View Single Post
Old 08-07-2018, 12:39 AM   #1
atl068
Member
atl068 could sell banana peel slippers to a Deveel.atl068 could sell banana peel slippers to a Deveel.atl068 could sell banana peel slippers to a Deveel.atl068 could sell banana peel slippers to a Deveel.atl068 could sell banana peel slippers to a Deveel.atl068 could sell banana peel slippers to a Deveel.atl068 could sell banana peel slippers to a Deveel.atl068 could sell banana peel slippers to a Deveel.atl068 could sell banana peel slippers to a Deveel.atl068 could sell banana peel slippers to a Deveel.atl068 could sell banana peel slippers to a Deveel.
 
Posts: 18
Karma: 3134
Join Date: Oct 2016
Device: Onyx Boox m96
Unable to use select_form in mechanized browser

My university uses OKTA for authentication, and then it allows me to access many journals, e.g. NEJM. So, for example, https://www.nejm.org.ezproxy.flinders.edu.au will allow me to read all of the journal articles in New England Journal of Medicine. NEJM has a great RSS feed and I like to use it with Calibre. Alas, the built-in recipe is for direct subscribers.

Thus, I tried to modify the script. However, I ran into a problem. For OKTA, the form doesn't have a "name" but has only an "id". So the br.select_form(name="login") in the built-in script will fail and I can't change it to any other name. The form id appears to be "form17". Looking at the mechanized browser in python, I should be able to use alternative methods like: looking for form with a class containing some names, or the Nth form using nr=17 etc. However, when I use anything other than select_form(name=xxxx), Calibre fails saying that it was unexpected. What am I doing wrong? I just can't figure it out. I tried google and forum searching for referring to the form id instead of name, etc... but to no avail.

Thanks.
atl068 is offline   Reply With Quote