def user_stats_data(uid, start=None, end=None):
"""
:param uid: int or something cast-able to int
:param start: datetime
:param end: datetime
:return:
"""
from sefaria.model.category import TOP_CATEGORIES
from sefaria.model.trend import Trend, TrendSet, read_in_category_key, reverse_read_in_category_key
from sefaria.sheets import user_sheets, sheet_list
uid = int(uid)
# todo: needs more thought. UserHistory.timeclause handles the Nones, but later usages in this method aren't so graceful.
timeclause = UserHistory.timeclause(start, end)
end = end or datetime.now()
start = start or datetime(2017, 12, 1) # start of Sefaria epoch
# All of user's sheets
usheets = user_sheets(uid)["sheets"]
usheet_ids = [s["id"] for s in usheets]
# Sheet views in this period
match_clause = {
"is_sheet": True,
"sheet_id": {"$in": usheet_ids},
}
match_clause.update(timeclause)
usheet_views = db.user_history.aggregate([
{"$match": match_clause},
{"$group": {
"_id": "$sheet_id",
"cnt": {"$sum": 1}}},
])
most_popular_sheet_ids = [s["_id"] for s in sorted(usheet_views, key=lambda o: o["cnt"], reverse=True)[:3]]
most_popular_sheets = [s for s in usheets if s["id"] in most_popular_sheet_ids]
sheets_this_period = [s for s in usheets if start <= datetime.strptime(s["created"], "%Y-%m-%dT%H:%M:%S.%f") <= end]
# Refs I viewed
match_clause = {
"uid": uid,
"secondary": False,
"is_sheet": False
}
match_clause.update(timeclause)
refs_viewed = db.user_history.aggregate([
{"$match": match_clause},
{"$group": {
"_id": "$ref",
"cnt": {"$sum": 1}}},
])
most_viewed_refs = [s["_id"] for s in sorted(refs_viewed, key=lambda o: o["cnt"], reverse=True) if s["cnt"] > 1 and "Genesis 1" not in s["_id"]][:10]
# Sheets I viewed
match_clause = {
"uid": uid,
"secondary": False,
"is_sheet": True
}
match_clause.update(timeclause)
sheets_viewed = db.user_history.aggregate([
{"$match": match_clause},
{"$group": {
"_id": "$sheet_id",
"cnt": {"$sum": 1}}},
])
most_viewed_sheets_ids = [s["_id"] for s in sorted(sheets_viewed, key=lambda o: o["cnt"], reverse=True) if s["cnt"] > 1 and s["_id"] not in usheet_ids][:10]
most_viewed_sheets = sheet_list({"id": {"$in":most_viewed_sheets_ids}})
# Construct returned data
d = public_user_data(uid)
sheetsReadQuery = {"is_sheet": True, "secondary": False, "uid": uid}
sheetsReadQuery.update(timeclause)
d["sheetsRead"] = UserHistorySet(sheetsReadQuery).count()
textsReadQuery = {"is_sheet": False, "secondary": False, "uid": uid}
textsReadQuery.update(timeclause)
d["textsRead"] = UserHistorySet(textsReadQuery).count()
d["categoriesRead"] = {reverse_read_in_category_key(t.name): t.value for t in TrendSet({"uid":uid, "name": {"$in": map(read_in_category_key, TOP_CATEGORIES)}})}
d["totalSheets"] = len(usheets)
d["publicSheets"] = len([s for s in usheets if s["status"] == "public"])
d["popularSheets"] = most_popular_sheets
d["sheetsThisPeriod"] = len(sheets_this_period)
d["mostViewedRefs"] = most_viewed_refs
d["mostViewedSheets"] = most_viewed_sheets
return d
def user_stats_data(uid):
"""
:param uid: int or something cast-able to int
:param start: datetime
:param end: datetime
:return:
"""
from sefaria.model.category import TOP_CATEGORIES
from sefaria.model.story import Story
from sefaria.sheets import user_sheets
uid = int(uid)
user_stats_dict = user_profile.public_user_data(uid)
# All of user's sheets
usheets = user_sheets(uid)["sheets"]
usheet_ids = [s["id"] for s in usheets]
for daterange in active_dateranges:
# Sheet views in this period
usheet_views = db.user_history.aggregate([
{"$match": daterange.update_match({
"is_sheet": True,
"sheet_id": {"$in": usheet_ids},
"uid": {"$ne": uid}
})},
{"$group": {
"_id": "$sheet_id",
"cnt": {"$sum": 1}}},
])
most_popular_sheet_ids = [s["_id"] for s in sorted(usheet_views, key=lambda o: o["cnt"], reverse=True)[:3]]
most_popular_sheets = []
for sheet_id in most_popular_sheet_ids:
most_popular_sheets += [s for s in usheets if s["id"] == sheet_id]
sheets_this_period = [s for s in usheets if daterange.contains(datetime.strptime(s["created"], "%Y-%m-%dT%H:%M:%S.%f"))]
# Refs I viewed
refs_viewed = db.user_history.aggregate([
{"$match": daterange.update_match({
"uid": uid,
"secondary": False,
"is_sheet": False
})},
{"$group": {
"_id": "$ref",
"cnt": {"$sum": 1}}}, # Using $num_times_read isn't reliable. It counts book views, but not text views.
])
most_viewed_trefs = [s["_id"] for s in sorted(refs_viewed, key=lambda o: o["cnt"], reverse=True) if s["cnt"] > 1 and "Genesis 1" not in s["_id"]][:9]
most_viewed_refs = [text.Ref(r) for r in most_viewed_trefs]
most_viewed_ref_dicts = [{"en": r.normal(), "he": r.he_normal(), "book": r.index.title} for r in most_viewed_refs]
# Sheets I viewed
sheets_viewed = db.user_history.aggregate([
{"$match": daterange.update_match({
"uid": uid,
"secondary": False,
"is_sheet": True
})},
{"$group": {
"_id": "$sheet_id",
"cnt": {"$sum": 1}}},
])
most_viewed_sheets_ids = [s["_id"] for s in sorted(sheets_viewed, key=lambda o: o["cnt"], reverse=True) if s["cnt"] > 1 and s["_id"] not in usheet_ids][:3]
most_viewed_sheets = [Story.sheet_metadata(i, return_id=True) for i in most_viewed_sheets_ids]
most_viewed_sheets = [a for a in most_viewed_sheets if a]
for sheet_dict in most_viewed_sheets:
sheet_dict.update(Story.publisher_metadata(sheet_dict["publisher_id"]))
# Construct returned data
user_stats_dict[daterange.key] = {
"sheetsRead": user_profile.UserHistorySet(daterange.update_match({"is_sheet": True, "secondary": False, "uid": uid})).hits(),
"textsRead": user_profile.UserHistorySet(daterange.update_match({"is_sheet": False, "secondary": False, "uid": uid})).hits(),
"categoriesRead": {reverse_read_in_category_key(t.name): t.value for t in TrendSet({"uid":uid, "period": daterange.key, "name": {"$in": list(map(read_in_category_key, TOP_CATEGORIES))}})},
"totalSheets": len(usheets),
"publicSheets": len([s for s in usheets if s["status"] == "public"]),
"popularSheets": most_popular_sheets,
"sheetsThisPeriod": len(sheets_this_period),
"mostViewedRefs": most_viewed_ref_dicts,
"mostViewedSheets": most_viewed_sheets
}
return user_stats_dict