Few things I mentioned in chat that should probably be in one place.
First of all, Fyborg is updated in a few ways as to how it works with 4get:
- If you want to paste an url from a chan site that you don't want 4get to grab, put a ! somewhere in the line in front of it. Preferably as the first character, then a space, then paste your url. That way people can still click it to load it up.
- Duplicate check works properly now, notifying you if the MD5 sum of a file matches one already downloaded. I thought this was working properly long ago, but one line out of place left it broken. There's over a thousand dupes on the site right now that I'm going to have to write a script to parse through and delete.
- Anytime you paste a duplicate, or do a random !4get, you get linked to the "view" page of an image now. It's basically a framed page with a little info box at the top, telling who linked the image, when, and the tag, and the image itself is displayed in a frame at the bottom. I decided to use this method so that whatever resizing preferences people were using in their browsers would be applied to the frame as if it were opened in a full window like before. Also it was the most cross-browser solution to implement quickly.
- Images are automatically renamed now upon being downloaded, that way filename conflicts are no longer a problem. Name-based checking was only necessary back when there was no MD5 comparison. Filename checking is obviously not a good way to test for dupes, since an image could easily be named "1.jpg" on a dozen sites and always be a different picture. MD5 matching has a pretty accurate detection rate, so that's all that's used now.
Related to 4get:
- The
beta page of the upcoming version of 4get is gradually being updated. IE sucks so images will open directly in the window like before, but Opera/Firefox/Safari will get the new fancy auto-resize display. Clicking the image when expanded makes it go away. Eventually I'll probably try to make this functionality work in IE too, but it's almost more trouble than it's worth.
- The
stats page was created to show how many people linked images, and then updated to let you sort based on name or image count. And for the record, "Unknown" is the name I applied when images were initially inserted into the database when I switched over to that method of storing 4get info, from before names were recorded.
Other:
- The main website template was updated to fix the horizontal scrollbar that kept showing up in Firefox.
I think that's all!