POC for --dry-run param in run command

pre_commit/commands/run.py

@@ -147,6 +147,7 @@ def _run_single_hook(
         diff_before: bytes,
         verbose: bool,
         use_color: bool,
+        dry_run: bool,
 ) -> tuple[bool, bytes]:
     filenames = tuple(classifier.filenames_for_hook(hook))
 
@@ -198,6 +199,7 @@ def _run_single_hook(
                 is_local=hook.src == 'local',
                 require_serial=hook.require_serial,
                 color=use_color,
+                dry_run=dry_run,
             )
         duration = round(time.monotonic() - time_before, 2) or 0
         diff_after = _get_diff()
@@ -296,6 +298,7 @@ def _run_hooks(
         current_retval, prior_diff = _run_single_hook(
             classifier, hook, skips, cols, prior_diff,
             verbose=args.verbose, use_color=args.color,
+            dry_run=args.dry_run,
         )
         retval |= current_retval
         if retval and (config['fail_fast'] or hook.fail_fast):

pre_commit/lang_base.py

@@ -55,6 +55,7 @@ class Language(Protocol):
             is_local: bool,
             require_serial: bool,
             color: bool,
+            dry_run: bool,
     ) -> tuple[int, bytes]:
         ...
 
@@ -158,6 +159,7 @@ def run_xargs(
         *,
         require_serial: bool,
         color: bool,
+        dry_run: bool,
 ) -> tuple[int, bytes]:
     if require_serial:
         jobs = 1
@@ -167,7 +169,7 @@ def run_xargs(
         # ordering.
         file_args = _shuffled(file_args)
         jobs = target_concurrency()
-    return xargs.xargs(cmd, file_args, target_concurrency=jobs, color=color)
+    return xargs.xargs(cmd, file_args, target_concurrency=jobs, color=color, dry_run=dry_run)
 
 
 def hook_cmd(entry: str, args: Sequence[str]) -> tuple[str, ...]:
@@ -183,10 +185,12 @@ def basic_run_hook(
         is_local: bool,
         require_serial: bool,
         color: bool,
+        dry_run: bool,
 ) -> tuple[int, bytes]:
     return run_xargs(
         hook_cmd(entry, args),
         file_args,
         require_serial=require_serial,
         color=color,
+        dry_run=dry_run,
     )

pre_commit/languages/docker.py

@@ -130,6 +130,7 @@ def run_hook(
         is_local: bool,
         require_serial: bool,
         color: bool,
+        dry_run: bool,
 ) -> tuple[int, bytes]:  # pragma: win32 no cover
     # Rebuild the docker image in case it has gone missing, as many people do
     # automated cleanup of docker images.
@@ -143,4 +144,5 @@ def run_hook(
         file_args,
         require_serial=require_serial,
         color=color,
+        dry_run=dry_run,
     )

pre_commit/languages/docker_image.py

@@ -22,6 +22,7 @@ def run_hook(
         is_local: bool,
         require_serial: bool,
         color: bool,
+        dry_run: bool,
 ) -> tuple[int, bytes]:  # pragma: win32 no cover
     cmd = docker_cmd() + lang_base.hook_cmd(entry, args)
     return lang_base.run_xargs(
@@ -29,4 +30,5 @@ def run_hook(
         file_args,
         require_serial=require_serial,
         color=color,
+        dry_run=dry_run,
     )

pre_commit/languages/pygrep.py

@@ -96,9 +96,10 @@ def run_hook(
         is_local: bool,
         require_serial: bool,
         color: bool,
+        dry_run: bool,
 ) -> tuple[int, bytes]:
     cmd = (sys.executable, '-m', __name__, *args, entry)
-    return xargs(cmd, file_args, color=color)
+    return xargs(cmd, file_args, color=color, dry_run=dry_run)
 
 
 def main(argv: Sequence[str] | None = None) -> int:

pre_commit/languages/r.py

@@ -185,6 +185,7 @@ def run_hook(
         is_local: bool,
         require_serial: bool,
         color: bool,
+        dry_run: bool,
 ) -> tuple[int, bytes]:
     cmd = _cmd_from_hook(prefix, entry, args, is_local=is_local)
     return lang_base.run_xargs(
@@ -192,4 +193,5 @@ def run_hook(
         file_args,
         require_serial=require_serial,
         color=color,
+        dry_run=dry_run,
     )

pre_commit/languages/script.py

@@ -21,6 +21,7 @@ def run_hook(
         is_local: bool,
         require_serial: bool,
         color: bool,
+        dry_run: bool,
 ) -> tuple[int, bytes]:
     cmd = lang_base.hook_cmd(entry, args)
     cmd = (prefix.path(cmd[0]), *cmd[1:])
@@ -29,4 +30,5 @@ def run_hook(
         file_args,
         require_serial=require_serial,
         color=color,
+        dry_run=dry_run,
     )

pre_commit/main.py

@@ -60,6 +60,12 @@ def _add_hook_type_option(parser: argparse.ArgumentParser) -> None:
 def _add_run_options(parser: argparse.ArgumentParser) -> None:
     parser.add_argument('hook', nargs='?', help='A single hook-id to run')
     parser.add_argument('--verbose', '-v', action='store_true', default=False)
+    parser.add_argument(
+        '--dry-run',
+        action='store_true',
+        help='Display the hook commands instead of run them (useful for debugging).',
+        default=False,
+    )
     mutex_group = parser.add_mutually_exclusive_group(required=False)
     mutex_group.add_argument(
         '--all-files', '-a', action='store_true', default=False,

pre_commit/xargs.py

@@ -134,6 +134,7 @@ def xargs(
         varargs: Sequence[str],
         *,
         color: bool = False,
+        dry_run: bool = False,
         target_concurrency: int = 1,
         _max_length: int = _get_platform_max_length(),
         **kwargs: Any,
@@ -172,14 +173,19 @@ def xargs(
         return cmd_fn(
             *run_cmd, check=False, stderr=subprocess.STDOUT, **kwargs,
         )
+    if dry_run:
+        for run_cmd in partitions:
+            print("DRY RUN EXEC: ", " ".join(run_cmd))
+        retcode = 0
+        stdout = b""
+    else:
+        threads = min(len(partitions), target_concurrency)
+        with _thread_mapper(threads) as thread_map:
+            results = thread_map(run_cmd_partition, partitions)
 
-    threads = min(len(partitions), target_concurrency)
-    with _thread_mapper(threads) as thread_map:
-        results = thread_map(run_cmd_partition, partitions)
-
-        for proc_retcode, proc_out, _ in results:
-            if abs(proc_retcode) > abs(retcode):
-                retcode = proc_retcode
-            stdout += proc_out
+            for proc_retcode, proc_out, _ in results:
+                if abs(proc_retcode) > abs(retcode):
+                    retcode = proc_retcode
+                stdout += proc_out
 
     return retcode, stdout