown.py
View File

@ -3,7 +3,9 @@ from __future__ import annotations
from typing import TYPE_CHECKING from typing import TYPE_CHECKING
if TYPE_CHECKING: if TYPE_CHECKING:
from ruff_ecosystem.projects import Project from unidiff import PatchSet
from ruff_ecosystem.projects import ClonedRepository, Project
def markdown_project_section( def markdown_project_section(

66
python/ruff-ecosystem/ruff_ecosystem/projects.py
View File

@ -1,3 +1,7 @@
"""
Abstractions and utilities for working with projects to run ecosystem checks on.
"""
from __future__ import annotations from __future__ import annotations
from asyncio import create_subprocess_exec from asyncio import create_subprocess_exec
@ -36,7 +40,7 @@ class ProjectSetupError(Exception):
@dataclass(frozen=True) @dataclass(frozen=True)
class Repository(Serializable): class Repository(Serializable):
""" """
A remote GitHub repository A remote GitHub repository.
""" """
owner: str owner: str
@ -75,13 +79,7 @@ class Repository(Serializable):
f"Failed to checkout {self.ref}: {stderr.decode()}" f"Failed to checkout {self.ref}: {stderr.decode()}"
) )
return ClonedRepository( return await ClonedRepository.from_path(checkout_dir, self)
name=self.name,
owner=self.owner,
ref=self.ref,
path=checkout_dir,
commit_hash=await self._get_head_commit(checkout_dir),
)
logger.debug(f"Cloning {self.owner}:{self.name} to {checkout_dir}") logger.debug(f"Cloning {self.owner}:{self.name} to {checkout_dir}")
command = [ command = [
@ -113,35 +111,13 @@ class Repository(Serializable):
logger.debug( logger.debug(
f"Finished cloning {self.fullname} with status {status_code}", f"Finished cloning {self.fullname} with status {status_code}",
) )
return ClonedRepository( return await ClonedRepository.from_path(checkout_dir, self)
name=self.name,
owner=self.owner,
ref=self.ref,
path=checkout_dir,
commit_hash=await self._get_head_commit(checkout_dir),
)
@staticmethod
async def _get_head_commit(checkout_dir: Path) -> str:
"""
Return the commit sha for the repository in the checkout directory.
"""
process = await create_subprocess_exec(
*["git", "rev-parse", "HEAD"],
cwd=checkout_dir,
stdout=PIPE,
)
stdout, _ = await process.communicate()
if await process.wait() != 0:
raise ProjectSetupError(f"Failed to retrieve commit sha at {checkout_dir}")
return stdout.decode().strip()
@dataclass(frozen=True) @dataclass(frozen=True)
class ClonedRepository(Repository, Serializable): class ClonedRepository(Repository, Serializable):
""" """
A cloned GitHub repository, which includes the hash of the cloned commit. A cloned GitHub repository, which includes the hash of the current commit.
""" """
commit_hash: str commit_hash: str
@ -166,3 +142,29 @@ class ClonedRepository(Repository, Serializable):
@property @property
def url(self: Self) -> str: def url(self: Self) -> str:
return f"https://github.com/{self.owner}/{self.name}@{self.commit_hash}" return f"https://github.com/{self.owner}/{self.name}@{self.commit_hash}"
@classmethod
async def from_path(cls, path: Path, repo: Repository):
return cls(
name=repo.name,
owner=repo.owner,
ref=repo.ref,
path=path,
commit_hash=await cls._get_head_commit(path),
)
@staticmethod
async def _get_head_commit(checkout_dir: Path) -> str:
"""
Return the commit sha for the repository in the checkout directory.
"""
process = await create_subprocess_exec(
*["git", "rev-parse", "HEAD"],
cwd=checkout_dir,
stdout=PIPE,
)
stdout, _ = await process.communicate()
if await process.wait() != 0:
raise ProjectSetupError(f"Failed to retrieve commit sha at {checkout_dir}")
return stdout.decode().strip()

15
python/ruff-ecosystem/ruff_ecosystem/types.py
View File

@ -46,7 +46,10 @@ class Diff(Serializable):
return len(self.removed) return len(self.removed)
@classmethod @classmethod
def new(cls, baseline: Sequence[str], comparison: Sequence[str]): def from_pair(cls, baseline: Sequence[str], comparison: Sequence[str]):
"""
Construct a diff from before and after.
"""
return cls(difflib.ndiff(baseline, comparison)) return cls(difflib.ndiff(baseline, comparison))
def jsonable(self) -> Any: def jsonable(self) -> Any:
@ -55,15 +58,23 @@ class Diff(Serializable):
@dataclass(frozen=True) @dataclass(frozen=True)
class Result(Serializable): class Result(Serializable):
"""
The result of an ecosystem check for a collection of projects.
"""
errored: list[tuple[Project, Exception]] errored: list[tuple[Project, Exception]]
completed: list[tuple[Project, Comparison]] completed: list[tuple[Project, Comparison]]
@dataclass(frozen=True) @dataclass(frozen=True)
class Comparison(Serializable): class Comparison(Serializable):
"""
The result of a completed ecosystem comparison for a single project.
"""
diff: Diff diff: Diff
repo: ClonedRepository repo: ClonedRepository
class RuffError(Exception): class RuffError(Exception):
"""An error reported by ruff.""" """An error reported by Ruff."""