I can provide documentation to support this,
But when I was discussing this with Vlad, I suggested that rather than the timestamp, a facility to allow us to name the report/PDF would be better.
If I am printing a customer invoice, then what can be more unique than The invoice number?
Example, if before it just created a document called customer invoice.PDF.
Then when you save it to your HD, that’s what it’s called.
And if the user wants to save it, they have to go manually rename it because there’s gonna be another one behind it called customer invoice (1).
And then customer invoice (2), this numbering being handled automatically by the Google download or operating system.
So if we know we’re printing an invoice for Acme tools, then why not just create the file called customer invoice – Acme tools.
Even better, customer invoice – acne tools – #12345.
This would avoid the cache issue because the next invoice would be customer invoice – Baker supply – number 12346.
Both of these would be unique names and would not fall Pray to the cache bug that popped up
And it would have the extra benefit of not making the end user have to manually rename it at all from its generic output name.
In my opinion, this would be an easy fix.
I’m glad it’s come to some visibility because usually there are issues here that at least when I post them, no one chimes in or seems to care, and I’ve always said that if no one‘s complaining about anything then why should Vlad put any effort into changing the behavior.
Going back to “the old way “is not really the solution as there’s definitely a bug and several circumstances when the name is the same over and over, we could literally see , as I said in my original post, the second invoice was not being downloaded because of cache, and you would not get the correct invoice shown on the screen if it was in a certain amount of time because of the cache expiration duration.
