![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to internal.nr CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [CSRG BSD Unix] / 43BSD / ingres / doc / access|
Return to internal.nr CVS log | Up to [CSRG BSD Unix] / 43BSD / ingres / doc / access |
1.1 root 1: .de am
2: .ne 8
3: .sp 4
4: \\$1 (\\$2)
5: ..
6: .th INTERNAL AM 1/19/79
7: .sh NAME
8: internal routine used by the access methods.
9: .sh SYNOPSIS
10: The functions listed in this section should not be
11: called directly by any higher level program.
12: These functions are all used to implement the
13: access method calls described elsewhere in this manual.
14: .am ACC_CLOSE 9/3/77
15: .sh NAME
16: acc_close - close relations opened by readadmin
17: .sh SYNOPSIS
18: acc_close()
19: .sh DESCRIPTION
20: ACC_CLOSE closes the relation and attribute relations
21: which were opened by READADMIN using the
22: Admin_rel and Admin_att descriptor structs.
23: ACC_CLOSE calls CLEANREL to flush and reset any
24: buffers associated with these two relations.
25: .sh DIAGNOSTICS
26: If an error occurs in CLEANREL
27: a SYSERR is generated.
28: .sh "SEE ALSO"
29: readadmin
30: .am ACC_ERR 9/13/77
31: .sh NAME
32: acc_err - set global Accerror indicator
33: .sh SYNOPSIS
34: .nf
35: acc_err(errnum)
36: int errnum;
37: .fi
38: .sh DESCRIPTION
39: ACC_ERR will set the global Accerror indicator to the value
40: of ERRNUM and then return that value.
41: The Accerror indicator is printed whenever a SYSERR occurs
42: and its value is non-zero.
43: .sh "TRACE FLAGS"
44: 85.0
45: .sh "SEE ALSO"
46: closer, get_attuples, get_page, openr, pageflush
47: .am ACC_INIT 8/31/77
48: .sh NAME
49: acc_init - initialize access methods
50: .sh SYNOPSIS
51: .nf
52: acc_init()
53: .fi
54: .sh DESCRIPTION
55: ACC_INIT initializes the access methods by setting up
56: all of the access method buffers and all the pointers
57: that link them together.
58: The buffers are organized so that only one buffer can be used
59: by any one relation.
60: This helps so that
61: "there is a little bit of each relation" available since
62: many programs go back and forth between two or more relations.
63: The buffers are allocated in a least recenlty used fashion.
64: ACC_INIT calls READADMIN to initialize the Admin,
65: Admin_rel and Admin_att structures and to
66: open the relation and attribute relations.
67: ACC_INIT also opens the lock device
68: used for concurrency control.
69: .sh DIAGNOSTICS
70: No value is returned.
71: If an error occurs a SYSERR is generated.
72: .sh "TRACE FLAGS"
73: 81.0, 81.1
74: .sh "SEE ALSO"
75: resetacc, readadmin
76: .am ADD_A_PAGE 9/3/77
77: .sh NAME
78: add_a_page - add a page to a relation
79: .sh SYNOPSIS
80: .nf
81: add_a_page(d, tid)
82: struct descriptor *d;
83: struct tup_id *tid;
84: .fi
85: .sh DESCRIPTION
86: ADD_A_PAGE adds a page to the given relation.
87: It is assumed that the current access method buffer
88: contains a page from the relation.
89: No linking is done to this page,
90: i.e. no other page is affected by adding a page
91: to a relation using this function.
92: The new page is not written but is marked for writing.
93: TID will contain the page id of the new page.
94: This function is used by MODIFY.
95: .sh DIAGNOSTICS
96: .nf
97: -1 -- pageflush failure on previous page in buffer
98: 0 -- success
99: .fi
100: .sh "SEE ALSO"
101: add_ovflo, add_prim
102: .am ADD_OVFLO 9/6/77
103: .sh NAME
104: add_ovflo - add overflow page to a relation
105: .sh SYNOPSIS
106: .nf
107: add_ovflo(d, tid)
108: struct descriptor *d;
109: struct tup_id *tid;
110: .fi
111: .sh DESCRIPTION
112: ADD_OVFLO will add an overflow page to the page
113: in the current access method buffer.
114: The page must be part of the relation described in D.
115: TID will be updated to the id of the new overflow page.
116: .sh DIAGNOSTICS
117: .nf
118: -5, -3 -- get_page failure
119: -4, -2, -1 -- pageflush failure
120: 0 -- success
121: .fi
122: .sh "SEE ALSO"
123: add_a_page, add_prim, findbest
124: .am ADD_PRIM 9/3/77
125: .sh NAME
126: add_prim - add a primary page to a relation
127: .sh SYNOPSIS
128: .nf
129: add_prim(d, tid)
130: struct descriptor *d;
131: struct tup_id *tid;
132: .fi
133: .sh DESCRIPTION
134: ADD_PRIM adds a primary page to the given relation.
135: It is assumed that the current access method buffer
136: contains the last primary page of the relation.
137: The new page is marked for writing.
138: TID will contain the page id of the new page.
139: ADD_PRIM is called by MODIFY.
140: .sh DIAGNOSTICS
141: .nf
142: -1, -2 -- pageflush failure
143: 0 -- success
144: .fi
145: .sh "SEE ALSO"
146: add_a_page, add_ovflo
147: .am ADDABYTE 9/7/77
148: .sh NAME
149: addabyte - add byte to hash key
150: .sh SYNOPSIS
151: .nf
152: addabyte(ch, word, knt)
153: char ch;
154: long *word;
155: int knt;
156: .fi
157: .sh DESCRIPTION
158: ADDABYTE takes a byte at a time from CH which is rotated and exclusive-ored
159: into WORD, a long integer.
160: This function is used by RHASH to form the hash key.
161: KNT determines how CH is rotated.
162: .sh DIAGNOSTICS
163: No value is returned.
164: .sh "SEE ALSO"
165: rhash
166: .am CANONICAL 9/13/77
167: .sh NAME
168: canonical - make tuple canonical
169: .sh SYNOPSIS
170: .nf
171: canonical (d, tuple)
172: struct descriptor *d;
173: char tuple[MAXTUP];
174: .fi
175: .sh DESCRIPTION
176: CANONICAL will make TUPLE canonical and return the new length
177: of the tuple.
178: If the relation specified by D is compressed,
179: TUPLE will be compressed and stored in the global Accanon.
180: If the relation is not compressed,
181: all character domains of TUPLE will be blank padded.
182: In either case the global pointer Acctuple
183: will point to the canonical tuple.
184: .sh DIAGNOSTICS
185: No error value is returned.
186: .sh "SEE ALSO"
187: comp_tup, insert, put_tuple, replace
188: .am CHOOSE_BUF 9/3/77
189: .sh NAME
190: choose_buf - select an access method buffer
191: .sh SYNOPSIS
192: .nf
193: struct accbuf *choose_buf()
194: .fi
195: .sh DESCRIPTION
196: CHOOSE_BUF will scan the access method buffers for a
197: free buffer and return a pointer to the buffer.
198: If there is no free buffer
199: the last buffer in the LRU list is flushed and reset
200: and a pointer to it is returned.
201: .sh DIAGNOSTICS
202: .nf
203: -1 -- pageflush or reset failure
204: .fi
205: .sh "TRACE FLAGS"
206: 82.8
207: .sh "SEE ALSO"
208: formatpg, get_page, resetacc
209: .am COMP_TUP 9/13/77
210: .sh NAME
211: comp_tup - compress a tuple
212: .sh SYNOPSIS
213: .nf
214: comp_tup(d, tuple)
215: struct descriptor *d;
216: char tuple[MAXTUP];
217: .fi
218: .sh DESCRIPTION
219: COMP_TUP will compress TUPLE into the global area Accanon.
220: The compression consists of replacing trailing blanks in character fields
221: with null string terminators.
222: No compression is done on numeric fields.
223: This compression method will never result in a tuple
224: becoming longer than the normal tuple.
225: COMP_TUP will return the length of the compressed tuple.
226: .sh DIAGNOSTICS
227: No error value is returned.
228: .sh "SEE ALSO"
229: canonical, put_tuple, uncomp_tup
230: .am DEL_TUPLE 9/6/77
231: .sh NAME
232: del_tuple - delete specified tuple from current page
233: .sh SYNOPSIS
234: .nf
235: del_tuple(tid, width)
236: struct tup_id *tid;
237: int width;
238: .fi
239: .sh DESCRIPTION
240: DEL_TUPLE deletes the tuple specified by TID from the page
241: in the current access method buffer.
242: WIDTH must equal the true width of the tuple in bytes
243: after any compression is performed.
244: No checking is done to guarantee that the page id in TID
245: matches the page in the access method buffer.
246: .sh DIAGNOSTICS
247: No value is returned.
248: .sh "SEE ALSO"
249: delete, replace
250: .am DUMPTID 9/14/77
251: .sh NAME
252: dumptid - print tuple id
253: .sh SYNOPSIS
254: .nf
255: dumptid(tid)
256: struct tup_id *tid;
257: .fi
258: .sh DESCRIPTION
259: DUMPTID will print the page and line ids of TID
260: on the UNIX standard output.
261: .sh DIAGNOSTICS
262: .nf
263: 0 -- success
264: .fi
265: .am DUP_CHECK 9/14/77
266: .sh NAME
267: dup_check - check page for duplicate tuple
268: .sh SYNOPSIS
269: .nf
270: dup_check(d, tuple)
271: struct descriptor *d;
272: char tuple[MAXTUP];
273: .fi
274: .sh DESCRIPTION
275: DUP_CHECK will scan the page in the current access method buffer
276: for a duplicate of TUPLE.
277: No checking is done to verify that the page in the current access
278: method buffer is part of the relation described by D.
279: .sh DIAGNOSTICS
280: .nf
281: 0 -- no duplicate found
282: 1 -- duplicate found
283: .fi
284: .sh "SEE ALSO"
285: findbest, scan_dups
286: .am FIND_BUF 9/3/77
287: .sh NAME
288: find_buf - find a buffer with a page from the given relation
289: .sh SYNOPSIS
290: .nf
291: struct accbuf *find_buf(d)
292: struct descriptor *d;
293: .fi
294: .sh DESCRIPTION
295: FIND_BUF will scan all the access method buffers until it finds
296: a buffer with a relation id that matches the relation id
297: in the descriptor struct,
298: and return a pointer to the buffer.
299: If no buffer is found a zero is returned.
300: If the access method buffers have not been initialized
301: by the time FIND_BUF is called,
302: ACC_INIT is called to initialize them.
303: .sh DIAGNOSTICS
304: If an error occurs in ACC_INIT a SYSERR is generated.
305: .sh "TRACE FLAGS"
306: 82.0
307: .sh "SEE ALSO"
308: acc_init, cleanrel, closer, find, findbest, inpcloser, noclose, get_page
309: .am FINDBEST 9/8/77
310: .sh NAME
311: findbest - determine best page to insert a tuple
312: .sh SYNOPSIS
313: .nf
314: findbest(d, tid, tuple, need, checkdups)
315: struct descriptor *d;
316: struct tup_id *tid;
317: char tuple[MAXTUP];
318: int need, checkdups;
319: .fi
320: .sh DESCRIPTION
321: FINDBEST will find the best page in the relation described
322: by D for TUPLE.
323: NEED indicates how much space is needed to store the tuple.
324: If the relation is stored as a heap,
325: FINDBEST will try to fit the tuple on the current page
326: in the access method buffers.
327: If there is no space,
328: or if the current page is not available,
329: the last page of the relation is selected.
330: For other storage structures,
331: FINDBEST will try to fit the tuple on the correct primary page
332: or on one of its overflow pages that is closest to the primary page.
333: If there is no free space a new overflow page is created.
334: If CHECKDUPS is true,
335: the primary page and all its overflow pages will be scanned
336: for a duplicate of the tuple.
337: If a duplicate is found
338: then the new tuple will not be inserted,
339: otherwise FINDBEST allocates an entry in the line number table
340: of the target page for the tuple being inserted.
341: TID will be set to the location of the new tuple.
342: .sh DIAGNOSTICS
343: .nf
344: -8 -- unidentifiable storage structure in find
345: -5, -3, -1 -- get_page failure
346: -4, -2, -1 -- pageflush failure
347: 1 -- duplicate found
348: .fi
349: .sh "TRACE FLAGS"
350: 88.2
351: .sh "SEE ALSO"
352: dup_check, find, insert, replace, scan_dups
353: .am FULLKEY 9/6/77
354: .sh NAME
355: fullkey - determine if SETKEY has been called
356: for every keyed domain of a relation.
357: .sh SYNOPSIS
358: .nf
359: fullkey(d)
360: struct descriptor *d;
361: .fi
362: .sh DESCRIPTION
363: FULLKEY will determine if SETKEY was called
364: for every keyed domain of a given relation.
365: This is important for the hashing function
366: since it can only accept keys where a value has
367: been provided for each keyed domain.
368: .sh DIAGNOSTICS
369: .nf
370: 0 -- only a partial key is provided
371: 1 -- a full key is provided
372: .fi
373: .sh "SEE ALSO"
374: find, rhash
375: .am GET_ADDR 9/12/77
376: .sh NAME
377: get_addr - get address of tuple within page
378: .sh SYNOPSIS
379: .nf
380: char *get_addr(tid)
381: struct tup_id *tid;
382: .fi
383: .sh DESCRIPTION
384: GET_ADDR computes and returns the character address of the tuple specified
385: by TID in the current access method buffer.
386: .sh DIAGNOSTICS
387: No error codes are returned.
388: If an error is detected
389: a SYSERR is generated.
390: .sh "SEE ALSO"
391: get_tuple, put_tuple
392: .am GET_ATTUPLES 9/13/77
393: .sh NAME
394: get_attuples - get tuples from attribute relation
395: .sh SYNOPSIS
396: .nf
397: get_attuples(d)
398: struct descriptor *d;
399: .fi
400: .sh DESCRIPTION
401: GET_ATTUPLES will fill the attribute template of the descriptor
402: struct D with data from tuples of the attribute relation.
403: The relation must be identified in D,
404: which can be done by calling GET_RELTUPLE.
405: .sh DIAGNOSTICS
406: .nf
407: -8 -- unidentifiable storage structure in find
408: -6 -- missing attributes from attribute relation
409: -2 -- pagelfush failure
410: -1 -- get_page failure or invalid error
411: 0 -- success
412: .fi
413: .sh "TRACE FLAGS"
414: 90.3
415: .sh "SEE ALSO"
416: get_reltuple, openr
417: .am GET_PAGE 4/5/77
418: .sh NAME
419: get_page - get a page from a relation
420: .sh SYNOPSIS
421: .nf
422: get_page(d, tid)
423: struct descriptor *d;
424: struct tup_id *tid;
425: .fi
426: .sh DESCRIPTION
427: GET_PAGE will retrieve a page from the UNIX file
428: which corresponds to a given relation,
429: unless the page is already available in one of
430: the access method buffers.
431: If the page has to be read in,
432: it will be placed in a buffer corresponding to the
433: relation specified by D.
434: If there is no such buffer,
435: the last buffer in the LRU buffer queue is flushed
436: and the page is placed in that buffer.
437: The buffer containing the requested page is then placed at the
438: top of the LRU buffer queue.
439: If the relation was opened in read/write mode,
440: a lock is set on the page to maintain proper concurrency control.
441: The current page size is 512 bytes of which
442: 498 bytes are available for tuple storage.
443: .sh DIAGNOSTICS
444: .nf
445: -2 -- pageflush error
446: -1 -- UNIX read error
447: 0 -- success
448: .fi
449: .sh "TRACE FLAGS"
450: 83.0, 83.1, 83.2
451: .sh "SEE ALSO"
452: choose_buf, find_buf, get, resetacc, pageflush, top_acc
453: .am GET_RELTUPLE 9/13/77
454: .sh NAME
455: get_reltuple - get tuple from relation relation
456: .sh SYNOPSIS
457: .nf
458: get_reltuple(d, name)
459: struct descriptor *d;
460: char name[MAXNAME + 1];
461: .fi
462: .sh DESCRIPTION
463: GET_RELTUPLE will get a tuple from the relation relation
464: for the relation specified by NAME
465: and place the tuple in the descriptor struct pointed to by D.
466: Two search attempts are made,
467: the first for a relation owned by the current user,
468: the second for a relation owned by the
469: data base administrator.
470: .sh DIAGNOSTICS
471: .nf
472: -8 -- unidentifiable storage structure in find
473: -2 -- pageflush failure
474: -1 -- get_page failure or invalid error
475: 0 -- success
476: 1 -- no tuples could be found
477: .fi
478: .sh "TRACE FLAGS"
479: 90.1
480: .sh "SEE ALSO"
481: openr, readadmin
482: .am GET_TUPLE 9/12/77
483: .sh NAME
484: get_tuple - get tuple from page
485: .sh SYNOPSIS
486: .nf
487: get_tuple(d, tid, tuple)
488: struct descriptor *d;
489: struct tup_id *tid;
490: char tuple[MAXTUP];
491: .fi
492: .sh DESCRIPTION
493: GET_TUPLE will move a tuple from the page in the current
494: access method buffer to TUPLE,
495: performing any necessary decompression.
496: The tuple is identified by TID,
497: no checking is done to verify that the current page in the
498: access method buffer corresponds to TID or the relation
499: specified by D.
500: .sh DIAGNOSTICS
501: No value is returned.
502: .sh "SEE ALSO"
503: get, uncomp_tup
504: .am INVALID 9/12/77
505: .sh NAME
506: invalid - check the validity of a tuple id
507: .sh SYNOPSIS
508: .nf
509: invalid(tid)
510: struct tup_id *tid;
511: .fi
512: .sh DESCRIPTION
513: INVALID checks if the tuple id represents a valid tuple
514: versus one that has been deleted.
515: INVALID assumes that the tuple id is for the
516: page in the current buffer.
517: .sh DIAGNOSTICS
518: .nf
519: -1 -- line id exceeded line number table
520: 0 -- tuple id is valid
521: 2 -- tuple was deleted
522: .fi
523: .sh "SEE ALSO"
524: delete, get, replace
525: .am LAST_PAGE 9/13/77
526: .sh NAME
527: last_page - compute tid for last page in relation
528: .sh SYNOPSIS
529: .nf
530: last_page(d, tid)
531: struct descriptor *d;
532: struct tup_id *tid;
533: .fi
534: .sh DESCRIPTION
535: LAST_PAGE computes a tuple id for the last page in the relation specified
536: by D and stores the id in TID.
537: The line id in TID is set to zero.
538: .sh DIAGNOSTICS
539: .nf
540: 0 -- success
541: .fi
542: If an error occurs a SYSERR is generated.
543: .sh "SEE ALSO"
544: add_ovflo, findbest, fstat (UNIX function)
545: .am NDXSEARCH 9/7/77
546: .sh NAME
547: ndxsearch - search ISAM directory
548: .sh SYNOPSIS
549: .nf
550: ndxsearch(d, tid, key, mode, keyok)
551: struct descriptor *d;
552: struct tup_id *tid;
553: char key[MAXTUP];
554: int mode, keyok;
555: .fi
556: .sh DESCRIPTION
557: NDXSEARCH will search the ISAM directory for the given relation.
558: It is assumed that the relation is stored as an ISAM file.
559: It is not necessary to provide values for all keyed domains,
560: but the only values used are those starting with the most
561: significant keyed domain
562: to the first keyed domain that does not have a value
563: provided by SETKEY.
564: If MODE is less than zero,
565: NDXSEARCH will determine the lower limit of a scan
566: according to the key.
567: The limit is returned in TID.
568: If MODE is greater than zero,
569: the upper scan limit is determined.
570: If KEYOK is true (not equal to zero),
571: it is assumed that values for all keyed domains
572: have been provided regardless of whether SETKEY was called.
573: .sh DIAGNOSTICS
574: .nf
575: -2 -- pageflush failure
576: -1 -- get_page failure
577: 0 -- success
578: .fi
579: .sh "SEE ALSO"
580: find, setkey
581: .am NEWLINO 9/13/77
582: .sh NAME
583: newlino - find free line number on page
584: .sh SYNOPSIS
585: .nf
586: newlino(len)
587: int len;
588: .fi
589: .sh DESCRIPTION
590: NEWLINO will scan the line number table in the page in the
591: current access method buffer,
592: looking for a free table entry.
593: If an entry is found it is allocated for a new tuple,
594: otherwise a new entry is created at the end of the
595: line number table and allocated for a new tuple.
596: Space in the page is also reserved for a tuple
597: of length LEN.
598: NEWLINO returns the line number allocated for a new tuple.
599: .sh DIAGNOSTICS
600: No error value is returned.
601: .sh "SEE ALSO"
602: findbest
603: .am PAGEFLUSH 9/13/77
604: .sh NAME
605: pageflush - flush access method buffers
606: .sh SYNOPSIS
607: .nf
608: pageflush(buf)
609: struct accbuf *buf;
610: .fi
611: .sh DESCRIPTION
612: PAGEFLUSH will write the access method buffer pointed to
613: by BUF onto disk if the buffer was tagged for writing.
614: PAGEFLUSH can be called with the parameter "Acc_head"
615: which is a global buffer pointer (declared in access.h)
616: which will flush the current access method buffer.
617: PAGEFLUSH can also be called with a 0 (i.e. a null pointer)
618: which will flush all buffers
619: tagged for writing.
620: If a page was locked,
621: it is unlocked after being written to disk.
622: .sh DIAGNOSTICS
623: .nf
624: -2 -- UNIX write error
625: 0 -- success
626: .fi
627: .sh "TRACE FLAGS"
628: 84.0, 84.1, 84.2
629: .sh "SEE ALSO"
630: get_page, resetacc
631: .am PLUCK_PAGE 9/14/77
632: .sh NAME
633: pluck_page - extract page id from tuple id
634: .sh SYNOPSIS
635: .nf
636: pluck_page(tid, pnum)
637: struct tup_id *tid;
638: long *pnum;
639: .fi
640: .sh DESCRIPTION
641: PLUCK_PAGE will extract the page id from TID and
642: store it in the long variable pointed to by PNUM.
643: .sh DIAGNOSTICS
644: .nf
645: 0 -- success
646: .fi
647: .sh "SEE ALSO"
648: stuff_page, tup_id
649: .am PUT_TUPLE 9/13/77
650: .sh NAME
651: put_tuple - put tuple in page
652: .sh SYNOPSIS
653: .nf
654: put_tuple(tid, tuple, length)
655: struct tup_id *tid;
656: char tuple[MAXTUP];
657: int length;
658: .fi
659: .sh DESCRIPTION
660: PUT_TUPLE will move TUPLE into the page in the current access method buffer
661: at the location specified by TID.
662: It is assumed that any required compression has already been done
663: and that the correct amount of space has been reserved on the page
664: for the tuple.
665: No checking is done to verify that the page identified by TID
666: corresponds to the page in the current access method buffer.
667: .sh DIAGNOSTICS
668: No value is returned.
669: .sh "TRACE FLAGS"
670: 87.0
671: .sh "SEE ALSO"
672: canonical, get_addr, insert, replace
673: .am READADMIN 9/3/77
674: .sh NAME
675: readadmin - read data from admin file
676: .sh SYNOPSIS
677: .nf
678: readadmin()
679: .fi
680: .sh DESCRIPTION
681: READADMIN will read data from the admin file.
682: The admin file contains two descriptor structs,
683: one for the relation relation and one for the
684: attribute relation.
685: These two descriptors are used by the access methods
686: whenever either of the relations has to be used
687: to retrieve or update information.
688: The two descriptors are globally declared as
689: Admin_rel and Admin_att.
690: The admin file also contains the user id for the data base administrator,
691: which is stored in the Admin_admin struct.
692: .sh DIAGNOSTICS
693: Nothing is returned.
694: If an error occurs a SYSERR is generated.
695: .sh "SEE ALSO"
696: acc_init
697: .am RESETACC 8/31/77
698: .sh NAME
699: resetacc - reset access method buffer
700: .sh SYNOPSIS
701: .nf
702: resetacc(buf)
703: struct accbuf *buf;
704: .fi
705: .sh DESCRIPTION
706: RESETACC will first flush the given buffer
707: if it contains a dirty page from some relation,
708: then clear the relation identifier and other fields in the buffer.
709: This has
710: the effect of erasing the buffer so that any attempt to access
711: the page that was contained in the buffer will fail unless
712: the page is read back from disk.
713: If the buf parameter equals zero,
714: the current access method buffer is reset.
715: .sh DIAGNOSTICS
716: .nf
717: 0 -- success
718: < 0 -- pageflush fail (see pageflush for error codes)
719: .fi
720: .sh "TRACE FLAGS"
721: 80.0
722: .sh "SEE ALSO"
723: acc_init, choose_buf, cleanrel, pageflush
724: .am RHASH 9/7/77
725: .sh NAME
726: rhash - compute hash key
727: .sh SYNOPSIS
728: .nf
729: long rhash(d, key)
730: struct descriptor *d;
731: char key[MAXTUP];
732: .fi
733: .sh DESCRIPTION
734: RHASH will perform a randomizing hash on the full key.
735: Values for the hash function are taken from every keyed domain of KEY.
736: It is important that every keyed domain have a correct value,
737: otherwise the hash result will be incorrect.
738: RHASH returns a page number in the form of a long integer.
739: .sh DIAGNOSTICS
740: None.
741: .sh "TRACE FLAGS"
742: 92.8
743: .sh "SEE ALSO"
744: addabyte, find
745: .am SCAN_DUPS 9/14/77
746: .sh NAME
747: scan_dups - scan for duplicate tuples
748: .sh SYNOPSIS
749: .nf
750: scan_dups(d, tid, tuple)
751: struct descriptor *d;
752: struct tup_id *tid;
753: char tuple[MAXTUP];
754: .fi
755: .sh DESCRIPTION
756: SCAN_DUPS will look for a duplicate of TUPLE
757: beginning with the page specified by TID.
758: SCAN_DUPS will also check all overflow pages linked
759: to the first page until a duplicate is found.
760: SCAN_DUPS will make sure that the page
761: that was in the current access method buffer upon entry
762: will be there upon exit,
763: as long as that page was part of the relation
764: described by D.
765: .sh DIAGNOSTICS
766: .nf
767: -2 -- pageflush failure
768: -1 -- get_page failure
769: 0 -- no duplicate found
770: 1 -- duplicate found
771: .fi
772: .sh "SEE ALSO"
773: dup_check, findbest, replace
774: .am SPACE_LEFT 9/13/77
775: .sh NAME
776: space_left - compute amount of free space on page
777: .sh DESCRIPTION
778: SPACE_LEFT will determine how much space is left on
779: the current page for tuple storage.
780: The return value is the amount of free space in bytes.
781: .sh DIAGNOSTICS
782: none
783: .sh "SEE ALSO"
784: findbest
785: .am STUFF_PAGE 9/14/77
786: .sh NAME
787: stuff_page - set page id of tuple id
788: .sh SYNOPSIS
789: .nf
790: stuff_page(tid, pnum)
791: struct tup_id *tid;
792: long *pnum;
793: .fi
794: .sh DESCRIPTION
795: STUFF_PAGE will take the page number pointed to by PNUM
796: and store it in TID without affecting the
797: line id in TID.
798: .sh DIAGNOSTICS
799: .nf
800: 0 -- success
801: .fi
802: .sh "SEE ALSO"
803: pluck_page, tup_id
804: .am TOP_ACC 9/3/77
805: .sh NAME
806: top_acc - move buffer to top of LRU list
807: .sh SYNOPSIS
808: .nf
809: top_acc(buf)
810: struct accbuf *buf
811: .fi
812: .sh DESCRIPTION
813: TOP_ACC unlinks the buffer from its current position
814: in the LRU linked list and relinks it at the top of the list.
815: TOP_ACC is called for a buffer every time it is accessed
816: to maintain the LRU sequence of the linked list.
817: .sh DIAGNOSTICS
818: .nf
819: 0 -- success
820: .fi
821: .sh "SEE ALSO"
822: acc_init
823: .am TUP_ID 9/14/77
824: .sh NAME
825: tup_id - tuple id structure
826: .sh SYNOPSIS
827: .nf
828: struct tup_id
829: {
830: char pg1, line_id;
831: int pg2;
832: } tid;
833: .fi
834: .sh DESCRIPTION
835: The TUP_ID structure is used to declare tuple id's
836: which are passed as parameters to some of the
837: access method routines.
838: The tuple id uniquely addresses a tuple of a relation.
839: Since tuples can move around within a relation,
840: tuple id's should NOT be saved.
841: It may be desirable to access the tuple id in some
842: special cases to gather statistics on disk usage.
843: At no time should a program modify a tuple id,
844: the access methods take care of initializing
845: and updating tuple id's as necessary.
846: .sh "SEE ALSO"
847: delete, find, get, insert, pluck_page, replace, stuff_page
848: .am TUP_LEN 9/14/77
849: .sh NAME
850: tup_len - determine length of tuple
851: .sh SYNOPSIS
852: .nf
853: tup_len(tid)
854: struct tup_id *tid;
855: .fi
856: .sh DESCRIPTION
857: TUP_LEN will determine how many bytes in the page in the current access method
858: buffer are used by the tuple
859: pointed to by TID,
860: and return the length.
861: No checking is done to verify that the page specified by TID
862: corresponds to the page in the current access method buffer.
863: .sh DIAGNOSTICS
864: No error value is returned.
865: .sh "TRACE FLAGS"
866: 89.1
867: .sh "SEE ALSO"
868: delete, replace
869: .am UNCOMP_TUP 9/12/77
870: .sh NAME
871: uncomp_tup - decompress tuple
872: .sh SYNOPSIS
873: .nf
874: uncomp_tup(d, cp, tuple)
875: struct descriptor *d;
876: char *cp, tuple[MAXTUP];
877: .fi
878: .sh DESCRIPTION
879: UNCOMP_TUP will decompress the tuple pointed to by CP
880: and store the result in TUPLE.
881: .sh DIAGNOSTICS
882: No value is returned.
883: .sh "SEE ALSO"
884: comp_tup, get_tuple
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.