Sequential (Text) Files

Because of their universally standard format, simple text files (also called sequential files) are often used as the storage method for information. The old DOS Edit program and the newer windows NotePad programs both create simple text files.

To simplify the handling of text files, Microsoft is working on a new set of features that will be implemented in an ActiveX object (the FileSystemObject). Until then, file handling will continue as the manual procedure that it is today.

The two statements you must use to access a text file are OPEN and CLOSE. Here's a quick example:

OPEN "filename" for INPUT as #1
CLOSE #1

In this example nothing was done but opening the file and closing it.

Here's a little more useful example:

OPEN "filename" FOR INPUT as #1
WHILE NOT Eof(1)
LINE INPUT #1, temp$
WEND
CLOSE #1

In this example, the code reads one line at a time, walking through the file until the end-of-file (EOF) is true. The line of data (minus the carriage return / line feed characters) is put into the string variable temp$. In your own applications you can save each line into an array, or concatenate them into a single string variable, or any other processing you want to do.

This next example shows how the text can be read as a single string variable and put into a textbox control.

OPEN "filename" FOR INPUT as #1
WHILE NOT EOF(1)
LINE INPUT #1, temp$
alltext$ = alltext$ & temp$ & vbcrlf
WEND
textbox1.text = alltext$
CLOSE #1

Note that I had to add back in the carriage return / line feed to maintain the same line break as was in the original file. If you want the file to be read simply as a long string, replace the vbcrlf with a space (" ") to keep separation between the word at the end of the line and the word which starts the next line.fx

The examples so far were to read a text file. Here's an example for writing to a text file:

OPEN "filename" FOR OUTPUT AS #1
For i = 1 to 10
PRINT #1, i
NEXT i
CLOSE #1

In this case, each PRINT operation goes to a different line in the text file (which means a crlf character is inserted into the byte stream). The PRINT statement provides a variety of options for putting the data into the output file. You can put all the data in one line, separate each piece by one or more characters, or format the numbers as they are written. The bottom line is that whatever is written to the file is written as text and when it's read back your program must convert it to numbers as needed.

One more example will be worthwhile. Suppose the data consists of columns, and you next to extract the numbers from those columns. How do you do it? Here's how it could be done for the case of having 3 numbers per line, 1 in each ten columns:

OPEN "filename" FOR INPUT as #1
WHILE NOT Eof(1)
LINE INPUT #1, temp$
Number1 = VAL (mid$(temp$, 1, 10))
Number2 = VAL (mid$(temp$, 11, 10))
Number3 = VAL (mid$(temp$, 21, 10))
WEND
CLOSE #1

Binary (Data) Files

One of the really great features of binary file access is that you can often avoid using the VB database features. This is important because the distribution files for database access are HUGE and I hate having to put them in my application distribution files.

Another especially nice feature of binary files is that you can PUT a complete user-defined variable in a single statement, and recover it just as easily. This greatly simplifies data storage in many cases.

Let's get right into some examples. To put data into a file, use this:

OPEN "filename" for BINARY as #1
PUT #1, var1, var2, var3
CLOSE #1

In this example, all we did was write three variables to the file. To read them back, use this:

OPEN "filename" for BINARY as #1
GET #1, var1, var2, var3
CLOSE #1

You'll notice that we didn't have to format the PUT or GET statements in any way.