diff options
| -rw-r--r-- | fuzzycat/utils.py | 4 | ||||
| -rw-r--r-- | fuzzycat/verify.py | 13 | ||||
| -rw-r--r-- | tests/test_verify.py | 5 |
3 files changed, 12 insertions, 10 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 - diff --git a/fuzzycat/verify.py b/fuzzycat/verify.py index d111871..d7b2d62 100644 --- a/fuzzycat/verify.py +++ b/fuzzycat/verify.py @@ -138,7 +138,8 @@ def compare(a, b): """ Compare two entities, return match status and reason. """ - if a.get("doi") and b.get("doi") and a.get("doi") == b.get("doi"): + if a.get("ext_ids", {}).get("doi") and b.get("ext_ids", {}).get("doi") and a.get( + "ext_ids", {}).get("doi") == b.get("ext_ids", {}).get("doi"): return (Status.EXACT, OK.DOI) if len(a.get("title", "")) < 5: return (Status.AMBIGUOUS, Miss.SHORT_TITLE) @@ -157,10 +158,11 @@ def compare(a, b): # TODO: figshare versions, "xxx.v1" FIGSHARE_PREFIX = "10.6084" - if a.get("doi") and b.get("doi") and a.get("doi").startswith(FIGSHARE_PREFIX + "/") and b.get( - "doi").startswith(FIGSHARE_PREFIX + "/"): - a_doi_v_stripped = re.sub(r"[.]v[0-9]+$", "", a.get("doi")) - b_doi_v_stripped = re.sub(r"[.]v[0-9]+$", "", a.get("doi")) + if a.get("ext_ids", {}).get("doi") and b.get("ext_ids", {}).get("doi") and a.get( + "ext_ids", {}).get("doi").startswith(FIGSHARE_PREFIX + "/") and b.get( + "ext_ids", {}).get("doi").startswith(FIGSHARE_PREFIX + "/"): + a_doi_v_stripped = re.sub(r"[.]v[0-9]+$", "", a.get("ext_ids", {}).get("doi", "")) + b_doi_v_stripped = re.sub(r"[.]v[0-9]+$", "", a.get("ext_ids", {}).get("doi", "")) if a_doi_v_stripped == b_doi_v_stripped: return (Status.STRONG, OK.FIGSHARE_VERSION) @@ -3542,4 +3544,3 @@ TITLE_BLACKLIST = set([ "週刊ダイヤモンド = diamond weekly 69(1)", "週刊ダイヤモンド = diamond weekly 別冊", ]) - diff --git a/tests/test_verify.py b/tests/test_verify.py index ebbb490..0c04d53 100644 --- a/tests/test_verify.py +++ b/tests/test_verify.py @@ -41,11 +41,12 @@ def test_compare(): pytest.fail("invalid test file, maybe missing a comma? {}".format(exc)) status, reason = compare(load_release_ident(a), load_release_ident(b)) if not expected_status or expected_status.lower() == "todo": - logger.warn( + logger.warning( "skipping test {base}/release/{a} {base}/release/{b} -- no result defined (we think {status}, {reason})" .format(a=a, b=b, base=FATCAT_BASE_URL, status=status, reason=reason)) assert status == status, "status: want {}, got {} for {} {}".format( expected_status, status, a, b) if expected_reason: - assert expected_reason.lower() == reason.lower(), "reason [{} {}]: want {}, got {}".format(a, b, expected_reason, reason) + assert expected_reason.lower() == reason.lower( + ), "reason [{} {}]: want {}, got {}".format(a, b, expected_reason, reason) logger.info("ran verification over {} cases (https://git.io/JkDgS)".format(i)) |