hash(3) [netbsd man page]
HASH(3) BSD Library Functions Manual HASH(3) NAME
hash -- hash database access method SYNOPSIS
#include <sys/types.h> #include <db.h> DESCRIPTION
The routine dbopen() is the library interface to database files. One of the supported file formats is hash files. The general description of the database access methods is in dbopen(3), this manual page describes only the hash specific information. The hash data structure is an extensible, dynamic hashing scheme. The access method specific data structure provided to dbopen() is defined in the <db.h> header as follows: typedef struct { u_int bsize; u_int ffactor; u_int nelem; u_int cachesize; uint32_t (*hash)(const void *, size_t); int lorder; } HASHINFO; The elements of this structure are as follows: bsize bsize defines the hash table bucket size, and defaults to 4096 for in-memory tables. If bsize is 0 (no bucket size is specified) a bucket size is chosen based on the underlying file system I/O block size. It may be preferable to increase the page size for disk-resident tables and tables with large data items. ffactor ffactor indicates a desired density within the hash table. It is an approximation of the number of keys allowed to accumulate in any one bucket, determining when the hash table grows or shrinks. The default value is 8. nelem nelem is an estimate of the final size of the hash table. If not set or set too low, hash tables will expand gracefully as keys are entered, although a slight performance degradation may be noticed. The default value is 1. cachesize A suggested maximum size, in bytes, of the memory cache. This value is only advisory, and the access method will allocate more memory rather than fail. hash hash is a user defined hash function. Since no hash function performs equally well on all possible data, the user may find that the built-in hash function does poorly on a particular data set. User specified hash functions must take two arguments (a pointer to a byte string and a length) and return a 32-bit quantity to be used as the hash value. lorder The byte order for integers in the stored database metadata. The number should represent the order as an integer; for example, big endian order would be the number 4,321. If lorder is 0 (no order is specified) the current host order is used. If the file already exists, the specified value is ignored and the value specified when the tree was created is used. If the file already exists (and the O_TRUNC flag is not specified), the values specified for the parameters bsize, ffactor, lorder, and nelem are ignored and the values specified when the tree was created are used. If a hash function is specified, hash_open() will attempt to determine if the hash function specified is the same as the one with which the database was created, and will fail if it is not. ERRORS
The hash access method routines may fail and set errno for any of the errors specified for the library routine dbopen(3). SEE ALSO
btree(3), dbopen(3), mpool(3), recno(3) Per-Ake Larson, "Dynamic Hash Tables", Communications of the ACM, Issue 4, Volume 31, April 1988. Margo Seltzer, "A New Hash Package for UNIX", Proceedings of the 1991 Winter USENIX Technical Conference, USENIX Association, http://www.usenix.org/publications/library/proceedings/seltzer2.pdf, 173-184, January 1991. BUGS
Only big and little endian byte order is supported. BSD
December 16, 2010 BSD
Check Out this Related Man Page
RECNO(3) BSD Library Functions Manual RECNO(3) NAME
recno -- record number database access method SYNOPSIS
#include <sys/types.h> #include <db.h> DESCRIPTION
The routine dbopen() is the library interface to database files. One of the supported file formats is record number files. The general description of the database access methods is in dbopen(3), this manual page describes only the recno specific information. The record number data structure is either variable or fixed-length records stored in a flat-file format, accessed by the logical record num- ber. The existence of record number five implies the existence of records one through four, and the deletion of record number one causes record number five to be renumbered to record number four, as well as the cursor, if positioned after record number one, to shift down one record. The recno access method specific data structure provided to dbopen() is defined in the <db.h> include file as follows: typedef struct { u_long flags; u_int cachesize; u_int psize; int lorder; size_t reclen; uint8_t bval; char *bfname; } RECNOINFO; The elements of this structure are defined as follows: flags The flag value is specified by or'ing any of the following values: R_FIXEDLEN The records are fixed-length, not byte delimited. The structure element reclen specifies the length of the record, and the structure element bval is used as the pad character. Any records, inserted into the database, that are less than reclen bytes long are automatically padded. R_NOKEY In the interface specified by dbopen(), the sequential record retrieval fills in both the caller's key and data structures. If the R_NOKEY flag is specified, the cursor routines are not required to fill in the key structure. This permits applications to retrieve records at the end of files without reading all of the intervening records. R_SNAPSHOT This flag requires that a snapshot of the file be taken when dbopen() is called, instead of permitting any unmodified records to be read from the original file. cachesize A suggested maximum size, in bytes, of the memory cache. This value is only advisory, and the access method will allocate more memory rather than fail. If cachesize is 0 (no size is specified) a default cache is used. psize The recno access method stores the in-memory copies of its records in a btree. This value is the size (in bytes) of the pages used for nodes in that tree. If psize is 0 (no page size is specified) a page size is chosen based on the underlying file system I/O block size. See btree(3) for more information. lorder The byte order for integers in the stored database metadata. The number should represent the order as an integer; for example, big endian order would be the number 4,321. If lorder is 0 (no order is specified) the current host order is used. reclen The length of a fixed-length record. bval The delimiting byte to be used to mark the end of a record for variable-length records, and the pad character for fixed-length records. If no value is specified, newlines (`` '') are used to mark the end of variable-length records and fixed-length records are padded with spaces. bfname The recno access method stores the in-memory copies of its records in a btree. If bfname is non-NULL, it specifies the name of the btree file, as if specified as the file name for a dbopen() of a btree file. The data part of the key/data pair used by the recno access method is the same as other access methods. The key is different. The data field of the key should be a pointer to a memory location of type recno_t, as defined in the <db.h> include file. This type is normally the largest unsigned integral type available to the implementation. The size field of the key should be the size of that type. Because there can be no meta-data associated with the underlying recno access method files, any changes made to the default values (e.g., fixed record length or byte separator value) must be explicitly specified each time the file is opened. In the interface specified by dbopen(), using the put interface to create a new record will cause the creation of multiple, empty records if the record number is more than one greater than the largest record currently in the database. ERRORS
The recno access method routines may fail and set errno for any of the errors specified for the library routine dbopen(3) or the following: EINVAL An attempt was made to add a record to a fixed-length database that was too large to fit. SEE ALSO
btree(3), dbopen(3), hash(3), mpool(3) Michael Stonebraker, Heidi Stettner, Joseph Kalash, Antonin Guttman, and Nadene Lynn, "Document Processing in a Relational Database System", Memorandum No. UCB/ERL M82/32, May 1982. BUGS
Only big and little endian byte order is supported. BSD
April 17, 2003 BSD