wrote on 2014-05-13:
Does the API implement any sort of PDF caching based on request URL?
I'm working on integrating pdfcrowd in development environment where I am making continual changes to figure out formatting issues, and I'm seeing cases where I don't see my changes after they are deployed. Further, I'm seeing cases where if I trigger the PDF to be rendered twice, the first time may have the changes and the second might not, suggesting there may be caching on a per-worker basis.
I've tried adding a nonce into the my URL as part of the query string (e.g. http://example.com/my/page?nonce=<random_number>) and that doesn't seem to help, but if I change the URL structure to http://example.com/<random_number>/my/page the caching problem goes away.
Can you comment on the caching behavior of pdfcrowd?
wrote on 2014-05-14:
There is no PDF caching in the API at all. An API call is a POST request so it should not be cached even by intermediary proxies.