```ts // the cache key from the cache URL const cacheKey = new Request(url.toString(), request); ```
can i ask what does including the request here do
Mso if you dont include it, you only cache by the url, but if you do include it what do you cache by
Murl and method ? surely not also headers right
Mits taken from here
MMlooks like in miniflare including the other parameter doesn't matter then?
https://github.com/cloudflare/workers-sdk/blob/f728503bd363b6421c35346c30d2d3e77025d49f/packages/miniflare/src/workers/cache/cache.worker.ts#L40
so im wondering if it even makes a difference at all or why its included this way in the example
https://github.com/cloudflare/workers-sdk/blob/f728503bd363b6421c35346c30d2d3e77025d49f/packages/miniflare/src/workers/cache/cache.worker.ts#L40
so im wondering if it even makes a difference at all or why its included this way in the example
CGoing to be honest, not sure if it does anything different and might just be a bad example.
Mok, thanks
Mwhat does
cf-cache-status: hit header mean in workers, i thought you had to handle caching on your own using the cache runtime api?Mi still get the log for the worker invocation if i enable realtime logs but the cf-cache-status header is "hit" so i dont really get it
CYou do. If you're fetching an origin behind CF you might be getting the cache status from the origin you're fetching
CWorkers will always be invocated, they run in front of cache
Minteresting
Mim not fetching anything though 
UWhats your code then? Either doing a fetch or adding the cf-cache-status header yourself is the only ways you'd get a cache status header returned from a worker
Mhttps://github.com/mastondzn/cdn/blob/main/src/routes/images.ts this is everything that happens on this route note the header is x-cache-status not cf-cache-status
GitHub
Personal image/file uploader, built with Cloudflare Workers & R2. To be used mainly with ShareX. - mastondzn/cdn
Mthis is a worker route
Myou can see the header right
UThat code is adding the header x-cache-status.
M
Myeah i see that which is intended
Mbut the cf-cache-status also appears
Mwhich is what i find odd
UIt might be automatically added by the cache api since you are using that 
EIf you return a response directly from the cache api then its going to give
cf-cache-status: HITMohh okay that makes sense
EBut if theres nothing in there then youre not going to get a MISS because Workers doesn't know what you intended, so it will just have no header
EAdding your own header is smart if you want to be sure whats happening, then you can add your own extra states as well if you find a need to
ESince you cant modify
cf- headersMokay thanks that makes sense actually
Mand also i wanna say i was using wrangler before v3 and now this is way better dx with the local dev mode great work guys 
Cminor but you can return your own cf-cache-status header from a worker, probably not a good idea, but I mean if you wanted to, could fake a miss
EFun, okay then
UYeah workers can add cf-<anything> just the rules stuff cant. or at the very least a few of the cf-<things> I tried worked
Ccan't override cf-ray but lots of other things you can fake
Mthis cache api is actually nice you can upload stuff and instantly insert in your cache aswell so even the first request can be cached, very cool
ODoes anyone have recommendations for a rate limiting library backed by D1 or KV?
UKV shouldnt really be used for rate limiting as theres a limit to writes on the same key to 1 per second. Not aware of any rate limiting libraries with D1 though, only rate limiter I am really aware of is this repo with the code for making one with durable objects: https://github.com/honzabit/durable-limiter
UOf course would highly recommend using cloudflare waf rate limiting rules if you can (they are free and unmetered now they used to be paid)
QQuick question around durable objects and setAlarm. Since Date.now() will only return the time of the last IO, the below example may set the alarm to a past time. Is this correct? Is the alarm still guaranteed to execute in that case?
DYour observation is correct, fortunately the documentation says that setting an alarm for a past date triggers immediately
DAha, found the docs: https://developers.cloudflare.com/durable-objects/api/transactional-storage-api/
If setAlarm() is called with a time equal to or before Date.now(), the alarm will be scheduled for asynchronous execution in the immediate future. If the alarm handler is currently executing in this case, it will not be canceled. Alarms can be set to millisecond granularity and will usually execute within a few milliseconds after the set time, but can be delayed by up to a minute due to maintenance or failures while failover takes place.
The Transactional Storage API allows you to achieve consistent key-value storage.
ESo is it still safe (from spoofing) to use MailChannels pages plugin so long as you have SPF, DKIM, and DMARC setup properly?
Qthank you so much for pointing this out, didn't see this part! Great to know
EHow do I pass a template to the mailchannels pages plugin? Is that possible? I see that it has a
content property, but I don't understand why it's an array.