base.Channel.readOctets

Read a given number of bytes and return as a character string.

Syntax

readOctets(
   length INTEGER)
  RETURNS STRING

length is the number of bytes to read, not the number of characters.

Usage

After opening the channel object, call the readOctets() method to read a given number of bytes from the channel. The bytes are returned as a character string.

Important: The bytes read with readOctets() must match the application locale. In multibyte encoding (UTF-8), if a sequence of bytes does represent a valid character, the resulting string will get ? question mark for invalid characters and data will be lost.

A valid use case of the method is the HTTP protocol. Reading HTML content with readLine() is not possible. The body consists of multiple lines, and the last line might not be terminated by a line-terminator, and the stream gets no EOF:

HTTP/1.0 200 OK
Date:  Wed, 16 Apr 2014 18:50:51 GMT
Content-Type: text/html
Content-Length: 1354

<html>
<body>
<h1>My title</h1>
  :
</body>
</html>

The readOctets() function returns NULL if end of file is reached.

Note: To distinguish empty lines from NULL, you must use the STRING data type. If you use a CHAR or VARCHAR, you will get NULL for empty lines. To properly detect end of file, use the isEof() method.

If readOctets() cannot read as many bytes as specified by the parameter, the method returns NULL. Before reading the actual bytes with a readOctets() call, find the number of data bytes to read from stream source, as shown in the example below.

Example

Note: This example uses UTF-8 encoding.

The main program:

MAIN
    DEFINE len INTEGER
    DEFINE chunk STRING
    DEFINE ch base.Channel
    LET ch = base.Channel.create()
    CALL ch.openFile("file.txt","r")
    WHILE TRUE
        LET len = ch.readOctets(3) -- 3 digits for length
        IF ch.isEof() THEN EXIT WHILE END IF
        LET chunk = ch.readOctets(len)
        DISPLAY len USING "##&"," ", NVL(chunk,"(NULL)")
    END WHILE
    CALL ch.close()
END MAIN

The data file (contains 3 digits for byte length of following string):

003abc006forêt000001x

Program output (note that forêt needs 3+2+1=6 bytes in UTF-8):

$ fglcomp -M readOctets.4gl  && fglrun readOctets.42m
  3 abc
  6 forêt
  0 
  1 x