Open stefanomasini opened 3 months ago
Hello!
Thanks for the PR.
Was testing it and it seems the Mexican one does not work so I kept the @dwelle's one. For turkey, I see there are 10 digits and TIN also has 11 digits.
The Chile changed regex is too forgiving: \b[0-9]{2}\.[0-9]{3}\.[0-9]{3}-[0-9kK]\b
this one should be it per: https://www.oecd.org/content/dam/oecd/en/topics/policy-issue-focus/aeoi/chile-tin.pdf
Also would be great to update examples so we can test it against it and let users know how it looks like
Thanks for your feedback.
I had a few thousand records to process with this library and the changes I submitted reflect what I had to do in order for it to work. I did a quick investigation on the ones that broke in my dataset and noticed that the regexps where not matching what was described in the links I posted.
This way I was able to process and manually fix a few records to make them match the updated regexps. Ultimately, the whole dataset went through Stripe with no issues, apart from a few obvious outliers, so I'm really happy about the outcome and I thank you very much for providing the library.
That said, I don't presume my data set to be comprehensive enough to represent a confirmation that my fix is 100% correct. I only used the updated regexp to reduce the number of those I would have needed to process by hand, and trusted Stripe to ultimately reject my work if the ids wouldn't comply. That didn't happen but, again, I only had a few thousand records.
Also, this was just a one shot job for me, so I'm offering my fixes as is. Unfortunately I don't have much time to work on this further.
See what you can do with this PR. And thanks again! Cheers Stefano