[clang-format] [NFC] Restructure getLineCommentIndentPrefix

[HazardyKnusperkeks]

Summary

When sorting the known prefixes after length the if in the loop will hit at most once, so we can return from there.

Also replace the inner loop with an algorithm, that makes it more readable.

[curdeius]

clang/lib/Format/BreakableToken.cpp

44

I'd rather put asserts instead of this comment.

[HazardyKnusperkeks]

Now with assert.

We could do a static_assert, but then we would have to write our own is_sorted, since the std is only constexpr in C++20.

[HazardyKnusperkeks]

Empty lines removed.

[curdeius]

Just an assert is ok IMO. We may fix it when LLVM will be compiled with C++20 but this code may change before it happens.

clang/lib/Format/BreakableToken.cpp

case '\r':
     return true;
   default:
     return false;
   }
 }
 
 static StringRef getLineCommentIndentPrefix(StringRef Comment,
                                             const FormatStyle &Style) {
   static constexpr StringRef KnownCStylePrefixes[] = {"///<", "//!<", "///",
-                                                      "//",   "//!",  "//:"};
-  static constexpr StringRef KnownTextProtoPrefixes[] = {"//", "#", "##", "###",
-                                                         "####"};
+                                                      "//!",  "//:",  "//"};
+  static constexpr StringRef KnownTextProtoPrefixes[] = {"####", "###", "##",
+                                                         "//", "#"};
   ArrayRef<StringRef> KnownPrefixes(KnownCStylePrefixes);
   if (Style.Language == FormatStyle::LK_TextProto)
     KnownPrefixes = KnownTextProtoPrefixes;
 
-  StringRef LongestPrefix;
+  assert(std::is_sorted(KnownPrefixes.begin(), KnownPrefixes.end(),
+                        [](StringRef Lhs, StringRef Rhs) noexcept {
+                          return Lhs.size() > Rhs.size();
+                        }));
+
   for (StringRef KnownPrefix : KnownPrefixes) {
     if (Comment.startswith(KnownPrefix)) {
-      size_t PrefixLength = KnownPrefix.size();
-      while (PrefixLength < Comment.size() && Comment[PrefixLength] == ' ')
-        ++PrefixLength;
-      if (PrefixLength > LongestPrefix.size())
-        LongestPrefix = Comment.substr(0, PrefixLength);
+      const auto PrefixLength =
+          Comment.find_first_not_of(' ', KnownPrefix.size());
+      return Comment.substr(0, PrefixLength);
     }
   }
-  return LongestPrefix;
+  return {};
 }
 
 static BreakableToken::Split
 getCommentSplit(StringRef Text, unsigned ContentStartColumn,
                 unsigned ColumnLimit, unsigned TabWidth,
                 encoding::Encoding Encoding, const FormatStyle &Style,
                 bool DecorationEndsWithStar = false) {
   LLVM_DEBUG(llvm::dbgs() << "Comment split: \"" << Text