include/llvm/TableGen/StringToOffsetTable.h

   1 //===- StringToOffsetTable.h - Emit a big concatenated string ---*- C++ -*-===//
   2 //
   3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
   4 // See https://llvm.org/LICENSE.txt for license information.
   5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
   6 //
   7 //===----------------------------------------------------------------------===//
   8
   9 #ifndef LLVM_TABLEGEN_STRINGTOOFFSETTABLE_H
  10 #define LLVM_TABLEGEN_STRINGTOOFFSETTABLE_H
  11
  12 #include "llvm/ADT/SmallString.h"
  13 #include "llvm/ADT/StringExtras.h"
  14 #include "llvm/ADT/StringMap.h"
  15 #include "llvm/Support/raw_ostream.h"
  16 #include <cctype>
  17
  18 namespace llvm {
  19
  20 /// StringToOffsetTable - This class uniques a bunch of nul-terminated strings
  21 /// and keeps track of their offset in a massive contiguous string allocation.
  22 /// It can then output this string blob and use indexes into the string to
  23 /// reference each piece.
  24 class StringToOffsetTable {
  25   StringMap<unsigned> StringOffset;
  26   std::string AggregateString;
  27
  28 public:
  29   bool Empty() const { return StringOffset.empty(); }
  30
  31   unsigned GetOrAddStringOffset(StringRef Str, bool appendZero = true) {
  32     auto IterBool =
  33         StringOffset.insert(std::make_pair(Str, AggregateString.size()));
  34     if (IterBool.second) {
  35       // Add the string to the aggregate if this is the first time found.
  36       AggregateString.append(Str.begin(), Str.end());
  37       if (appendZero)
  38         AggregateString += '\0';
  39     }
  40
  41     return IterBool.first->second;
  42   }
  43
  44   void EmitString(raw_ostream &O) {
  45     // Escape the string.
  46     SmallString<256> Str;
  47     raw_svector_ostream(Str).write_escaped(AggregateString);
  48     AggregateString = Str.str();
  49
  50     O << "    \"";
  51     unsigned CharsPrinted = 0;
  52     for (unsigned i = 0, e = AggregateString.size(); i != e; ++i) {
  53       if (CharsPrinted > 70) {
  54         O << "\"\n    \"";
  55         CharsPrinted = 0;
  56       }
  57       O << AggregateString[i];
  58       ++CharsPrinted;
  59
  60       // Print escape sequences all together.
  61       if (AggregateString[i] != '\\')
  62         continue;
  63
  64       assert(i + 1 < AggregateString.size() && "Incomplete escape sequence!");
  65       if (isdigit(AggregateString[i + 1])) {
  66         assert(isdigit(AggregateString[i + 2]) &&
  67                isdigit(AggregateString[i + 3]) &&
  68                "Expected 3 digit octal escape!");
  69         O << AggregateString[++i];
  70         O << AggregateString[++i];
  71         O << AggregateString[++i];
  72         CharsPrinted += 3;
  73       } else {
  74         O << AggregateString[++i];
  75         ++CharsPrinted;
  76       }
  77     }
  78     O << "\"";
  79   }
  80
  81   /// Emit the string using character literals. MSVC has a limitation that
  82   /// string literals cannot be longer than 64K.
  83   void EmitCharArray(raw_ostream &O) {
  84     assert(AggregateString.find(')') == std::string::npos &&
  85            "can't emit raw string with closing parens");
  86     int Count = 0;
  87     O << ' ';
  88     for (char C : AggregateString) {
  89       O << " \'";
  90       O.write_escaped(StringRef(&C, 1));
  91       O << "\',";
  92       Count++;
  93       if (Count > 14) {
  94         O << "\n ";
  95         Count = 0;
  96       }
  97     }
  98     O << '\n';
  99   }
 100 };
 101
 102 } // end namespace llvm
 103
 104 #endif