|
|
1.1 root 1: .th REPLACE AM 9/14/77
2: .sh NAME
3: replace - modify an existing tuple
4: .sh SYNOPSIS
5: .nf
6: replace(d, tid, tuple, checkdups)
7: struct descriptor *d;
8: struct tup_id *tid;
9: char tuple[MAXTUP];
10: int checkdups;
11: .fi
12: .sh DESCRIPTION
13: REPLACE will replace the current value of the tuple identified
14: by TID with the new value in TUPLE.
15: CHECKDUPS has the same meaning for REPLACE as it does for INSERT.
16: See INSERT for description.
17: .sp
18: REPLACE will normally put the new tuple in the same
19: location occupied by the old tuple except if the
20: relation is compressed or if the value of a keyed
21: domain is changed.
22: This is necessary in the former case because compression
23: may give a tuple size different from the previous tuple,
24: while in the latter case a new key value for the tuple
25: may force the tuple to be located in a different section
26: of the relation so that it can be found by FIND.
27: The two exceptions are handled by calling DELETE
28: to remove the old tuple and calling INSERT to
29: add the new tuple.
30: .sh DIAGNOSTICS
31: .nf
32: AMFIND_ERR -- unidentifiable storage structure in find
33: AMWRITE_ERR -- write error
34: AMREAD_ERR -- read error
35: 0 -- success
36: 1 -- new tuple was a duplicate
37: Tid is set to the tuple id of the duplicate
38: 2 -- tuple identified by tid has been deleted
39: .fi
40: .sh "TRACE FLAGS"
41: 24.4, 24.5, 24.6
42: .sh "SEE ALSO"
43: closer, delete, get, insert, openr
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.