Suggested upgrade to the SerpApiClient.get_html() method #69
Conversation
|
I'm not sure if the |
|
@gutoarraes Sorry about that! You should be able to grad directly the HTML from just the ID. No need to download the JSON first. The URL should be always structured like this: https://serpapi.com/searches/5b50d58a304bda2fca30bac9.json?api_key=f0776d63edb8dd607154f126eb2c07049ef4bc06caf6f621d5065be2458a607d. Do you want to update your PR in this way or should we do it? Thank you! |
|
Hey @hartator, thanks for clarifying! If you wouldn't mind I'd like to give it a crack. |
|
Cool, thank you! Meant https://serpapi.com/searches/5b50d58a304bda2fca30bac9.html?api_key=f0776d63edb8dd607154f126eb2c07049ef4bc06caf6f621d5065be2458a607d instead of https://serpapi.com/searches/5b50d58a304bda2fca30bac9.json?api_key=f0776d63edb8dd607154f126eb2c07049ef4bc06caf6f621d5065be2458a607d in my earlier comment of course. |
|
I was cracking my head a bit about this just to realize I had already changed the PR but didn't update the description here. I removed the accessing of the JSON and simply appendended Let me know if there is a better approach, but I believe this may be what you're talking about. |
There was a problem hiding this comment.
@gutoarraes Thank you for your help!
Let's fix the test for get_html.
gutoarraes
force-pushed
the
augusto-get-html
branch
from
643c415
to
6816b81
Compare
There was a problem hiding this comment.
Thank you @ilyazub for the suggestions. I've applied everything and it's now ready for review.
Will start looking for the my next contribution. Cheers!
My suggestion to fix the open issue 66.
Since the returned object from
requests.getisn't necessarily an HTML, I approached this by accessing the JSON returned fromget_dictionaryand performing anotherrequests.geton theraw_html_file.Also added an assertion in
test_google_search.pyto confirm there is a<html>tag present in the returned data.