I have been using DevonAgent to quickly acquire data, to enter into Devonthink.
I might not use DevonAgent in the most efficient way, but at least it is in a simple enough way to understand for me .
My usage scenario was:
I would like to download all the PDFs in a web-published repository of paper (papers of students). Links to these paper were distributed over 20 linked HTML pages.
My ideal way of work would have been:
Open page 1 in DevonAgent
Use the ‘magic hat’ and select all linked PDFs for download
go to next page, and repeat from step 2 until last page is reached
Wait for all the requested PDFs to download to complete.
My experienced usage was:
Open page 1 in DevonAgent
Use the ‘magic hat’ and select all linked PDFs for download
Wait for all the requested PDFs to download of a single page to complete. (as tab navigation was blocked by the requested downloads)
go to next page, and repeat from step 2 until last page is reached
I could not identify a quick way to prioritise navigation in the tabs over downloading files.
Actually downloading shouldn’t have (almost) no impact on the responsiveness of the user interface. Are you able to reproduce this? What’s the URL of the webpages?
A small clarification: with ‘blocking’ I meant only that the page was not loading further. (I do not mean that the UI becomes non-responsive)
A few more tests:
In the office I could not reproduce the issue, however at home I could.
(I would assume the main factor would be the less available bandwidth at home ± 15Mbps)
A second test with a (new) random site with PDFs, showed that DevonAgent is not blocking tabs by default indeed.
However, the site for which I was using it, still shows this behaviour, when DevonAgent was busy downloading PDFs from the same site.
The tab-browsing did not show ‘blocking’ behaviour If DevonAgents download queue was still busy with downloading PDFs from other sites (hence only blocking when PDFs were downloaded from the site itself.
I can share the url of the site by private message (if this feature can be enabled for my profile).
Downloads and webpages don’t block each other. But maybe the server limits the amount of concurrent connections. How many downloads did you add to the queue?
The issue has been resolved. I was not able to consistently reproduce it.
The nr. of concurrent connections seems to be the most likely explanation.
I was downloading BSc theses from a repository as reference material. A single page contained approx. 10 links to PDFs, and in total I needed to visit ± 37 pages (thus approx. 370 PDFs of theses to be downloaded).
The issue has been resolved for me. Thank you for your feedback!
.