From c2011c5def9154c9a48f5e7e17d48d840aad675d Mon Sep 17 00:00:00 2001 From: erdgeist <> Date: Thu, 29 Apr 2004 19:42:16 +0000 Subject: Start --- src/dumpindex2.c | 33 ++++++++++++++ src/dumppointers2.c | 127 ++++++++++++++++++++++++++++++++++++++++++++++++++++ src/mystdlib.c | 54 ++++++++++++++++++++++ src/mystdlib.h | 32 +++++++++++++ src/sortindex.c | 81 +++++++++++++++++++++++++++++++++ 5 files changed, 327 insertions(+) create mode 100644 src/dumpindex2.c create mode 100644 src/dumppointers2.c create mode 100644 src/mystdlib.c create mode 100644 src/mystdlib.h create mode 100644 src/sortindex.c (limited to 'src') diff --git a/src/dumpindex2.c b/src/dumpindex2.c new file mode 100644 index 0000000..f49a329 --- /dev/null +++ b/src/dumpindex2.c @@ -0,0 +1,33 @@ +#include "mystdlib.h" +#include +#include + +static int indexed = -1; + +int main( int argc, char **argv ) { + MAP index = NULL; + int i,j; + char out[50]; + + if( argc != 3 ) + { fputs( "Syntax: sortindex ", stderr); exit( 1 ); } + + if( ( indexed = open( argv[1], O_RDONLY ) ) == -1 ) + { fprintf( stderr, "Could not open file: %s\n", argv[1] ); exit( 1 ); } + + if( !(index = map_file( argv[2], 0 ) ) ) exit( 1 ); + + for( i = 0; i < index->size; i+= 16 ) { + unsigned char *x = i + (unsigned char*)index->addr; + unsigned long p = *(unsigned long*)x; + + pread( indexed, out, 40, (off_t)p ); + for( j=0;j<40;++j) if( out[j] == '\t' || out[j] == '\n' ) out[j] = 0; out[j] = 0; + puts( out ); + } + + unmap_file( &index ); + close( indexed ); + + return 0; +} diff --git a/src/dumppointers2.c b/src/dumppointers2.c new file mode 100644 index 0000000..80a4b61 --- /dev/null +++ b/src/dumppointers2.c @@ -0,0 +1,127 @@ +#include +#include +#include + +#define BUFSIZE 0x1000000 +#define BUFLOW 0x400 +static int infile = -1; +static unsigned char nblock[ BUFSIZE + 16 ]; +static unsigned char*inblock = nblock + 16; +static unsigned long inblockoffs = 0; + +static unsigned long infilesize = 0; +static unsigned long infileoffs = 0; +static unsigned char to_lower[]; +static unsigned char run = 1; + +static void buffer_lowwatermark( void ) +{ + int i; + if( infilesize - infileoffs < BUFSIZE ) + return; + memmove( inblock, inblock + inblockoffs, BUFSIZE - inblockoffs ); + read( infile, inblock + (BUFSIZE - inblockoffs), inblockoffs); + for ( i = BUFSIZE - inblockoffs; i BUFSIZE - BUFLOW ) + buffer_lowwatermark(); + if( infilesize <= infileoffs + inblockoffs) + return 0; + if( !(mycount++ & 255) ) + printf( "%cSo far: %010.6f%% (%010lu / %010lu)", 13, 100.00 * (double)(infileoffs+inblockoffs) / (double)infilesize, infileoffs+inblockoffs, infilesize ); + // Fall through + case 1 : + use = (1< 1 ) ) + { + if( !fd[ b | (a*256) ][currrow] ) + nowfd = fd[ b | (a*256) ][currrow] = makefile( currrow, a, b); + else + nowfd = fd[ b | (a*256) ][currrow]; + + *(unsigned long*)(inblock+inblockoffs-4) = infileoffs + inblockoffs - 1 - inrow; + inblock[inblockoffs] = inrow++; + write( nowfd, inblock+inblockoffs-4, 16); + inblock[inblockoffs] = b; + break; + } + } + } + + return 0; +} diff --git a/src/mystdlib.c b/src/mystdlib.c new file mode 100644 index 0000000..2deda22 --- /dev/null +++ b/src/mystdlib.c @@ -0,0 +1,54 @@ +#include +#include +#include +#include +#include +#include + +#include "mystdlib.h" + +MAP map_file( char *filename, int readonly ) +{ + struct stat fstatus; + MAP map = (MAP)malloc( sizeof( *map )); + + if( map ) + { + memset( map, 0, sizeof( *map )); + + if( ( map->fh = open( filename, readonly ? O_RDONLY : O_RDWR ) ) >= 0 ) + { + fstat( map->fh, &fstatus ); + if( ( map->addr = mmap( NULL, map->size = (size_t)fstatus.st_size, + PROT_READ | ( readonly ? 0 : PROT_WRITE), MAP_NOCORE | (readonly ? 0 : MAP_SHARED), map->fh, 0) ) == MAP_FAILED ) + { + fprintf( stderr, "Mapping file '%s' failed\n", filename ); + close( map->fh ); free( map ); map = NULL; + } + } else { + fprintf( stderr, "Couldn't open file: '%s'\n", filename ); + free( map ); map = NULL; + } + } else { + fputs( "Couldn't allocate memory", stderr ); + } + + return map; +} + +void unmap_file ( MAP *pMap ) +{ + if( !pMap || !*pMap ) return; + munmap( (*pMap)->addr, (*pMap)->size); + close( (*pMap)->fh); + free( *pMap ); *pMap = NULL; +} + +int getfilesize( int fd, unsigned long *size) +{ + struct stat sb; + int ret; + if( fstat( fd, &sb )) return -1; + *size = sb.st_size; + return 0; +} diff --git a/src/mystdlib.h b/src/mystdlib.h new file mode 100644 index 0000000..2e9499f --- /dev/null +++ b/src/mystdlib.h @@ -0,0 +1,32 @@ +#include +#include + +typedef struct { int fh; unsigned char *addr; size_t size; } *MAP; + +/* Mapps a file into memory + returns pointer to the mapping struct, + containing the file's size, the mapped + address and its file handle. + + If readonly is true, the file will be + opened and mapped read only. File is + opened and mapped writable, if false. + + Returns NULL if memory could not be + allocated, file could not be opened or + mapped. Gives out an diagnostic message + on stderr +*/ +MAP map_file( char *filename, int readonly ); + +/* Unmapps a file from memory. NULL pointer + checks are being done, so this is safe + to be called from cleanup without knowing + whether there actually is a map. +*/ +void unmap_file ( MAP *pMap ); + +/* Gets file size of open file + returns != 0 in case of error */ +inline int getfilesize( int fd, unsigned long *size ); + diff --git a/src/sortindex.c b/src/sortindex.c new file mode 100644 index 0000000..b3b3bfe --- /dev/null +++ b/src/sortindex.c @@ -0,0 +1,81 @@ +#include "mystdlib.h" +#include +#include + +static int indexed = -1; +static unsigned char to_lower[] = { +0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, 0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x1F, +0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x2D, 0x2E, 0x2F, +0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x3A, 0x3B, 0x3C, 0x3D, 0x3E, 0x3F, +0x40, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, +0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0x5B, 0x5C, 0x5D, 0x5E, 0x5F, +0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, +0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0x7B, 0x7C, 0x7D, 0x7E, 0x7F, +0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x8D, 0x8E, 0x8F, +0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, 0x98, 0x99, 0x9A, 0x9B, 0x9C, 0x9D, 0x9E, 0x9F, +0xA0, 0xA1, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7, 0xA8, 0xA9, 0xAA, 0xAB, 0xAC, 0xAD, 0xAE, 0xAF, +0xB0, 0xB1, 0xB2, 0xB3, 0xB4, 0xB5, 0xB6, 0xB7, 0xB8, 0xB9, 0xBA, 0xBB, 0xBC, 0xBD, 0xBE, 0xBF, +0xC0, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7, 0xC8, 0xC9, 0xCA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF, +0xD0, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xF6, 0xD7, 0xD8, 0xD9, 0xDA, 0xDB, 0xFC, 0xDD, 0xDE, 0xDF, +0xE0, 0xE1, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7, 0xE8, 0xE9, 0xEA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF, +0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7, 0xF8, 0xF9, 0xFA, 0xFB, 0xFC, 0xFD, 0xFE, 0xFF +}; + +int +mystrcasecmp(const unsigned char *s1, const unsigned char *s2) { + while (to_lower[*s1] == to_lower[*s2++]) + if ( to_lower[*s1++] == '\0' ) return (0); + return (to_lower[*s1] - to_lower[*--s2]); +} + +int +mystrcasecmp2(const unsigned char *s1, const unsigned char *s2) { + int i = 0; + while (i++<11 && to_lower[*s1] == to_lower[*s2++]) + if ( to_lower[*s1++] == '\0' ) return (0); + if( i != 11 ) + return (to_lower[*s1] - to_lower[*--s2]); +} + +int cb_compare( const void* a, const void* b) { + int result; + + result = mystrcasecmp2( 5+(unsigned char*)a, 5+(unsigned char*)b ); + if( result != 0x1000 ) + return result; + + { + void *myptr1 = mmap( NULL, 1024, PROT_READ, MAP_NOCORE, indexed, 16+(off_t)*(unsigned long*)a ); + void *myptr2 = mmap( NULL, 1024, PROT_READ, MAP_NOCORE, indexed, 16+(off_t)*(unsigned long*)b ); + + if( myptr1 && myptr2) + result = mystrcasecmp( myptr1, myptr2 ); + else { + fprintf( stderr, "Mapping during sort failed.\n" ); + exit( 1 ); + } + munmap( myptr1, 1024 ); + munmap( myptr2, 1024 ); + } + return result; +} + +int main( int argc, char **argv ) { + MAP index = NULL; + + if( argc != 3 ) + { fputs( "Syntax: sortindex ", stderr); exit( 1 ); } + + if( ( indexed = open( argv[1], O_RDONLY ) ) == -1 ) + { fprintf( stderr, "Could not open file: %s\n", argv[1] ); exit( 1 ); } + + if( !(index = map_file( argv[2], 0 ) ) ) exit( 1 ); + + qsort( index->addr, index->size / 16, 16, cb_compare ); + + unmap_file( &index ); + close( indexed ); + + return 0; +} -- cgit v1.2.3