Storage and processing of data

File organization and access

The LASER-DOS allows you to save data on the diskette from a BASIC program and then process them again.

This data is stored in special data files with the type code "D".

The storage form offered is “sequential”. Sequential means that the data in the file is stored one after the other, like on a cassette. Reading or writing data always begins at the beginning of the file, further read and write calls access the subsequent positions of the file.

In contrast to this is the “direct” type of access (random access), with which any data in a file can be accessed directly. Unfortunately, the LASER-DOS does not support this type of memory as standard. However, it can easily be reproduced with a little knowledge of assembler/machine language and the help routines described in the last chapter.

Sequential access is data flow oriented, i.e. the number of characters for a write or read process can vary. This is also referred to as data records of variable length, where a data record is the sum of the data elements that are written to or read from the diskette with a write or read call.

Sequential files represent the simplest form of data storage and retrieval. They are ideal for storing raw data without wasting a lot of space between each data element. Data is read back in the same order as it was written.

In order to be able to access a data file, it must first be opened. A special OPEN call is available for this purpose. With the opening you also specify the type of access, whether data should be written or read.

After completing the data manipulations, each data file should be closed with a CLOSE call.

When editing data files, there are a few important points to keep in mind:

If a file that does not exist is opened for writing, it is created anew and positioned at the start of the file.
If an existing file is opened for writing, it is positioned at the end of the file, i.e. the file is extended with the following write calls.
If you want to rewrite an existing file from the beginning, you must first delete it.
After opening, reading always begins at the beginning of the file. If you are looking for data within a file, you must read over the preceding ones.
To update a sequential file, read in the source file and write the updated data to a new file.
When reading the file, the exact structure of the data record to be read must be known. This does not refer to the length of the sentence and the individual elements; However, you must know the number of elements and the format of each element (string, integer, etc.) and provide an appropriate receiving field for each element.
Within the file, the data is stored exclusively in ASCII format. The individual elements are separated by commas. For example, the number 1.2345 takes up 8 bytes of storage, including a space for the sign at the beginning and another space at the end. The text “ROBERT MAIER” takes up 12 bytes on the disk.
A data record should not be longer than 200 bytes, otherwise problems with the internal data structure of BASIC will occur when reading.

The 200 bytes count
- the individual characters
- the commas as element separators
- for numbers, the sign and an additional space
- a final RETURN (CR) at the end of the sentence
- A maximum of two files can be opened at the same time, whereby the types of access can be the same or mixed.
Warning

If your system reports DISK BASIC V1.0, work with one file at a time to be on the safe side. The management of two open files is still incorrect there and can lead to significant data loss.
The DOS does not tell you when the end of the file is reached when reading, you have to determine this yourself, for example by writing a specific end identifier as the last in the file.

Example of sequential output:

We want to store a table of English to metric conversion data.

English unit	Metric unit
1 Inch	2.54001cm
1 Mile	1.60935 km
1 Acre	4046.86 qm
1 Cubic Inch	0.01639 ltr
1 U.S. Gallon	3.785 ltr
1 Liquid Quart	0.9463 ltr
1 lbs	0.45359 kg

The data should be structured as follows on the diskette and entered in a data file called “ENG>MET”:

“English unit -> Metric unit” , conversion factor

e.g. “IN->CM”, 2.54001

The following program creates such a file.

OPEN "ENG>MET",1
FOR I% = 1 TO 7
READ E$,F
PR# "ENG>MET",E$,F
NEXT
CLOSE "ENG>MET"
DATA "IN->CM",2.54001,"MI->KM",1.60935,"ACRE->QKM",4046.86E-6
DATA "CU.IN->LTR",1.638716E-2,"GAL->LTR",3.785
DATA "LIQ.QT->LTR",0.9463,"LB->KG",0.45359
END

Line 10 creates the file “ENG>MET” and opens it for writing.
In line 40, one data record is written to the file.
Line 50 closes the “ENG>MET” file again.

Example of sequential input:

The following program reads the “ENG>MET” file into two parallel matrices and then asks about conversion problems.

CLEAR 1000
DIM E$(6),F(6)
OPEN "ENG>MET",0
FOR I% = 0 TO 6
IN# "ENG>MET",E$(I%),F(I%)
NEXT
CLOSE "ENG>MET"
CLS: PRINT "CONVERSION ENGLISH=>METRIC"
PRINT: FOR I%=0 TO 6
PRINT TAB(4); USING "(## ) % % ";I%,E$(I%)
NEXT
PRINT @320, "WHICH CONVERSION (0-6)";
INPUT W%: IF W% > 6 THEN 190
INPUT "ENGLISH VALUE";V
PRINT "THE METRIC VALUE IS" V*F(W%)
INPUT "CONTINUE WITH <RETURN>";X
GOTO 100

Line 30 opens the file for input. Reading begins at the beginning of the file. In line 50, a data set with the elements E$ (unit) and F (factor) is read and distributed to the matrices.

Note that the variable list when reading in is the same as the write command in the previous program.

In line 70 the file is closed again.

Updating a file

If you want to add one or more records to an existing file, open this file for writing and simply enter additional data records with PR#, which will be appended to the existing database.

If you want to change data within a file, we recommend the following procedure (not with DISK BASIC V1.0).

Open the file to be edited for reading.

Open a second new file for writing

Read a record and edit the data

Write the record to the new file

Repeat points 3 and 4 to the end of the file

Close both files

Delete the source file

Rename the new file to the original file

With DISK BASIC V1.0, the only solution is to read the file to be processed completely іп the memory, process it and write it completely into the new file. However, this limits the size of the file to the available memory.

OPEN - Open a file.

Syntax:

OPEN "name",n
"name" - File/program name, max. 8 characters,
         enclosed in quotation marks.
n - type of access
    0 - Read
    1 - Write

Permitted only in program mode.

The OPEN command opens a data file (type = D) for writing or reading.

The OPEN command creates a file control block internally for each open file, which contains function codes and pointers.

Furthermore, the following is positioned on the data according to the access code: * When reading, always at the beginning of the file * When writing to a new file, to the beginning of the file * When writing to an existing file at the end of the file,

Since there are only two file control blocks in the system, only two files can be open at a time. The type of access is irrelevant, both can be opened for writing, both for reading or one for reading and the second for writing (see restriction DISK BASIC V1.0 on the previous pages).

Example:

OPEN "TEST",0

The “TEST” data file is opened for reading.

A data file can only be opened once at a time. Attempting to open the same file again results in an error message.

Since the file control blocks (File Control Blocks (FCB)) are located outside the BASIC programs, a file remains open if the calling program was aborted before the CLOSE call due to an error or by pressing the BREAK key and is perhaps no longer in memory. Such a file can no longer be opened without further ado.

If it happen that a BASIC program is aborted without properly closing its files, you should do so with a direct command (CLOSE “filename”).

Possible Errors:

?ILLEGAL DIRECT An attempt was made to execute the OPEN command in direct mode.

?SYNTAX ERROR

one or both parameters are missing
no comma as separator
filename not in quotes
access type not 0 or 1

?FILE ALREADY OPEN File is already open, if necessary close it with the direct command CLOSE.

?FILE TYPE MISMATCH The file addressed in the OPEN command is not a data file

?FILE NOT FOUND A file to be opened for reading does not exist on the diskette.

?DISK BUFFER FULL Two files are already open and no more file control block is available.

?DISK I/O ERROR An error occurred while reading from the floppy disk.

PR# - Writing records to a file

Syntax:

PR# "name",item list
"name" - File/program name, max. 8 characters,
         enclosed in quotation marks.
item list - List of variables and values to be written to the file.
            The individual elements are to be separated by commas

Permitted only in program mode.

Assembles a data record from the values in the element list and causes it to be written to the data file.

This must first have been opened for writing with an OPEN command.

Example:

A1 = -40.456: B$ = "STRING-VALUE"
OPEN "TEST",1
PR# "TEST",A1,B$,"THE VAR'S"
CLOSE "TEST"
END

After opening the “TEST” file in line 210, a data record is compiled in line 220 and written to this file.

The data record contains the current values of Al and B$ and also the character string “THE VAR’S”. The values can later be read in again with an IN# command.

It must be ensured that the element list of the IN# command is the same as that of the PR# command with regard to the number and type of elements.

The values represented by the item list should not exceed 200 characters in total. In addition to the values themselves, this also includes all separators (commas) between the values, in the case of numeric values the sign position and a trailing space and finally the end of data record identifier (CR).

The record in the previous example would be 31 characters long

-40,456 ,STRING VALUE, THAT’S IT

Unfortunately, when creating the element list, one often does not know exactly how large the individual variables will be at the time of storage. Then only careful estimation helps. Always stay on the safe side and, if in doubt, split your element list into several PR# commands.

Unfortunately, the PR# command does not notice when a data record is too long. This is simply written to the diskette in its entirety. Reading in with the IN# command then causes problems, whereby in the simplest case “only” data is lost.

Possible Errors:

?ILLEGAL DIRECT An attempt was made to execute the PR# command in direct mode.

?SYNTAX ERROR

no file name specified
Filename not in quotes
no item in the list
no comma as separator

?FILE NOT OPEN File was not previously opened.

?ILLEGAL WRITE The file has been opened for reading.

?DISK WRITE PROTECTED The disk’s write-protect notch is taped over.

?DISK FULL No more free sectors could be found on the diskette.

?DISK I/O ERROR An error occurred while reading or writing to the diskette.

Warning

If one of these errors occurs, the program is terminated with the corresponding error message. Please note that this file was not closed afterwards, you should do this manually.

IN# - Reading records from a file

Syntax:

IN# "name",item list
"name" - File/program name, max. 8 characters,
         enclosed in quotation marks.
item list - List of variables and values to be written to the file.
            The individual elements are to be separated by commas

Permitted only in program mode.

IN# reads a record from the specified file and assigns the elements of that record to the specified variables.

The file must first have been opened for reading with an OPEN command.

Example:

200 OPEN "TEST",0
210 IN# "TEST",X,A$,B$
220 CLOSE "TEST"
...
...

This example refers to the data set created in the example of the PR# command in the “TEST” file. The data stored there are assigned to the variables of the IN# command in sequence.

After executing line 210, the variables contain the following values:

X = -40.456
A$ = "STRING-VALUE"
B$ = "THE VAR'S"

The element list of the IN# command must correspond to that of the PR# command with regard to the number and type of variables. Likewise, the order must be observed for different types, the naming is irrelevant.

If records are read continuously from a file with IN#, it is difficult to recognize the end of the file at the right time. There is no special “END OF FILE” identifier for LASER-DOS.

There are various possible solutions:

the number of records is known, they are counted with a counter in the reading program,
a second small file contains the sentence counter for the main file.
A short label consisting of only one alphanumeric character (e.g. PR# “name”, “A”) is written in front of each correct record.

In the reading program, this identifier is first read before each reading of a data record (e.g., IN# “name”, A$) If the receiving string variable is then empty, the end of the file has been reached.

Possible Errors:

?ILLEGAL DIRECT An attempt was made to execute the IN# command in direct mode.

?SYNTAX ERROR

no file name specified
Filename not in quotes
no item in the list
no comma as separator

?FILE NOT OPEN File was not previously opened.

?ILLEGAL WRITE The file has been opened for reading.

?ILLEGAL READ The file was opened for writing.

?DISK I/O ERROR An error occurred while reading or writing to the diskette.

?REDO The type of one of the specified variables does not match the data read in from the diskette. The program continues to run, the variable remains empty.

?EXTRA IGNORED In the variable list of the IN# command fewer variables are given than values are present in the data set, the numbered values are ignored, the program continues.

?? The variable list contains more variables than there are values in the data set. The frogram now expects the missing values to be entered via the keyboard.

Warning

If one of these errors occurs (except REDO, EXTRA IGNORED and ??), the program is terminated after the corresponding message has been output. Please note that this file was not closed, you should do this manually.

CLOSE - Closing a data file

Syntax:

CLOSE "name"
"name" - File/program name, max. 8 characters,
         enclosed in quotation marks.

Allowed as direct command and in program mode.

A previously processed data file is closed with the CLOSE command..

If a file is open for reading or an inactive file (i.e. the last file access was not to this file) or in direct mode, only the file control block (File Control Blocks (FCB) = File Control Block) is released again. Disk access does not take place.

However, if the CLOSE command is given in program mode and the file to be closed is open for writing and is currently active, the last sector in the buffer is also written back to the diskette so that no data is lost.

It is good programmer practice to close any open file after use. However, it is essential for output files, unless you accept the possibility of data loss.

Example:

CLOSE “MAILBOX”

The “MAILBOX” data file is closed.

It is always necessary to close and reopen a file if you want to change the type of access (e.g. from writing to reading).

If the file to be closed is not open at all, i.e. there is no open file control block for this file, the CLOSE command is skipped without any error message. This is especially useful for closing all files used in a program prophylactically at the end without checking which ones are currently open.

Possible Errors:

?SYNTAX ERROR

no file name specified
Filename not in quotes

?DISK WRITE PROTECTED The disk’s write-protect notch is taped over.

?DISK I/O ERROR An error occurred while reading or writing to the diskette.