tldm-universe/Ardent/UV/APP.PROGS/ACCT.DUMP

651 lines
14 KiB
Plaintext
Raw Permalink Normal View History

2024-09-09 21:51:08 +00:00
******************************************************************************
*
* Program to provide a means of dumping uniVerse files to tape
* and loading these files on a Pick or Information system
*
* ACCT.DUMP Version %I% Date %H%
*
* (c) Copyright 1998 Ardent Software Inc. - All Rights Reserved
* This is unpublished proprietary source code of Ardent Software Inc.
* The copyright notice above does not evidence any actual or intended
* publication of such source code.
*
******************************************************************************
*
* Date.... GTAR# WHO Description.........................................
* 07/25/88 - - Maintenence log purged at 5.2.1, see release 5.1.10.
*
******************************************************************************
$OPTIONS DEFAULT
$INCLUDE UNIVERSE.INCLUDE FILENAMES.H
MAX.NAME = 12
DATA.START = MAX.NAME - 3
DATA.LIMIT = MAX.NAME
open '','VOC' to voc.file else print 'Cannot open VOC file'; stop
savemod = inmat()
savetype = status()
open '','NEWACC' to newacc.file else print 'Cannot open NEWACC file'; stop
open '','VOCLIB' to voclib.file else print 'Cannot open VOCLIB file'; stop
x = 0
flavors = ''
pick.dict = 0
pick.flag = 0
part.voc = 0
dfnames = ''
fmods = ''
ftypes = ''
lfnames = ''
pathnames = ''
fno = 0
lno = 0
* look for format keyword *
acct.format = 0
dim line(6)
matparse line from trim(@sentence),' '
i = 2
loop
until line(i) = '' or i = 6 do
keyword = upcase(line(i))
read item from voc.file,keyword then
if item<1> = 'K' then
begin case
case item<2> = 205
if acct.format = 0 then
acct.format = 1; * REALITY.FORMAT *
dump.format = keyword
end else
print 'Account format keyword previously specified'
stop
end
case item<2> = 206
if acct.format = 0 then
acct.format = 2; * INFORMATION.FORMAT *
dump.format = keyword
end else
print 'Account format keyword previously specified'
stop
end
case item<2> = 211
if acct.format = 0 then
acct.format = 3; * PICK.FORMAT *
dump.format = keyword
end else
print 'Account format keyword previously specified'
stop
end
end case
end
end
i += 1
repeat
* get valid account types *
openpath UV.NEWACC to file then
sselect file to 1
10*
readnext fid from 1 else goto 20
x += 1
flavors<x> = fid
goto 10
end else
print 'Unable to open ':UV.NEWACC:' file'
stop
end
20*
* get account type *
readv uv.type from voc.file,'RELLEVEL',3 else uv.type = ''
if uv.type = '' then
30*
msg = 'Enter uniVerse account type '
print msg:
input uv.type
uv.type = upcase(uv.type)
if uv.type = '' or uv.type = 'QUIT' then print 'ACCT.DUMP aborted!'; stop
if uv.type = 'IDEAL' then uv.type = 'NEWACC'
end
fnd = 0
for i = 1 to x while fnd = 0
if trim(uv.type) = flavors<i> then fnd = 1
next x
40*
if fnd = 0 then
print 'Type "':uv.type:'"':' is not a valid uniVerse account type'
print
print 'Valid account types are: '
for i = 1 to 32 until flavors<i> = ''
print flavors<i>
next i
goto 30
end
* default to current account type, if format keyword is not specified *
if acct.format = 0 then
if uv.type = 'PICK' then
acct.format = 3; dump.format = 'PICK.FORMAT'
end else
if uv.type = 'REALITY' then
acct.format = 1; dump.format = 'REALITY.FORMAT'
end else
acct.format = 2; dump.format = 'INFORMATION.FORMAT'
end
end
end
print
* prompt for voc file handling *
msg = "Do you want to dump 'ALL' VOC entries or just those 'NOT' in NEWACC? (A/N)"
50*
print msg:
input ans
ans = upcase(ans)
if ans = 'QUIT' then print 'ACCT.DUMP aborted!'; stop
if ans # 'A' and ans # 'N' then goto 50
if ans # 'A' then part.voc = 1
* prompt for dictionary file handling *
if acct.format # 2 then
60*
msg = "Do you want to dump the 'DICT' or the 'PDICT' files? (D/P)"
print msg:
input ans
ans = upcase(ans)
if ans = 'QUIT' then print 'ACCT.DUMP aborted!'; stop
if ans # 'D' and ans # 'P' then goto 60
if ans = 'P' then pick.dict = 1
end
* check VOCLIB file for ACCT.DUMP *
print
dump.name = 'ACCT.DUMP'
read item from voclib.file,dump.name then
70*
msg = dump.name:' exists in VOCLIB file. OK to overwrite? (Y/N)'
print msg:
input ans
ans = upcase(ans)
if ans = 'QUIT' then print 'ACCT.DUMP aborted!'; stop
if ans # 'Y' and ans # 'N' then goto 70
if ans # 'Y' then print 'ACCT.DUMP aborted!'; stop
end
* check VOC file for ACCT.DUMP *
read item from voc.file,dump.name then
80*
msg = dump.name:' exists in VOC file. OK to overwrite? (Y/N)'
print msg:
input ans
ans = upcase(ans)
if ans = 'QUIT' then print 'ACCT.DUMP aborted!'; stop
if ans # 'Y' and ans # 'N' then goto 80
if ans # 'Y' then print 'ACCT.DUMP aborted!'; stop
end
* check VOCLIB file for ACCT.LOAD *
load.name = 'ACCT.LOAD'
read item from voclib.file,load.name then
90*
msg = load.name:' exists in VOCLIB file. OK to overwrite? (Y/N)'
print msg:
input ans
ans = upcase(ans)
if ans = 'QUIT' then print 'ACCT.DUMP aborted!'; stop
if ans # 'Y' and ans # 'N' then goto 90
if ans # 'Y' then print 'ACCT.DUMP aborted!'; stop
end
* get VOC entries to dump *
voc.list = '&&ACCT.DUMP':@USERNO
if part.voc = 0 then
perform 'SSELECT VOC'
stmt = 'SAVE.LIST ':voc.list
perform stmt
end else
perform 'SSELECT VOC'
perform 'NSELECT NEWACC'
stmt = 'SAVE.LIST ':voc.list
perform stmt
end
* handle VOC file *
read rec from voc.file,'VOC' else
print "Cannot read VOC record 'VOC'"
stop
end
path = rec<3>
if path # '' then ins path before pathnames<-1>
mod = 1
type = 3
fno += 1
lfnames<fno> = 'DICT UV.VOC'
dfnames<fno> = ''
fmods<fno> = mod
ftypes<fno> = type
path = rec<2>
if path # '' then ins path before pathnames<-1>
mod = savemod
type = savetype
fno += 1
lfnames<fno> = 'DATA UV.VOC'
dfnames<fno> = ''
fmods<fno> = mod
ftypes<fno> = type
* get filenames to dump *
stmt = \SSELECT VOC BY F3 WITH TYPE EQ "F" AND F2 NOT.MATCHING ".../..."\
stmt := ' AND F2 NOT.MATCHING "...\..."'
stmt := \ AND F2 NE "."\
stmt := \ AND @ID NOT.MATCHING "...'.O'"\
stmt := \ AND @ID NE "VOC"\
stmt := \ TO 1\
perform stmt
100*
readnext id from 1 else goto 200
read rec from voc.file,id else
print 'Warning - Cannot read VOC record ':id
goto 100
end
* get DICT filename *
no.pdict = 0
path = rec<3>
dictword = 'DICT '
if pick.dict then
if rec<5> # '' then path = rec<5>; dictword = 'PDICT ' else no.pdict = 1
end
if path # '' then
* ignore remote dict files *
if path matches ".../..." or path matches "...\..." then goto 105
* ignore null file names - query will not handle *
if path = "D_?" or path = "P_?" then goto 105
gosub eftoif
savepath = path
path = path[3,999]
gosub matchid
path = savepath
if match.flag then
fnd = 1
locate path in pathnames setting val else fnd = 0
if fnd = 0 then
open dictword,id then
mod = inmat()
type = status()
ins path before pathnames<-1>
name = dictword:id
if type = 1 and acct.format # 2 then gosub type.chk
fno += 1
if no.pdict then dfnames<fno> = '' else dfnames<fno> = name
lfnames<fno> = 'DICT ':id
fmods<fno> = mod
ftypes<fno> = type
close file
end else
name = dictword:id
print 'WARNING - Unable to open ':name
end
end else
*** must be a dict qpointer ***
end
end else
*** must be a dict qpointer ***
end
end
105*
* get DATA filename(s) *
path = rec<2>
if path # '' and path # rec<3> then
* ignore null filename - query will not handle *
if path = "?" then goto 120
gosub eftoif
gosub matchid
* handle multi-level data files *
if upcase(rec<4>[1,1]) = 'M' then
if match.flag then
fnd = 1
locate path in pathnames setting val else fnd = 0
if fnd = 0 then
if rec<3> = '' then
print 'WARNING - Ignoring file ':id:' - data file without a dictionary file'
goto 120
end
ins path before pathnames<-1>
openpath path to file then
sselect file to 2
110*
readnext fid from 2 else goto 120
open '',id:',':fid to file then
if acct.format = 2 then
print 'WARNING - ':id:',':fid:' is not a valid name on Prime Information'
end
mod = inmat()
type = status()
name = 'DATA ':id:',':fid
if type = 1 and acct.format # 2 then gosub type.chk
fno += 1
dfnames<fno> = name
lfnames<fno> = name
fmods<fno> = mod
ftypes<fno> = type
close file
end else
name = 'DATA ':id:',':fid
print 'WARNING - Unable to open ':name
end
goto 110
end else
print 'WARNING - Unable to open path ':path
end
end else
*** must be a multi level qpointer ***
end
end
end else
* handle regular data files *
if match.flag then
fnd = 1
locate path in pathnames setting val else fnd = 0
if fnd = 0 then
if rec<3> = '' then
print 'WARNING - Ignoring file ':id:' - data file without a dictionary file'
goto 120
end
open '',id to file then
mod = inmat()
type = status()
ins path before pathnames<-1>
name = 'DATA ':id
if type = 1 and acct.format # 2 then gosub type.chk
fno += 1
dfnames<fno> = name
lfnames<fno> = name
fmods<fno> = mod
ftypes<fno> = type
close file
end else
name = 'DATA ':id
print 'WARNING - Unable to open ':name
end
end else
*** must be a data qpointer ***
end
end else
** must be a data qpointer ***
end
end
end
120*
goto 100
200*
print
print 'The following files will dumped to tape:'
print
for j = 1 to fno
if dfnames<j> # '' then print dfnames<j>
next j
210*
msg = "OK to continue? (Y/N)"
print msg:
input ans
ans = upcase(ans)
if ans = 'QUIT' then print 'ACCT.DUMP aborted!'; stop
if ans # 'Y' and ans # 'N' then goto 210
if ans # 'Y' then print 'ACCT.DUMP aborted!'; stop
* setup ACCT.LOAD *
sep = 1
if acct.format = 2 then sep = 4
dfnames<2> = 'DATA UV.VOC'
pa = ''
lno = 0
if acct.format = 2 then
lno += 1
pa<lno> = 'PA'
lno += 1
pa<lno> = 'T.ATT'
lno += 1
pa<lno> = 'T.REW'
lno += 1
pa<lno> = 'T.FWD'
end else
lno += 1
pa<lno> = 'PQ'
lno += 1
pa<lno> = 'H T-ATT (8192)'
lno += 1
pa<lno> = 'P'
lno += 1
pa<lno> = 'H T-REW'
lno += 1
pa<lno> = 'P'
lno += 1
pa<lno> = 'H T-FWD'
lno += 1
pa<lno> = 'P'
end
for j = 1 to fno
if acct.format = 2 then
cmd = 'CREATE.FILE '
cmd := lfnames<j>
cmd := ' '
cmd := ftypes<j>
if ftypes<j> # 1 then
cmd := ' '
cmd := fmods<j>
cmd := ' '
cmd := sep
end
lno += 1
pa<lno> = cmd
if dfnames<j> # '' then
name = lfnames<j>
if name[1,4] = 'DATA' then name = name[6,999999]
cmd = 'T.LOAD '
cmd := name
lno += 1
pa<lno> = cmd
end
end else
cmd = 'H CREATE-FILE ('
cmd := lfnames<j>
cmd := ' '
cmd := fmods<j>
cmd := ','
cmd := sep
lno += 1
pa<lno> = cmd
lno += 1
pa<lno> = 'P'
if dfnames<j> # '' then
name = lfnames<j>
if name[1,4] = 'DATA' then name = name[6,999999]
cmd = 'H T-LOAD '
cmd := name
lno += 1
pa<lno> = cmd
lno += 1
pa<lno> = 'P'
end
end
next j
if acct.format = 2 then
cmd = 'T.DET'
lno += 1
pa<lno> = cmd
end else
cmd = 'H T-DET'
lno += 1
pa<lno> = cmd
cmd = 'P'
lno += 1
pa<lno> = cmd
end
write pa on voclib.file,load.name
* setup ACCT.DUMP *
dfnames<2> = ''
pa = ''
lno = 0
lno += 1
pa<lno> = 'PA'
lno += 1
pa<lno> = 'T.ATT'
if acct.format # 2 then pa<lno> = pa<lno>:' BLK 8192'
lno += 1
pa<lno> = "T.DUMP VOCLIB '":load.name:"' ":dump.format
lno += 1
pa<lno> = 'GET.LIST ':voc.list
lno += 1
pa<lno> = 'T.DUMP VOC ':dump.format
for j = 1 to fno
name = dfnames<j>
if name # '' then
if name[1,4] = 'DATA' then name = name[6,999999]
lno += 1
pa<lno> = 'T.DUMP ':name:' ':dump.format
end
next j
lno += 1
pa<lno> = 'T.DET'
write pa on voclib.file,dump.name
pa = ''
pa<1> = 'R'
pa<2> = 'VOCLIB'
pa<3> = dump.name
write pa on voc.file,dump.name
* now ready to dump files *
perform dump.name
print "ACCT.DUMP completed"
stop
* subroutine to calc modulo for type 1 files *
type.chk:
cmd = 'COUNT '
if name[1,4] = 'DATA' then
cmd := name[6,999999]
end else
cmd := name
end
perform cmd
mod = @SYSTEM.RETURN.CODE
if mod = 0 then mod = 1 else gosub prime.chk
return
* subroutine to make modulo a prime number *
prime.chk:
pno = mod
if pno < 0 then pno = 1
if (pno > 2 and mod(pno,2) = 0) then pno += 1
ptest:
limit = sqrt(pno + 0.0) + .5
for test = 3 to limit
if mod(pno,test) = 0 then
pno += 2
goto ptest
end
next test
mod = pno
return
* subroutine to see if pathname matches record id *
matchid:
match.flag = 0
if uid = path then
match.flag = 1
end else
if len(uid) > MAX.NAME then
if uid[1,DATA.START] = path[1,DATA.START] then
if num(path[DATA.START+1,DATA.LIMIT-DATA.START]) then
match.flag = 1
end
end
end
end
return
* subroutine to convert record id to unix filename *
eftoif:
if id = '' then
uid = '?'
end else
if id[1,1] = '.' then
uid = '?.'; m = 2
end else
uid = ''; m = 1
end
l = len(id)
for j = m to l
c = id[j,1]
begin case
case c = '?'; uid := '??'
case c = '/'; uid := '?\'
case c = '\'; uid := '?\'
case c = char(0); uid := '?0'
case 1; uid := c
end case
next j
end
return
end