[UpdateCCTestChecks] Fix --replace-value-regex across RUN lines

[jdenny]

Summary

Without this patch, llvm/utils/update_cc_test_checks.py fails to
perform --replace-value-regex replacements when two RUN lines
produce the same output and use the same single FileCheck prefix. The
problem is that replacements in a RUN line's output are not performed
until after comparing against previous RUN lines' output, where
replacements have already been performed. This patch fixes that.

[MaskRay]

LG

[jdenny]

In D104566#2831315, @MaskRay wrote:

LG

Thanks.

clang/test/utils/update_cc_test_checks/Inputs/replace-value-regex-across-runs.c

+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -fopenmp -emit-llvm -o - %s | \
+// RUN:     FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -fopenmp -emit-llvm -o - %s | \
+// RUN:     FileCheck %s
+
+void foo(void) {
+  #pragma omp target
+  ;
+}

clang/test/utils/update_cc_test_checks/Inputs/replace-value-regex-across-runs.c.expected

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --replace-value-regex "__omp_offloading_[0-9a-z]+_[0-9a-z]+"
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -fopenmp -emit-llvm -o - %s | \
+// RUN:     FileCheck %s
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -fopenmp -emit-llvm -o - %s | \
+// RUN:     FileCheck %s
+
+// CHECK-LABEL: @foo(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_foo_l7() #[[ATTR2:[0-9]+]]
+// CHECK-NEXT:    ret void
+//
+void foo(void) {
+  #pragma omp target
+  ;
+}

clang/test/utils/update_cc_test_checks/replace-value-regex-across-runs.test

+# Test that --replace-value-regex is applied correctly when multiple RUN lines
+# use the same FileCheck prefix and have the same output.
+
+RUN: cp %S/Inputs/replace-value-regex-across-runs.c %t.c
+RUN: %update_cc_test_checks %t.c \
+RUN:     --replace-value-regex '__omp_offloading_[0-9a-z]+_[0-9a-z]+'
+RUN: diff -u %S/Inputs/replace-value-regex-across-runs.c.expected %t.c

llvm/utils/UpdateTestChecks/common.py

for m in function_re.finditer(raw_tool_output):
       if func.startswith('stress'):
         # We only use the last line of the function body for stress tests.
         scrubbed_body = '\n'.join(scrubbed_body.splitlines()[-1:])
       if self._verbose:
         print('Processing function: ' + func, file=sys.stderr)
         for l in scrubbed_body.splitlines():
           print('  ' + l, file=sys.stderr)
       for prefix in prefixes:
-        if func in self._func_dict[prefix]:
-          if (self._func_dict[prefix][func] is None or
-              str(self._func_dict[prefix][func]) != scrubbed_body or
-              self._func_dict[prefix][func].args_and_sig != args_and_sig or
-                  self._func_dict[prefix][func].attrs != attrs):
-            if (self._func_dict[prefix][func] is not None and
-                self._func_dict[prefix][func].is_same_except_arg_names(
-                scrubbed_extra,
-                args_and_sig,
-                attrs)):
-              self._func_dict[prefix][func].scrub = scrubbed_extra
-              self._func_dict[prefix][func].args_and_sig = args_and_sig
-              continue
-            else:
-              # This means a previous RUN line produced a body for this function
-              # that is different from the one produced by this current RUN line,
-              # so the body can't be common accross RUN lines. We use None to
-              # indicate that.
-              self._func_dict[prefix][func] = None
-              continue
-
         # Replace function names matching the regex.
         for regex in self._replace_value_regex:
           # Pattern that matches capture groups in the regex in leftmost order.
           group_regex = re.compile('\(.*?\)')
           # Replace function name with regex.
           match = re.match(regex, func)
           if match:
             func_repl = regex
             # Replace any capture groups with their matched strings.
             for g in match.groups():
               func_repl = group_regex.sub(re.escape(g), func_repl, count=1)
             func = re.sub(func_repl, '{{' + func_repl + '}}', func)
 
           # Replace all calls to regex matching functions.
           matches = re.finditer(regex, scrubbed_body)
           for match in matches:
             func_repl = regex
             # Replace any capture groups with their matched strings.
             for g in match.groups():
                 func_repl = group_regex.sub(re.escape(g), func_repl, count=1)
             # Substitute function call names that match the regex with the same
             # capture groups set.
-            scrubbed_body = re.sub(func_repl, '{{' + func_repl + '}}', scrubbed_body)
+            scrubbed_body = re.sub(func_repl, '{{' + func_repl + '}}',
+                                   scrubbed_body)
+
+        if func in self._func_dict[prefix]:
+          if (self._func_dict[prefix][func] is None or
+              str(self._func_dict[prefix][func]) != scrubbed_body or
+              self._func_dict[prefix][func].args_and_sig != args_and_sig or
+                  self._func_dict[prefix][func].attrs != attrs):
+            if (self._func_dict[prefix][func] is not None and
+                self._func_dict[prefix][func].is_same_except_arg_names(
+                scrubbed_extra,
+                args_and_sig,
+                attrs)):
+              self._func_dict[prefix][func].scrub = scrubbed_extra
+              self._func_dict[prefix][func].args_and_sig = args_and_sig
+              continue
+            else:
+              # This means a previous RUN line produced a body for this function
+              # that is different from the one produced by this current RUN line,
+              # so the body can't be common accross RUN lines. We use None to
+              # indicate that.
+              self._func_dict[prefix][func] = None
+              continue
 
         self._func_dict[prefix][func] = function_body(
             scrubbed_body, scrubbed_extra, args_and_sig, attrs)
         self._func_order[prefix].append(func)
 
   def _get_failed_prefixes(self):
     # This returns the list of those prefixes that failed to match any function,
     # because there were conflicting bodies produced by different RUN lines, in