Regex Solution not being accepted as correct answer

Q&A DQ Courses
,
#1

Screen Link:
https://app.dataquest.io/m/369/advanced-regular-expressions/8/extracting-domains-from-urls

My Code: 

#pattern = r'https?://([\w-]+\.[\w-]+\.?[\w-]?)' (Please ignore this)
Edit - r'https?://([\w-]+\.[\w-]+\.?[\w-]*)' 
test_urls_clean = test_urls.str.extract(pattern, flags=re.I)
domains = hn['url'].str.extract(pattern, flags=re.I)
top_domains = domains.value_counts().head(5)

What I expected to happen:
The solution above gives the correct results and should be accepted.

What actually happened:

image
image888×487 13.3 KB

No error message is shown but until we give the regex pattern as in the solution answer, it is not allowing to submit and pass the exercise.

#2

Your code shouldn’t be given you those results.

Here’s what I see on my end:

image
image824×769 64.4 KB

#3

Hi Bruno,
I am sorry, I posted the other pattern, this is the one I was trying -
pattern = r’https{0,1}://([\w-]{1,}.[\w-]{1,}.{0,1}[\w-]{0,})’
It is same as r’https?://([\w-]+.[\w-]+.?[\w-]*)’ In the topic post I mistakenly wrote ? instead of * at the end.

#4

The pattern https{0,1}://([\w-]{1,}.[\w-]{1,}.{0,1}[\w-]{0,}) doesn’t yield those results either.

1 Like
#5

Hi @Bruno, somehow posting the reply removed the backslashes present before the ‘.’ character in the pattern as the code was not between backticks, apologies for the confusion, I have edited the pattern in the main post. Here it is again -

pattern = r'https?://([\w-]+\.[\w-]+\.?[\w-]*)'

image
image923×856 40.2 KB

#6

Here’s a sample of the differences between Dataquest’s solution and yours:

Dataquest Sohamdey
www.cam.ac.uk www.cam.ac
www.independent.co.uk www.independent.co
mint.lc.intuit.com mint.lc.intuit
www.bbc.co.uk www.bbc.co
www.eecs.berkeley.edu www.eecs.berkeley
1 Like
#7

Hi Bruno, Thanks a lot for pointing it out. So, if I put another set of ‘\.[\w-]*’ at the end,
i.e

https?://([\w-]+\.[\w-]+\.?[\w-]*\.?[\w-]*)

it would cater these cases, but would make it more complex.
Is it better to avoid this completely?

#8

Ok, I found out, again this will not be matched - www.keith.seas.harvard.edu

#9

Yes, I don’t think this is going in the right direction. Take a look at what happens with http://ftp://ftp.tcl.tk/pub/incoming/p15/RichardHipp/microoptimization/paper.html. Your technique will never match this.

To be fair, however, Dataquest’s solution also fails as it says the domain is ftp, when it seems it should be ftp:.

1 Like