]> git.madduck.net Git - etc/vim.git/blobdiff - src/black_primer/lib.py

madduck's git repository

Every one of the projects in this repository is available at the canonical URL git://git.madduck.net/madduck/pub/<projectpath> — see each project's metadata for the exact URL.

All patches and comments are welcome. Please squash your changes to logical commits before using git-format-patch and git-send-email to patches@git.madduck.net. If you'd read over the Git project's submission guidelines and adhered to them, I'd be especially grateful.

SSH access, as well as push access can be individually arranged.

If you use my repositories frequently, consider adding the following snippet to ~/.gitconfig and using the third clone URL listed for each project:

[url "git://git.madduck.net/madduck/"]
  insteadOf = madduck:

Remove `language_version` for pre-commit (#2430)
[etc/vim.git] / src / black_primer / lib.py
index 4f929f128ce1fd7b949441317f4b89e22bc1b4af..784134bf6523900690a616f54424749b2e00cd6d 100644 (file)
@@ -1,15 +1,18 @@
-#!/usr/bin/env python3
-
 import asyncio
+import errno
 import json
 import logging
+import os
+import stat
 import sys
+from functools import partial
 from pathlib import Path
 from platform import system
 from shutil import rmtree, which
 from subprocess import CalledProcessError
 from sys import version_info
-from typing import Any, Dict, NamedTuple, Optional, Sequence, Tuple
+from tempfile import TemporaryDirectory
+from typing import Any, Callable, Dict, NamedTuple, Optional, Sequence, Tuple
 from urllib.parse import urlparse
 
 import click
@@ -17,7 +20,7 @@ import click
 
 WINDOWS = system() == "Windows"
 BLACK_BINARY = "black.exe" if WINDOWS else "black"
-GIT_BIANRY = "git.exe" if WINDOWS else "git"
+GIT_BINARY = "git.exe" if WINDOWS else "git"
 LOG = logging.getLogger(__name__)
 
 
@@ -36,24 +39,31 @@ class Results(NamedTuple):
 
 async def _gen_check_output(
     cmd: Sequence[str],
-    timeout: float = 30,
+    timeout: float = 600,
     env: Optional[Dict[str, str]] = None,
     cwd: Optional[Path] = None,
+    stdin: Optional[bytes] = None,
 ) -> Tuple[bytes, bytes]:
     process = await asyncio.create_subprocess_exec(
         *cmd,
+        stdin=asyncio.subprocess.PIPE,
         stdout=asyncio.subprocess.PIPE,
         stderr=asyncio.subprocess.STDOUT,
         env=env,
         cwd=cwd,
     )
     try:
-        (stdout, stderr) = await asyncio.wait_for(process.communicate(), timeout)
+        (stdout, stderr) = await asyncio.wait_for(process.communicate(stdin), timeout)
     except asyncio.TimeoutError:
         process.kill()
         await process.wait()
         raise
 
+    # A non-optional timeout was supplied to asyncio.wait_for, guaranteeing
+    # a timeout or completed process.  A terminated Python process will have a
+    # non-empty returncode value.
+    assert process.returncode is not None
+
     if process.returncode != 0:
         cmd_str = " ".join(cmd)
         raise CalledProcessError(
@@ -104,32 +114,69 @@ def analyze_results(project_count: int, results: Results) -> int:
 
 
 async def black_run(
-    repo_path: Path, project_config: Dict[str, Any], results: Results
+    project_name: str,
+    repo_path: Optional[Path],
+    project_config: Dict[str, Any],
+    results: Results,
+    no_diff: bool = False,
 ) -> None:
     """Run Black and record failures"""
+    if not repo_path:
+        results.stats["failed"] += 1
+        results.failed_projects[project_name] = CalledProcessError(
+            69, [], f"{project_name} has no repo_path: {repo_path}".encode(), b""
+        )
+        return
+
+    stdin_test = project_name.upper() == "STDIN"
     cmd = [str(which(BLACK_BINARY))]
     if "cli_arguments" in project_config and project_config["cli_arguments"]:
-        cmd.extend(*project_config["cli_arguments"])
-    cmd.extend(["--check", "--diff", "."])
-
-    try:
-        _stdout, _stderr = await _gen_check_output(cmd, cwd=repo_path)
-    except asyncio.TimeoutError:
-        results.stats["failed"] += 1
-        LOG.error(f"Running black for {repo_path} timed out ({cmd})")
-    except CalledProcessError as cpe:
-        # TODO: Tune for smarter for higher signal
-        # If any other return value than 1 we raise - can disable project in config
-        if cpe.returncode == 1:
-            if not project_config["expect_formatting_changes"]:
+        cmd.extend(project_config["cli_arguments"])
+    cmd.append("--check")
+    if not no_diff:
+        cmd.append("--diff")
+
+    # Workout if we should read in a python file or search from cwd
+    stdin = None
+    if stdin_test:
+        cmd.append("-")
+        stdin = repo_path.read_bytes()
+    else:
+        cmd.append(".")
+
+    with TemporaryDirectory() as tmp_path:
+        # Prevent reading top-level user configs by manipulating environment variables
+        env = {
+            **os.environ,
+            "XDG_CONFIG_HOME": tmp_path,  # Unix-like
+            "USERPROFILE": tmp_path,  # Windows (changes `Path.home()` output)
+        }
+
+        cwd_path = repo_path.parent if stdin_test else repo_path
+        try:
+            _stdout, _stderr = await _gen_check_output(
+                cmd, cwd=cwd_path, env=env, stdin=stdin
+            )
+        except asyncio.TimeoutError:
+            results.stats["failed"] += 1
+            LOG.error(f"Running black for {repo_path} timed out ({cmd})")
+        except CalledProcessError as cpe:
+            # TODO: Tune for smarter for higher signal
+            # If any other return value than 1 we raise - can disable project in config
+            if cpe.returncode == 1:
+                if not project_config["expect_formatting_changes"]:
+                    results.stats["failed"] += 1
+                    results.failed_projects[repo_path.name] = cpe
+                else:
+                    results.stats["success"] += 1
+                return
+            elif cpe.returncode > 1:
                 results.stats["failed"] += 1
                 results.failed_projects[repo_path.name] = cpe
-            else:
-                results.stats["success"] += 1
-            return
+                return
 
-        LOG.error(f"Unknown error with {repo_path}")
-        raise
+            LOG.error(f"Unknown error with {repo_path}")
+            raise
 
     # If we get here and expect formatting changes something is up
     if project_config["expect_formatting_changes"]:
@@ -150,7 +197,7 @@ async def git_checkout_or_rebase(
     depth: int = 1,
 ) -> Optional[Path]:
     """git Clone project or rebase"""
-    git_bin = str(which(GIT_BIANRY))
+    git_bin = str(which(GIT_BINARY))
     if not git_bin:
         LOG.error("No git binary found")
         return None
@@ -176,6 +223,30 @@ async def git_checkout_or_rebase(
     return repo_path
 
 
+def handle_PermissionError(
+    func: Callable, path: Path, exc: Tuple[Any, Any, Any]
+) -> None:
+    """
+    Handle PermissionError during shutil.rmtree.
+
+    This checks if the erroring function is either 'os.rmdir' or 'os.unlink', and that
+    the error was EACCES (i.e. Permission denied). If true, the path is set writable,
+    readable, and executable by everyone. Finally, it tries the error causing delete
+    operation again.
+
+    If the check is false, then the original error will be reraised as this function
+    can't handle it.
+    """
+    excvalue = exc[1]
+    LOG.debug(f"Handling {excvalue} from {func.__name__}... ")
+    if func in (os.rmdir, os.unlink) and excvalue.errno == errno.EACCES:
+        LOG.debug(f"Setting {path} writable, readable, and executable by everyone... ")
+        os.chmod(path, stat.S_IRWXU | stat.S_IRWXG | stat.S_IRWXO)  # chmod 0777
+        func(path)  # Try the error causing delete operation again
+    else:
+        raise
+
+
 async def load_projects_queue(
     config_path: Path,
 ) -> Tuple[Dict[str, Any], asyncio.Queue]:
@@ -202,6 +273,7 @@ async def project_runner(
     long_checkouts: bool = False,
     rebase: bool = False,
     keep: bool = False,
+    no_diff: bool = False,
 ) -> None:
     """Check out project and run Black on it + record result"""
     loop = asyncio.get_event_loop()
@@ -212,6 +284,7 @@ async def project_runner(
         except asyncio.QueueEmpty:
             LOG.debug(f"project_runner {idx} exiting")
             return
+        LOG.debug(f"worker {idx} working on {project_name}")
 
         project_config = config["projects"][project_name]
 
@@ -236,14 +309,22 @@ async def project_runner(
             LOG.debug(f"Skipping {project_name} as it's configured as a long checkout")
             continue
 
-        repo_path = await git_checkout_or_rebase(work_path, project_config, rebase)
-        if not repo_path:
-            continue
-        await black_run(repo_path, project_config, results)
+        repo_path: Optional[Path] = Path(__file__)
+        stdin_project = project_name.upper() == "STDIN"
+        if not stdin_project:
+            repo_path = await git_checkout_or_rebase(work_path, project_config, rebase)
+            if not repo_path:
+                continue
+        await black_run(project_name, repo_path, project_config, results, no_diff)
 
-        if not keep:
+        if not keep and not stdin_project:
             LOG.debug(f"Removing {repo_path}")
-            await loop.run_in_executor(None, rmtree, repo_path)
+            rmtree_partial = partial(
+                rmtree, path=repo_path, onerror=handle_PermissionError
+            )
+            await loop.run_in_executor(None, rmtree_partial)
+
+        LOG.info(f"Finished {project_name}")
 
 
 async def process_queue(
@@ -253,6 +334,7 @@ async def process_queue(
     keep: bool = False,
     long_checkouts: bool = False,
     rebase: bool = False,
+    no_diff: bool = False,
 ) -> int:
     """
     Process the queue with X workers and evaluate results
@@ -280,7 +362,15 @@ async def process_queue(
     await asyncio.gather(
         *[
             project_runner(
-                i, config, queue, work_path, results, long_checkouts, rebase, keep
+                i,
+                config,
+                queue,
+                work_path,
+                results,
+                long_checkouts,
+                rebase,
+                keep,
+                no_diff,
             )
             for i in range(workers)
         ]