diff options
| author | Bryan Newbold <bnewbold@robocracy.org> | 2019-09-03 14:13:37 -0700 |
|---|---|---|
| committer | Bryan Newbold <bnewbold@robocracy.org> | 2019-09-03 14:13:37 -0700 |
| commit | 7690ee86e58d258b8f14ad33927037d7f06e2c67 (patch) | |
| tree | 2e3e851bb53ce4844091dfabf992fff9c71d1823 /python/fatcat_tools/importers | |
| parent | a24ca520d1eafae302762a7b0f6eea0b01ef186a (diff) | |
| download | fatcat-7690ee86e58d258b8f14ad33927037d7f06e2c67.tar.gz fatcat-7690ee86e58d258b8f14ad33927037d7f06e2c67.zip | |
more careful chocula import counts; don't re-update empty URLs
Diffstat (limited to 'python/fatcat_tools/importers')
| -rw-r--r-- | python/fatcat_tools/importers/chocula.py | 8 |
1 files changed, 6 insertions, 2 deletions
diff --git a/python/fatcat_tools/importers/chocula.py b/python/fatcat_tools/importers/chocula.py index 89b0736d..25edccee 100644 --- a/python/fatcat_tools/importers/chocula.py +++ b/python/fatcat_tools/importers/chocula.py @@ -80,9 +80,11 @@ class ChoculaImporter(EntityImporter): except fatcat_client.rest.ApiException as err: if err.status != 404: raise err + self.counts['exists'] += 1 self.counts['exists-not-found'] += 1 return False if existing.state != 'active': + self.counts['exists'] += 1 self.counts['exists-inactive'] += 1 return False @@ -94,6 +96,7 @@ class ChoculaImporter(EntityImporter): if err.status != 404: raise err if existing: + self.counts['exists'] += 1 self.counts['exists-by-issnl'] += 1 return False # doesn't exist, always create @@ -103,7 +106,7 @@ class ChoculaImporter(EntityImporter): do_update = False if not existing.extra: existing.extra = dict() - if set(ce.extra.get('urls', [])) != set(existing.extra.get('urls', [])): + if ce.extra.get('urls') and set(ce.extra.get('urls', [])) != set(existing.extra.get('urls', [])): do_update = True if ce.publisher and not existing.publisher: do_update = True @@ -128,7 +131,8 @@ class ChoculaImporter(EntityImporter): self.counts['update'] += 1 return False else: - self.counts['skip-update'] += 1 + self.counts['exists'] += 1 + self.counts['exists-skip-update'] += 1 return False # if we got this far, it's a bug |