[UpdateTestChecks] Emit warning when invalid test paths

[xbolva00]

Summary

Recently I ran into the following issue:

./update_test_checks.py /path/not-existing-file.ll

The script was silent and I was suprised why the real test file hadn't been updated.

Solution:
Emit warning if we detect this problem.

[jdoerfert]

LGTM.

[xbolva00]

Improve code, emit warning instead of error.

[xbolva00]

In D64572#1580897, @jdoerfert wrote:

LGTM.

Do you think hard error is better? I switched to warning since if somebody use

./update_test_checks.py fileA fileB fileC

and fileC does not exists, the script will exit.

New behaviour, we process fileA and fileB and we emit warnign for fileC.

[lebedev.ri]

I personally think hard error may be worse.

[xbolva00]

Maybe @spatel would like to leave a comment too?

[nikic]

Your current variant with the warning looks good to me.

[lebedev.ri]

(didn't see i'm on the review list here)

[spatel]

LGTM

[xbolva00]

Thank you all!

llvm/trunk/utils/update_test_checks.py

def main():
   autogenerated_note = (ADVERT + 'utils/' + os.path.basename(__file__))
 
   opt_basename = os.path.basename(args.opt_binary)
   if not re.match(r'^opt(-\d+)?$', opt_basename):
     print('ERROR: Unexpected opt name: ' + opt_basename, file=sys.stderr)
     sys.exit(1)
   opt_basename = 'opt'
 
-  test_paths = [test for pattern in args.tests for test in glob.glob(pattern)]
+  test_paths = []
+  for test in args.tests:
+    if not glob.glob(test):
+      print('WARNING: Test file \'%s\' was not found. Ignoring it.' % (test,), file=sys.stderr)
+      continue
+    test_paths.append(test)
+
   for test in test_paths:
     if args.verbose:
       print('Scanning for RUN lines in test file: %s' % (test,), file=sys.stderr)
     with open(test) as f:
       input_lines = [l.rstrip() for l in f]
 
     raw_lines = [m.group(1)
                  for m in [common.RUN_LINE_RE.match(l) for l in input_lines] if m]