Mag Gam wrote: Please don't top-post.
> Sorry if I wasn't clear before. > > While reading my csv file, notice I am putting the content in an array. That's already in the code you posted. What's missing is the value of "mtypes". I really meant it when I asked you to provide a self-contained example, i. e. one that I can run on my computer. > If lets say, row[5] has nothing in it No, it containes an empty string as MRAB already inferred. > , python gives an exception. What exception? What traceback? Please take the time to read http://catb.org/esr/faqs/smart-questions.html > Instead of the exception, I would like to assign 'NULL' to row[5]. The string "NULL"? That would be # untested for s, row in enumerate(reader): row = tuple(c if c else "NULL" for c in row) t = np.array([row], dtype=mtype) ... I'm not a numpy expert, but creating a lot of temporary arrays does look wasteful... > Does that help? Sorry, no. I (or someone else who doesn't bother to guess) could have come up with a better answer in a shorter time if you had actually answered my questions. Peter -- http://mail.python.org/mailman/listinfo/python-list
