Good idea, but does it actually work?
Steemit seem to proxy all the images, so your tracking pixel becomes:
https://img1.steemit.com/0x0/https://steem-pixel.herokuapp.com/@kurtrohlandt/introducing-steem-pixel-track-your-views
and returns an error...
You are viewing a single comment's thread from:
Just realised that my first attempts to fix the issue worked but img1.steemit.com caches heavily (including errors lol). Event If you load that link now you will get a 500 error. I added an extra 's' end and now it works crystal. *facepalm.
Yeah, I'd assume it would cache the result.
So what's the veredict? No go because of the cache?
Lol the verdict is It's fixed it but didnt know because I was looking at a cached response and expecting to see the changes from my code reflecting in it. I have put full level logging on the system so if something like this happens again i'll know in 10 minutes unless I am sleeping. The only time it will be come tricky is in this exact case - where steemit is now caching the error but even that can be worked around by adding a query string parameter to the end of the url i.e. http://steempixel.com/@user/broken-link-address?revised. The revised will drop off into a different object.
I noticed that an hour ago and have been working on fixing it. The proxy just doesn't like data being return - the image is written out with a header - and probably rightly so too since it is a common method to attack servers... I am still working an a patch which involves switching it out with static content. It does still work though, the data is saved when I receive the request and the error is return when its parses my response. I think we have just found a potential minor vulnerability - when I wrote the code I tested in the editor, just checked again to be sure. The editor does not proxy the image.