clang/tools/driver/driver.cpp

//===-- driver.cpp - Clang GCC-Compatible Driver --------------------------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This is the entry point to the clang driver; it is a thin wrapper
// for functionality in the Driver clang library.
//
//===----------------------------------------------------------------------===//

#include "clang/Driver/Compilation.h"
#include "clang/Driver/Driver.h"
#include "clang/Driver/Option.h"
#include "clang/Frontend/DiagnosticOptions.h"
#include "clang/Frontend/TextDiagnosticPrinter.h"

#include "llvm/ADT/SmallString.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/OwningPtr.h"
#include "llvm/Config/config.h"
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/ManagedStatic.h"
#include "llvm/Support/MemoryBuffer.h"
#include "llvm/Support/PrettyStackTrace.h"
#include "llvm/Support/Regex.h"
#include "llvm/Support/Timer.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/System/Host.h"
#include "llvm/System/Path.h"
#include "llvm/System/Signals.h"
using namespace clang;
using namespace clang::driver;

llvm::sys::Path GetExecutablePath(const char *Argv0, bool CanonicalPrefixes) {
  if (!CanonicalPrefixes)
    return llvm::sys::Path(Argv0);

  // This just needs to be some symbol in the binary; C++ doesn't
  // allow taking the address of ::main however.
  void *P = (void*) (intptr_t) GetExecutablePath;
  return llvm::sys::Path::GetMainExecutable(Argv0, P);
}

static const char *SaveStringInSet(std::set<std::string> &SavedStrings,
                                   llvm::StringRef S) {
  return SavedStrings.insert(S).first->c_str();
}

/// ApplyQAOverride - Apply a list of edits to the input argument lists.
///
/// The input string is a space separate list of edits to perform,
/// they are applied in order to the input argument lists. Edits
/// should be one of the following forms:
///
///  '#': Silence information about the changes to the command line arguments.
///
///  '^': Add FOO as a new argument at the beginning of the command line.
///
///  '+': Add FOO as a new argument at the end of the command line.
///
///  's/XXX/YYY/': Substitute the regular expression XXX with YYY in the command
///  line.
///
///  'xOPTION': Removes all instances of the literal argument OPTION.
///
///  'XOPTION': Removes all instances of the literal argument OPTION,
///  and the following argument.
///
///  'Ox': Removes all flags matching 'O' or 'O[sz0-9]' and adds 'Ox'
///  at the end of the command line.
///
/// \param OS - The stream to write edit information to.
/// \param Args - The vector of command line arguments.
/// \param Edit - The override command to perform.
/// \param SavedStrings - Set to use for storing string representations.
static void ApplyOneQAOverride(llvm::raw_ostream &OS,
                               llvm::SmallVectorImpl<const char*> &Args,
                               llvm::StringRef Edit,
                               std::set<std::string> &SavedStrings) {
  // This does not need to be efficient.

  if (Edit[0] == '^') {
    const char *Str =
      SaveStringInSet(SavedStrings, Edit.substr(1));
    OS << "### Adding argument " << Str << " at beginning\n";
    Args.insert(Args.begin() + 1, Str);
  } else if (Edit[0] == '+') {
    const char *Str =
      SaveStringInSet(SavedStrings, Edit.substr(1));
    OS << "### Adding argument " << Str << " at end\n";
    Args.push_back(Str);
  } else if (Edit[0] == 's' && Edit[1] == '/' && Edit.endswith("/") &&
             Edit.slice(2, Edit.size()-1).find('/') != llvm::StringRef::npos) {
    llvm::StringRef MatchPattern = Edit.substr(2).split('/').first;
    llvm::StringRef ReplPattern = Edit.substr(2).split('/').second;
    ReplPattern = ReplPattern.slice(0, ReplPattern.size()-1);

    for (unsigned i = 1, e = Args.size(); i != e; ++i) {
      std::string Repl = llvm::Regex(MatchPattern).sub(ReplPattern, Args[i]);

      if (Repl != Args[i]) {
        OS << "### Replacing '" << Args[i] << "' with '" << Repl << "'\n";
        Args[i] = SaveStringInSet(SavedStrings, Repl);
      }
    }
  } else if (Edit[0] == 'x' || Edit[0] == 'X') {
    std::string Option = Edit.substr(1, std::string::npos);
    for (unsigned i = 1; i < Args.size();) {
      if (Option == Args[i]) {
        OS << "### Deleting argument " << Args[i] << '\n';
        Args.erase(Args.begin() + i);
        if (Edit[0] == 'X') {
          if (i < Args.size()) {
            OS << "### Deleting argument " << Args[i] << '\n';
            Args.erase(Args.begin() + i);
          } else
            OS << "### Invalid X edit, end of command line!\n";
        }
      } else
        ++i;
    }
  } else if (Edit[0] == 'O') {
    for (unsigned i = 1; i < Args.size();) {
      const char *A = Args[i];
      if (A[0] == '-' && A[1] == 'O' &&
          (A[2] == '\0' ||
           (A[3] == '\0' && (A[2] == 's' || A[2] == 'z' ||
                             ('0' <= A[2] && A[2] <= '9'))))) {
        OS << "### Deleting argument " << Args[i] << '\n';
        Args.erase(Args.begin() + i);
      } else
        ++i;
    }
    OS << "### Adding argument " << Edit << " at end\n";
    Args.push_back(SaveStringInSet(SavedStrings, '-' + Edit.str()));
  } else {
    OS << "### Unrecognized edit: " << Edit << "\n";
  }
}

/// ApplyQAOverride - Apply a comma separate list of edits to the
/// input argument lists. See ApplyOneQAOverride.
static void ApplyQAOverride(llvm::SmallVectorImpl<const char*> &Args,
                            const char *OverrideStr,
                            std::set<std::string> &SavedStrings) {
  llvm::raw_ostream *OS = &llvm::errs();

  if (OverrideStr[0] == '#') {
    ++OverrideStr;
    OS = &llvm::nulls();
  }

  *OS << "### QA_OVERRIDE_GCC3_OPTIONS: " << OverrideStr << "\n";

  // This does not need to be efficient.

  const char *S = OverrideStr;
  while (*S) {
    const char *End = ::strchr(S, ' ');
    if (!End)
      End = S + strlen(S);
    if (End != S)
      ApplyOneQAOverride(*OS, Args, std::string(S, End), SavedStrings);
    S = End;
    if (*S != '\0')
      ++S;
  }
}

extern int cc1_main(const char **ArgBegin, const char **ArgEnd,
                    const char *Argv0, void *MainAddr);
extern int cc1as_main(const char **ArgBegin, const char **ArgEnd,
                      const char *Argv0, void *MainAddr);

static void ExpandArgsFromBuf(const char *Arg,
                              llvm::SmallVectorImpl<const char*> &ArgVector,
                              std::set<std::string> &SavedStrings) {
  const char *FName = Arg + 1;
  llvm::MemoryBuffer *MemBuf = llvm::MemoryBuffer::getFile(FName);
  if (!MemBuf) {
    ArgVector.push_back(SaveStringInSet(SavedStrings, Arg));
    return;
  }

  const char *Buf = MemBuf->getBufferStart();
  char InQuote = ' ';
  std::string CurArg;

  for (const char *P = Buf; ; ++P) {
    if (*P == '\0' || (isspace(*P) && InQuote == ' ')) {
      if (!CurArg.empty()) {

        if (CurArg[0] != '@') {
          ArgVector.push_back(SaveStringInSet(SavedStrings, CurArg));
        } else {
          ExpandArgsFromBuf(CurArg.c_str(), ArgVector, SavedStrings);
        }

        CurArg = "";
      }
      if (*P == '\0')
        break;
      else
        continue;
    }

    if (isspace(*P)) {
      if (InQuote != ' ')
        CurArg.push_back(*P);
      continue;
    }

    if (*P == '"' || *P == '\'') {
      if (InQuote == *P)
        InQuote = ' ';
      else if (InQuote == ' ')
        InQuote = *P;
      else
        CurArg.push_back(*P);
      continue;
    }

    if (*P == '\\') {
      ++P;
      if (*P != '\0')
        CurArg.push_back(*P);
      continue;
    }
    CurArg.push_back(*P);
  }
  delete MemBuf;
}

static void ExpandArgv(int argc, const char **argv,
                       llvm::SmallVectorImpl<const char*> &ArgVector,
                       std::set<std::string> &SavedStrings) {
  for (int i = 0; i < argc; ++i) {
    const char *Arg = argv[i];
    if (Arg[0] != '@') {
      ArgVector.push_back(SaveStringInSet(SavedStrings, std::string(Arg)));
      continue;
    }

    ExpandArgsFromBuf(Arg, ArgVector, SavedStrings);
  }
}

int main(int argc_, const char **argv_) {
  llvm::sys::PrintStackTraceOnErrorSignal();
  llvm::PrettyStackTraceProgram X(argc_, argv_);

  std::set<std::string> SavedStrings;
  llvm::SmallVector<const char*, 256> argv;

  ExpandArgv(argc_, argv_, argv, SavedStrings);

  // Handle -cc1 integrated tools.
  if (argv.size() > 1 && llvm::StringRef(argv[1]).startswith("-cc1")) {
    llvm::StringRef Tool = argv[1] + 4;

    if (Tool == "")
      return cc1_main(argv.data()+2, argv.data()+argv.size(), argv[0],
                      (void*) (intptr_t) GetExecutablePath);
    if (Tool == "as")
      return cc1as_main(argv.data()+2, argv.data()+argv.size(), argv[0],
                      (void*) (intptr_t) GetExecutablePath);

    // Reject unknown tools.
    llvm::errs() << "error: unknown integrated tool '" << Tool << "'\n";
    return 1;
  }

  bool CanonicalPrefixes = true;
  for (int i = 1, size = argv.size(); i < size; ++i) {
    if (llvm::StringRef(argv[i]) == "-no-canonical-prefixes") {
      CanonicalPrefixes = false;
      break;
    }
  }

  llvm::sys::Path Path = GetExecutablePath(argv[0], CanonicalPrefixes);

  TextDiagnosticPrinter DiagClient(llvm::errs(), DiagnosticOptions());
  DiagClient.setPrefix(Path.getBasename());

  Diagnostic Diags(&DiagClient);

#ifdef CLANG_IS_PRODUCTION
  const bool IsProduction = true;
#  ifdef CLANGXX_IS_PRODUCTION
  const bool CXXIsProduction = true;
#  else
  const bool CXXIsProduction = false;
#  endif
#else
  const bool IsProduction = false;
  const bool CXXIsProduction = false;
#endif
  Driver TheDriver(Path.str(), llvm::sys::getHostTriple(),
                   "a.out", IsProduction, CXXIsProduction,
                   Diags);

  // Check for ".*++" or ".*++-[^-]*" to determine if we are a C++
  // compiler. This matches things like "c++", "clang++", and "clang++-1.1".
  //
  // Note that we intentionally want to use argv[0] here, to support "clang++"
  // being a symlink.
  //
  // We use *argv instead of argv[0] to work around a bogus g++ warning.
  const char *progname = argv_[0];
  std::string ProgName(llvm::sys::Path(progname).getBasename());
  if (llvm::StringRef(ProgName).endswith("++") ||
      llvm::StringRef(ProgName).rsplit('-').first.endswith("++")) {
    TheDriver.CCCIsCXX = true;
    TheDriver.CCCGenericGCCName = "g++";
  }

  // Handle CC_PRINT_OPTIONS and CC_PRINT_OPTIONS_FILE.
  TheDriver.CCPrintOptions = !!::getenv("CC_PRINT_OPTIONS");
  if (TheDriver.CCPrintOptions)
    TheDriver.CCPrintOptionsFilename = ::getenv("CC_PRINT_OPTIONS_FILE");

  // Handle QA_OVERRIDE_GCC3_OPTIONS and CCC_ADD_ARGS, used for editing a
  // command line behind the scenes.
  if (const char *OverrideStr = ::getenv("QA_OVERRIDE_GCC3_OPTIONS")) {
    // FIXME: Driver shouldn't take extra initial argument.
    ApplyQAOverride(argv, OverrideStr, SavedStrings);
  } else if (const char *Cur = ::getenv("CCC_ADD_ARGS")) {
    // FIXME: Driver shouldn't take extra initial argument.
    std::vector<const char*> ExtraArgs;

    for (;;) {
      const char *Next = strchr(Cur, ',');

      if (Next) {
        ExtraArgs.push_back(SaveStringInSet(SavedStrings,
                                            std::string(Cur, Next)));
        Cur = Next + 1;
      } else {
        if (*Cur != '\0')
          ExtraArgs.push_back(SaveStringInSet(SavedStrings, Cur));
        break;
      }
    }

    argv.insert(&argv[1], ExtraArgs.begin(), ExtraArgs.end());
  }

  llvm::OwningPtr<Compilation> C(TheDriver.BuildCompilation(argv.size(),
                                                            &argv[0]));
  int Res = 0;
  if (C.get())
    Res = TheDriver.ExecuteCompilation(*C);
  
  // If any timers were active but haven't been destroyed yet, print their
  // results now.  This happens in -disable-free mode.
  llvm::TimerGroup::printAll(llvm::errs());
  
  llvm::llvm_shutdown();

  return Res;
}