Problem reading UDT as strings from binary file

[Chris Holbrook]

Duh!

forgetting that mid$(s,1) copies from the first character of the string, not the second. Will they ever learn?

[Chris Holbrook]

I have loaded each row of each array into a string (not asciz) , preceded by an identifying character

like this:

'        "C" is an array identifier in this case
        sTemp(n) = "C" + PEEK$(VARPTR(gCA(i)), SIZEOF(tCPH))
'       etc, then
....
        put# hfile, sTemp()

To dig out the code from the application (which is what I've resorted to) is taking a few moments...

[Michael Mattias]

OPEN sAppName FOR [B]BINARY[/B] AS #hfile
    FILESCAN #hfile, RECORDS TO nrecs
    DIM sTemp(0 TO nrecs - 1)  ' sTemp is a string array
    GET #hfile, 1, [B]sTemp()[/B]

GET help:

With a dynamic string array, it is assumed the file was written in the PowerBASIC and/or VB packed string format using PUT of an entire string array. If a string is shorter than 65535 bytes, a 2-byte length WORD is followed by the string data.

FILESCAN has a similar caveat.

You may be running into length words and/or array bound violations, the latter of course can cause all kinds of weird things to happen.

Of course, "CODE [which stores the data] NOT SHOWN" so this may not be a consideration at all.

[Chris Holbrook]

Searching for a way of storing a variable number of different array (both strings & UDTs) and a couple of non-array UDTs, I have loaded each row of each array into a string (not asciz) , preceded by an identifying character. The file resulting from this looks convincing.

Reloading the file looks like this:

    hfile = FREEFILE
    OPEN sAppName FOR BINARY AS #hfile
    FILESCAN #hfile, RECORDS TO nrecs
    DIM sTemp(0 TO nrecs - 1)  ' sTemp is a string array
    GET #hfile, 1, sTemp()
    CLOSE #hfile
    DIM gCA(%lotsandlots) AS GLOBAL tCPH
    DIM gEB(%lotsandlots) AS GLOBAL STRING
        FOR i = 0 TO UBOUND(sTemp)
        SELECT CASE LEFT$(sTemp(i),1)
            CASE "F"
                POKE$ VARPTR(Fstuff), MID$(sTemp(i),1) ' Fstuff is a UDT
            CASE "C"
                POKE$ VARPTR(gCA(lcx)), MID$(sTemp(i),1)
                INCR lcx
            CASE "E"
                gEB(lex) = MID$(sTemp(i),1)
                INCR lex
        END SELECT
    NEXT
    REDIM PRESERVE gCA( 0 TO lcx) AS GLOBAL tCPH
    REDIM PRESERVE gEB( 0 TO lex) AS GLOBAL STRING

and it does just what I wanted, except that everything is offset "left" by one byte, so that a string member of a UDT array will have a leading zero byte, for example.

Where am I going wrong?