fvieira / coursera_resources_downloader Goto Github PK

Since the JSON data where the script get's the info from the courses changed format, the command is broken.

./coursera list-course-ids command gives "modelthinker" as id; should be "modelthinking"

Old XPath for course_title_element no longer matches. New XPath is course_title_element = tree.xpath('//div[@class="course-logo-name"]/text()').
The error message for not course_title_element is strange. 'Failed to get the course title at {0}' might be better.

Looks like the previous issue is back, due to change in the site. Please see if you can fix it again. Thanks.

For the past week or so, the program fails with the following message:

Trying to open lecture index page for the vision course
Done

ERROR: Failed to get the course title at https://class.coursera.org/vision/lecture/index
Please make sure that you are enrolled in the vision course and that it has already started.

Something has changed in all the coursera web pages and now the following line does not give back the expected result:
course_title_element = tree.xpath('//div[@Class="course-logo-name"]/nobr/text()')

I'm assuming there are always 4 resources per lecture and that they're order is pdf, pptx, subs and video.
The algo course has two cases where this is wrong, in one the subtitles are missing and in the other there are two pdfs and no subtitles.
I should be looking at the url to decide what the resources are and name them accordingly.

It would be better to be able to pass the user and password to the script and have it handling the authentication, at least these wouldn't change each time you log out and are more user friendly than having to somehow get the cookie from the browser.