RE: new error checking

> If we want the normal process to fail first, it would have to > continue into the write_file() routine, then return a failure, > then unravel the typically deep directory structure it was trying > to write, figure out the appropriate error, etc. Is it not true that you can't know that there is a problem until you actually try to write the file since the directory structure may not exist yet? So you would have to do your checks, then if they fail, you'd have to TRY to write the directories and then the file. If THOSE fail, then you are back to having to say that the process failed again. So why not just attempt the process and then report the status? One way or another, you may have to unwind part of the operation. > Also, I think the is_writable() information will already be cached by > PHP at this point since the file was already attempated to be read > from there. I could be wrong though :-) I think you are right, which I think implies that you'd have to clear the PHP stat cache for all files to get the correct result in the case that the directory didn't exist yet. ASIDE: Is there any reason that use_sub_dirs logic can't be changed to the following: - if create sub dirs fails issue a but silently set use_sub_dirs = false and try again. __________________________________ Do you Yahoo!? Yahoo! Calendar - Free online calendar with sync to Outlook(TM). http://calendar.yahoo.com