Hello!, and thank you for your PR.

Regarding unit tests:

• I have not written tests and need help to write them

I’ve pushed a branch with unit tests for your changes (diff can be seen here; I could create a PR if you’d like).

Some of the tests are commented out, since they would fail with the current implementation. Starting with the least complex, these are:

1. validate_isbn13 properly strips dashes from the value, but validate_isbn10 doesn’t.

2. validate_isbn10 doesn’t detect invalid checksum characters, e.g. 012345678Z (very minor; could be ignored).

3. ISBNs include a “check digit” that guards against typos (brief summary: the last digit is always ‘computed’ from the rest of the ISBN, so it can be used to detect whether a mistake was made when typing the number).

   It would be nice, I think, if form validation detected this as well, in addition to malformed values. Wikipedia describes check digits in ISBNs, and our isbn_10_to_13 and isbn_13_to_10 functions include code to compute it.

   • NOTE: fixing this item involves a fair amount of work, so please do not feel obligated to implement it. And the original bug report (An invalid ISBN causes the book page to crash (quite problematic) #3564) is unaffected by it, I think.

Hi,

Thanks for the test-examples and comments, those help a lot! Actually looking the models/book.py file, I could validate isbn10 by converting it to isbn13 and back and check result to match. For isbn13 similar should work if I just extend isbn13_to_10 also accept '979' prefix.

But please open PR for those tests and I'll take them in and work on those failing cases you spotted.

Realised that I can make the PR with your link directly, so took it in, thanks!