I am working on a dictionary utility and I need to read in data from disk, the database is a just a list of words, but with the 1st letter missing, and no delimiters, i.e.

APPLE
ACKLE
ADDLE

would be stored as a string PPLECKLEDDLE

With it so far ?

OK, to read in the data, we know that this particular file contains 5 letter words beginning with 'A', so we start with 'A' and read in the first 4 bytes and add that to the 'A', then start with 'A' again, read next 4 bytes and so on.

The reading is being done in a loop, doing a BGET# for as many characters we need to read, now this works perfectly, but it's quite slow as you need to loop reading just one byte at a time.

Is there any way in BBC Basic, machine code call, hack etc to read a known number of bytes, i.e. a string in one operation, i.e. something like

in$=BGET# file,number of chars to read

I cant use INPUT# as that expects a certain format, and it does not know where the string ends ?

Any ideas ?

Take a look at the OSGBPB call in the Advanced User Guide. The Electron version of the guide has this description of the call in section 5.5: "This routine enables the transfer of a group of bytes to or from an open file. This routine is implemented particularly for filing systems which have a high time overhead associated with each data transfer e.g. the Econet."

Thanks, that looks like it might well do the job, I've read up on it, and found some example code. Looks very promising, I'll give it a go

Just a quick update: Thanks DavidB, I got the OSGBPB call working

It speeded up that part of the code as I thought it might

Good to know it was useful.