Home | History | Annotate | Line # | Download | only in hash
README revision 1.2 
      1 #	@(#)README	8.1 (Berkeley) 6/4/93
      2 
      3 This package implements a superset of the hsearch and dbm/ndbm libraries.
      4 
      5 Test Programs:
      6 	All test programs which need key/data pairs expect them entered
      7 	with key and data on separate lines
      8 
      9 	tcreat3.c	
     10 		Takes 
     11 			bucketsize (bsize), 
     12 			fill factor (ffactor), and
     13 			initial number of elements (nelem).  
     14 		Creates a hash table named hashtest containing the 
     15 		keys/data pairs entered from standard in.
     16 	thash4.c
     17 		Takes
     18 			bucketsize (bsize), 
     19 			fill factor (ffactor), 
     20 			initial number of elements (nelem)
     21 			bytes of cache (ncached), and
     22 			file from which to read data  (fname)
     23 		Creates a table from the key/data pairs on standard in and
     24 		then does a read of each key/data in fname
     25 	tdel.c
     26 		Takes
     27 			bucketsize (bsize), and
     28 			fill factor (ffactor).
     29 			file from which to read data (fname)
     30 		Reads each key/data pair from fname and deletes the
     31 		key from the hash table hashtest
     32 	tseq.c
     33 		Reads the key/data pairs in the file hashtest and writes them
     34 		to standard out.
     35 	tread2.c
     36 		Takes
     37 			butes of cache (ncached).
     38 		Reads key/data pairs from standard in and looks them up
     39 		in the file hashtest.
     40 	tverify.c
     41 		Reads key/data pairs from standard in, looks them up
     42 		in the file hashtest, and verifies that the data is
     43 		correct.
     44 
     45 NOTES:
     46 
     47 The file search.h is provided for using the hsearch compatible interface
     48 on BSD systems.  On System V derived systems, search.h should appear in 
     49 /usr/include.
     50 
     51 The man page ../man/db.3 explains the interface to the hashing system.
     52 The file hash.ps is a postscript copy of a paper explaining
     53 the history, implementation, and performance of the hash package.
     54 
     55 "bugs" or idiosyncracies
     56 
     57 If you have a lot of overflows, it is possible to run out of overflow
     58 pages.  Currently, this will cause a message to be printed on stderr.
     59 Eventually, this will be indicated by a return error code.
     60 
     61 If you are using the ndbm interface and exit without flushing or closing the
     62 file, you may lose updates since the package buffers all writes.  Also,
     63 the db interface only creates a single database file.  To avoid overwriting
     64 the user's original file, the suffix ".db" is appended to the file name
     65 passed to dbm_open.  Additionally, if your code "knows" about the historic
     66 .dir and .pag files, it will break.  
     67 
     68 There is a fundamental difference between this package and the old hsearch.
     69 Hsearch requires the user to maintain the keys and data in the application's
     70 allocated memory while hash takes care of all storage management.  The down
     71 side is that the byte strings passed in the ENTRY structure must be null
     72 terminated (both the keys and the data).
     73