reference, declarationdefinition
definition → references, declarations, derived classes, virtual overrides
reference to multiple definitions → definitions
unreferenced
    1
    2
    3
    4
    5
    6
    7
    8
    9
   10
   11
   12
   13
   14
   15
   16
   17
   18
   19
   20
   21
   22
   23
   24
   25
   26
   27
   28
   29
   30
   31
   32
   33
   34
   35
   36
   37
   38
   39
   40
   41
   42
   43
   44
   45
   46
   47
   48
   49
   50
   51
   52
   53
   54
   55
   56
   57
   58
   59
   60
   61
   62
   63
   64
   65
   66
   67
   68
   69
   70
   71
   72
   73
   74
   75
   76
   77
   78
   79
   80
   81
   82
   83
   84
   85
   86
   87
   88
   89
   90
   91
   92
   93
   94
   95
   96
   97
   98
   99
  100
  101
  102
  103
  104
  105
  106
  107
  108
  109
  110
  111
  112
  113
  114
  115
  116
  117
  118
  119
  120
  121
  122
  123
  124
  125
  126
  127
  128
  129
  130
  131
  132
  133
  134
  135
  136
  137
  138
  139
  140
  141
  142
  143
  144
  145
  146
  147
  148
  149
  150
  151
  152
  153
  154
  155
  156
  157
  158
  159
  160
  161
  162
  163
  164
  165
  166
  167
  168
  169
  170
  171
  172
  173
  174
  175
  176
  177
  178
  179
  180
  181
  182
  183
  184
  185
  186
  187
  188
  189
  190
  191
  192
  193
  194
  195
  196
  197
  198
  199
  200
  201
  202
  203
  204
  205
  206
  207
  208
  209
  210
  211
  212
  213
  214
  215
  216
  217
  218
  219
  220
  221
  222
  223
  224
  225
  226
  227
  228
  229
  230
  231
  232
  233
  234
  235
  236
  237
  238
  239
  240
  241
//===--- UppercaseLiteralSuffixCheck.cpp - clang-tidy ---------------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//

#include "UppercaseLiteralSuffixCheck.h"
#include "../utils/ASTUtils.h"
#include "../utils/OptionsUtils.h"
#include "clang/AST/ASTContext.h"
#include "clang/ASTMatchers/ASTMatchFinder.h"
#include "llvm/ADT/Optional.h"
#include "llvm/ADT/SmallString.h"

using namespace clang::ast_matchers;

namespace clang {
namespace tidy {
namespace readability {

namespace {

struct IntegerLiteralCheck {
  using type = clang::IntegerLiteral;
  static constexpr llvm::StringLiteral Name = llvm::StringLiteral("integer");
  // What should be skipped before looking for the Suffixes? (Nothing here.)
  static constexpr llvm::StringLiteral SkipFirst = llvm::StringLiteral("");
  // Suffix can only consist of 'u' and 'l' chars, and can be a complex number
  // ('i', 'j'). In MS compatibility mode, suffixes like i32 are supported.
  static constexpr llvm::StringLiteral Suffixes =
      llvm::StringLiteral("uUlLiIjJ");
};
constexpr llvm::StringLiteral IntegerLiteralCheck::Name;
constexpr llvm::StringLiteral IntegerLiteralCheck::SkipFirst;
constexpr llvm::StringLiteral IntegerLiteralCheck::Suffixes;

struct FloatingLiteralCheck {
  using type = clang::FloatingLiteral;
  static constexpr llvm::StringLiteral Name =
      llvm::StringLiteral("floating point");
  // C++17 introduced hexadecimal floating-point literals, and 'f' is both a
  // valid hexadecimal digit in a hex float literal and a valid floating-point
  // literal suffix.
  // So we can't just "skip to the chars that can be in the suffix".
  // Since the exponent ('p'/'P') is mandatory for hexadecimal floating-point
  // literals, we first skip everything before the exponent.
  static constexpr llvm::StringLiteral SkipFirst = llvm::StringLiteral("pP");
  // Suffix can only consist of 'f', 'l', "f16", 'h', 'q' chars,
  // and can be a complex number ('i', 'j').
  static constexpr llvm::StringLiteral Suffixes =
      llvm::StringLiteral("fFlLhHqQiIjJ");
};
constexpr llvm::StringLiteral FloatingLiteralCheck::Name;
constexpr llvm::StringLiteral FloatingLiteralCheck::SkipFirst;
constexpr llvm::StringLiteral FloatingLiteralCheck::Suffixes;

struct NewSuffix {
  SourceRange LiteralLocation;
  StringRef OldSuffix;
  llvm::Optional<FixItHint> FixIt;
};

llvm::Optional<SourceLocation> GetMacroAwareLocation(SourceLocation Loc,
                                                     const SourceManager &SM) {
  // Do nothing if the provided location is invalid.
  if (Loc.isInvalid())
    return llvm::None;
  // Look where the location was *actually* written.
  SourceLocation SpellingLoc = SM.getSpellingLoc(Loc);
  if (SpellingLoc.isInvalid())
    return llvm::None;
  return SpellingLoc;
}

llvm::Optional<SourceRange> GetMacroAwareSourceRange(SourceRange Loc,
                                                     const SourceManager &SM) {
  llvm::Optional<SourceLocation> Begin =
      GetMacroAwareLocation(Loc.getBegin(), SM);
  llvm::Optional<SourceLocation> End = GetMacroAwareLocation(Loc.getEnd(), SM);
  if (!Begin || !End)
    return llvm::None;
  return SourceRange(*Begin, *End);
}

llvm::Optional<std::string>
getNewSuffix(llvm::StringRef OldSuffix,
             const std::vector<std::string> &NewSuffixes) {
  // If there is no config, just uppercase the entirety of the suffix.
  if (NewSuffixes.empty())
    return OldSuffix.upper();
  // Else, find matching suffix, case-*insensitive*ly.
  auto NewSuffix = llvm::find_if(
      NewSuffixes, [OldSuffix](const std::string &PotentialNewSuffix) {
        return OldSuffix.equals_lower(PotentialNewSuffix);
      });
  // Have a match, return it.
  if (NewSuffix != NewSuffixes.end())
    return *NewSuffix;
  // Nope, I guess we have to keep it as-is.
  return llvm::None;
}

template <typename LiteralType>
llvm::Optional<NewSuffix>
shouldReplaceLiteralSuffix(const Expr &Literal,
                           const std::vector<std::string> &NewSuffixes,
                           const SourceManager &SM, const LangOptions &LO) {
  NewSuffix ReplacementDsc;

  const auto &L = cast<typename LiteralType::type>(Literal);

  // The naive location of the literal. Is always valid.
  ReplacementDsc.LiteralLocation = L.getSourceRange();

  // Was this literal fully spelled or is it a product of macro expansion?
  bool RangeCanBeFixed =
      utils::rangeCanBeFixed(ReplacementDsc.LiteralLocation, &SM);

  // The literal may have macro expansion, we need the final expanded src range.
  llvm::Optional<SourceRange> Range =
      GetMacroAwareSourceRange(ReplacementDsc.LiteralLocation, SM);
  if (!Range)
    return llvm::None;

  if (RangeCanBeFixed)
    ReplacementDsc.LiteralLocation = *Range;
  // Else keep the naive literal location!

  // Get the whole literal from the source buffer.
  bool Invalid;
  const StringRef LiteralSourceText = Lexer::getSourceText(
      CharSourceRange::getTokenRange(*Range), SM, LO, &Invalid);
  assert(!Invalid && "Failed to retrieve the source text.");

  size_t Skip = 0;

  // Do we need to ignore something before actually looking for the suffix?
  if (!LiteralType::SkipFirst.empty()) {
    // E.g. we can't look for 'f' suffix in hexadecimal floating-point literals
    // until after we skip to the exponent (which is mandatory there),
    // because hex-digit-sequence may contain 'f'.
    Skip = LiteralSourceText.find_first_of(LiteralType::SkipFirst);
    // We could be in non-hexadecimal floating-point literal, with no exponent.
    if (Skip == StringRef::npos)
      Skip = 0;
  }

  // Find the beginning of the suffix by looking for the first char that is
  // one of these chars that can be in the suffix, potentially starting looking
  // in the exponent, if we are skipping hex-digit-sequence.
  Skip = LiteralSourceText.find_first_of(LiteralType::Suffixes, /*From=*/Skip);

  // We can't check whether the *Literal has any suffix or not without actually
  // looking for the suffix. So it is totally possible that there is no suffix.
  if (Skip == StringRef::npos)
    return llvm::None;

  // Move the cursor in the source range to the beginning of the suffix.
  Range->setBegin(Range->getBegin().getLocWithOffset(Skip));
  // And in our textual representation too.
  ReplacementDsc.OldSuffix = LiteralSourceText.drop_front(Skip);
  assert(!ReplacementDsc.OldSuffix.empty() &&
         "We still should have some chars left.");

  // And get the replacement suffix.
  llvm::Optional<std::string> NewSuffix =
      getNewSuffix(ReplacementDsc.OldSuffix, NewSuffixes);
  if (!NewSuffix || ReplacementDsc.OldSuffix == *NewSuffix)
    return llvm::None; // The suffix was already the way it should be.

  if (RangeCanBeFixed)
    ReplacementDsc.FixIt = FixItHint::CreateReplacement(*Range, *NewSuffix);

  return ReplacementDsc;
}

} // namespace

UppercaseLiteralSuffixCheck::UppercaseLiteralSuffixCheck(
    StringRef Name, ClangTidyContext *Context)
    : ClangTidyCheck(Name, Context),
      NewSuffixes(
          utils::options::parseStringList(Options.get("NewSuffixes", ""))),
      IgnoreMacros(Options.getLocalOrGlobal("IgnoreMacros", 1) != 0) {}

void UppercaseLiteralSuffixCheck::storeOptions(
    ClangTidyOptions::OptionMap &Opts) {
  Options.store(Opts, "NewSuffixes",
                utils::options::serializeStringList(NewSuffixes));
  Options.store(Opts, "IgnoreMacros", IgnoreMacros);
}

void UppercaseLiteralSuffixCheck::registerMatchers(MatchFinder *Finder) {
  // Sadly, we can't check whether the literal has sufix or not.
  // E.g. i32 suffix still results in 'BuiltinType::Kind::Int'.
  // And such an info is not stored in the *Literal itself.
  Finder->addMatcher(
      stmt(eachOf(integerLiteral().bind(IntegerLiteralCheck::Name),
                  floatLiteral().bind(FloatingLiteralCheck::Name)),
           unless(anyOf(hasParent(userDefinedLiteral()),
                        hasAncestor(isImplicit()),
                        hasAncestor(substNonTypeTemplateParmExpr())))),
      this);
}

template <typename LiteralType>
bool UppercaseLiteralSuffixCheck::checkBoundMatch(
    const MatchFinder::MatchResult &Result) {
  const auto *Literal =
      Result.Nodes.getNodeAs<typename LiteralType::type>(LiteralType::Name);
  if (!Literal)
    return false;

  // We won't *always* want to diagnose.
  // We might have a suffix that is already uppercase.
  if (auto Details = shouldReplaceLiteralSuffix<LiteralType>(
          *Literal, NewSuffixes, *Result.SourceManager, getLangOpts())) {
    if (Details->LiteralLocation.getBegin().isMacroID() && IgnoreMacros)
      return true;
    auto Complaint = diag(Details->LiteralLocation.getBegin(),
                          "%0 literal has suffix '%1', which is not uppercase")
                     << LiteralType::Name << Details->OldSuffix;
    if (Details->FixIt) // Similarly, a fix-it is not always possible.
      Complaint << *(Details->FixIt);
  }

  return true;
}

void UppercaseLiteralSuffixCheck::check(
    const MatchFinder::MatchResult &Result) {
  if (checkBoundMatch<IntegerLiteralCheck>(Result))
    return; // If it *was* IntegerLiteral, don't check for FloatingLiteral.
  checkBoundMatch<FloatingLiteralCheck>(Result);
}

} // namespace readability
} // namespace tidy
} // namespace clang