BGI/BASE/src/2bwt/TextConverter.h

   1 /*
   2
   3    TextConverter.h              Text Converter
   4
   5    This module contains miscellaneous text conversion functions.
   6
   7    Copyright (C) 2004, Wong Chi Kwong.
   8
   9    This program is free software; you can redistribute it and/or
  10    modify it under the terms of the GNU General Public License
  11    as published by the Free Software Foundation; either version 2
  12    of the License, or (at your option) any later version.
  13
  14    This program is distributed in the hope that it will be useful,
  15    but WITHOUT ANY WARRANTY; without even the implied warranty of
  16    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  17    GNU General Public License for more details.
  18
  19    You should have received a copy of the GNU General Public License
  20    along with this program; if not, write to the Free Software
  21    Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  22
  23 */
  24
  25 #ifndef __TEXTCONVERTOR_H__
  26 #define __TEXTCONVERTOR_H__
  27
  28 #include "TypeNLimit.h"
  29 #include "MemManager.h"
  30
  31 #define INVALID_CHAR 0xFF
  32 #define CHAR_MAP_SIZE 256
  33 #define PACKED_BUFFER_SIZE                      (PACKED_BUFFER_SIZE_IN_WORD * BYTES_IN_WORD)
  34 #define PACKED_BUFFER_SIZE_IN_WORD      65536
  35 #define MAX_SEQ_NAME_LENGTH                     256
  36 #define RANDOM_SUBSTITUTE                       'R'
  37
  38 // charMap is a char array of size 256. The index of the array is the input text value
  39 // and the content of the array is the output text value. e.g. A -> 0, C -> 1
  40 // If the value of an entry = INVALID_CHAR, the indexed text value is an invalid input
  41
  42 // Retrieve word packed text
  43 unsigned long long GetWordPackedText(const unsigned int *packedText, const unsigned long long index, const unsigned long long shift, const unsigned long long numberOfBit, const unsigned long long vacantBit);
  44
  45 // Character map functions
  46 unsigned long long ReadCharMap(unsigned char *charMap, const char *inputFileName, const unsigned char defaultMapping);
  47 void GenerateReverseCharMap(const unsigned char *charMap, unsigned char *reverseCharMap);
  48
  49 // Word packed text functions
  50 unsigned int BitPerWordPackedChar(const unsigned int alphabetSize);
  51 unsigned long long TextLengthFromWordPacked(unsigned long long wordPackedLength, unsigned int bitPerChar, unsigned int lastWordLength);
  52 unsigned long long WordPackedLengthFromText(unsigned long long textLength, unsigned int bitPerChar);
  53 unsigned int LastWordLength(unsigned long long textLength, unsigned int bitPerChar);
  54
  55 // Byte packed text functions
  56 unsigned int BitPerBytePackedChar(const unsigned int alphabetSize);
  57 unsigned long long TextLengthFromBytePacked(unsigned long long bytePackedLength, unsigned int bitPerChar, unsigned int lastByteLength);
  58 unsigned long long BytePackedLengthFromText(unsigned long long textLength, unsigned int bitPerChar);
  59 unsigned char LastByteLength(unsigned long long textLength, unsigned int bitPerChar);
  60
  61 // Conversion functions
  62 void ConvertTextToWordPacked(const unsigned char *input, unsigned int *output, const unsigned char *charMap, const unsigned int alphabetSize, const unsigned long long textLength);
  63 void ConvertTextToBytePacked(const unsigned char *input, unsigned char *output, const unsigned char *charMap, const unsigned int alphabetSize, const unsigned long long textLength);
  64 void ConvertWordPackedToText(const unsigned int *input, unsigned char *output, const unsigned char *reverseCharMap, const unsigned int alphabetSize, const unsigned long long textLength);
  65 void ConvertBytePackedToText(const unsigned char *input, unsigned char *output, const unsigned char *reverseCharMap, const unsigned int alphabetSize, const unsigned long long textLength);
  66 void ConvertBytePackedToCode(const unsigned char *input, unsigned char *output, const unsigned int alphabetSize, const unsigned long long textLength);
  67 void ConvertWordPackedToBytePacked(const unsigned int *input, unsigned char *output, const unsigned int alphabetSize, const unsigned long long textLength);
  68 void ConvertBytePackedToWordPacked(const unsigned char *input, unsigned int *output, const unsigned int alphabetSize, const unsigned long long textLength);
  69 void ConvertTextToCode(const unsigned char *input, unsigned char *output, const unsigned char *charMap, const unsigned long long textLength);
  70 void ConvertCodeToText(const unsigned char *input, unsigned char *output, const unsigned char *reverseCharMap, const unsigned long long textLength);
  71
  72 // Pack text with all shift
  73 void PackTextWithAllShift(const unsigned char *input, unsigned int **output, const unsigned char *charMap, const unsigned int alphabetSize, const unsigned long long textLength);
  74
  75 // Full load function
  76 unsigned long long ReadTextAsWordPacked(const char *inputFileName, const unsigned char *charMap, const unsigned int alphabetSize, unsigned int *targetAddress, const unsigned long long maxTextLength);
  77 unsigned long long ReadBytePackedAsWordPacked(const char *inputFileName, const unsigned int alphabetSize, unsigned int *targetAddress, const unsigned long long maxTextLength);
  78 void *DNALoadPacked(const char *inputFileName, unsigned long long *textLength, const unsigned int convertToWordPacked, const unsigned int trailerBufferInWord);
  79 void DNAFreePacked(void* packedDna, const unsigned long long textLength, const unsigned int trailerBufferInWord);
  80
  81 // Save functions
  82 void SaveText(const char *outputFileName, const unsigned char *text, const unsigned long long textLength);
  83 void SaveBytePacked(const char *outputFileName, const unsigned char *wordPacked, const unsigned long long textLength, const unsigned int alphabetSize);
  84 void SaveWordPacked(const char *outputFileName, const unsigned int *wordPacked, const unsigned long long textLength, const unsigned int alphabetSize);
  85
  86 // Incremental load functions (start from end of text)
  87 FILE *InitialLoadPackedIncFromEnd(const char* inputFileName, unsigned char *packedOutput, const unsigned int alphabetSize, const unsigned long long packedLengthPerLoad, unsigned long long *textLength, unsigned long long *textLengthForThisLoad);
  88 void LoadPackedIncFromEnd(FILE *packedFile, unsigned char *packedOutput, const unsigned long long packedLengthPerLoad);
  89 FILE *InitialLoadTextIncFromEnd(const char* inputFileName, unsigned char *textOutput, const unsigned long long textLengthPerLoad, unsigned long long *textLength, unsigned long long *textLengthForThisLoad);
  90 void LoadTextIncFromEnd(FILE *textFile, unsigned char *textOutput, const unsigned long long textLengthPerLoad);
  91
  92
  93 #endif