diff options
| author | Martin Czygan <martin.czygan@gmail.com> | 2020-11-26 11:20:35 +0100 |
|---|---|---|
| committer | Martin Czygan <martin.czygan@gmail.com> | 2020-11-26 11:20:35 +0100 |
| commit | 47761357e341b7355c98778ec1665ae73e5d6fe3 (patch) | |
| tree | 14c0af2f9a6cc52bf8fd784d1ad8ab477c9eaef3 /fuzzycat/utils.py | |
| parent | 67589a0bc9217f3259f0f093b5283b9e92828d0f (diff) | |
| download | fuzzycat-47761357e341b7355c98778ec1665ae73e5d6fe3.tar.gz fuzzycat-47761357e341b7355c98778ec1665ae73e5d6fe3.zip | |
figshare fix
Diffstat (limited to 'fuzzycat/utils.py')
| -rw-r--r-- | fuzzycat/utils.py | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/fuzzycat/utils.py b/fuzzycat/utils.py index 1cac668..4d1325d 100644 --- a/fuzzycat/utils.py +++ b/fuzzycat/utils.py @@ -1,13 +1,14 @@ import io import itertools -import string import re +import string printable_no_punct = string.digits + string.ascii_letters + string.whitespace # More correct: https://www.johndcook.com/blog/2016/02/04/regular-expression-to-match-a-chemical-element/ CHEM_FORMULA = re.compile(r"([A-Z]{1,2}[0-9]{1,2})+") + def slugify_string(s: str) -> str: """ Keeps ascii chars and single whitespace only. @@ -89,4 +90,3 @@ def contains_chemical_formula(s): for token in s.split(): if CHEM_FORMULA.search(token): return True - |