1 //===--- MemoryBuffer.h - Memory Buffer Interface ---------------*- C++ -*-===//
2 //
3 //                     The LLVM Compiler Infrastructure
4 //
5 // This file is distributed under the University of Illinois Open Source
6 // License. See LICENSE.TXT for details.
7 //
8 //===----------------------------------------------------------------------===//
9 //
10 //  This file defines the MemoryBuffer interface.
11 //
12 //===----------------------------------------------------------------------===//
13 
14 #ifndef LLVM_SUPPORT_MEMORYBUFFER_H
15 #define LLVM_SUPPORT_MEMORYBUFFER_H
16 
17 #include "llvm-c/Types.h"
18 #include "llvm/ADT/ArrayRef.h"
19 #include "llvm/ADT/StringRef.h"
20 #include "llvm/ADT/Twine.h"
21 #include "llvm/Support/CBindingWrapping.h"
22 #include "llvm/Support/ErrorOr.h"
23 #include "llvm/Support/FileSystem.h"
24 #include <cstddef>
25 #include <cstdint>
26 #include <memory>
27 
28 namespace llvm {
29 
30 class MemoryBufferRef;
31 
32 /// This interface provides simple read-only access to a block of memory, and
33 /// provides simple methods for reading files and standard input into a memory
34 /// buffer.  In addition to basic access to the characters in the file, this
35 /// interface guarantees you can read one character past the end of the file,
36 /// and that this character will read as '\0'.
37 ///
38 /// The '\0' guarantee is needed to support an optimization -- it's intended to
39 /// be more efficient for clients which are reading all the data to stop
40 /// reading when they encounter a '\0' than to continually check the file
41 /// position to see if it has reached the end of the file.
42 class MemoryBuffer {
43   const char *BufferStart; // Start of the buffer.
44   const char *BufferEnd;   // End of the buffer.
45 
46 protected:
47   MemoryBuffer() = default;
48 
49   void init(const char *BufStart, const char *BufEnd,
50             bool RequiresNullTerminator);
51 
52   static constexpr sys::fs::mapped_file_region::mapmode Mapmode =
53       sys::fs::mapped_file_region::readonly;
54 
55 public:
56   MemoryBuffer(const MemoryBuffer &) = delete;
57   MemoryBuffer &operator=(const MemoryBuffer &) = delete;
58   virtual ~MemoryBuffer();
59 
getBufferStart()60   const char *getBufferStart() const { return BufferStart; }
getBufferEnd()61   const char *getBufferEnd() const   { return BufferEnd; }
getBufferSize()62   size_t getBufferSize() const { return BufferEnd-BufferStart; }
63 
getBuffer()64   StringRef getBuffer() const {
65     return StringRef(BufferStart, getBufferSize());
66   }
67 
68   /// Return an identifier for this buffer, typically the filename it was read
69   /// from.
getBufferIdentifier()70   virtual StringRef getBufferIdentifier() const { return "Unknown buffer"; }
71 
72   /// Open the specified file as a MemoryBuffer, returning a new MemoryBuffer
73   /// if successful, otherwise returning null. If FileSize is specified, this
74   /// means that the client knows that the file exists and that it has the
75   /// specified size.
76   ///
77   /// \param IsVolatile Set to true to indicate that the contents of the file
78   /// can change outside the user's control, e.g. when libclang tries to parse
79   /// while the user is editing/updating the file or if the file is on an NFS.
80   static ErrorOr<std::unique_ptr<MemoryBuffer>>
81   getFile(const Twine &Filename, int64_t FileSize = -1,
82           bool RequiresNullTerminator = true, bool IsVolatile = false);
83 
84   /// Read all of the specified file into a MemoryBuffer as a stream
85   /// (i.e. until EOF reached). This is useful for special files that
86   /// look like a regular file but have 0 size (e.g. /proc/cpuinfo on Linux).
87   static ErrorOr<std::unique_ptr<MemoryBuffer>>
88   getFileAsStream(const Twine &Filename);
89 
90   /// Given an already-open file descriptor, map some slice of it into a
91   /// MemoryBuffer. The slice is specified by an \p Offset and \p MapSize.
92   /// Since this is in the middle of a file, the buffer is not null terminated.
93   static ErrorOr<std::unique_ptr<MemoryBuffer>>
94   getOpenFileSlice(int FD, const Twine &Filename, uint64_t MapSize,
95                    int64_t Offset, bool IsVolatile = false);
96 
97   /// Given an already-open file descriptor, read the file and return a
98   /// MemoryBuffer.
99   ///
100   /// \param IsVolatile Set to true to indicate that the contents of the file
101   /// can change outside the user's control, e.g. when libclang tries to parse
102   /// while the user is editing/updating the file or if the file is on an NFS.
103   static ErrorOr<std::unique_ptr<MemoryBuffer>>
104   getOpenFile(int FD, const Twine &Filename, uint64_t FileSize,
105               bool RequiresNullTerminator = true, bool IsVolatile = false);
106 
107   /// Open the specified memory range as a MemoryBuffer. Note that InputData
108   /// must be null terminated if RequiresNullTerminator is true.
109   static std::unique_ptr<MemoryBuffer>
110   getMemBuffer(StringRef InputData, StringRef BufferName = "",
111                bool RequiresNullTerminator = true);
112 
113   static std::unique_ptr<MemoryBuffer>
114   getMemBuffer(MemoryBufferRef Ref, bool RequiresNullTerminator = true);
115 
116   /// Open the specified memory range as a MemoryBuffer, copying the contents
117   /// and taking ownership of it. InputData does not have to be null terminated.
118   static std::unique_ptr<MemoryBuffer>
119   getMemBufferCopy(StringRef InputData, const Twine &BufferName = "");
120 
121   /// Read all of stdin into a file buffer, and return it.
122   static ErrorOr<std::unique_ptr<MemoryBuffer>> getSTDIN();
123 
124   /// Open the specified file as a MemoryBuffer, or open stdin if the Filename
125   /// is "-".
126   static ErrorOr<std::unique_ptr<MemoryBuffer>>
127   getFileOrSTDIN(const Twine &Filename, int64_t FileSize = -1,
128                  bool RequiresNullTerminator = true);
129 
130   /// Map a subrange of the specified file as a MemoryBuffer.
131   static ErrorOr<std::unique_ptr<MemoryBuffer>>
132   getFileSlice(const Twine &Filename, uint64_t MapSize, uint64_t Offset,
133                bool IsVolatile = false);
134 
135   //===--------------------------------------------------------------------===//
136   // Provided for performance analysis.
137   //===--------------------------------------------------------------------===//
138 
139   /// The kind of memory backing used to support the MemoryBuffer.
140   enum BufferKind {
141     MemoryBuffer_Malloc,
142     MemoryBuffer_MMap
143   };
144 
145   /// Return information on the memory mechanism used to support the
146   /// MemoryBuffer.
147   virtual BufferKind getBufferKind() const = 0;
148 
149   MemoryBufferRef getMemBufferRef() const;
150 };
151 
152 /// This class is an extension of MemoryBuffer, which allows copy-on-write
153 /// access to the underlying contents.  It only supports creation methods that
154 /// are guaranteed to produce a writable buffer.  For example, mapping a file
155 /// read-only is not supported.
156 class WritableMemoryBuffer : public MemoryBuffer {
157 protected:
158   WritableMemoryBuffer() = default;
159 
160   static constexpr sys::fs::mapped_file_region::mapmode Mapmode =
161       sys::fs::mapped_file_region::priv;
162 
163 public:
164   using MemoryBuffer::getBuffer;
165   using MemoryBuffer::getBufferEnd;
166   using MemoryBuffer::getBufferStart;
167 
168   // const_cast is well-defined here, because the underlying buffer is
169   // guaranteed to have been initialized with a mutable buffer.
getBufferStart()170   char *getBufferStart() {
171     return const_cast<char *>(MemoryBuffer::getBufferStart());
172   }
getBufferEnd()173   char *getBufferEnd() {
174     return const_cast<char *>(MemoryBuffer::getBufferEnd());
175   }
getBuffer()176   MutableArrayRef<char> getBuffer() {
177     return {getBufferStart(), getBufferEnd()};
178   }
179 
180   static ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
181   getFile(const Twine &Filename, int64_t FileSize = -1,
182           bool IsVolatile = false);
183 
184   /// Map a subrange of the specified file as a WritableMemoryBuffer.
185   static ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
186   getFileSlice(const Twine &Filename, uint64_t MapSize, uint64_t Offset,
187                bool IsVolatile = false);
188 
189   /// Allocate a new MemoryBuffer of the specified size that is not initialized.
190   /// Note that the caller should initialize the memory allocated by this
191   /// method. The memory is owned by the MemoryBuffer object.
192   static std::unique_ptr<WritableMemoryBuffer>
193   getNewUninitMemBuffer(size_t Size, const Twine &BufferName = "");
194 
195   /// Allocate a new zero-initialized MemoryBuffer of the specified size. Note
196   /// that the caller need not initialize the memory allocated by this method.
197   /// The memory is owned by the MemoryBuffer object.
198   static std::unique_ptr<WritableMemoryBuffer>
199   getNewMemBuffer(size_t Size, const Twine &BufferName = "");
200 
201 private:
202   // Hide these base class factory function so one can't write
203   //   WritableMemoryBuffer::getXXX()
204   // and be surprised that he got a read-only Buffer.
205   using MemoryBuffer::getFileAsStream;
206   using MemoryBuffer::getFileOrSTDIN;
207   using MemoryBuffer::getMemBuffer;
208   using MemoryBuffer::getMemBufferCopy;
209   using MemoryBuffer::getOpenFile;
210   using MemoryBuffer::getOpenFileSlice;
211   using MemoryBuffer::getSTDIN;
212 };
213 
214 /// This class is an extension of MemoryBuffer, which allows write access to
215 /// the underlying contents and committing those changes to the original source.
216 /// It only supports creation methods that are guaranteed to produce a writable
217 /// buffer.  For example, mapping a file read-only is not supported.
218 class WriteThroughMemoryBuffer : public MemoryBuffer {
219 protected:
220   WriteThroughMemoryBuffer() = default;
221 
222   static constexpr sys::fs::mapped_file_region::mapmode Mapmode =
223       sys::fs::mapped_file_region::readwrite;
224 
225 public:
226   using MemoryBuffer::getBuffer;
227   using MemoryBuffer::getBufferEnd;
228   using MemoryBuffer::getBufferStart;
229 
230   // const_cast is well-defined here, because the underlying buffer is
231   // guaranteed to have been initialized with a mutable buffer.
getBufferStart()232   char *getBufferStart() {
233     return const_cast<char *>(MemoryBuffer::getBufferStart());
234   }
getBufferEnd()235   char *getBufferEnd() {
236     return const_cast<char *>(MemoryBuffer::getBufferEnd());
237   }
getBuffer()238   MutableArrayRef<char> getBuffer() {
239     return {getBufferStart(), getBufferEnd()};
240   }
241 
242   static ErrorOr<std::unique_ptr<WriteThroughMemoryBuffer>>
243   getFile(const Twine &Filename, int64_t FileSize = -1);
244 
245   /// Map a subrange of the specified file as a ReadWriteMemoryBuffer.
246   static ErrorOr<std::unique_ptr<WriteThroughMemoryBuffer>>
247   getFileSlice(const Twine &Filename, uint64_t MapSize, uint64_t Offset);
248 
249 private:
250   // Hide these base class factory function so one can't write
251   //   WritableMemoryBuffer::getXXX()
252   // and be surprised that he got a read-only Buffer.
253   using MemoryBuffer::getFileAsStream;
254   using MemoryBuffer::getFileOrSTDIN;
255   using MemoryBuffer::getMemBuffer;
256   using MemoryBuffer::getMemBufferCopy;
257   using MemoryBuffer::getOpenFile;
258   using MemoryBuffer::getOpenFileSlice;
259   using MemoryBuffer::getSTDIN;
260 };
261 
262 class MemoryBufferRef {
263   StringRef Buffer;
264   StringRef Identifier;
265 
266 public:
267   MemoryBufferRef() = default;
MemoryBufferRef(MemoryBuffer & Buffer)268   MemoryBufferRef(MemoryBuffer& Buffer)
269       : Buffer(Buffer.getBuffer()), Identifier(Buffer.getBufferIdentifier()) {}
MemoryBufferRef(StringRef Buffer,StringRef Identifier)270   MemoryBufferRef(StringRef Buffer, StringRef Identifier)
271       : Buffer(Buffer), Identifier(Identifier) {}
272 
getBuffer()273   StringRef getBuffer() const { return Buffer; }
274 
getBufferIdentifier()275   StringRef getBufferIdentifier() const { return Identifier; }
276 
getBufferStart()277   const char *getBufferStart() const { return Buffer.begin(); }
getBufferEnd()278   const char *getBufferEnd() const { return Buffer.end(); }
getBufferSize()279   size_t getBufferSize() const { return Buffer.size(); }
280 };
281 
282 // Create wrappers for C Binding types (see CBindingWrapping.h).
283 DEFINE_SIMPLE_CONVERSION_FUNCTIONS(MemoryBuffer, LLVMMemoryBufferRef)
284 
285 } // end namespace llvm
286 
287 #endif // LLVM_SUPPORT_MEMORYBUFFER_H
288