There are a number of reasons why altering the code isn't a good idea. It's nothing to do with "imgupr is ok, let everyone use that".
First is that we'd have to modify core WW code. All of our custom work is sensibly split off into a UKVAC branch, so we don't break upgrades.
If we just appended a random number to the file name we'd still have a chance of file name collisions. The likelihood would increase as more photos were uploaded and so then we'd end up with a much more random failure.
One way around that would be to add a hash to the file name, but then we're going to be adding 100+ bytes to each file name. Does WW have any short file name length buffers? And short string manipulation routines? Hopefully not but hope isn't generally something I look forward to in a code release
and a full code audit is impractical. Even if we just used a hash as the entire file name we'd make it impossible for anyone to recognise a file by name so they'd have to click through File Manager to look for some old pic.
And once the uploaded had been modified to hash and concatenate we'd have to check the following code path that ties image upload to insertion plus any other part of the File Manager. By which point we'd have touched a non-trivial percentage of the WW code so that when they release a supported patch, we can't apply it because we're running a weirdly non-standard hacked up version.
It's honestly not that we can't be arsed - it's that the available solutions all have a degree of complexity and the imgupr has the least when considered against the requirements of all forum members.
