Register now to learn Fabric in free live sessions led by the best Microsoft experts. From Apr 16 to May 9, in English and Spanish.
Hi all,
The main purpose is to find a way to scrap all source code from a web page. So far, "From web" method doesn't work reliably for me when imported as csv, text or html because some of the content is not imported.
I was trying to scrap page ids from facebook pages. In the source code, one occurence always comes after the string "/?id=", as such:
I first tried through the get "From web" option of powerquery, but the result is rather disappointing; it either times out or imports (very) incomplete source code, without most of the <head> part:
I the imported it as text and wrote the following function which worked:
let page_id = (local_FacebookPage) => let Source = Table.FromColumns({Lines.FromBinary(Web.Contents(local_FacebookPage))}), #"Kept First Rows" = Table.FirstN(Source,4), #"Split Column by Delimiter" = Table.SplitColumn(#"Kept First Rows", "Column1", Splitter.SplitTextByEachDelimiter({"/?id="}, QuoteStyle.None, false), {"Column1.1", "Column1.2"}), #"Split Column by Delimiter1" = Table.SplitColumn(#"Split Column by Delimiter", "Column1.2", Splitter.SplitTextByEachDelimiter({""""}, QuoteStyle.None, false), {"Column1.2.1", "Column1.2.2"}), #"Removed Columns" = Table.RemoveColumns(#"Split Column by Delimiter1",{"Column1.1", "Column1.2.2"}), extract_page_id = Text.Combine(Table.Column(#"Removed Columns", "Column1.2.1")) in extract_page_id in page_id
A day ago it stopped working on my test page: https://www.facebook.com/groups/TalkPowerBI/.
The explanation is that even imported as text, it shows only part of the content now, and not the part where my function is looking for the id.
Somehow though, it still works on other pages like: https://www.facebook.com/PowerBIGuru/.
Antoine
Solved! Go to Solution.
I think the reason is that the Guru page is public while the Talk page is not. When I open a Chrome window in incognito mode, the Talk page redirects me to login. I think that's the same thing happening for Power BI desktop. It actually scarps the login page so that's why no "/?id" found.
I don't find that "page/?id" even when I view the source code from Chrome for the TalkPowerBI page.
For the page PowerBIGuru I can find the "page/?id" and your PowerQuery captured the expected id.
So I think Power BI is just working fine. The issue is the facebook pages seems not always containing the page id.
By the way, in my opinion, Power BI is a good tool for web scarping.
Hi @Eric_Zhang,
Thank you for your reply.
The string to look for is "/?id=" and not "page/?id" (facebook has different page types, including "page/?id=", "group/?id=" and such).
So in your example, for talkPowerBI, it looks like this :
And the id is "661419710731947".
And this one does not work in my function, this part is not imported.
Antoine
Anyone ?
I'd really like to understand why powerquery doesn't import all content with the From.Web connector.
I get that importing it as html doesn't work well, even though i think it's a shame. But even imported as text, it's incomplete on my tests.
Antoine
I think the reason is that the Guru page is public while the Talk page is not. When I open a Chrome window in incognito mode, the Talk page redirects me to login. I think that's the same thing happening for Power BI desktop. It actually scarps the login page so that's why no "/?id" found.
Hi @Eric_Zhang,
That seems to be it ! Only thing I can't figure out is why it used to work on that page and doesn't anymore. Maybe because I had a access token that I removed at one point ?
One last thing : any idea why the content is incomplete when using "html" parameter instead of "text" ?
Anyways, thanks !
Antoine
Covering the world! 9:00-10:30 AM Sydney, 4:00-5:30 PM CET (Paris/Berlin), 7:00-8:30 PM Mexico City
Check out the April 2024 Power BI update to learn about new features.
User | Count |
---|---|
109 | |
98 | |
77 | |
66 | |
54 |
User | Count |
---|---|
144 | |
104 | |
100 | |
86 | |
64 |