one solution could probably be made using: glob(), sort() and/or filemtime(). btw: I assume your aware that the filenames are not very unique (i.e. twice a day, everyday a given file will potentially be overwritten) and that this is intentional Ashley M. Kirchner wrote: > > I have files that are named in the following manner: > > file.####.jpg (where #### is an hhmm timestamp [hour and minutes]) > > What's the best way to take the current hhmm timestamp and run > backwards finding the previous 5 files, and checking whether they > actually exist? There's a possibility one or more might not exist and > rather than returning a blank value, I want to continue further back > till I find the next. > > So for example, say I have: > > file 1122.jpg > file 1121.jpg > file 1119.jpg > file 1117.jpg > file 1116.jpg > file 1114.jpg > > Notice the gaps, 1120, 1118, and 1115 are missing. But I still want > to return five, thus 1122, 1121, 1119, 1117, and 1116. > > I thought of taking the current timestamp and loop five times, > subtracting from it every time, but that's assuming all five are there. > If one's missing then I only have four images. > > -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
