So if someone has already created a build I don't have two different short urls map to the same long
So if someone has already created a build I don't have two different short urls map to the same long url
RThough that could probably be minimized with the cache
IStorage is more expensive in D1, but not much.
RI'm more scared of hitting the 2GB limit
IOh, I assumed you hashed the build to create the shortened link.
RThan the price
IIt should hopefully increase a bit more during the beta before GA 
RRAnd that becomes the ID in the table
II see. I would consider doing the hashing, though. It would avoid you adding an additional index on a pretty long column.
RWouldn't I run the risk of collisions there?
RWould be pretty bad
ITechnically yes, practically? I really doubt it. Provided the hash is long enough.
RSo just keep a third column with the hash?
RAnd index that instead of the long one?
IYes, one option.
RAh, that's a good idea
IOr just use the hash as the link
RWell, the hash would be bigger than I want though, I feel like 1-LnFi0Jy6sbF is a pretty nice length
IYou can do lower lenght hashes, too.
IOr just use a calculated column with the hash, SQLite should support the hashing function.
IOh, maybe not.
RHmm, not familiar with this, I'll have to look into it
RThanks!
II do know there are DB engines that do hash calculations automatically, SQLite may not so you might need to add it manually.
IBy the way, here are some stats on the hash collision probability, @Rabid
https://stackoverflow.com/questions/62664761/probability-of-hash-collision
https://stackoverflow.com/questions/62664761/probability-of-hash-collision
Stack Overflow
I am looking for some precise math on the likelihood of collisions for MD5, SHA1, and SHA256 based on the birthday paradox.
I am looking for something like a graph that says "If you have 10^8 ...
I am looking for something like a graph that says "If you have 10^8 ...
IHow many options have you got on that tree? How many possibilities? 
II'm honestly curious to calculate the chances here...
RWould be hard to calculate honestly.
RPeople can choose "paragon" skills:
Rmany many possibilities
R58 choices for skills to allocate
RPlus gear choices etc
II see...
IDo you know how many one can select and how many are there?
ISame here?
IAnd here?
I(Ignoring the fact some can only be chosen if the adjacent is, like total skill points given vs. total possible options)
R225 max selections for the paragon boards, 6 boards, up to 441 tiles per board
RBoards can be rotated
RPositioned differently lol
Rthey have slots where you can socket "glyphs" there are like 50 of those
IOh, ok. It's gonna get too involed 
RMost classes have about 12 gear choices, plus gems per gear slot haha
IIf the tables were fixed positions, it might be done easily ahah
RYeah hah. Well I was mostly worried about collisions if I was generating a shortened version of it that was only like 10 characters long
RBut the full MD5 should be fine, going to make that change, thanks for the suggestion! Should help keep my table much smaller
ÉHey, I currently use D1 via the API. Where can I request a limit increase for my key? Starting to get 429s :P
WIs there a slick way in D1 (generated columns, maybe?) to support updatedAt columns? Currently, I'm using the following.
COnly if you are enterprise. The recommended way would be to run an HTTP proxy worker so you donāt have a rate limit
CSomething like https://github.com/elithrar/http-api-d1-example
GitHub
An example HTTP API for Cloudflare's D1 database: https://developers.cloudflare.com/d1/
- GitHub - elithrar/http-api-d1-example: An example HTTP API for Cloudflare's D1 database: ht...