tldextract.update() strange behaviour

[commutecat]

I just try to make a script to find out whether a domain make use of ccTLD, and differentiate between public or private domains.

So I initiate 2 explicit tldextract class that deal with this with explicit include_psl_private_domains switch. However, I notice the class behave differently if you "trigger" the extract ASAP than later on.

Code #1 show include_psl_private_domains=True works if I "test" the class function immediately.

# Trigger the extract immeidately 
import tldextract
url = "www.blogspot.co.za"
expub = tldextract.TLDExtract(include_psl_private_domains=False)
expub.update()
expub(url)
# result : ExtractResult(subdomain='www', domain='blogspot', suffix='co.za')

expri = tldextract.TLDExtract(include_psl_private_domains=True)
expri.update()
expri(url)
# result : ExtractResult(subdomain='', domain='www', suffix='blogspot.co.za')

url2 = "another.blogspot.co.uk"
expub(url2)
# result : ExtractResult(subdomain='another', domain='blogspot', suffix='co.uk')
expri(url2)
# result : ExtractResult(subdomain='', domain='another', suffix='blogspot.co.uk')

In following code, include_psl_private_domains=True doesn't work as intended

import tldextract
url = "www.blogspot.co.za"
expub = tldextract.TLDExtract(include_psl_private_domains=False)
expub.update()
expri = tldextract.TLDExtract(include_psl_private_domains=True)
expri.update()

expub(url)
# result : ExtractResult(subdomain='www', domain='blogspot', suffix='co.za')
expri(url) 
# result : ExtractResult(subdomain='www', domain='blogspot', suffix='co.za')

# Try another domain 
url2 = "another.blogspot.co.uk"
expub(url2)
# result : ExtractResult(subdomain='another', domain='blogspot', suffix='co.uk')
expri(url2) 
# result : ExtractResult(subdomain='another', domain='blogspot', suffix='co.uk')

[john-kurkowski]

I think this is covered by #66.

[floer32]

Yes, duplicate of #66.