diff --git a/.gitignore b/.gitignore index 013b7bb..002d812 100644 --- a/.gitignore +++ b/.gitignore @@ -1,2 +1,2 @@ -/pg_filedump.o +/*.o /pg_filedump diff --git a/Makefile b/Makefile index 7894a49..6835ddd 100644 --- a/Makefile +++ b/Makefile @@ -14,16 +14,25 @@ PGSQL_LIB_DIR=$(shell $(PG_CONFIG) --libdir) PGSQL_BIN_DIR=$(shell $(PG_CONFIG) --bindir) DISTFILES= README.pg_filedump Makefile Makefile.contrib \ - pg_filedump.h pg_filedump.c + pg_filedump.h pg_filedump.c decode.h decode.c stringinfo.c pg_lzcompress.c all: pg_filedump -pg_filedump: pg_filedump.o - ${CC} ${PGSQL_LDFLAGS} ${LDFLAGS} -o pg_filedump pg_filedump.o -L${PGSQL_LIB_DIR} -lpgport +pg_filedump: pg_filedump.o decode.o stringinfo.o pg_lzcompress.o + ${CC} ${PGSQL_LDFLAGS} ${LDFLAGS} -o pg_filedump pg_filedump.o decode.o stringinfo.o pg_lzcompress.o -L${PGSQL_LIB_DIR} -lpgport pg_filedump.o: pg_filedump.c ${CC} ${PGSQL_CFLAGS} ${CFLAGS} -I${PGSQL_INCLUDE_DIR} pg_filedump.c -c +decode.o: decode.c + ${CC} ${PGSQL_CFLAGS} ${CFLAGS} -I${PGSQL_INCLUDE_DIR} decode.c -c + +stringinfo.o: stringinfo.c + ${CC} ${PGSQL_CFLAGS} ${CFLAGS} -I${PGSQL_INCLUDE_DIR} stringinfo.c -c + +pg_lzcompress.o: pg_lzcompress.c + ${CC} ${PGSQL_CFLAGS} ${CFLAGS} -I${PGSQL_INCLUDE_DIR} pg_lzcompress.c -c + dist: rm -rf pg_filedump-${FD_VERSION} pg_filedump-${FD_VERSION}.tar.gz mkdir pg_filedump-${FD_VERSION} diff --git a/README.pg_filedump b/README.pg_filedump index 1c950ae..723ae55 100644 --- a/README.pg_filedump +++ b/README.pg_filedump @@ -59,7 +59,7 @@ not require any manual adjustments of the Makefile. ------------------------------------------------------------------------ Invocation: -pg_filedump [-abcdfhikxy] [-R startblock [endblock]] [-S blocksize] [-s segsize] [-n segnumber] file +pg_filedump [-abcdfhikxy] [-R startblock [endblock]] [-D attrlist] [-S blocksize] [-s segsize] [-n segnumber] file Defaults are: relative addressing, range of the entire file, block size as listed on block 0 in the file @@ -71,6 +71,8 @@ The following options are valid for heap and index files: off all formatting options) -d Display formatted block content dump (Option will turn off all other formatting options) + -D Try to decode tuples using provided list of attribute types. + [attrlist] should be something like int,timestamp,bool,uuid -f Display formatted block content dump along with interpretation -h Display this information -i Display interpreted item details diff --git a/decode.c b/decode.c new file mode 100644 index 0000000..4da4c7c --- /dev/null +++ b/decode.c @@ -0,0 +1,814 @@ +#include "postgres.h" +#include "decode.h" +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define ATTRTYPES_STR_MAX_LEN (1024-1) + +typedef int (*decode_callback_t)(const char* buffer, unsigned int buff_size, + unsigned int* out_size); + +static int +decode_smallint(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_int(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_bigint(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_time(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_timetz(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_date(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_timestamp(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_float4(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_float8(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_bool(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_uuid(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_macaddr(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int +decode_string(const char* buffer, unsigned int buff_size, unsigned int* out_size); + +static int ncallbacks = 0; +static decode_callback_t callbacks[ATTRTYPES_STR_MAX_LEN / 2] = { NULL }; + +typedef struct { + char* name; + decode_callback_t callback; +} ParseCallbackTableItem; + +static ParseCallbackTableItem callback_table[] = { + { "smallserial", &decode_smallint }, + { "smallint", &decode_smallint }, + { "int", &decode_int }, + { "serial", &decode_int }, + { "bigint", &decode_bigint }, + { "bigserial", &decode_bigint }, + { "time", &decode_time }, + { "timetz", &decode_timetz }, + { "date", &decode_date }, + { "timestamp", &decode_timestamp }, + { "float4", &decode_float4 }, + { "float8", &decode_float8 }, + { "float", &decode_float8 }, + { "bool", &decode_bool }, + { "uuid", &decode_uuid }, + { "macaddr", &decode_macaddr }, + + /* internally all string types are stored the same way */ + { "char", &decode_string }, + { "varchar", &decode_string }, + { "text", &decode_string }, + { "json", &decode_string }, + { "xml", &decode_string }, + { NULL, NULL}, +}; + +static StringInfoData copyString; +static bool copyStringInitDone = false; + +/* + * Temporary buffer for storing decompressed data. + * + * 64K should be enough in most cases. If it's not user can manually change + * this limit. Unfortunately there is no way to know how much memory user + * is willing to allocate. + */ +static char decompress_tmp_buff[64*1024]; + +/* Used by some PostgreSQL macro definitions */ +void +ExceptionalCondition(const char *conditionName, + const char *errorType, + const char *fileName, + int lineNumber) +{ + printf("Exceptional condition: name = %s, type = %s, fname = %s, line = %d\n", + conditionName ? conditionName : "(NULL)", + errorType ? errorType : "(NULL)", + fileName ? fileName : "(NULL)", + lineNumber); + exit(1); +} + +/* Append given string to current COPY line */ +static void +CopyAppend(const char* str) +{ + if(!copyStringInitDone) + { + initStringInfo(©String); + copyStringInitDone = true; + } + + /* Caller probably wanted just to init copyString */ + if(str == NULL) + return; + + if(copyString.data[0] != '\0') + appendStringInfoString(©String, "\t"); + + appendStringInfoString(©String, str); +} + +/* + * Append given string to current COPY line and encode special symbols + * like \r, \n, \t and \\. + */ +static void +CopyAppendEncode(const char* str, int orig_len) +{ + /* + * Should be enough in most cases. If it's not user can manually change + * this limit. Unfortunately there is no way to know how much memory user + * is willing to allocate. + */ + static char tmp_buff[64*1024]; + /* Reserve one byte for a trailing zero. */ + const int max_offset = sizeof(tmp_buff) - 2; + int curr_offset = 0; + int len = orig_len; + + while(len > 0) + { + /* + * Make sure there is enough free space for at least one special symbol + * and a trailing zero. + */ + if(curr_offset > max_offset - 2) + { + printf("ERROR: Unable to properly encode a string since it's too " + "large (%d bytes). Try to increase tmp_buff size in CopyAppendEncode " + "procedure.\n", orig_len); + exit(1); + } + + /* + * Since we are working with potentially corrupted data we can encounter + * \0 as well. + */ + if(*str == '\0') + { + tmp_buff[curr_offset] = '\\'; + tmp_buff[curr_offset+1] = '0'; + curr_offset += 2; + } + else if(*str == '\r') + { + tmp_buff[curr_offset] = '\\'; + tmp_buff[curr_offset+1] = 'r'; + curr_offset += 2; + } + else if(*str == '\n') + { + tmp_buff[curr_offset] = '\\'; + tmp_buff[curr_offset+1] = 'n'; + curr_offset += 2; + } + else if(*str == '\t') + { + tmp_buff[curr_offset] = '\\'; + tmp_buff[curr_offset+1] = 'r'; + curr_offset += 2; + } + else if(*str == '\\') + { + tmp_buff[curr_offset] = '\\'; + tmp_buff[curr_offset+1] = '\\'; + curr_offset += 2; + } + else + { + /* It's a regular symbol. */ + tmp_buff[curr_offset] = *str; + curr_offset++; + } + + str++; + len--; + } + + tmp_buff[curr_offset] = '\0'; + CopyAppend(tmp_buff); +} + +/* CopyAppend version with format string support */ +#define CopyAppendFmt(fmt, ...) do { \ + char __copy_format_buff[512]; \ + snprintf(__copy_format_buff, sizeof(__copy_format_buff), fmt, ##__VA_ARGS__); \ + CopyAppend(__copy_format_buff); \ + } while(0) + +/* Discard accumulated COPY line */ +static void +CopyClear(void) +{ + /* Make sure init is done */ + CopyAppend(NULL); + + resetStringInfo(©String); +} + +/* Output and then clear accumulated COPY line */ +static void +CopyFlush(void) +{ + /* Make sure init is done */ + CopyAppend(NULL); + + printf("COPY: %s\n", copyString.data); + CopyClear(); +} + +/* + * Add a callback to `callbacks` table for given type name + * + * Arguments: + * type - name of a single type, always lowercase + * + * Return value is: + * == 0 - no error + * < 0 - invalid type name + */ +static int +AddTypeCallback(const char* type) +{ + int idx = 0; + + if(*type == '\0') /* ignore empty strings */ + return 0; + + while(callback_table[idx].name != NULL) + { + if(strcmp(callback_table[idx].name, type) == 0) + { + callbacks[ncallbacks] = callback_table[idx].callback; + ncallbacks++; + return 0; + } + idx++; + } + + printf("Error: type <%s> doesn't exist or is not currently supported\n", type); + printf("Full list of known types: "); + idx = 0; + while(callback_table[idx].name != NULL) + { + printf("%s ", callback_table[idx].name); + idx++; + } + printf("\n"); + return -1; +} + +/* + * Decode attribute types string like "int,timestamp,bool,uuid" + * + * Arguments: + * str - types string + * Return value is: + * == 0 - if string is valid + * < 0 - if string is invalid + */ +int +ParseAttributeTypesString(const char* str) +{ + char *curr_type, *next_type; + char attrtypes[ATTRTYPES_STR_MAX_LEN+1]; + int i, len = strlen(str); + + if(len > ATTRTYPES_STR_MAX_LEN) + { + printf("Error: attribute types string is longer then %u characters!\n", + ATTRTYPES_STR_MAX_LEN); + return -1; + } + + strcpy(attrtypes, str); + for(i = 0; i < len; i++) + attrtypes[i] = tolower(attrtypes[i]); + + curr_type = attrtypes; + while(curr_type) + { + next_type = strstr(curr_type, ","); + if(next_type) + { + *next_type = '\0'; + next_type++; + } + + if(AddTypeCallback(curr_type) < 0) + return -1; + + curr_type = next_type; + } + + return 0; +} + +/* + * Convert Julian day number (JDN) to a date. + * Copy-pasted from src/backend/utils/adt/datetime.c + */ +static void +j2date(int jd, int *year, int *month, int *day) +{ + unsigned int julian; + unsigned int quad; + unsigned int extra; + int y; + + julian = jd; + julian += 32044; + quad = julian / 146097; + extra = (julian - quad * 146097) * 4 + 3; + julian += 60 + quad * 3 + extra / 146097; + quad = julian / 1461; + julian -= quad * 1461; + y = julian * 4 / 1461; + julian = ((y != 0) ? ((julian + 305) % 365) : ((julian + 306) % 366)) + + 123; + y += quad * 4; + *year = y - 4800; + quad = julian * 2141 / 65536; + *day = julian - 7834 * quad / 256; + *month = (quad + 10) % MONTHS_PER_YEAR + 1; +} + +/* Decode a smallint type */ +static int +decode_smallint(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + const char* new_buffer = (const char*)TYPEALIGN(sizeof(int16), (uintptr_t)buffer); + unsigned int delta = (unsigned int)( (uintptr_t)new_buffer - (uintptr_t)buffer ); + + if(buff_size < delta) + return -1; + + buff_size -= delta; + buffer = new_buffer; + + if(buff_size < sizeof(int16)) + return -2; + + CopyAppendFmt("%d", (int)(*(int16*)buffer)); + *out_size = sizeof(int16) + delta; + return 0; +} + + +/* Decode an int type */ +static int +decode_int(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + const char* new_buffer = (const char*)TYPEALIGN(sizeof(int32), (uintptr_t)buffer); + unsigned int delta = (unsigned int)( (uintptr_t)new_buffer - (uintptr_t)buffer ); + + if(buff_size < delta) + return -1; + + buff_size -= delta; + buffer = new_buffer; + + if(buff_size < sizeof(int32)) + return -2; + + CopyAppendFmt("%d", *(int32*)buffer); + *out_size = sizeof(int32) + delta; + return 0; +} + +/* Decode a bigint type */ +static int +decode_bigint(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + const char* new_buffer = (const char*)TYPEALIGN(sizeof(int64), (uintptr_t)buffer); + unsigned int delta = (unsigned int)( (uintptr_t)new_buffer - (uintptr_t)buffer ); + + if(buff_size < delta) + return -1; + + buff_size -= delta; + buffer = new_buffer; + + if(buff_size < sizeof(int64)) + return -2; + + CopyAppendFmt("%ld", *(int64*)buffer); + *out_size = sizeof(int64) + delta; + return 0; +} + +/* Decode a time type */ +static int +decode_time(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + const char* new_buffer = (const char*)TYPEALIGN(sizeof(int64), (uintptr_t)buffer); + unsigned int delta = (unsigned int)( (uintptr_t)new_buffer - (uintptr_t)buffer ); + int64 timestamp, timestamp_sec; + + if(buff_size < delta) + return -1; + + buff_size -= delta; + buffer = new_buffer; + + if(buff_size < sizeof(int64)) + return -2; + + timestamp = *(int64*)buffer; + timestamp_sec = timestamp / 1000000; + *out_size = sizeof(int64) + delta; + + CopyAppendFmt("%02ld:%02ld:%02ld.%06ld", + timestamp_sec / 60 / 60, (timestamp_sec / 60) % 60, timestamp_sec % 60, + timestamp % 1000000); + + return 0; +} + +/* Decode a timetz type */ +static int +decode_timetz(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + const char* new_buffer = (const char*)TYPEALIGN(sizeof(int64), (uintptr_t)buffer); + unsigned int delta = (unsigned int)( (uintptr_t)new_buffer - (uintptr_t)buffer ); + int64 timestamp, timestamp_sec; + int32 tz_sec, tz_min; + + if(buff_size < delta) + return -1; + + buff_size -= delta; + buffer = new_buffer; + + if(buff_size < (sizeof(int64) + sizeof(int32))) + return -2; + + timestamp = *(int64*)buffer; + tz_sec = *(int32*)(buffer + sizeof(int64)); + timestamp_sec = timestamp / 1000000; + tz_min = - (tz_sec / 60); + *out_size = sizeof(int64) + sizeof(int32) + delta; + + CopyAppendFmt("%02ld:%02ld:%02ld.%06ld%c%02d:%02d", + timestamp_sec / 60 / 60, (timestamp_sec / 60) % 60, timestamp_sec % 60, + timestamp % 1000000, (tz_min > 0 ? '+' : '-'), abs(tz_min / 60), abs(tz_min % 60)); + + return 0; +} + +/* Decode a date type */ +static int +decode_date(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + const char* new_buffer = (const char*)TYPEALIGN(sizeof(int32), (uintptr_t)buffer); + unsigned int delta = (unsigned int)( (uintptr_t)new_buffer - (uintptr_t)buffer ); + int32 jd, year, month, day; + + if(buff_size < delta) + return -1; + + buff_size -= delta; + buffer = new_buffer; + + if(buff_size < sizeof(int32)) + return -2; + + *out_size = sizeof(int32) + delta; + + jd = *(int32*)buffer + POSTGRES_EPOCH_JDATE; + j2date(jd, &year, &month, &day); + + CopyAppendFmt("%04d-%02d-%02d%s", (year <= 0) ? -year + 1 : year, month, day, (year <= 0) ? " BC" : ""); + + return 0; +} + +/* Decode a timestamp type */ +static int +decode_timestamp(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + const char* new_buffer = (const char*)TYPEALIGN(sizeof(int64), (uintptr_t)buffer); + unsigned int delta = (unsigned int)( (uintptr_t)new_buffer - (uintptr_t)buffer ); + int64 timestamp, timestamp_sec; + int32 jd, year, month, day; + + if(buff_size < delta) + return -1; + + buff_size -= delta; + buffer = new_buffer; + + if(buff_size < sizeof(int64)) + return -2; + + *out_size = sizeof(int64) + delta; + timestamp = *(int64*)buffer; + + jd = timestamp / USECS_PER_DAY; + if (jd != 0) + timestamp -= jd * USECS_PER_DAY; + + if (timestamp < INT64CONST(0)) + { + timestamp += USECS_PER_DAY; + jd -= 1; + } + + /* add offset to go from J2000 back to standard Julian date */ + jd += POSTGRES_EPOCH_JDATE; + + j2date(jd, &year, &month, &day); + timestamp_sec = timestamp / 1000000; + + CopyAppendFmt("%04d-%02d-%02d %02ld:%02ld:%02ld.%06ld%s", + (year <= 0) ? -year + 1 : year, month, day, + timestamp_sec / 60 / 60, (timestamp_sec / 60) % 60, timestamp_sec % 60, + timestamp % 1000000, + (year <= 0) ? " BC" : ""); + + return 0; +} + +/* Decode a float4 type */ +static int +decode_float4(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + const char* new_buffer = (const char*)TYPEALIGN(sizeof(float), (uintptr_t)buffer); + unsigned int delta = (unsigned int)( (uintptr_t)new_buffer - (uintptr_t)buffer ); + + if(buff_size < delta) + return -1; + + buff_size -= delta; + buffer = new_buffer; + + if(buff_size < sizeof(float)) + return -2; + + CopyAppendFmt("%.12f", *(float*)buffer); + *out_size = sizeof(float) + delta; + return 0; +} + +/* Decode a float8 type */ +static int +decode_float8(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + const char* new_buffer = (const char*)TYPEALIGN(sizeof(double), (uintptr_t)buffer); + unsigned int delta = (unsigned int)( (uintptr_t)new_buffer - (uintptr_t)buffer ); + + if(buff_size < delta) + return -1; + + buff_size -= delta; + buffer = new_buffer; + + if(buff_size < sizeof(double)) + return -2; + + CopyAppendFmt("%.12lf", *(double*)buffer); + *out_size = sizeof(double) + delta; + return 0; +} + +/* Decode an uuid type */ +static int +decode_uuid(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + unsigned char uuid[16]; + + if(buff_size < sizeof(uuid)) + return -1; + + memcpy(uuid, buffer, sizeof(uuid)); + CopyAppendFmt("%02x%02x%02x%02x-%02x%02x-%02x%02x-%02x%02x-%02x%02x%02x%02x%02x%02x", + uuid[0], uuid[1], uuid[2], uuid[3], uuid[4], uuid[5], uuid[6], uuid[7], + uuid[8], uuid[9], uuid[10], uuid[11], uuid[12], uuid[13], uuid[14], uuid[15] + ); + *out_size = sizeof(uuid); + return 0; +} + +/* Decode a macaddr type */ +static int +decode_macaddr(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + unsigned char macaddr[6]; + const char* new_buffer = (const char*)TYPEALIGN(sizeof(int32), (uintptr_t)buffer); + unsigned int delta = (unsigned int)( (uintptr_t)new_buffer - (uintptr_t)buffer ); + + if(buff_size < delta) + return -1; + + buff_size -= delta; + buffer = new_buffer; + + if(buff_size < sizeof(macaddr)) + return -2; + + memcpy(macaddr, buffer, sizeof(macaddr)); + CopyAppendFmt("%02x:%02x:%02x:%02x:%02x:%02x", + macaddr[0], macaddr[1], macaddr[2], macaddr[3], macaddr[4], macaddr[5] + ); + *out_size = sizeof(macaddr) + delta; + return 0; +} + +/* Decode a bool type */ +static int +decode_bool(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + if(buff_size < sizeof(bool)) + return -1; + + CopyAppend(*(bool*)buffer ? "t" : "f"); + *out_size = sizeof(bool); + return 0; +} + +/* Decode char(N), varchar(N), text, json or xml types */ +static int +decode_string(const char* buffer, unsigned int buff_size, unsigned int* out_size) +{ + int padding = 0; + + /* Skip padding bytes. */ + while(*buffer == 0x00) + { + if(buff_size == 0) + return -1; + + buff_size--; + buffer++; + padding++; + } + + if(VARATT_IS_1B_E(buffer)) + { + /* + * 00000001 1-byte length word, unaligned, TOAST pointer + */ + uint8 tag = VARTAG_1B_E(buffer); + uint32 len = VARTAG_SIZE(tag); + if(len > buff_size) + return -1; + + CopyAppend("(TOASTED)"); + *out_size = padding + len; + return 0; + } + + if(VARATT_IS_1B(buffer)) + { + /* + * xxxxxxx1 1-byte length word, unaligned, uncompressed data (up to 126b) + * xxxxxxx is 1 + string length + */ + uint8 len = VARSIZE_1B(buffer); + if(len > buff_size) + return -1; + + CopyAppendEncode(buffer + 1, len - 1); + *out_size = padding + len; + return 0; + } + + if(VARATT_IS_4B_U(buffer) && buff_size >= 4) + { + /* + * xxxxxx00 4-byte length word, aligned, uncompressed data (up to 1G) + */ + uint32 len = VARSIZE_4B(buffer); + if(len > buff_size) + return -1; + + CopyAppendEncode(buffer + 4, len - 4); + *out_size = padding + len; + return 0; + } + + if(VARATT_IS_4B_C(buffer) && buff_size >= 8) + { + /* + * xxxxxx10 4-byte length word, aligned, *compressed* data (up to 1G) + */ + int decompress_ret; + uint32 len = VARSIZE_4B(buffer); + uint32 decompressed_len = VARRAWSIZE_4B_C(buffer); + + if(len > buff_size) + return -1; + + if(decompressed_len > sizeof(decompress_tmp_buff)) + { + printf("WARNING: Unable to decompress a string since it's too " + "large (%d bytes after decompressing). Consider increasing " + "decompress_tmp_buff size.\n", decompressed_len); + + CopyAppend("(COMPRESSED)"); + *out_size = padding + len; + return 0; + } + + decompress_ret = pglz_decompress(VARDATA_4B_C(buffer), len - 2*sizeof(uint32), + decompress_tmp_buff, decompressed_len); + if((decompress_ret != decompressed_len) || (decompress_ret < 0)) + { + printf("WARNING: Unable to decompress a string. Data is corrupted.\n"); + CopyAppend("(COMPRESSED)"); + *out_size = padding + len; + return 0; + } + + CopyAppendEncode(decompress_tmp_buff, decompressed_len); + *out_size = padding + len; + return 0; + } + + return -9; +} + +/* + * Try to decode a tuple using a types string provided previously. + * + * Arguments: + * tupleData - pointer to the tuple data + * tupleSize - tuple size in bytes + */ +void +FormatDecode(const char* tupleData, unsigned int tupleSize) +{ + HeapTupleHeader header = (HeapTupleHeader)tupleData; + const char* data = tupleData + header->t_hoff; + unsigned int size = tupleSize - header->t_hoff; + int curr_attr; + + CopyClear(); + + for(curr_attr = 0; curr_attr < ncallbacks; curr_attr++) + { + int ret; + unsigned int processed_size = 0; + + if( (header->t_infomask & HEAP_HASNULL) && att_isnull(curr_attr, header->t_bits) ) + { + CopyAppend("\\N"); + continue; + } + + if(size <= 0) + { + printf("Error: unable to decode a tuple, no more bytes left. Partial data: %s\n", + copyString.data); + return; + } + + ret = callbacks[curr_attr](data, size, &processed_size); + if(ret < 0) + { + printf("Error: unable to decode a tuple, callback #%d returned %d. Partial data: %s\n", + curr_attr+1, ret, copyString.data); + return; + } + + size -= processed_size; + data += processed_size; + } + + if(size != 0) + { + printf("Error: unable to decode a tuple, %d bytes left, 0 expected. Partial data: %s\n", + size, copyString.data); + return; + } + + CopyFlush(); +} diff --git a/decode.h b/decode.h new file mode 100644 index 0000000..a91b883 --- /dev/null +++ b/decode.h @@ -0,0 +1,10 @@ +#ifndef _PG_FILEDUMP_DECODE_H_ +#define _PG_FILEDUMP_DECODE_H_ + +int +ParseAttributeTypesString(const char* str); + +void +FormatDecode(const char* tupleData, unsigned int tupleSize); + +#endif diff --git a/pg_filedump.c b/pg_filedump.c index ba55711..fc02cc6 100644 --- a/pg_filedump.c +++ b/pg_filedump.c @@ -32,6 +32,7 @@ #include "storage/checksum.h" #include "storage/checksum_impl.h" +#include "decode.h" /*** * Global variables for ease of use mostly @@ -104,7 +105,7 @@ DisplayOptions(unsigned int validOptions) FD_VERSION, FD_PG_VERSION); printf - ("\nUsage: pg_filedump [-abcdfhikxy] [-R startblock [endblock]] [-S blocksize] [-s segsize] [-n segnumber] file\n\n" + ("\nUsage: pg_filedump [-abcdfhikxy] [-R startblock [endblock]] [-D attrlist] [-S blocksize] [-s segsize] [-n segnumber] file\n\n" "Display formatted contents of a PostgreSQL heap/index/control file\n" "Defaults are: relative addressing, range of the entire file, block\n" " size as listed on block 0 in the file\n\n" @@ -115,6 +116,8 @@ DisplayOptions(unsigned int validOptions) " off all formatting options)\n" " -d Display formatted block content dump (Option will turn off\n" " all other formatting options)\n" + " -D Try to decode tuples using provided list of attribute types.\n" + " [attrlist] should be something like int,timestamp,bool,uuid\n" " -f Display formatted block content dump along with interpretation\n" " -h Display this information\n" " -i Display interpreted item details\n" @@ -308,6 +311,36 @@ ConsumeOptions(int numOptions, char **options) break; } } + /* Check for the special case where the user forces tuples decoding. */ + else if((optionStringLength == 2) + && (strcmp(optionString, "-D") == 0)) + { + SET_OPTION(blockOptions, BLOCK_DECODE, 'D'); + /* Only accept the decode option once */ + if (rc == OPT_RC_DUPLICATE) + break; + + /* The token immediately following -D is attrubute types string */ + if (x >= (numOptions - 2)) + { + rc = OPT_RC_INVALID; + printf("Error: Missing attribute types string.\n"); + exitCode = 1; + break; + } + + /* Next option encountered must be attribute types string */ + optionString = options[++x]; + + if(ParseAttributeTypesString(optionString) < 0) + { + rc = OPT_RC_INVALID; + printf("Error: Invalid attribute types string <%s>.\n", + optionString); + exitCode = 1; + break; + } + } /* Check for the special case where the user forces a segment number */ /* instead of having the tool determine it by file name. */ else if ((optionStringLength == 2) @@ -968,6 +1001,10 @@ FormatItemBlock(Page page) if (blockOptions & BLOCK_FORMAT) FormatBinary(itemSize, itemOffset); + /* Decode tuple data */ + if(blockOptions & BLOCK_DECODE) + FormatDecode(&buffer[itemOffset], itemSize); + if (x == maxOffset) printf("\n"); } diff --git a/pg_filedump.h b/pg_filedump.h index d0634a8..345f6c0 100644 --- a/pg_filedump.h +++ b/pg_filedump.h @@ -52,7 +52,8 @@ typedef enum blockSwitches BLOCK_FORCED = 0x00000008, /* -S: Block size forced */ BLOCK_NO_INTR = 0x00000010, /* -d: Dump straight blocks */ BLOCK_RANGE = 0x00000020, /* -R: Specific block range to dump */ - BLOCK_CHECKSUMS = 0x00000040 /* -k: verify block checksums */ + BLOCK_CHECKSUMS = 0x00000040, /* -k: verify block checksums */ + BLOCK_DECODE = 0x00000080 /* -D: Try to decode tuples */ } blockSwitches; /* Segment-related options */ diff --git a/pg_lzcompress.c b/pg_lzcompress.c new file mode 100644 index 0000000..5ec93ec --- /dev/null +++ b/pg_lzcompress.c @@ -0,0 +1,778 @@ +/* ---------- + * pg_lzcompress.c - + * + * This is an implementation of LZ compression for PostgreSQL. + * It uses a simple history table and generates 2-3 byte tags + * capable of backward copy information for 3-273 bytes with + * a max offset of 4095. + * + * Entry routines: + * + * int32 + * pglz_compress(const char *source, int32 slen, char *dest, + * const PGLZ_Strategy *strategy); + * + * source is the input data to be compressed. + * + * slen is the length of the input data. + * + * dest is the output area for the compressed result. + * It must be at least as big as PGLZ_MAX_OUTPUT(slen). + * + * strategy is a pointer to some information controlling + * the compression algorithm. If NULL, the compiled + * in default strategy is used. + * + * The return value is the number of bytes written in the + * buffer dest, or -1 if compression fails; in the latter + * case the contents of dest are undefined. + * + * int32 + * pglz_decompress(const char *source, int32 slen, char *dest, + * int32 rawsize) + * + * source is the compressed input. + * + * slen is the length of the compressed input. + * + * dest is the area where the uncompressed data will be + * written to. It is the callers responsibility to + * provide enough space. + * + * The data is written to buff exactly as it was handed + * to pglz_compress(). No terminating zero byte is added. + * + * rawsize is the length of the uncompressed data. + * + * The return value is the number of bytes written in the + * buffer dest, or -1 if decompression fails. + * + * The decompression algorithm and internal data format: + * + * It is made with the compressed data itself. + * + * The data representation is easiest explained by describing + * the process of decompression. + * + * If compressed_size == rawsize, then the data + * is stored uncompressed as plain bytes. Thus, the decompressor + * simply copies rawsize bytes to the destination. + * + * Otherwise the first byte tells what to do the next 8 times. + * We call this the control byte. + * + * An unset bit in the control byte means, that one uncompressed + * byte follows, which is copied from input to output. + * + * A set bit in the control byte means, that a tag of 2-3 bytes + * follows. A tag contains information to copy some bytes, that + * are already in the output buffer, to the current location in + * the output. Let's call the three tag bytes T1, T2 and T3. The + * position of the data to copy is coded as an offset from the + * actual output position. + * + * The offset is in the upper nibble of T1 and in T2. + * The length is in the lower nibble of T1. + * + * So the 16 bits of a 2 byte tag are coded as + * + * 7---T1--0 7---T2--0 + * OOOO LLLL OOOO OOOO + * + * This limits the offset to 1-4095 (12 bits) and the length + * to 3-18 (4 bits) because 3 is always added to it. To emit + * a tag of 2 bytes with a length of 2 only saves one control + * bit. But we lose one byte in the possible length of a tag. + * + * In the actual implementation, the 2 byte tag's length is + * limited to 3-17, because the value 0xF in the length nibble + * has special meaning. It means, that the next following + * byte (T3) has to be added to the length value of 18. That + * makes total limits of 1-4095 for offset and 3-273 for length. + * + * Now that we have successfully decoded a tag. We simply copy + * the output that occurred bytes back to the current + * output location in the specified . Thus, a + * sequence of 200 spaces (think about bpchar fields) could be + * coded in 4 bytes. One literal space and a three byte tag to + * copy 199 bytes with a -1 offset. Whow - that's a compression + * rate of 98%! Well, the implementation needs to save the + * original data size too, so we need another 4 bytes for it + * and end up with a total compression rate of 96%, what's still + * worth a Whow. + * + * The compression algorithm + * + * The following uses numbers used in the default strategy. + * + * The compressor works best for attributes of a size between + * 1K and 1M. For smaller items there's not that much chance of + * redundancy in the character sequence (except for large areas + * of identical bytes like trailing spaces) and for bigger ones + * our 4K maximum look-back distance is too small. + * + * The compressor creates a table for lists of positions. + * For each input position (except the last 3), a hash key is + * built from the 4 next input bytes and the position remembered + * in the appropriate list. Thus, the table points to linked + * lists of likely to be at least in the first 4 characters + * matching strings. This is done on the fly while the input + * is compressed into the output area. Table entries are only + * kept for the last 4096 input positions, since we cannot use + * back-pointers larger than that anyway. The size of the hash + * table is chosen based on the size of the input - a larger table + * has a larger startup cost, as it needs to be initialized to + * zero, but reduces the number of hash collisions on long inputs. + * + * For each byte in the input, its hash key (built from this + * byte and the next 3) is used to find the appropriate list + * in the table. The lists remember the positions of all bytes + * that had the same hash key in the past in increasing backward + * offset order. Now for all entries in the used lists, the + * match length is computed by comparing the characters from the + * entries position with the characters from the actual input + * position. + * + * The compressor starts with a so called "good_match" of 128. + * It is a "prefer speed against compression ratio" optimizer. + * So if the first entry looked at already has 128 or more + * matching characters, the lookup stops and that position is + * used for the next tag in the output. + * + * For each subsequent entry in the history list, the "good_match" + * is lowered by 10%. So the compressor will be more happy with + * short matches the farer it has to go back in the history. + * Another "speed against ratio" preference characteristic of + * the algorithm. + * + * Thus there are 3 stop conditions for the lookup of matches: + * + * - a match >= good_match is found + * - there are no more history entries to look at + * - the next history entry is already too far back + * to be coded into a tag. + * + * Finally the match algorithm checks that at least a match + * of 3 or more bytes has been found, because that is the smallest + * amount of copy information to code into a tag. If so, a tag + * is omitted and all the input bytes covered by that are just + * scanned for the history add's, otherwise a literal character + * is omitted and only his history entry added. + * + * Acknowledgements: + * + * Many thanks to Adisak Pochanayon, who's article about SLZ + * inspired me to write the PostgreSQL compression this way. + * + * Jan Wieck + * + * Copyright (c) 1999-2017, PostgreSQL Global Development Group + * + * src/common/pg_lzcompress.c + * ---------- + */ +#ifndef FRONTEND +#include "postgres.h" +#else +#include "postgres_fe.h" +#endif + +#include + +#include "common/pg_lzcompress.h" + + +/* ---------- + * Local definitions + * ---------- + */ +#define PGLZ_MAX_HISTORY_LISTS 8192 /* must be power of 2 */ +#define PGLZ_HISTORY_SIZE 4096 +#define PGLZ_MAX_MATCH 273 + + +/* ---------- + * PGLZ_HistEntry - + * + * Linked list for the backward history lookup + * + * All the entries sharing a hash key are linked in a doubly linked list. + * This makes it easy to remove an entry when it's time to recycle it + * (because it's more than 4K positions old). + * ---------- + */ +typedef struct PGLZ_HistEntry +{ + struct PGLZ_HistEntry *next; /* links for my hash key's list */ + struct PGLZ_HistEntry *prev; + int hindex; /* my current hash key */ + const char *pos; /* my input position */ +} PGLZ_HistEntry; + + +/* ---------- + * The provided standard strategies + * ---------- + */ +static const PGLZ_Strategy strategy_default_data = { + 32, /* Data chunks less than 32 bytes are not + * compressed */ + INT_MAX, /* No upper limit on what we'll try to + * compress */ + 25, /* Require 25% compression rate, or not worth + * it */ + 1024, /* Give up if no compression in the first 1KB */ + 128, /* Stop history lookup if a match of 128 bytes + * is found */ + 10 /* Lower good match size by 10% at every loop + * iteration */ +}; +const PGLZ_Strategy *const PGLZ_strategy_default = &strategy_default_data; + + +static const PGLZ_Strategy strategy_always_data = { + 0, /* Chunks of any size are compressed */ + INT_MAX, + 0, /* It's enough to save one single byte */ + INT_MAX, /* Never give up early */ + 128, /* Stop history lookup if a match of 128 bytes + * is found */ + 6 /* Look harder for a good match */ +}; +const PGLZ_Strategy *const PGLZ_strategy_always = &strategy_always_data; + + +/* ---------- + * Statically allocated work arrays for history + * ---------- + */ +static int16 hist_start[PGLZ_MAX_HISTORY_LISTS]; +static PGLZ_HistEntry hist_entries[PGLZ_HISTORY_SIZE + 1]; + +/* + * Element 0 in hist_entries is unused, and means 'invalid'. Likewise, + * INVALID_ENTRY_PTR in next/prev pointers mean 'invalid'. + */ +#define INVALID_ENTRY 0 +#define INVALID_ENTRY_PTR (&hist_entries[INVALID_ENTRY]) + +/* ---------- + * pglz_hist_idx - + * + * Computes the history table slot for the lookup by the next 4 + * characters in the input. + * + * NB: because we use the next 4 characters, we are not guaranteed to + * find 3-character matches; they very possibly will be in the wrong + * hash list. This seems an acceptable tradeoff for spreading out the + * hash keys more. + * ---------- + */ +#define pglz_hist_idx(_s,_e, _mask) ( \ + ((((_e) - (_s)) < 4) ? (int) (_s)[0] : \ + (((_s)[0] << 6) ^ ((_s)[1] << 4) ^ \ + ((_s)[2] << 2) ^ (_s)[3])) & (_mask) \ + ) + + +/* ---------- + * pglz_hist_add - + * + * Adds a new entry to the history table. + * + * If _recycle is true, then we are recycling a previously used entry, + * and must first delink it from its old hashcode's linked list. + * + * NOTE: beware of multiple evaluations of macro's arguments, and note that + * _hn and _recycle are modified in the macro. + * ---------- + */ +#define pglz_hist_add(_hs,_he,_hn,_recycle,_s,_e, _mask) \ +do { \ + int __hindex = pglz_hist_idx((_s),(_e), (_mask)); \ + int16 *__myhsp = &(_hs)[__hindex]; \ + PGLZ_HistEntry *__myhe = &(_he)[_hn]; \ + if (_recycle) { \ + if (__myhe->prev == NULL) \ + (_hs)[__myhe->hindex] = __myhe->next - (_he); \ + else \ + __myhe->prev->next = __myhe->next; \ + if (__myhe->next != NULL) \ + __myhe->next->prev = __myhe->prev; \ + } \ + __myhe->next = &(_he)[*__myhsp]; \ + __myhe->prev = NULL; \ + __myhe->hindex = __hindex; \ + __myhe->pos = (_s); \ + /* If there was an existing entry in this hash slot, link */ \ + /* this new entry to it. However, the 0th entry in the */ \ + /* entries table is unused, so we can freely scribble on it. */ \ + /* So don't bother checking if the slot was used - we'll */ \ + /* scribble on the unused entry if it was not, but that's */ \ + /* harmless. Avoiding the branch in this critical path */ \ + /* speeds this up a little bit. */ \ + /* if (*__myhsp != INVALID_ENTRY) */ \ + (_he)[(*__myhsp)].prev = __myhe; \ + *__myhsp = _hn; \ + if (++(_hn) >= PGLZ_HISTORY_SIZE + 1) { \ + (_hn) = 1; \ + (_recycle) = true; \ + } \ +} while (0) + + +/* ---------- + * pglz_out_ctrl - + * + * Outputs the last and allocates a new control byte if needed. + * ---------- + */ +#define pglz_out_ctrl(__ctrlp,__ctrlb,__ctrl,__buf) \ +do { \ + if ((__ctrl & 0xff) == 0) \ + { \ + *(__ctrlp) = __ctrlb; \ + __ctrlp = (__buf)++; \ + __ctrlb = 0; \ + __ctrl = 1; \ + } \ +} while (0) + + +/* ---------- + * pglz_out_literal - + * + * Outputs a literal byte to the destination buffer including the + * appropriate control bit. + * ---------- + */ +#define pglz_out_literal(_ctrlp,_ctrlb,_ctrl,_buf,_byte) \ +do { \ + pglz_out_ctrl(_ctrlp,_ctrlb,_ctrl,_buf); \ + *(_buf)++ = (unsigned char)(_byte); \ + _ctrl <<= 1; \ +} while (0) + + +/* ---------- + * pglz_out_tag - + * + * Outputs a backward reference tag of 2-4 bytes (depending on + * offset and length) to the destination buffer including the + * appropriate control bit. + * ---------- + */ +#define pglz_out_tag(_ctrlp,_ctrlb,_ctrl,_buf,_len,_off) \ +do { \ + pglz_out_ctrl(_ctrlp,_ctrlb,_ctrl,_buf); \ + _ctrlb |= _ctrl; \ + _ctrl <<= 1; \ + if (_len > 17) \ + { \ + (_buf)[0] = (unsigned char)((((_off) & 0xf00) >> 4) | 0x0f); \ + (_buf)[1] = (unsigned char)(((_off) & 0xff)); \ + (_buf)[2] = (unsigned char)((_len) - 18); \ + (_buf) += 3; \ + } else { \ + (_buf)[0] = (unsigned char)((((_off) & 0xf00) >> 4) | ((_len) - 3)); \ + (_buf)[1] = (unsigned char)((_off) & 0xff); \ + (_buf) += 2; \ + } \ +} while (0) + + +/* ---------- + * pglz_find_match - + * + * Lookup the history table if the actual input stream matches + * another sequence of characters, starting somewhere earlier + * in the input buffer. + * ---------- + */ +static inline int +pglz_find_match(int16 *hstart, const char *input, const char *end, + int *lenp, int *offp, int good_match, int good_drop, int mask) +{ + PGLZ_HistEntry *hent; + int16 hentno; + int32 len = 0; + int32 off = 0; + + /* + * Traverse the linked history list until a good enough match is found. + */ + hentno = hstart[pglz_hist_idx(input, end, mask)]; + hent = &hist_entries[hentno]; + while (hent != INVALID_ENTRY_PTR) + { + const char *ip = input; + const char *hp = hent->pos; + int32 thisoff; + int32 thislen; + + /* + * Stop if the offset does not fit into our tag anymore. + */ + thisoff = ip - hp; + if (thisoff >= 0x0fff) + break; + + /* + * Determine length of match. A better match must be larger than the + * best so far. And if we already have a match of 16 or more bytes, + * it's worth the call overhead to use memcmp() to check if this match + * is equal for the same size. After that we must fallback to + * character by character comparison to know the exact position where + * the diff occurred. + */ + thislen = 0; + if (len >= 16) + { + if (memcmp(ip, hp, len) == 0) + { + thislen = len; + ip += len; + hp += len; + while (ip < end && *ip == *hp && thislen < PGLZ_MAX_MATCH) + { + thislen++; + ip++; + hp++; + } + } + } + else + { + while (ip < end && *ip == *hp && thislen < PGLZ_MAX_MATCH) + { + thislen++; + ip++; + hp++; + } + } + + /* + * Remember this match as the best (if it is) + */ + if (thislen > len) + { + len = thislen; + off = thisoff; + } + + /* + * Advance to the next history entry + */ + hent = hent->next; + + /* + * Be happy with lesser good matches the more entries we visited. But + * no point in doing calculation if we're at end of list. + */ + if (hent != INVALID_ENTRY_PTR) + { + if (len >= good_match) + break; + good_match -= (good_match * good_drop) / 100; + } + } + + /* + * Return match information only if it results at least in one byte + * reduction. + */ + if (len > 2) + { + *lenp = len; + *offp = off; + return 1; + } + + return 0; +} + + +/* ---------- + * pglz_compress - + * + * Compresses source into dest using strategy. Returns the number of + * bytes written in buffer dest, or -1 if compression fails. + * ---------- + */ +int32 +pglz_compress(const char *source, int32 slen, char *dest, + const PGLZ_Strategy *strategy) +{ + unsigned char *bp = (unsigned char *) dest; + unsigned char *bstart = bp; + int hist_next = 1; + bool hist_recycle = false; + const char *dp = source; + const char *dend = source + slen; + unsigned char ctrl_dummy = 0; + unsigned char *ctrlp = &ctrl_dummy; + unsigned char ctrlb = 0; + unsigned char ctrl = 0; + bool found_match = false; + int32 match_len; + int32 match_off; + int32 good_match; + int32 good_drop; + int32 result_size; + int32 result_max; + int32 need_rate; + int hashsz; + int mask; + + /* + * Our fallback strategy is the default. + */ + if (strategy == NULL) + strategy = PGLZ_strategy_default; + + /* + * If the strategy forbids compression (at all or if source chunk size out + * of range), fail. + */ + if (strategy->match_size_good <= 0 || + slen < strategy->min_input_size || + slen > strategy->max_input_size) + return -1; + + /* + * Limit the match parameters to the supported range. + */ + good_match = strategy->match_size_good; + if (good_match > PGLZ_MAX_MATCH) + good_match = PGLZ_MAX_MATCH; + else if (good_match < 17) + good_match = 17; + + good_drop = strategy->match_size_drop; + if (good_drop < 0) + good_drop = 0; + else if (good_drop > 100) + good_drop = 100; + + need_rate = strategy->min_comp_rate; + if (need_rate < 0) + need_rate = 0; + else if (need_rate > 99) + need_rate = 99; + + /* + * Compute the maximum result size allowed by the strategy, namely the + * input size minus the minimum wanted compression rate. This had better + * be <= slen, else we might overrun the provided output buffer. + */ + if (slen > (INT_MAX / 100)) + { + /* Approximate to avoid overflow */ + result_max = (slen / 100) * (100 - need_rate); + } + else + result_max = (slen * (100 - need_rate)) / 100; + + /* + * Experiments suggest that these hash sizes work pretty well. A large + * hash table minimizes collision, but has a higher startup cost. For a + * small input, the startup cost dominates. The table size must be a power + * of two. + */ + if (slen < 128) + hashsz = 512; + else if (slen < 256) + hashsz = 1024; + else if (slen < 512) + hashsz = 2048; + else if (slen < 1024) + hashsz = 4096; + else + hashsz = 8192; + mask = hashsz - 1; + + /* + * Initialize the history lists to empty. We do not need to zero the + * hist_entries[] array; its entries are initialized as they are used. + */ + memset(hist_start, 0, hashsz * sizeof(int16)); + + /* + * Compress the source directly into the output buffer. + */ + while (dp < dend) + { + /* + * If we already exceeded the maximum result size, fail. + * + * We check once per loop; since the loop body could emit as many as 4 + * bytes (a control byte and 3-byte tag), PGLZ_MAX_OUTPUT() had better + * allow 4 slop bytes. + */ + if (bp - bstart >= result_max) + return -1; + + /* + * If we've emitted more than first_success_by bytes without finding + * anything compressible at all, fail. This lets us fall out + * reasonably quickly when looking at incompressible input (such as + * pre-compressed data). + */ + if (!found_match && bp - bstart >= strategy->first_success_by) + return -1; + + /* + * Try to find a match in the history + */ + if (pglz_find_match(hist_start, dp, dend, &match_len, + &match_off, good_match, good_drop, mask)) + { + /* + * Create the tag and add history entries for all matched + * characters. + */ + pglz_out_tag(ctrlp, ctrlb, ctrl, bp, match_len, match_off); + while (match_len--) + { + pglz_hist_add(hist_start, hist_entries, + hist_next, hist_recycle, + dp, dend, mask); + dp++; /* Do not do this ++ in the line above! */ + /* The macro would do it four times - Jan. */ + } + found_match = true; + } + else + { + /* + * No match found. Copy one literal byte. + */ + pglz_out_literal(ctrlp, ctrlb, ctrl, bp, *dp); + pglz_hist_add(hist_start, hist_entries, + hist_next, hist_recycle, + dp, dend, mask); + dp++; /* Do not do this ++ in the line above! */ + /* The macro would do it four times - Jan. */ + } + } + + /* + * Write out the last control byte and check that we haven't overrun the + * output size allowed by the strategy. + */ + *ctrlp = ctrlb; + result_size = bp - bstart; + if (result_size >= result_max) + return -1; + + /* success */ + return result_size; +} + + +/* ---------- + * pglz_decompress - + * + * Decompresses source into dest. Returns the number of bytes + * decompressed in the destination buffer, or -1 if decompression + * fails. + * ---------- + */ +int32 +pglz_decompress(const char *source, int32 slen, char *dest, + int32 rawsize) +{ + const unsigned char *sp; + const unsigned char *srcend; + unsigned char *dp; + unsigned char *destend; + + sp = (const unsigned char *) source; + srcend = ((const unsigned char *) source) + slen; + dp = (unsigned char *) dest; + destend = dp + rawsize; + + while (sp < srcend && dp < destend) + { + /* + * Read one control byte and process the next 8 items (or as many as + * remain in the compressed input). + */ + unsigned char ctrl = *sp++; + int ctrlc; + + for (ctrlc = 0; ctrlc < 8 && sp < srcend; ctrlc++) + { + if (ctrl & 1) + { + /* + * Otherwise it contains the match length minus 3 and the + * upper 4 bits of the offset. The next following byte + * contains the lower 8 bits of the offset. If the length is + * coded as 18, another extension tag byte tells how much + * longer the match really was (0-255). + */ + int32 len; + int32 off; + + len = (sp[0] & 0x0f) + 3; + off = ((sp[0] & 0xf0) << 4) | sp[1]; + sp += 2; + if (len == 18) + len += *sp++; + + /* + * Check for output buffer overrun, to ensure we don't clobber + * memory in case of corrupt input. Note: we must advance dp + * here to ensure the error is detected below the loop. We + * don't simply put the elog inside the loop since that will + * probably interfere with optimization. + */ + if (dp + len > destend) + { + dp += len; + break; + } + + /* + * Now we copy the bytes specified by the tag from OUTPUT to + * OUTPUT. It is dangerous and platform dependent to use + * memcpy() here, because the copied areas could overlap + * extremely! + */ + while (len--) + { + *dp = dp[-off]; + dp++; + } + } + else + { + /* + * An unset control bit means LITERAL BYTE. So we just copy + * one from INPUT to OUTPUT. + */ + if (dp >= destend) /* check for buffer overrun */ + break; /* do not clobber memory */ + + *dp++ = *sp++; + } + + /* + * Advance the control bit + */ + ctrl >>= 1; + } + } + + /* + * Check we decompressed the right amount. + */ + if (dp != destend || sp != srcend) + return -1; + + /* + * That's it. + */ + return rawsize; +} diff --git a/stringinfo.c b/stringinfo.c new file mode 100644 index 0000000..4cdc8c5 --- /dev/null +++ b/stringinfo.c @@ -0,0 +1,170 @@ +/* + * Code mostly borrowed from PostgreSQL's stringinfo.c + * palloc replaced to malloc, etc. + */ + +#include "postgres.h" +#include +#include +#include + +/* 64 Kb - until pg_filedump doesn't support TOAST it doesn't need more */ +#define MaxAllocSize ((Size) (64*1024)) + +/*------------------------- + * StringInfoData holds information about an extensible string. + * data is the current buffer for the string. + * len is the current string length. There is guaranteed to be + * a terminating '\0' at data[len], although this is not very + * useful when the string holds binary data rather than text. + * maxlen is the allocated size in bytes of 'data', i.e. the maximum + * string size (including the terminating '\0' char) that we can + * currently store in 'data' without having to reallocate + * more space. We must always have maxlen > len. + * cursor is initialized to zero by makeStringInfo or initStringInfo, + * but is not otherwise touched by the stringinfo.c routines. + * Some routines use it to scan through a StringInfo. + *------------------------- + */ + +/* + * initStringInfo + * + * Initialize a StringInfoData struct (with previously undefined contents) + * to describe an empty string. + */ +void +initStringInfo(StringInfo str) +{ + int size = 1024; /* initial default buffer size */ + + str->data = (char *) malloc(size); + str->maxlen = size; + resetStringInfo(str); +} + +/* + * resetStringInfo + * + * Reset the StringInfo: the data buffer remains valid, but its + * previous content, if any, is cleared. + */ +void +resetStringInfo(StringInfo str) +{ + str->data[0] = '\0'; + str->len = 0; + str->cursor = 0; +} + +/* + * appendStringInfoString + * + * Append a null-terminated string to str. + */ +void +appendStringInfoString(StringInfo str, const char *s) +{ + appendBinaryStringInfo(str, s, strlen(s)); +} + +/* + * appendBinaryStringInfo + * + * Append arbitrary binary data to a StringInfo, allocating more space + * if necessary. + */ +void +appendBinaryStringInfo(StringInfo str, const char *data, int datalen) +{ + assert(str != NULL); + + /* Make more room if needed */ + enlargeStringInfo(str, datalen); + + /* OK, append the data */ + memcpy(str->data + str->len, data, datalen); + str->len += datalen; + + /* + * Keep a trailing null in place, even though it's probably useless for + * binary data. (Some callers are dealing with text but call this because + * their input isn't null-terminated.) + */ + str->data[str->len] = '\0'; +} + +/* + * enlargeStringInfo + * + * Make sure there is enough space for 'needed' more bytes + * ('needed' does not include the terminating null). + * + * External callers usually need not concern themselves with this, since + * all stringinfo.c routines do it automatically. However, if a caller + * knows that a StringInfo will eventually become X bytes large, it + * can save some malloc overhead by enlarging the buffer before starting + * to store data in it. + */ +void +enlargeStringInfo(StringInfo str, int needed) +{ + Size newlen; + Size limit; + char* old_data; + + limit = MaxAllocSize; + + /* + * Guard against out-of-range "needed" values. Without this, we can get + * an overflow or infinite loop in the following. + */ + if (needed < 0) /* should not happen */ + { + printf("Error: invalid string enlargement request size: %d", needed); + exit(1); + } + + if (((Size) needed) >= (limit - (Size) str->len)) + { + printf("Error: cannot enlarge string buffer containing %d bytes by %d more bytes.", + str->len, needed); + exit(1); + } + + needed += str->len + 1; /* total space required now */ + + /* Because of the above test, we now have needed <= limit */ + + if (needed <= str->maxlen) + return; /* got enough space already */ + + /* + * We don't want to allocate just a little more space with each append; + * for efficiency, double the buffer size each time it overflows. + * Actually, we might need to more than double it if 'needed' is big... + */ + newlen = 2 * str->maxlen; + while (needed > newlen) + newlen = 2 * newlen; + + /* + * Clamp to the limit in case we went past it. Note we are assuming here + * that limit <= INT_MAX/2, else the above loop could overflow. We will + * still have newlen >= needed. + */ + if (newlen > limit) + newlen = limit; + + old_data = str->data; + str->data = (char *) realloc(str->data, (Size) newlen); + if(str->data == NULL) + { + free(old_data); + printf("Error: realloc() failed!\n"); + exit(1); + } + + str->maxlen = newlen; +} +