DataExtractor.h revision 0fea0512e47d1820dd78de2748b874c81eea77d6
1//===-- DataExtractor.h -----------------------------------------*- 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#ifndef liblldb_DataExtractor_h_
11#define liblldb_DataExtractor_h_
12#if defined (__cplusplus)
13
14
15#include "lldb/lldb-private.h"
16#include <limits.h>
17#include <stdint.h>
18#include <string.h>
19
20namespace lldb_private {
21
22//----------------------------------------------------------------------
23/// @class DataExtractor DataExtractor.h "lldb/Core/DataExtractor.h"
24/// @brief An data extractor class.
25///
26/// DataExtractor is a class that can extract data (swapping if needed)
27/// from a data buffer. The data buffer can be caller owned, or can be
28/// shared data that can be shared between multiple DataExtractor
29/// instances. Multiple DataExtractor objects can share the same data,
30/// yet extract values in different address sizes and byte order modes.
31/// Each object can have a unique position in the shared data and extract
32/// data from different offsets.
33///
34/// @see DataBuffer
35//----------------------------------------------------------------------
36class DataExtractor
37{
38public:
39    //------------------------------------------------------------------
40    /// @typedef DataExtractor::Type
41    /// @brief Type enumerations used in the dump routines.
42    /// @see DataExtractor::Dump()
43    /// @see DataExtractor::DumpRawHexBytes()
44    //------------------------------------------------------------------
45    typedef enum
46    {
47        TypeUInt8,      ///< Format output as unsigned 8 bit integers
48        TypeChar,       ///< Format output as characters
49        TypeUInt16,     ///< Format output as unsigned 16 bit integers
50        TypeUInt32,     ///< Format output as unsigned 32 bit integers
51        TypeUInt64,     ///< Format output as unsigned 64 bit integers
52        TypePointer,    ///< Format output as pointers
53        TypeULEB128,    ///< Format output as ULEB128 numbers
54        TypeSLEB128     ///< Format output as SLEB128 numbers
55    } Type;
56
57    static void
58    DumpHexBytes (Stream *s,
59                  const void *src,
60                  size_t src_len,
61                  uint32_t bytes_per_line,
62                  lldb::addr_t base_addr); // Pass LLDB_INVALID_ADDRESS to not show address at start of line
63    //------------------------------------------------------------------
64    /// Default constructor.
65    ///
66    /// Initialize all members to a default empty state.
67    //------------------------------------------------------------------
68    DataExtractor ();
69
70    //------------------------------------------------------------------
71    /// Construct with a buffer that is owned by the caller.
72    ///
73    /// This constructor allows us to use data that is owned by the
74    /// caller. The data must stay around as long as this object is
75    /// valid.
76    ///
77    /// @param[in] data
78    ///     A pointer to caller owned data.
79    ///
80    /// @param[in] data_length
81    ///     The length in bytes of \a data.
82    ///
83    /// @param[in] byte_order
84    ///     A byte order of the data that we are extracting from.
85    ///
86    /// @param[in] addr_size
87    ///     A new address byte size value.
88    //------------------------------------------------------------------
89    DataExtractor (const void* data, uint32_t data_length, lldb::ByteOrder byte_order, uint8_t addr_size);
90
91    //------------------------------------------------------------------
92    /// Construct with shared data.
93    ///
94    /// Copies the data shared pointer which adds a reference to the
95    /// contained in \a data_sp. The shared data reference is reference
96    /// counted to ensure the data lives as long as anyone still has a
97    /// valid shared pointer to the data in \a data_sp.
98    ///
99    /// @param[in] data_sp
100    ///     A shared pointer to data.
101    ///
102    /// @param[in] byte_order
103    ///     A byte order of the data that we are extracting from.
104    ///
105    /// @param[in] addr_size
106    ///     A new address byte size value.
107    //------------------------------------------------------------------
108    DataExtractor (const lldb::DataBufferSP& data_sp, lldb::ByteOrder byte_order, uint8_t addr_size);
109
110    //------------------------------------------------------------------
111    /// Construct with a subset of \a data.
112    ///
113    /// Initialize this object with a subset of the data bytes in \a
114    /// data. If \a data contains shared data, then a reference to the
115    /// shared data will be added to ensure the shared data stays around
116    /// as long as any objects have references to the shared data. The
117    /// byte order value and the address size settings are copied from \a
118    /// data. If \a offset is not a valid offset in \a data, then no
119    /// reference to the shared data will be added. If there are not
120    /// \a length bytes available in \a data starting at \a offset,
121    /// the length will be truncated to contain as many bytes as
122    /// possible.
123    ///
124    /// @param[in] data
125    ///     Another DataExtractor object that contains data.
126    ///
127    /// @param[in] offset
128    ///     The offset into \a data at which the subset starts.
129    ///
130    /// @param[in] length
131    ///     The length in bytes of the subset of data.
132    //------------------------------------------------------------------
133    DataExtractor (const DataExtractor& data, uint32_t offset, uint32_t length);
134
135    DataExtractor (const DataExtractor& rhs);
136    //------------------------------------------------------------------
137    /// Assignment operator.
138    ///
139    /// Copies all data, byte order and address size settings from \a rhs into
140    /// this object. If \a rhs contains shared data, a reference to that
141    /// shared data will be added.
142    ///
143    /// @param[in] rhs
144    ///     Another DataExtractor object to copy.
145    ///
146    /// @return
147    ///     A const reference to this object.
148    //------------------------------------------------------------------
149    const DataExtractor&
150    operator= (const DataExtractor& rhs);
151
152    //------------------------------------------------------------------
153    /// Destructor
154    ///
155    /// If this object contains a valid shared data reference, the
156    /// reference count on the data will be decremented, and if zero,
157    /// the data will be freed.
158    //------------------------------------------------------------------
159    ~DataExtractor ();
160
161    //------------------------------------------------------------------
162    /// Clears the object state.
163    ///
164    /// Clears the object contents back to a default invalid state, and
165    /// release any references to shared data that this object may
166    /// contain.
167    //------------------------------------------------------------------
168    void
169    Clear ();
170
171    //------------------------------------------------------------------
172    /// Dumps the binary data as \a type objects to stream \a s (or to
173    /// Log() if \a s is NULL) starting \a offset bytes into the data
174    /// and stopping after dumping \a length bytes. The offset into the
175    /// data is displayed at the beginning of each line and can be
176    /// offset by base address \a base_addr. \a num_per_line objects
177    /// will be displayed on each line.
178    ///
179    /// @param[in] s
180    ///     The stream to dump the output to. If NULL the output will
181    ///     be dumped to Log().
182    ///
183    /// @param[in] offset
184    ///     The offset into the data at which to start dumping.
185    ///
186    /// @param[in] length
187    ///     The number of bytes to dump.
188    ///
189    /// @param[in] base_addr
190    ///     The base address that gets added to the offset displayed on
191    ///     each line.
192    ///
193    /// @param[in] num_per_line
194    ///     The number of \a type objects to display on each line.
195    ///
196    /// @param[in] type
197    ///     The type of objects to use when dumping data from this
198    ///     object. See DataExtractor::Type.
199    ///
200    /// @param[in] type_format
201    ///     The optional format to use for the \a type objects. If this
202    ///     is NULL, the default format for the \a type will be used.
203    ///
204    /// @return
205    ///     The offset at which dumping ended.
206    //------------------------------------------------------------------
207    uint32_t
208    PutToLog (Log *log,
209              uint32_t offset,
210              uint32_t length,
211              uint64_t base_addr,
212              uint32_t num_per_line,
213              Type type,
214              const char *type_format = NULL) const;
215
216    //------------------------------------------------------------------
217    /// Dumps \a item_count objects into the stream \a s.
218    ///
219    /// Dumps \a item_count objects using \a item_format, each of which
220    /// are \a item_byte_size bytes long starting at offset \a offset
221    /// bytes into the contained data, into the stream \a s. \a
222    /// num_per_line objects will be dumped on each line before a new
223    /// line will be output. If \a base_addr is a valid address, then
224    /// each new line of output will be prededed by the address value
225    /// plus appropriate offset, and a colon and space. Bitfield values
226    /// can be dumped by calling this function multiple times with the
227    /// same start offset, format and size, yet differing \a
228    /// item_bit_size and \a item_bit_offset values.
229    ///
230    /// @param[in] s
231    ///     The stream to dump the output to. This value can not be NULL.
232    ///
233    /// @param[in] offset
234    ///     The offset into the data at which to start dumping.
235    ///
236    /// @param[in] item_format
237    ///     The format to use when dumping each item.
238    ///
239    /// @param[in] item_byte_size
240    ///     The byte size of each item.
241    ///
242    /// @param[in] item_count
243    ///     The number of items to dump.
244    ///
245    /// @param[in] num_per_line
246    ///     The number of items to display on each line.
247    ///
248    /// @param[in] base_addr
249    ///     The base address that gets added to the offset displayed on
250    ///     each line if the value is valid. Is \a base_addr is
251    ///     LLDB_INVALID_ADDRESS then no address values will be prepended
252    ///     to any lines.
253    ///
254    /// @param[in] item_bit_size
255    ///     If the value to display is a bitfield, this value should
256    ///     be the number of bits that the bitfield item has within the
257    ///     item's byte size value. This function will need to be called
258    ///     multiple times with identical \a offset and \a item_byte_size
259    ///     values in order to display multiple bitfield values that
260    ///     exist within the same integer value. If the items being
261    ///     displayed are not bitfields, this value should be zero.
262    ///
263    /// @param[in] item_bit_offset
264    ///     If the value to display is a bitfield, this value should
265    ///     be the offset in bits, or shift right amount, that the
266    ///     bitfield item occupies within the item's byte size value.
267    ///     This function will need to be called multiple times with
268    ///     identical \a offset and \a item_byte_size values in order
269    ///     to display multiple bitfield values that exist within the
270    ///     same integer value. If the items being displayed are not
271    ///     bitfields, this value should be zero.
272    ///
273    /// @return
274    ///     The offset at which dumping ended.
275    //------------------------------------------------------------------
276    uint32_t
277    Dump (Stream *s,
278          uint32_t offset,
279          lldb::Format item_format,
280          uint32_t item_byte_size,
281          uint32_t item_count,
282          uint32_t num_per_line,
283          uint64_t base_addr,
284          uint32_t item_bit_size,
285          uint32_t item_bit_offset,
286          ExecutionContextScope *exe_scope = NULL) const;
287
288    //------------------------------------------------------------------
289    /// Dump a UUID value at \a offset.
290    ///
291    /// Dump a UUID starting at \a offset bytes into this object's data.
292    /// If the stream \a s is NULL, the output will be sent to Log().
293    ///
294    /// @param[in] s
295    ///     The stream to dump the output to. If NULL the output will
296    ///     be dumped to Log().
297    ///
298    /// @param[in] offset
299    ///     The offset into the data at which to extract and dump a
300    ///     UUID value.
301    //------------------------------------------------------------------
302    void
303    DumpUUID (Stream *s, uint32_t offset) const;
304
305    //------------------------------------------------------------------
306    /// Extract an arbitrary number of bytes in the specified byte
307    /// order.
308    ///
309    /// Attemps to extract \a length bytes starting at \a offset bytes
310    /// into this data in the requested byte order (\a dst_byte_order)
311    /// and place the results in \a dst. \a dst must be at least \a
312    /// length bytes long.
313    ///
314    /// @param[in] offset
315    ///     The offset in bytes into the contained data at which to
316    ///     start extracting.
317    ///
318    /// @param[in] length
319    ///     The number of bytes to extract.
320    ///
321    /// @param[in] dst_byte_order
322    ///     A byte order of the data that we want when the value in
323    ///     copied to \a dst.
324    ///
325    /// @param[out] dst
326    ///     The buffer that will receive the extracted value if there
327    ///     are enough bytes available in the current data.
328    ///
329    /// @return
330    ///     The number of bytes that were extracted which will be \a
331    ///     length when the value is successfully extracted, or zero
332    ///     if there aren't enough bytes at the specified offset.
333    //------------------------------------------------------------------
334    size_t
335    ExtractBytes (uint32_t offset, uint32_t length, lldb::ByteOrder dst_byte_order, void *dst) const;
336
337    //------------------------------------------------------------------
338    /// Extract an address from \a *offset_ptr.
339    ///
340    /// Extract a single address from the data and update the offset
341    /// pointed to by \a offset_ptr. The size of the extracted address
342    /// comes from the \a m_addr_size member variable and should be
343    /// set correctly prior to extracting any address values.
344    ///
345    /// @param[in,out] offset_ptr
346    ///     A pointer to an offset within the data that will be advanced
347    ///     by the appropriate number of bytes if the value is extracted
348    ///     correctly. If the offset is out of bounds or there are not
349    ///     enough bytes to extract this value, the offset will be left
350    ///     unmodified.
351    ///
352    /// @return
353    ///     The extracted address value.
354    //------------------------------------------------------------------
355    uint64_t
356    GetAddress (uint32_t *offset_ptr) const;
357
358    uint64_t
359    GetAddress_unchecked (uint32_t *offset_ptr) const;
360
361    //------------------------------------------------------------------
362    /// Get the current address size.
363    ///
364    /// Return the size in bytes of any address values this object will
365    /// extract.
366    ///
367    /// @return
368    ///     The size in bytes of address values that will be extracted.
369    //------------------------------------------------------------------
370    uint8_t
371    GetAddressByteSize () const
372    {
373        return m_addr_size;
374    }
375
376    //------------------------------------------------------------------
377    /// Get the number of bytes contained in this object.
378    ///
379    /// @return
380    ///     The total number of bytes of data this object refers to.
381    //------------------------------------------------------------------
382    size_t
383    GetByteSize () const
384    {
385        return m_end - m_start;
386    }
387
388    //------------------------------------------------------------------
389    /// Extract a C string from \a *offset_ptr.
390    ///
391    /// Returns a pointer to a C String from the data at the offset
392    /// pointed to by \a offset_ptr. A variable length NULL terminated C
393    /// string will be extracted and the \a offset_ptr will be
394    /// updated with the offset of the byte that follows the NULL
395    /// terminator byte.
396    ///
397    /// @param[in,out] offset_ptr
398    ///     A pointer to an offset within the data that will be advanced
399    ///     by the appropriate number of bytes if the value is extracted
400    ///     correctly. If the offset is out of bounds or there are not
401    ///     enough bytes to extract this value, the offset will be left
402    ///     unmodified.
403    ///
404    /// @return
405    ///     A pointer to the C string value in the data. If the offset
406    ///     pointed to by \a offset_ptr is out of bounds, or if the
407    ///     offset plus the length of the C string is out of bounds,
408    ///     NULL will be returned.
409    //------------------------------------------------------------------
410    const char *
411    GetCStr (uint32_t *offset_ptr) const;
412
413    //------------------------------------------------------------------
414    /// Extract \a length bytes from \a *offset_ptr.
415    ///
416    /// Returns a pointer to a bytes in this object's data at the offset
417    /// pointed to by \a offset_ptr. If \a length is zero or too large,
418    /// then the offset pointed to by \a offset_ptr will not be updated
419    /// and NULL will be returned.
420    ///
421    /// @param[in,out] offset_ptr
422    ///     A pointer to an offset within the data that will be advanced
423    ///     by the appropriate number of bytes if the value is extracted
424    ///     correctly. If the offset is out of bounds or there are not
425    ///     enough bytes to extract this value, the offset will be left
426    ///     unmodified.
427    ///
428    /// @param[in] length
429    ///     The optional length of a string to extract. If the value is
430    ///     zero, a NULL terminated C string will be extracted.
431    ///
432    /// @return
433    ///     A pointer to the bytes in this object's data if the offset
434    ///     and length are valid, or NULL otherwise.
435    //------------------------------------------------------------------
436    const void*
437    GetData (uint32_t *offset_ptr, uint32_t length) const;
438
439    //------------------------------------------------------------------
440    /// Copy \a dst_len bytes from \a *offset_ptr and ensure the copied
441    /// data is treated as a value that can be swapped to match the
442    /// specified byte order.
443    ///
444    /// For values that are larger than the supported integer sizes,
445    /// this function can be used to extract data in a specified byte
446    /// order. It can also be used to copy a smaller integer value from
447    /// to a larger value. The extra bytes left over will be padded
448    /// correctly according to the byte order of this object and the
449    /// \a dst_byte_order. This can be very handy when say copying a
450    /// partial data value into a register.
451    ///
452    /// @param[in] src_offset
453    ///     The offset into this data from which to start copying an
454    ///     endian entity
455    ///
456    /// @param[in] src_len
457    ///     The length of the endian data to copy from this object
458    ///     into the \a dst object
459    ///
460    /// @param[out] dst
461    ///     The buffer where to place the endian data. The data might
462    ///     need to be byte swapped (and appropriately padded with
463    ///     zeroes if \a src_len != \a dst_len) if \a dst_byte_order
464    ///     does not match the byte order in this object.
465    ///
466    /// @param[in] dst_len
467    ///     The length number of bytes that the endian value will
468    ///     occupy is \a dst.
469    ///
470    /// @param[in] byte_order
471    ///     The byte order that the endian value should be in the \a dst
472    ///     buffer.
473    ///
474    /// @return
475    ///     Returns the number of bytes that were copied, or zero if
476    ///     anything goes wrong.
477    //------------------------------------------------------------------
478    uint32_t
479    CopyByteOrderedData (uint32_t src_offset,
480                         uint32_t src_len,
481                         void *dst,
482                         uint32_t dst_len,
483                         lldb::ByteOrder dst_byte_order) const;
484
485    //------------------------------------------------------------------
486    /// Get the data end pointer.
487    ///
488    /// @return
489    ///     Returns a pointer to the next byte contained in this
490    ///     object's data, or NULL of there is no data in this object.
491    //------------------------------------------------------------------
492    const uint8_t *
493    GetDataEnd () const
494    {
495        return m_end;
496    }
497
498    //------------------------------------------------------------------
499    /// Get the shared data offset.
500    ///
501    /// Get the offset of the first byte of data in the shared data (if
502    /// any).
503    ///
504    /// @return
505    ///     If this object contains shared data, this function returns
506    ///     the offset in bytes into that shared data, zero otherwise.
507    //------------------------------------------------------------------
508    size_t
509    GetSharedDataOffset () const;
510
511    //------------------------------------------------------------------
512    /// Get a the data start pointer.
513    ///
514    /// @return
515    ///     Returns a pointer to the first byte contained in this
516    ///     object's data, or NULL of there is no data in this object.
517    //------------------------------------------------------------------
518    const uint8_t *
519    GetDataStart () const
520    {
521        return m_start;
522    }
523
524
525    //------------------------------------------------------------------
526    /// Extract a float from \a *offset_ptr.
527    ///
528    /// Extract a single float value.
529    ///
530    /// @param[in,out] offset_ptr
531    ///     A pointer to an offset within the data that will be advanced
532    ///     by the appropriate number of bytes if the value is extracted
533    ///     correctly. If the offset is out of bounds or there are not
534    ///     enough bytes to extract this value, the offset will be left
535    ///     unmodified.
536    ///
537    /// @return
538    ///     The floating value that was extracted, or zero on failure.
539    //------------------------------------------------------------------
540    float
541    GetFloat (uint32_t *offset_ptr) const;
542
543    double
544    GetDouble (uint32_t *offset_ptr) const;
545
546    long double
547    GetLongDouble (uint32_t *offset_ptr) const;
548
549    //------------------------------------------------------------------
550    /// Extract a GNU encoded pointer value from \a *offset_ptr.
551    ///
552    /// @param[in,out] offset_ptr
553    ///     A pointer to an offset within the data that will be advanced
554    ///     by the appropriate number of bytes if the value is extracted
555    ///     correctly. If the offset is out of bounds or there are not
556    ///     enough bytes to extract this value, the offset will be left
557    ///     unmodified.
558    ///
559    /// @param[in] eh_ptr_enc
560    ///     The GNU pointer encoding type.
561    ///
562    /// @param[in] pc_rel_addr
563    ///     The PC relative address to use when the encoding is
564    ///     \c DW_GNU_EH_PE_pcrel.
565    ///
566    /// @param[in] text_addr
567    ///     The text (code) relative address to use when the encoding is
568    ///     \c DW_GNU_EH_PE_textrel.
569    ///
570    /// @param[in] data_addr
571    ///     The data relative address to use when the encoding is
572    ///     \c DW_GNU_EH_PE_datarel.
573    ///
574    /// @return
575    ///     The extracted GNU encoded pointer value.
576    //------------------------------------------------------------------
577    uint64_t
578    GetGNUEHPointer (uint32_t *offset_ptr, uint32_t eh_ptr_enc, lldb::addr_t pc_rel_addr, lldb::addr_t text_addr, lldb::addr_t data_addr);
579
580    //------------------------------------------------------------------
581    /// Extract an integer of size \a byte_size from \a *offset_ptr.
582    ///
583    /// Extract a single integer value and update the offset pointed to
584    /// by \a offset_ptr. The size of the extracted integer is specified
585    /// by the \a byte_size argument. \a byte_size should have a value
586    /// >= 1 and <= 4 since the return value is only 32 bits wide. Any
587    /// \a byte_size values less than 1 or greater than 4 will result in
588    /// nothing being extracted, and zero being returned.
589    ///
590    /// @param[in,out] offset_ptr
591    ///     A pointer to an offset within the data that will be advanced
592    ///     by the appropriate number of bytes if the value is extracted
593    ///     correctly. If the offset is out of bounds or there are not
594    ///     enough bytes to extract this value, the offset will be left
595    ///     unmodified.
596    ///
597    /// @param[in] byte_size
598    ///     The size in byte of the integer to extract.
599    ///
600    /// @return
601    ///     The integer value that was extracted, or zero on failure.
602    //------------------------------------------------------------------
603    uint32_t
604    GetMaxU32 (uint32_t *offset_ptr, uint32_t byte_size) const;
605
606    //------------------------------------------------------------------
607    /// Extract an unsigned integer of size \a byte_size from \a
608    /// *offset_ptr.
609    ///
610    /// Extract a single unsigned integer value and update the offset
611    /// pointed to by \a offset_ptr. The size of the extracted integer
612    /// is specified by the \a byte_size argument. \a byte_size should
613    /// have a value greater than or equal to one and less than or equal
614    /// to eight since the return value is 64 bits wide. Any
615    /// \a byte_size values less than 1 or greater than 8 will result in
616    /// nothing being extracted, and zero being returned.
617    ///
618    /// @param[in,out] offset_ptr
619    ///     A pointer to an offset within the data that will be advanced
620    ///     by the appropriate number of bytes if the value is extracted
621    ///     correctly. If the offset is out of bounds or there are not
622    ///     enough bytes to extract this value, the offset will be left
623    ///     unmodified.
624    ///
625    /// @param[in] byte_size
626    ///     The size in byte of the integer to extract.
627    ///
628    /// @return
629    ///     The unsigned integer value that was extracted, or zero on
630    ///     failure.
631    //------------------------------------------------------------------
632    uint64_t
633    GetMaxU64 (uint32_t *offset_ptr, uint32_t byte_size) const;
634
635    uint64_t
636    GetMaxU64_unchecked (uint32_t *offset_ptr, uint32_t byte_size) const;
637
638    //------------------------------------------------------------------
639    /// Extract an signed integer of size \a byte_size from \a *offset_ptr.
640    ///
641    /// Extract a single signed integer value (sign extending if required)
642    /// and update the offset pointed to by \a offset_ptr. The size of
643    /// the extracted integer is specified by the \a byte_size argument.
644    /// \a byte_size should have a value greater than or equal to one
645    /// and less than or equal to eight since the return value is 64
646    /// bits wide. Any \a byte_size values less than 1 or greater than
647    /// 8 will result in nothing being extracted, and zero being returned.
648    ///
649    /// @param[in,out] offset_ptr
650    ///     A pointer to an offset within the data that will be advanced
651    ///     by the appropriate number of bytes if the value is extracted
652    ///     correctly. If the offset is out of bounds or there are not
653    ///     enough bytes to extract this value, the offset will be left
654    ///     unmodified.
655    ///
656    /// @param[in] byte_size
657    ///     The size in byte of the integer to extract.
658    ///
659    /// @return
660    ///     The sign extended signed integer value that was extracted,
661    ///     or zero on failure.
662    //------------------------------------------------------------------
663    int64_t
664    GetMaxS64 (uint32_t *offset_ptr, uint32_t size) const;
665
666    //------------------------------------------------------------------
667    /// Extract an unsigned integer of size \a byte_size from \a
668    /// *offset_ptr, then extract the bitfield from this value if
669    /// \a bitfield_bit_size is non-zero.
670    ///
671    /// Extract a single unsigned integer value and update the offset
672    /// pointed to by \a offset_ptr. The size of the extracted integer
673    /// is specified by the \a byte_size argument. \a byte_size should
674    /// have a value greater than or equal to one and less than or equal
675    /// to 8 since the return value is 64 bits wide. Any
676    /// \a byte_size values less than 1 or greater than 8 will result in
677    /// nothing being extracted, and zero being returned.
678    ///
679    /// @param[in,out] offset_ptr
680    ///     A pointer to an offset within the data that will be advanced
681    ///     by the appropriate number of bytes if the value is extracted
682    ///     correctly. If the offset is out of bounds or there are not
683    ///     enough bytes to extract this value, the offset will be left
684    ///     unmodified.
685    ///
686    /// @param[in] byte_size
687    ///     The size in byte of the integer to extract.
688    ///
689    /// @param[in] bitfield_bit_size
690    ///     The size in bits of the bitfield value to extract, or zero
691    ///     to just extract the entire integer value.
692    ///
693    /// @param[in] bitfield_bit_offset
694    ///     The bit offset of the bitfield value in the extracted
695    ///     integer (the number of bits to shift the integer to the
696    ///     right).
697    ///
698    /// @return
699    ///     The unsigned bitfield integer value that was extracted, or
700    ///     zero on failure.
701    //------------------------------------------------------------------
702    uint64_t
703    GetMaxU64Bitfield (uint32_t *offset_ptr, uint32_t size, uint32_t bitfield_bit_size, uint32_t bitfield_bit_offset) const;
704
705    //------------------------------------------------------------------
706    /// Extract an signed integer of size \a byte_size from \a
707    /// *offset_ptr, then extract and signe extend the bitfield from
708    /// this value if \a bitfield_bit_size is non-zero.
709    ///
710    /// Extract a single signed integer value (sign extending if required)
711    /// and update the offset pointed to by \a offset_ptr. The size of
712    /// the extracted integer is specified by the \a byte_size argument.
713    /// \a byte_size should have a value greater than or equal to one
714    /// and less than or equal to eight since the return value is 64
715    /// bits wide. Any \a byte_size values less than 1 or greater than
716    /// 8 will result in nothing being extracted, and zero being returned.
717    ///
718    /// @param[in,out] offset_ptr
719    ///     A pointer to an offset within the data that will be advanced
720    ///     by the appropriate number of bytes if the value is extracted
721    ///     correctly. If the offset is out of bounds or there are not
722    ///     enough bytes to extract this value, the offset will be left
723    ///     unmodified.
724    ///
725    /// @param[in] byte_size
726    ///     The size in bytes of the integer to extract.
727    ///
728    /// @param[in] bitfield_bit_size
729    ///     The size in bits of the bitfield value to extract, or zero
730    ///     to just extract the entire integer value.
731    ///
732    /// @param[in] bitfield_bit_offset
733    ///     The bit offset of the bitfield value in the extracted
734    ///     integer (the number of bits to shift the integer to the
735    ///     right).
736    ///
737    /// @return
738    ///     The signed bitfield integer value that was extracted, or
739    ///     zero on failure.
740    //------------------------------------------------------------------
741    int64_t
742    GetMaxS64Bitfield (uint32_t *offset_ptr, uint32_t size, uint32_t bitfield_bit_size, uint32_t bitfield_bit_offset) const;
743
744    //------------------------------------------------------------------
745    /// Extract an pointer from \a *offset_ptr.
746    ///
747    /// Extract a single pointer from the data and update the offset
748    /// pointed to by \a offset_ptr. The size of the extracted pointer
749    /// comes from the \a m_addr_size member variable and should be
750    /// set correctly prior to extracting any pointer values.
751    ///
752    /// @param[in,out] offset_ptr
753    ///     A pointer to an offset within the data that will be advanced
754    ///     by the appropriate number of bytes if the value is extracted
755    ///     correctly. If the offset is out of bounds or there are not
756    ///     enough bytes to extract this value, the offset will be left
757    ///     unmodified.
758    ///
759    /// @return
760    ///     The extracted pointer value as a 64 integer.
761    //------------------------------------------------------------------
762    uint64_t
763    GetPointer (uint32_t *offset_ptr) const;
764
765    //------------------------------------------------------------------
766    /// Get the current byte order value.
767    ///
768    /// @return
769    ///     The current byte order value from this object's internal
770    ///     state.
771    //------------------------------------------------------------------
772    lldb::ByteOrder
773    GetByteOrder() const
774    {
775        return m_byte_order;
776    }
777
778    //------------------------------------------------------------------
779    /// Extract a uint8_t value from \a *offset_ptr.
780    ///
781    /// Extract a single uint8_t from the binary data at the offset
782    /// pointed to by \a offset_ptr, and advance the offset on success.
783    ///
784    /// @param[in,out] offset_ptr
785    ///     A pointer to an offset within the data that will be advanced
786    ///     by the appropriate number of bytes if the value is extracted
787    ///     correctly. If the offset is out of bounds or there are not
788    ///     enough bytes to extract this value, the offset will be left
789    ///     unmodified.
790    ///
791    /// @return
792    ///     The extracted uint8_t value.
793    //------------------------------------------------------------------
794    uint8_t
795    GetU8 ( uint32_t *offset_ptr) const;
796
797    uint8_t
798    GetU8_unchecked (uint32_t *offset_ptr) const
799    {
800        uint8_t val = m_start[*offset_ptr];
801        *offset_ptr += 1;
802        return val;
803    }
804
805    uint16_t
806    GetU16_unchecked (uint32_t *offset_ptr) const;
807
808    uint32_t
809    GetU32_unchecked (uint32_t *offset_ptr) const;
810
811    uint64_t
812    GetU64_unchecked (uint32_t *offset_ptr) const;
813    //------------------------------------------------------------------
814    /// Extract \a count uint8_t values from \a *offset_ptr.
815    ///
816    /// Extract \a count uint8_t values from the binary data at the
817    /// offset pointed to by \a offset_ptr, and advance the offset on
818    /// success. The extracted values are copied into \a dst.
819    ///
820    /// @param[in,out] offset_ptr
821    ///     A pointer to an offset within the data that will be advanced
822    ///     by the appropriate number of bytes if the value is extracted
823    ///     correctly. If the offset is out of bounds or there are not
824    ///     enough bytes to extract this value, the offset will be left
825    ///     unmodified.
826    ///
827    /// @param[out] dst
828    ///     A buffer to copy \a count uint8_t values into. \a dst must
829    ///     be large enough to hold all requested data.
830    ///
831    /// @param[in] count
832    ///     The number of uint8_t values to extract.
833    ///
834    /// @return
835    ///     \a dst if all values were properly extracted and copied,
836    ///     NULL otherise.
837    //------------------------------------------------------------------
838    void *
839    GetU8 ( uint32_t *offset_ptr, void *dst, uint32_t count) const;
840
841    //------------------------------------------------------------------
842    /// Extract a uint16_t value from \a *offset_ptr.
843    ///
844    /// Extract a single uint16_t from the binary data at the offset
845    /// pointed to by \a offset_ptr, and update the offset on success.
846    ///
847    /// @param[in,out] offset_ptr
848    ///     A pointer to an offset within the data that will be advanced
849    ///     by the appropriate number of bytes if the value is extracted
850    ///     correctly. If the offset is out of bounds or there are not
851    ///     enough bytes to extract this value, the offset will be left
852    ///     unmodified.
853    ///
854    /// @return
855    ///     The extracted uint16_t value.
856    //------------------------------------------------------------------
857    uint16_t
858    GetU16 (uint32_t *offset_ptr) const;
859
860    //------------------------------------------------------------------
861    /// Extract \a count uint16_t values from \a *offset_ptr.
862    ///
863    /// Extract \a count uint16_t values from the binary data at the
864    /// offset pointed to by \a offset_ptr, and advance the offset on
865    /// success. The extracted values are copied into \a dst.
866    ///
867    /// @param[in,out] offset_ptr
868    ///     A pointer to an offset within the data that will be advanced
869    ///     by the appropriate number of bytes if the value is extracted
870    ///     correctly. If the offset is out of bounds or there are not
871    ///     enough bytes to extract this value, the offset will be left
872    ///     unmodified.
873    ///
874    /// @param[out] dst
875    ///     A buffer to copy \a count uint16_t values into. \a dst must
876    ///     be large enough to hold all requested data.
877    ///
878    /// @param[in] count
879    ///     The number of uint16_t values to extract.
880    ///
881    /// @return
882    ///     \a dst if all values were properly extracted and copied,
883    ///     NULL otherise.
884    //------------------------------------------------------------------
885    void *
886    GetU16 (uint32_t *offset_ptr, void *dst, uint32_t count) const;
887
888    //------------------------------------------------------------------
889    /// Extract a uint32_t value from \a *offset_ptr.
890    ///
891    /// Extract a single uint32_t from the binary data at the offset
892    /// pointed to by \a offset_ptr, and update the offset on success.
893    ///
894    /// @param[in,out] offset_ptr
895    ///     A pointer to an offset within the data that will be advanced
896    ///     by the appropriate number of bytes if the value is extracted
897    ///     correctly. If the offset is out of bounds or there are not
898    ///     enough bytes to extract this value, the offset will be left
899    ///     unmodified.
900    ///
901    /// @return
902    ///     The extracted uint32_t value.
903    //------------------------------------------------------------------
904    uint32_t
905    GetU32 (uint32_t *offset_ptr) const;
906
907    //------------------------------------------------------------------
908    /// Extract \a count uint32_t values from \a *offset_ptr.
909    ///
910    /// Extract \a count uint32_t values from the binary data at the
911    /// offset pointed to by \a offset_ptr, and advance the offset on
912    /// success. The extracted values are copied into \a dst.
913    ///
914    /// @param[in,out] offset_ptr
915    ///     A pointer to an offset within the data that will be advanced
916    ///     by the appropriate number of bytes if the value is extracted
917    ///     correctly. If the offset is out of bounds or there are not
918    ///     enough bytes to extract this value, the offset will be left
919    ///     unmodified.
920    ///
921    /// @param[out] dst
922    ///     A buffer to copy \a count uint32_t values into. \a dst must
923    ///     be large enough to hold all requested data.
924    ///
925    /// @param[in] count
926    ///     The number of uint32_t values to extract.
927    ///
928    /// @return
929    ///     \a dst if all values were properly extracted and copied,
930    ///     NULL otherise.
931    //------------------------------------------------------------------
932    void *
933    GetU32 (uint32_t *offset_ptr, void *dst, uint32_t count) const;
934
935    //------------------------------------------------------------------
936    /// Extract a uint64_t value from \a *offset_ptr.
937    ///
938    /// Extract a single uint64_t from the binary data at the offset
939    /// pointed to by \a offset_ptr, and update the offset on success.
940    ///
941    /// @param[in,out] offset_ptr
942    ///     A pointer to an offset within the data that will be advanced
943    ///     by the appropriate number of bytes if the value is extracted
944    ///     correctly. If the offset is out of bounds or there are not
945    ///     enough bytes to extract this value, the offset will be left
946    ///     unmodified.
947    ///
948    /// @return
949    ///     The extracted uint64_t value.
950    //------------------------------------------------------------------
951    uint64_t
952    GetU64 (uint32_t *offset_ptr) const;
953
954    //------------------------------------------------------------------
955    /// Extract \a count uint64_t values from \a *offset_ptr.
956    ///
957    /// Extract \a count uint64_t values from the binary data at the
958    /// offset pointed to by \a offset_ptr, and advance the offset on
959    /// success. The extracted values are copied into \a dst.
960    ///
961    /// @param[in,out] offset_ptr
962    ///     A pointer to an offset within the data that will be advanced
963    ///     by the appropriate number of bytes if the value is extracted
964    ///     correctly. If the offset is out of bounds or there are not
965    ///     enough bytes to extract this value, the offset will be left
966    ///     unmodified.
967    ///
968    /// @param[out] dst
969    ///     A buffer to copy \a count uint64_t values into. \a dst must
970    ///     be large enough to hold all requested data.
971    ///
972    /// @param[in] count
973    ///     The number of uint64_t values to extract.
974    ///
975    /// @return
976    ///     \a dst if all values were properly extracted and copied,
977    ///     NULL otherise.
978    //------------------------------------------------------------------
979    void *
980    GetU64 ( uint32_t *offset_ptr, void *dst, uint32_t count) const;
981
982    //------------------------------------------------------------------
983    /// Extract a signed LEB128 value from \a *offset_ptr.
984    ///
985    /// Extracts an signed LEB128 number from this object's data
986    /// starting at the offset pointed to by \a offset_ptr. The offset
987    /// pointed to by \a offset_ptr will be updated with the offset of
988    /// the byte following the last extracted byte.
989    ///
990    /// @param[in,out] offset_ptr
991    ///     A pointer to an offset within the data that will be advanced
992    ///     by the appropriate number of bytes if the value is extracted
993    ///     correctly. If the offset is out of bounds or there are not
994    ///     enough bytes to extract this value, the offset will be left
995    ///     unmodified.
996    ///
997    /// @return
998    ///     The extracted signed integer value.
999    //------------------------------------------------------------------
1000    int64_t
1001    GetSLEB128 (uint32_t *offset_ptr) const;
1002
1003    //------------------------------------------------------------------
1004    /// Extract a unsigned LEB128 value from \a *offset_ptr.
1005    ///
1006    /// Extracts an unsigned LEB128 number from this object's data
1007    /// starting at the offset pointed to by \a offset_ptr. The offset
1008    /// pointed to by \a offset_ptr will be updated with the offset of
1009    /// the byte following the last extracted byte.
1010    ///
1011    /// @param[in,out] offset_ptr
1012    ///     A pointer to an offset within the data that will be advanced
1013    ///     by the appropriate number of bytes if the value is extracted
1014    ///     correctly. If the offset is out of bounds or there are not
1015    ///     enough bytes to extract this value, the offset will be left
1016    ///     unmodified.
1017    ///
1018    /// @return
1019    ///     The extracted unsigned integer value.
1020    //------------------------------------------------------------------
1021    uint64_t
1022    GetULEB128 (uint32_t *offset_ptr) const;
1023
1024    lldb::DataBufferSP &
1025    GetSharedDataBuffer ()
1026    {
1027        return m_data_sp;
1028    }
1029
1030    //------------------------------------------------------------------
1031    /// Peek at a C string at \a offset.
1032    ///
1033    /// Peeks at a string in the contained data. No verification is done
1034    /// to make sure the entire string lies within the bounds of this
1035    /// object's data, only \a offset is verified to be a valid offset.
1036    ///
1037    /// @param[in] offset
1038    ///     An offset into the data.
1039    ///
1040    /// @return
1041    ///     A non-NULL C string pointer if \a offset is a valid offset,
1042    ///     NULL otherwise.
1043    //------------------------------------------------------------------
1044    const char *
1045    PeekCStr (uint32_t offset) const;
1046
1047    //------------------------------------------------------------------
1048    /// Peek at a bytes at \a offset.
1049    ///
1050    /// Returns a pointer to \a length bytes at \a offset as long as
1051    /// there are \a length bytes available starting at \a offset.
1052    ///
1053    /// @return
1054    ///     A non-NULL data pointer if \a offset is a valid offset and
1055    ///     there are \a length bytes available at that offset, NULL
1056    ///     otherwise.
1057    //------------------------------------------------------------------
1058    const uint8_t*
1059    PeekData (uint32_t offset, uint32_t length) const;
1060
1061    //------------------------------------------------------------------
1062    /// Set the address byte size.
1063    ///
1064    /// Set the size in bytes that will be used when extracting any
1065    /// address and pointer values from data contained in this object.
1066    ///
1067    /// @param[in] addr_size
1068    ///     The size in bytes to use when extracting addresses.
1069    //------------------------------------------------------------------
1070    void
1071    SetAddressByteSize (uint8_t addr_size)
1072    {
1073        m_addr_size = addr_size;
1074    }
1075
1076    //------------------------------------------------------------------
1077    /// Set data with a buffer that is caller owned.
1078    ///
1079    /// Use data that is owned by the caller when extracting values.
1080    /// The data must stay around as long as this object, or any object
1081    /// that copies a subset of this object's data, is valid. If \a
1082    /// bytes is NULL, or \a length is zero, this object will contain
1083    /// no data.
1084    ///
1085    /// @param[in] bytes
1086    ///     A pointer to caller owned data.
1087    ///
1088    /// @param[in] length
1089    ///     The length in bytes of \a bytes.
1090    ///
1091    /// @param[in] byte_order
1092    ///     A byte order of the data that we are extracting from.
1093    ///
1094    /// @return
1095    ///     The number of bytes that this object now contains.
1096    //------------------------------------------------------------------
1097    uint32_t
1098    SetData (const void *bytes, uint32_t length, lldb::ByteOrder byte_order);
1099
1100    //------------------------------------------------------------------
1101    /// Adopt a subset of \a data.
1102    ///
1103    /// Set this object's data to be a subset of the data bytes in \a
1104    /// data. If \a data contains shared data, then a reference to the
1105    /// shared data will be added to ensure the shared data stays around
1106    /// as long as any objects have references to the shared data. The
1107    /// byte order and the address size settings are copied from \a
1108    /// data. If \a offset is not a valid offset in \a data, then no
1109    /// reference to the shared data will be added. If there are not
1110    /// \a length bytes available in \a data starting at \a offset,
1111    /// the length will be truncated to contains as many bytes as
1112    /// possible.
1113    ///
1114    /// @param[in] data
1115    ///     Another DataExtractor object that contains data.
1116    ///
1117    /// @param[in] offset
1118    ///     The offset into \a data at which the subset starts.
1119    ///
1120    /// @param[in] length
1121    ///     The length in bytes of the subset of \a data.
1122    ///
1123    /// @return
1124    ///     The number of bytes that this object now contains.
1125    //------------------------------------------------------------------
1126    uint32_t
1127    SetData (const DataExtractor& data, uint32_t offset, uint32_t length);
1128
1129    //------------------------------------------------------------------
1130    /// Adopt a subset of shared data in \a data_sp.
1131    ///
1132    /// Copies the data shared pointer which adds a reference to the
1133    /// contained in \a data_sp. The shared data reference is reference
1134    /// counted to ensure the data lives as long as anyone still has a
1135    /// valid shared pointer to the data in \a data_sp. The byte order
1136    /// and address byte size settings remain the same. If
1137    /// \a offset is not a valid offset in \a data_sp, then no reference
1138    /// to the shared data will be added. If there are not \a length
1139    /// bytes available in \a data starting at \a offset, the length
1140    /// will be truncated to contains as many bytes as possible.
1141    ///
1142    /// @param[in] data_sp
1143    ///     A shared pointer to data.
1144    ///
1145    /// @param[in] offset
1146    ///     The offset into \a data_sp at which the subset starts.
1147    ///
1148    /// @param[in] length
1149    ///     The length in bytes of the subset of \a data_sp.
1150    ///
1151    /// @return
1152    ///     The number of bytes that this object now contains.
1153    //------------------------------------------------------------------
1154    uint32_t
1155    SetData (const lldb::DataBufferSP& data_sp, uint32_t offset = 0, uint32_t length = UINT32_MAX);
1156
1157    //------------------------------------------------------------------
1158    /// Set the byte_order value.
1159    ///
1160    /// Sets the byte order of the data to extract. Extracted values
1161    /// will be swapped if necessary when decoding.
1162    ///
1163    /// @param[in] byte_order
1164    ///     The byte order value to use when extracting data.
1165    //------------------------------------------------------------------
1166    void
1167    SetByteOrder (lldb::ByteOrder byte_order)
1168    {
1169        m_byte_order = byte_order;
1170    }
1171
1172    //------------------------------------------------------------------
1173    /// Skip an LEB128 number at \a *offset_ptr.
1174    ///
1175    /// Skips a LEB128 number (signed or unsigned) from this object's
1176    /// data starting at the offset pointed to by \a offset_ptr. The
1177    /// offset pointed to by \a offset_ptr will be updated with the
1178    /// offset of the byte following the last extracted byte.
1179    ///
1180    /// @param[in,out] offset_ptr
1181    ///     A pointer to an offset within the data that will be advanced
1182    ///     by the appropriate number of bytes if the value is extracted
1183    ///     correctly. If the offset is out of bounds or there are not
1184    ///     enough bytes to extract this value, the offset will be left
1185    ///     unmodified.
1186    ///
1187    /// @return
1188    //      The number of bytes consumed during the extraction.
1189    //------------------------------------------------------------------
1190    uint32_t
1191    Skip_LEB128 (uint32_t *offset_ptr) const;
1192
1193    //------------------------------------------------------------------
1194    /// Test the validity of \a offset.
1195    ///
1196    /// @return
1197    ///     \b true if \a offset is a valid offset into the data in this
1198    ///     object, \b false otherwise.
1199    //------------------------------------------------------------------
1200    bool
1201    ValidOffset (uint32_t offset) const
1202    {
1203        return offset < GetByteSize();
1204    }
1205
1206    //------------------------------------------------------------------
1207    /// Test the availability of \a length bytes of data from \a offset.
1208    ///
1209    /// @return
1210    ///     \b true if \a offset is a valid offset and there are \a
1211    ///     length bytes available at that offset, \b false otherwise.
1212    //------------------------------------------------------------------
1213    bool
1214    ValidOffsetForDataOfSize (uint32_t offset, uint32_t length) const;
1215
1216    size_t
1217    Copy (DataExtractor& dest_data) const;
1218
1219    bool
1220    Append (DataExtractor& rhs);
1221
1222    bool
1223    Append (void* bytes, uint32_t length);
1224
1225protected:
1226    //------------------------------------------------------------------
1227    // Member variables
1228    //------------------------------------------------------------------
1229    const uint8_t * m_start;        ///< A pointer to the first byte of data.
1230    const uint8_t * m_end;          ///< A pointer to the byte that is past the end of the data.
1231    lldb::ByteOrder m_byte_order;   ///< The byte order of the data we are extracting from.
1232    uint8_t         m_addr_size;    ///< The address size to use when extracting pointers or addresses
1233    mutable lldb::DataBufferSP m_data_sp; ///< The shared pointer to data that can be shared among multilple instances
1234};
1235
1236} // namespace lldb_private
1237
1238#endif  // #if defined (__cplusplus)
1239#endif  // #ifndef liblldb_DataExtractor_h_
1240