def dep_counts(name):
commentators = model.IndexSet({"categories.0": "Commentary"}).distinct("title")
ref_patterns = {
'alone': r'^{} \d'.format(re.escape(name)),
'commentor': r'{} on'.format(re.escape(name)),
'commentee': r'^({}) on {} \d'.format("|".join(commentators), re.escape(name))
}
commentee_title_pattern = r'^({}) on {} \d'.format("|".join(commentators), re.escape(name))
ret = {
'version title exact match': model.VersionSet({"title": name}).count(),
'version title match commentor': model.VersionSet({"title": {"$regex": ref_patterns["commentor"]}}).count(),
'version title match commentee': model.VersionSet({"title": {"$regex": commentee_title_pattern}}).count(),
'history title exact match': model.HistorySet({"title": name}).count(),
'history title match commentor': model.HistorySet({"title": {"$regex": ref_patterns["commentor"]}}).count(),
'history title match commentee': model.HistorySet({"title": {"$regex": commentee_title_pattern}}).count(),
}
for pname, pattern in ref_patterns.items():
ret.update({
'note match ' + pname: model.NoteSet({"ref": {"$regex": pattern}}).count(),
'link match ' + pname: model.LinkSet({"refs": {"$regex": pattern}}).count(),
'history refs match ' + pname: model.HistorySet({"ref": {"$regex": pattern}}).count(),
'history new refs match ' + pname: model.HistorySet({"new.refs": {"$regex": pattern}}).count()
})
return ret
def dep_counts(name, indx):
def construct_query(attribute, queries):
query_list = [{attribute: {'$regex': query}} for query in queries]
return {'$or': query_list}
from sefaria.model.text import prepare_index_regex_for_dependency_process
patterns = prepare_index_regex_for_dependency_process(indx, as_list=True)
patterns = [
pattern.replace(re.escape(indx.title), re.escape(name))
for pattern in patterns
]
ret = {
'version title exact match':
model.VersionSet({
"title": name
}, sort=[('title', 1)]).count(),
'history title exact match':
model.HistorySet({
"title": name
}, sort=[('title', 1)]).count(),
'note match ':
model.NoteSet(construct_query("ref", patterns),
sort=[('ref', 1)]).count(),
'link match ':
model.LinkSet(construct_query("refs", patterns)).count(),
'history refs match ':
model.HistorySet(construct_query("ref", patterns),
sort=[('ref', 1)]).count(),
'history new refs match ':
model.HistorySet(construct_query("new.refs", patterns),
sort=[('new.refs', 1)]).count()
}
return ret
from sefaria.clean import remove_old_counts
from sefaria.counts import update_counts
# Remove duplicate 'Sefer Abudraham'
db.index.remove({"title": "Sefer Abudraham "})
db.index.remove({"title": "Tiferet Yisrael "})
db.index.remove({"title": "Igrot Moshe "})
db.index.remove({"title": "The Sabbath, Heschel "})
db.index.remove({"title": "Sifre Devarim "})
remove_old_counts()
texts = model.IndexSet({})
for t in texts:
if t.title != t.title.strip():
t.title = t.title.strip()
t.save()
ns = model.NoteSet({"public": {"$exists": False}})
for n in ns:
if not getattr(n, "owner", None):
n.owner = 1
ns.update({"public": False})
# Remove "sectionCounts" field form sectionCounts
db.counts.update({}, {"$unset": {"sectionCounts": ""}}, multi=True)
# Rebuild counts docs, so they get the allVersionCounts field
update_counts()
