-Option defaults to a day
-Cookie is saved to keep the last time used

-Added Title to scraps and links
-Links flow
-Loading scraps gets the hash from the struct just created, rather than the file attempted (for logging to be saner)
-Total count loaded shown after loading the scraps on startup
-Header sends no-cache

-Changed text to none (which really doesn't seem to change anything, but might be correct for Prism)
-Added background to pre blocks to make plain text stand out properly
-Removed a few newlines in the template to make the source look nicer in the browser's view source option

-Janitor launches once per scrap after the loader determines it's not expired yet
-Scrap.delete() is called after the delay

-The most recently used language is stored in a cookie and set as default for the next two weeks

-Even better styling (again)
-Username cookie is stored and retrieved (duration: 2 weeks)
-Argument to the index handler renamed to avoid confusion with context parameters

-Pointers in struct funcs
-Delete scraps from global list when deleting files

-Some styling changes
-Recent posts show up, max 10
-Newly created scraps were not added to the global list

-Posted scraps are shown when their hashes are added to the index URL
-Prism.js adds syntax highlighting
-Small bugfixes

-Removed unused template generation
-Corrected variable showing languages

-Website with form to post a scrap of code
-Necessary folders are created on startup if they don't exist
-Index is saved with hash as name, containing the language and expiry timestamp
-Actual data is saved in "./data/<language>/<hash>" (TODO: make datapath configurable)
-Scrap struct has save, load, check for expiry and delete funcs
-Checking for expiry also deletes the two associated files if it has expired

TODO: Get a list of 10 recent posts and display them on the front page
TODO: Store username with cookies
TODO: Store username in index files
TODO: Timers to delete expired files during runtime, instead of relying on startup