You can download the XML list of datasheets which is what I use to determine if anything has been added or removed. That gives a list of where the PDF files are and one can determine if they have changed. So those parts can be automated.... your only option is to download the pdf every day and compare it by handThat doesn't however resolve the issue of knowing when documentation has been updated or what has been changed.
The big challenge is in determining what's changed between PDF releases. It is possible to extract the text from each and compare that. But narrowing down what the actual changes are is tricky. And, where there are huge changes, it's not easy to distil that into a simpler "Added details of Feature X" and the like.
Each document's Revision History does give some idea of what's changed but listed items can be rather vague 'things changed' without any detail of exactly what or where. And recently, and rather annoyingly, a document's Revision History can now indicate what has changed across a document set, not what has changed within any specific document.
A document's Revision History may suggest a lot of things have changed when that document hasn't changed at all apart from its Revision History. That also means you don't know which documents have actually changed, have to compare multiple PDF to discover if anything has changed in each.
I ran into that problem when tracking RP2040 related changes - viewtopic.php?t=370551
I don't like doing that, can't encourage it.or just open the pdf in the browser right from the url, and never save it
It goes against my principle of not downloading when you shouldn't have to and my belief that everyone should keep local copies instead.
While the energy consumed and the environmental impact may be small for individual cases it all adds up at scale. It may not be significant against other downloads but I do believe in "every little helps", see it as a point of principle. I haven't yet been convinced by "don't worry about the resources you are consuming".
Second it can be a PITA finding the datasheet you want in the on-line list of what's available. It is possible to create an index which is easier to use but once on that path you might as well automate as much as you can.
Statistics: Posted by hippy — Mon Jul 01, 2024 12:17 pm