CP/M/CPM_CDROM.iso/simtel/cpmug/cpmug028.ark/DATABASE.DOC

Text File | 1984-04-29 | 5.2 KB | 117 lines

    TARBELL DATABASE MANAGEMENT SYSTEM       March 23, 1978

The main theme of this system is to provide a common set of programs
that help the user create, modify, and access data files for a variety
of needs.  In this way, the system can be better tailored for a particular
situation, and yet the different parts of it can also be much more
compatible.  The following programs are provided:

DBSETUP - This program is used to set up database files.  It is an
    interactive program, which asks the operator a series of
    questions concerning the structure of the file.  Field
    names and types are entered, and it is decided whether the
    file will be random or sequential, whether it will have
    an associated index file (.IND), and whether there will be
    links between each data record.
DBENTRY - This program is also an interactive program, which is used
    to enter items into a file.  It may be used for initial entries
    or to append additional entries on the end of a file.

DBQUERY - This program is used to ask questions of the database,
    and may also be used to enter new items or make changes.  Reports
    of different types may be generated.

DBMAIN - This program is a transaction entry program.  The main idea
    is to update several different files by entering each transaction
    only once.  This program is more specialized than any of the others,
    as it is oriented toward one user's situation:  in this case a small
    manufacturing company (mine).  Each set of transactions is entered by
    first typing in a one-letter command.  The computer program then
    asks the operator a series of questions to obtain the data for that
    transaction.  The responses normally go into a buffer until all
    transactions of that type are entered.  Then the responses in the
    buffer are used to update the appropriate files.  The commands and
    the files they update are listed below:

    Commands:    Transaction    Files Updated

        R    Received items:  RLOG, PINV, PAYABLE, PURCHASE
        S    Shipped items:  SLOG, PINV, RECEIVBL, ORDERS
        A    Assembled or packaged items:  PLOG, PINV
        O    Order Received:  OLOG, ORDERS
        P    We Paid someone:  PLOG, PAYABLE
        C    Someone paid us:  CLOG, RECEIVBL
        M    An order was made by us to someone:  MLOG, PURCHASE
        Q    A query is to be made:  all files
        D    Set the Date:  no files

        Filename    Description

        RLOG        Receiving log
        PINV        Production Inventory
        PAYABLE        Accounts Payable
        PURCHASE    Open Purchase Orders
        SLOG        Shipping log
        RECEIVBL    Accounts Receiveable
        ORDERS        Open Orders to us
        PLOG        Payment Log (like checkbook)
        OLOG        Order (Sales) log
        CLOG        Cash, checks, or money orders received log
        MLOG        Purchase orders by us log

The log files are sequential;  they are added to as time goes along by
each command.  They may be used to recreate the other files if something
goes wrong.  The other files are generally divided into two parts:  an
index, which is sequential, and the main file, which is random.  The
index contains a list of keys which are used to access the records in
the random file more quickly than a search of the entire file would
allow.

These programs only represent a start.  Tarbell Electronics would
appreciate any additions that people would like to contribute.
Close attention has been paid to making the system as extensible as
possible, with the hope that it will continue to grow and change to
meet different user's needs.

It may also be noticed that certain features explained or referred to
by the programs or documentation are not actually implemented.  This
is because the programs are in a continuous state of flux.  It is
felt, however, that the programs are already to a certain level of
usefulness, which is why they are being made available.

Notes about the file formats:

Each file has some general information at the beginning, called a
Header Block.  This block contains information about the structure
of the rest of the file, and may be several records long.  The
first record is called Header 1.  The format of this record follows:
    RECLEN        number of bytes per record, zero if sequential.
    NO.OF.FIELDS    number of fields per record (1 or more).
    NHEAD        number of heading lines for the file.
    NIND        number of indexes for the file (0 or 1).
    FILETYPE    0=sequential,1=random,3=indexed,7=linked
    CODE        restrictions: 0=none,1=read,2=write,3=r/w.
    SPARE        for future expansion.

After header 1, there may be as many lines of headings as are desired.
Then the field definition records start, with the following format:
    FIELD        field name in ASCII.
    TYPE        0=numeric, 1=alphanumeric.
    WIDE        bytes for this field, 0=variable.
    FEED        0 means same line, 1 means next line.
    MIN        minimum value if numeric (0 is default).
    MAX        maximum value if numeric (0 is default).
    SPARE        for future expansion.

The data then starts, with the following format:
    field1!field2!field3!...!fieldn!

All of the above formats are generated automatically by the program
in response to operator answers and input.  The sequence is normally
to first run the SETUP program to define the name and structure of the
file;  then run the ENTRY program to make the initial entries; then
run the QUERY program to access the files.

Don Tarbell