class TradegoodAction:
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def j_yaml(cls):
filepath = os.path.join(FILE_DIR, "action.yaml")
return YAMLTool.filepath2j(filepath)
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def p_command(cls, ):
return KhalaAction.j_yaml2p_command(cls.j_yaml())
@classmethod
def text_body2match(cls, text_body):
return KhalaAction.pattern_text2match(cls.p_command(), text_body)
@classmethod
def respond(cls, packet):
from henrique.main.entity.tradegood.subaction.tradegood_subactions import TradegoodTradegoodSubaction
text = KhalaPacket.packet2text(packet)
tradegood_entity_list = TradegoodEntity.text2entity_list(text)
str_list = lmap(
lambda p: TradegoodTradegoodSubaction.tradegood_entity2response(
p, packet), tradegood_entity_list)
str_out = "\n\n".join(str_list)
return KhalaResponse.Builder.str2j_response(str_out)
class TradegoodGooglesheets:
@classmethod
def spreadsheetId(cls):
return "1XgTitp7h-oeAIzaxlLkQx1KX4c2uk4-izwn6W5ke290"
@classmethod
@cached(cache=TTLCache(maxsize=2, ttl=60 * 10))
def dict_sheetname2data_ll(cls, ):
sheetname_list = [NameskoSheet.NAME, NamesenSheet.NAME, TradegoodtypeSheet.NAME]
return GooglesheetsTool.sheet_ranges2dict_range2data_ll(HenriqueGoogleapi.credentials(),
cls.spreadsheetId(),
sheetname_list,
)
@classmethod
def sheetname2data_ll(cls, sheetname):
return cls.dict_sheetname2data_ll()[sheetname]
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@cached(cache=TTLCache(maxsize=2, ttl=60 * 10))
def dict_codename2tradegood(cls):
tradegood_list_all = cls.tradegood_list_all()
h = merge_dicts([{Tradegood.tradegood2codename(tradegood): tradegood} for tradegood in tradegood_list_all],
vwrite=DictTool.VWrite.f_vwrite2f_hvwrite(vwrite_no_duplicate_key),
)
return h
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@cached(cache=TTLCache(maxsize=2, ttl=60 * 10))
def tradegood_list_all(cls):
h_codename2aliases_en = NamesenSheet.dict_codename2aliases()
h_codename2aliases_ko = NameskoSheet.dict_codename2aliases()
h_codename2tradegoodtype = TradegoodtypeSheet.dict_codename2tradegoodtype()
# raise Exception({"h_codename2product_list":h_codename2product_list})
codename_list = luniq(chain(h_codename2aliases_en.keys(),
h_codename2aliases_ko.keys(),
)
)
def codename2port(codename):
aliases = DictTool.filter(lambda k, v: v,
{"en": h_codename2aliases_en.get(codename),
"ko": h_codename2aliases_ko.get(codename),
})
tradegoodtype = h_codename2tradegoodtype.get(codename)
port = DictTool.filter(lambda k, v: bool(v),
{Tradegood.Field.CODENAME: codename,
Tradegood.Field.TRADEGOODTYPE: tradegoodtype,
Tradegood.Field.ALIASES: aliases,
}
)
return DictTool.filter(lambda k, v: v, port)
return lmap(codename2port, codename_list)
class Product:
class Field:
PORT = "port"
TRADEGOOD = "tradegood"
PRICE = "price"
@classmethod
def product2port(cls, product):
return product.get(cls.Field.PORT)
@classmethod
def product2tradegood(cls, product):
return product.get(cls.Field.TRADEGOOD)
@classmethod
def product2tradegoodtype(cls, product):
from henrique.main.document.tradegood.tradegood import Tradegood
tradegood = Tradegood.codename2tradegood(cls.product2tradegood(product))
return Tradegood.tradegood2tradegoodtype(tradegood)
@classmethod
def product2price(cls, product):
return product.get(cls.Field.PRICE)
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def list_all(cls):
return lchain(*map(Port.port2products, Port.list_all()))
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def _dict_port2products(cls):
return dict_groupby_tree(cls.list_all(), [Product.product2port])
@classmethod
def port2products(cls, port_codename):
return cls._dict_port2products().get(port_codename) or []
# @classmethod
# @FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
# def dict_tradegood2products(cls):
# return dict_groupby_tree(cls.list_all(), [Product.product2tradegood])
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def _dict_tradegoodtype2products(cls):
return dict_groupby_tree(cls.list_all(), [Product.product2tradegoodtype])
@classmethod
def tradegoodtype2products(cls, tradegoodtype_codename):
return cls._dict_tradegoodtype2products().get(tradegoodtype_codename) or []
class SkillGooglesheets:
@classmethod
def spreadsheetId(cls):
return "18D67KgdOwq1RbDP5mgIS8FzFAAOkQPCHbD8zcFRyoyQ"
@classmethod
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def dict_sheetname2data_ll(cls, ):
sheetname_list = [AliasesEn.NAME, AliasesKo.NAME]
return GooglesheetsTool.sheet_ranges2dict_range2data_ll(
HenriqueGoogleapi.credentials(),
cls.spreadsheetId(),
sheetname_list,
)
@classmethod
def sheetname2data_ll(cls, sheetname):
return cls.dict_sheetname2data_ll()[sheetname]
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def dict_lang2codename2aliases(cls):
return {
"en": AliasesEn.dict_codename2aliases(),
"ko": AliasesKo.dict_codename2aliases(),
}
# WARMER.warmup()
class ChatroomCollection:
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def collection(cls, *_, **__):
db = HenriqueMongodb.db()
return db.get_collection("chatroom", *_, **__)
class ChannelUserDocCache:
class Constant:
MAXSIZE = 200
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
def warmer(cls, ):
collection = ChannelUserCollection.collection()
docs = map(MongoDBTool.bson2json, collection.find())
ChannelUserDoc._docs2cache(docs)
class NanbanSkillError:
class Codename:
NO_PREV_NANBAN_ERROR = "NO_PREV_NANBAN_ERROR"
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def yaml(cls):
filepath = os.path.join(FILE_DIR, "error.yaml")
j = YAMLTool.filepath2j(filepath)
return j
@classmethod
def codename_lang2text(cls, codename, lang):
return JsonTool.down(cls.yaml(), [codename, lang])
class Sign:
class Constant:
PLUS = "+"
MINUS = "-"
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def pattern(cls):
return re.compile(RegexTool.rstr2wordbounded("[+-]"))
@classmethod
def sign2int(cls, sign):
h = {cls.Constant.PLUS: 1, cls.Constant.MINUS: -1}
return h[sign]
class PortReference:
class YAML:
NAME = "name"
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def j_yaml(cls):
filepath = os.path.join(FILE_DIR, "port_reference.yaml")
j = YAMLTool.filepath2j(filepath)
return j
@classmethod
def lang2name(cls, lang):
j_yaml = cls.j_yaml()
return jdown(j_yaml, [cls.YAML.NAME, lang])
class ServerGooglesheets:
@classmethod
def spreadsheetId(cls):
return "1z_8oCBFUj5ArGr8WvQFio3-uoQgbAOC87BupNNAF0_M"
@classmethod
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def dict_sheetname2data_ll(cls, ):
sheetname_list = [NameskoSheet.NAME, NamesenSheet.NAME]
return GooglesheetsTool.sheet_ranges2dict_range2data_ll(
HenriqueGoogleapi.credentials(),
cls.spreadsheetId(),
sheetname_list,
)
@classmethod
def sheetname2data_ll(cls, sheetname):
return cls.dict_sheetname2data_ll()[sheetname]
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def server_list_all(cls):
h_codename2aliases_en = NamesenSheet.dict_codename2aliases()
h_codename2aliases_ko = NameskoSheet.dict_codename2aliases()
codename_list = luniq(
chain(
h_codename2aliases_en.keys(),
h_codename2aliases_ko.keys(),
))
def codename2server(codename):
aliases = DictTool.filter(
lambda k, v: v, {
"en": h_codename2aliases_en.get(codename),
"ko": h_codename2aliases_ko.get(codename),
})
culture = {
Server.Field.CODENAME: codename,
Server.Field.ALIASES: aliases,
}
return DictTool.filter(lambda k, v: v, culture)
return lmap(codename2server, codename_list)
class HenriqueCommand:
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def pattern_prefix(cls):
return re.compile(r"^\s*\?", re.I)
@classmethod
def packet2skill_code(cls, packet):
text_in = KhalaPacket.packet2text(packet)
config = HenriqueEntity.Config.packet2config(packet)
return cls._text_config2skill_code(text_in, config)
@classmethod
@CacheTool.cache2hashable(cache=lru_cache(maxsize=HenriqueEntity.Cache.DEFAULT_SIZE),
f_pair=CacheTool.JSON.func_pair(), )
def _text_config2skill_code(cls, text_in, config):
pattern_prefix = cls.pattern_prefix()
match_list_prefix = list(pattern_prefix.finditer(text_in))
if not match_list_prefix:
return None
entity_list = SkillEntity.text2entity_list(text_in, config=config)
if not entity_list:
return None
spans_list = [lmap(lambda m: m.span(), match_list_prefix),
lmap(FoxylibEntity.entity2span, entity_list)
]
gap2is_valid = partial(StringTool.str_span2match_blank_or_nullstr, text_in)
indextuple_list = ContextfreeTool.spans_list2reducible_indextuple_list(spans_list, gap2is_valid)
assert_in(len(indextuple_list), [0, 1])
if not indextuple_list:
return None
index_entity = l_singleton2obj(indextuple_list)[1]
entity = entity_list[index_entity]
return SkillEntity.entity2skill_codename(entity)
@classmethod
def text2is_query(cls, text):
return bool(cls.pattern_prefix().match(text))
class NanbanTimedeltaSuffix:
class Key:
SUFFIX = "suffix"
I_DONT_KNOW = "i_dont_know"
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def yaml(cls):
filepath = os.path.join(FILE_DIR, "suffix.yaml")
j_yaml = YAMLTool.filepath2j(filepath, yaml.SafeLoader)
return j_yaml
@classmethod
@VersionTool.inactive
def str_timedelta2relativetimedelta(cls, str_timedelta, lang):
suffix_format = JsonTool.down(cls.yaml(), [cls.Key.SUFFIX, lang])
return suffix_format.format(str_timedelta)
@classmethod
def lang2str_idk(cls, lang):
return JsonTool.down(cls.yaml(), [cls.Key.I_DONT_KNOW, lang])
class ChatroomKakaotalk:
class Constant:
LOCALE = "ko-KR"
NAME = "uwo"
@classmethod
def codename(cls):
return Chatroom.Constant.DELIM.join([Channel.Codename.KAKAOTALK_UWO, cls.Constant.NAME])
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def chatroom(cls, ):
logger = KhalaLogger.func_level2logger(cls.chatroom, logging.DEBUG)
chatroom = {Chatroom.Field.CHANNEL: Channel.Codename.KAKAOTALK_UWO,
Chatroom.Field.CODENAME: cls.codename(),
# Chatroom.Field.EXTRA: ChatroomDiscord.Extra.message2extra(message),
Chatroom.Field.LOCALE: cls.Constant.LOCALE,
}
logger.debug({"chatroom": chatroom,
})
return chatroom
class PortGooglesheets:
@classmethod
def spreadsheetId(cls):
return "1DxaBuSsOvAf4nsy4n2XNwcmPVqBLRvWgCbs5Y8AHFtE"
@classmethod
def _dict_sheetname2data_ll(cls, ):
sheetname_list = [NameskoSheet.NAME, NamesenSheet.NAME,
CommentsKoSheet.NAME,
CultureSheet.NAME, ProductSheet.NAME,
]
return GooglesheetsTool.sheet_ranges2dict_range2data_ll(HenriqueGoogleapi.credentials(),
cls.spreadsheetId(),
sheetname_list,
)
@classmethod
@CacheManager.attach_cachedmethod(self2cache=lambda x: LRUCache(maxsize=2),)
def dict_sheetname2data_ll(cls,):
return cls._dict_sheetname2data_ll()
@classmethod
def sheetname2data_ll(cls, sheetname):
return cls.dict_sheetname2data_ll()[sheetname]
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@cached(cache=TTLCache(maxsize=2, ttl=60 * 10))
# @FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def dict_codename2port(cls):
port_list_all = cls.port_list_all()
h = merge_dicts([{Port.port2codename(port): port} for port in port_list_all],
vwrite=DictTool.VWrite.f_vwrite2f_hvwrite(vwrite_no_duplicate_key),
)
return h
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@cached(cache=TTLCache(maxsize=2, ttl=60 * 10))
# @FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def port_list_all(cls):
h_codename2aliases_en = NamesenSheet.dict_codename2aliases()
h_codename2aliases_ko = NameskoSheet.dict_codename2aliases()
h_codename2culture = CultureSheet.dict_codename2culture()
h_codename2product_list = ProductSheet.dict_codename2products()
h_codename2comments_ko = CommentsKoSheet.dict_codename2comments()
# raise Exception({"h_codename2product_list":h_codename2product_list})
codename_list = luniq(chain(h_codename2aliases_en.keys(),
h_codename2aliases_ko.keys(),
)
)
def codename2port(codename):
aliases = DictTool.filter(lambda k, v: v,
{"en": h_codename2aliases_en.get(codename),
"ko": h_codename2aliases_ko.get(codename),
})
comments = DictTool.filter(lambda k, v: v,
{"ko": h_codename2comments_ko.get(codename),
})
port = {Port.Field.CODENAME: codename,
Port.Field.CULTURE: h_codename2culture[codename],
Port.Field.ALIASES: aliases,
Port.Field.PRODUCTS: h_codename2product_list.get(codename),
Port.Field.COMMENTS: comments,
}
return DictTool.filter(lambda k, v: v, port)
return lmap(codename2port, codename_list)
class TimedeltaEntityUnit:
class Value:
YEAR = "year"
MONTH = "month"
WEEK = "week"
DAY = "day"
HOUR = "hour"
MINUTE = "minute"
SECOND = "second"
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def yaml(cls):
filepath = os.path.join(FILE_DIR, "timedelta_unit.yaml")
j_yaml = YAMLTool.filepath2j(filepath, yaml.SafeLoader)
return j_yaml
@classmethod
def normalize(cls, text):
return str2lower(text)
@classmethod
def timedelta2unit(cls, td):
h = {
timedelta(days=7): cls.Value.WEEK,
timedelta(days=1): cls.Value.DAY,
timedelta(seconds=3600): cls.Value.HOUR,
timedelta(seconds=60): cls.Value.MINUTE,
timedelta(seconds=1): cls.Value.SECOND,
}
return h[td]
@classmethod
def v_unit_lang2str(cls, v, unit, lang):
j_yaml = cls.yaml()
str_unit = JsonTool.down(j_yaml, [unit, lang])[0]
if lang in {"en"}:
return " ".join([str(v), str_unit])
if lang in {"ko"}:
return "".join([str(v), str_unit])
raise Exception("Invalid language: {}".format(lang))
@classmethod
def unit2plural(cls, unit):
return "{}s".format(unit)
@classmethod
def gazetteer_all(cls, ):
gazetteer = {k: lchain(*j.values()) for k, j in cls.yaml().items()}
return gazetteer
@classmethod
def langs2gazetteer(cls, langs):
gazetteer = {
k: lchain(*[j.get(lang, []) for lang in langs])
for k, j in cls.yaml().items()
}
return gazetteer
@classmethod
def langs2matcher(cls, langs):
return cls._langs2matcher(frozenset(langs))
@classmethod
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=4))
def _langs2matcher(cls, langs):
logger = HenriqueLogger.func_level2logger(cls._langs2matcher,
logging.DEBUG)
gazetteer = cls.langs2gazetteer(langs)
def texts2pattern(texts):
rstr_raw = RegexTool.rstr_iter2or(map(re.escape, texts))
left_bounds = lchain(
RegexTool.bounds2suffixed(RegexTool.left_wordbounds(), "\d"),
RegexTool.left_wordbounds(),
)
right_bounds = RegexTool.right_wordbounds()
rstr = RegexTool.rstr2bounded(rstr_raw, left_bounds, right_bounds)
logger.debug({
"rstr": rstr,
"rstr_raw": rstr_raw,
})
return re.compile(rstr, re.I)
config = {
GazetteerMatcher.Config.Key.TEXTS2PATTERN: texts2pattern,
GazetteerMatcher.Config.Key.NORMALIZER: cls.normalize,
}
matcher = GazetteerMatcher(gazetteer, config=config)
return matcher
@classmethod
def warmer(cls):
cls.langs2matcher({"ko", "en"})
cls.langs2matcher({"en"})
class ChatroomuserGooglesheets:
@classmethod
def shorturl(cls):
return "https://bit.ly/2TRYkvi"
@classmethod
def spreadsheetId(cls):
return "1HcW7Im6SWy2T8g6POFDTOQXzH1jeTBjPxJn7KZ3RHB4"
@classmethod
def _dict_sheetname2data_ll(cls, ):
sheetname_list = [
AliasesSheet.NAME,
CommentsSheet.NAME,
]
return GooglesheetsTool.sheet_ranges2dict_range2data_ll(
HenriqueGoogleapi.credentials(),
cls.spreadsheetId(),
sheetname_list,
)
@classmethod
@CacheManager.attach_cachedmethod(
self2cache=lambda x: LRUCache(maxsize=2), )
def dict_sheetname2data_ll(cls, ):
return cls._dict_sheetname2data_ll()
@classmethod
def sheetname2data_ll(cls, sheetname):
return cls.dict_sheetname2data_ll()[sheetname]
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def dict_codename2chatroomuser(cls):
chatroomuser_list_all = cls.chatroomuser_list_all()
h = merge_dicts(
[{
Chatroomuser.chatroomuser2codename(chatroomuser): chatroomuser
} for chatroomuser in chatroomuser_list_all],
vwrite=DictTool.VWrite.f_vwrite2f_hvwrite(vwrite_no_duplicate_key),
)
return h
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def chatroomuser_list_all(cls):
h_codename2aliases = AliasesSheet.dict_codename2aliases()
h_codename2comments = CommentsSheet.dict_codename2comments()
# raise Exception({"h_codename2product_list":h_codename2product_list})
codename_list = list(h_codename2aliases.keys())
def codename2chatroomuser(codename):
aliases = h_codename2aliases.get(codename) or []
comments = h_codename2comments.get(codename)
chatroomuser = {
Chatroomuser.Field.CODENAME: codename,
Chatroomuser.Field.COMMENTS: comments,
Chatroomuser.Field.ALIASES: aliases,
}
return DictTool.filter(lambda k, v: v, chatroomuser)
return lmap(codename2chatroomuser, codename_list)
class TradegoodEntitySpecialcase:
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def pattern_ko(cls):
return re.compile(RegexTool.rstr2wordbounded(r"육메(?:크|클)?"))
@classmethod
def text2entity_list(cls, text_in, config=None):
locale = HenriqueEntity.Config.config2locale(
config) or HenriqueLocale.DEFAULT
lang = LocaleTool.locale2lang(locale)
langs_recognizable = HenriqueLocale.lang2langs_recognizable(lang)
if "ko" not in langs_recognizable:
return []
match_list = list(cls.pattern_ko().finditer(text_in))
def match2entity_list(match):
span = match.span()
assert_in(SpanTool.span2len(span), (2, 3))
entity_list = []
s, e = span
span_nutmeg = (s, s + 1)
entity_nutmeg = {
FoxylibEntity.Field.SPAN:
span_nutmeg,
FoxylibEntity.Field.TEXT:
StringTool.str_span2substr(text_in, span_nutmeg),
FoxylibEntity.Field.VALUE:
"Nutmeg",
FoxylibEntity.Field.TYPE:
TradegoodEntity.entity_type(),
}
entity_list.append(entity_nutmeg)
span_mace = (s + 1, s + 2)
entity_mace = {
FoxylibEntity.Field.SPAN:
span_mace,
FoxylibEntity.Field.TEXT:
StringTool.str_span2substr(text_in, span_mace),
FoxylibEntity.Field.VALUE:
"Mace",
FoxylibEntity.Field.TYPE:
TradegoodEntity.entity_type(),
}
entity_list.append(entity_mace)
if SpanTool.span2len(span) == 3:
span_clove = (s + 2, s + 3)
entity_cloves = {
FoxylibEntity.Field.SPAN:
span_clove,
FoxylibEntity.Field.TEXT:
StringTool.str_span2substr(text_in, span_clove),
FoxylibEntity.Field.VALUE:
"Cloves",
FoxylibEntity.Field.TYPE:
TradegoodEntity.entity_type(),
}
entity_list.append(entity_cloves)
return entity_list
entity_list = [
entity for m in match_list for entity in match2entity_list(m)
]
return entity_list
class Port:
class Field:
CODENAME = "codename"
CULTURE = "culture"
ALIASES = "aliases"
PRODUCTS = "products"
COMMENTS = "comments"
# @classmethod
# @FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
# def _dict_codename2port_all_OLD(cls):
# from henrique.main.document.port.mongodb.port_doc import PortDoc
# h_mongo = PortDoc.dict_codename2port()
#
# from henrique.main.document.port.googlesheets.port_googlesheets import PortGooglesheets
# h_googlesheets = PortGooglesheets.dict_codename2port()
#
# codename_list = luniq(chain(h_googlesheets.keys(), h_mongo.keys(),))
#
# def codename2port(codename):
# port = merge_dicts([h_mongo.get(codename) or {},
# h_googlesheets.get(codename) or {},
# ], vwrite=vwrite_update_if_identical,
# )
# return port
#
# dict_codename2port = merge_dicts([{codename: codename2port(codename)}
# for codename in codename_list],
# vwrite=vwrite_no_duplicate_key, )
# return dict_codename2port
@classmethod
def _dict_codename2port_all(cls):
from henrique.main.document.port.googlesheets.port_googlesheets import PortGooglesheets
return PortGooglesheets.dict_codename2port()
@classmethod
def list_all(cls):
return list(cls._dict_codename2port_all().values())
@classmethod
def port2codename(cls, port):
return port[cls.Field.CODENAME]
@classmethod
def port2culture(cls, port):
return port[cls.Field.CULTURE]
@classmethod
def port2products(cls, port):
return port.get(cls.Field.PRODUCTS) or []
@classmethod
def port_lang2aliases(cls, port, lang):
return JsonTool.down(port, [cls.Field.ALIASES, lang])
@classmethod
def port_lang2comments(cls, port, lang):
return JsonTool.down(port, [cls.Field.COMMENTS, lang])
@classmethod
def port_langs2aliases(cls, port, langs):
return luniq(chain(*[cls.port_lang2aliases(port, lang) for lang in langs]))
@classmethod
def port_lang2name(cls, port, lang):
return IterTool.first(cls.port_lang2aliases(port, lang))
@classmethod
def codename2port(cls, codename):
return cls._dict_codename2port_all().get(codename)
# @classmethod
# def port_tradegood2is_sold(cls, port, tg_codename):
# products = cls.port2products(port)
#
# for product in products:
# tg_codename_product = Product.product2tradegood(product)
# if tg_codename == tg_codename_product:
# return True
# return False
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def _dict_tradegood2ports(cls,):
def h_tradegood2ports_iter():
port_list = cls.list_all()
# raise Exception({"port_list":port_list})
for port in port_list:
for product in cls.port2products(port):
yield {Product.product2tradegood(product): [port]}
h_tg2ports_list = list(h_tradegood2ports_iter())
h_tg2ports = merge_dicts(h_tg2ports_list, vwrite=DictTool.VWrite.extend)
return h_tg2ports
@classmethod
def tradegood2ports(cls, tg_codename):
return cls._dict_tradegood2ports().get(tg_codename) or []
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def _dict_culture2ports(cls,):
h_culture2ports = merge_dicts([{cls.port2culture(port): [port]} for port in cls.list_all()],
vwrite=DictTool.VWrite.extend)
return h_culture2ports
# return GroupbyTool.dict_groupby_tree(culture_port_iter(), [ig(0)])
@classmethod
def culture2ports(cls, culture_codename):
return cls._dict_culture2ports().get(culture_codename) or []
class TradegoodtypeGooglesheets:
@classmethod
def spreadsheetId(cls):
return "1tCXSXrjzOdR8URx8SavC9feUrgSELuB87V5IvzdFsPE"
@classmethod
@cached(cache=TTLCache(maxsize=2, ttl=60 * 10))
# @FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def dict_sheetname2data_ll(cls, ):
sheetname_list = [
NameskoSheet.NAME, NamesenSheet.NAME, CategorySheet.NAME
]
return GooglesheetsTool.sheet_ranges2dict_range2data_ll(
HenriqueGoogleapi.credentials(),
cls.spreadsheetId(),
sheetname_list,
)
@classmethod
def sheetname2data_ll(cls, sheetname):
return cls.dict_sheetname2data_ll()[sheetname]
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@cached(cache=TTLCache(maxsize=2, ttl=60 * 10))
def dict_codename2tradegoodtype(cls):
tradegoodtype_list_all = cls.tradegoodtype_list_all()
h = merge_dicts(
[{
Tradegoodtype.tradegoodtype2codename(tgt): tgt
} for tgt in tradegoodtype_list_all],
vwrite=DictTool.VWrite.f_vwrite2f_hvwrite(vwrite_no_duplicate_key),
)
return h
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@cached(cache=TTLCache(maxsize=2, ttl=60 * 10))
def tradegoodtype_list_all(cls):
h_codename2aliases_en = NamesenSheet.dict_codename2aliases()
h_codename2aliases_ko = NameskoSheet.dict_codename2aliases()
h_codename2category = CategorySheet.dict_codename2tradegoodtype()
# raise Exception({"h_codename2product_list":h_codename2product_list})
codename_list = luniq(
chain(
h_codename2aliases_en.keys(),
h_codename2aliases_ko.keys(),
))
def codename2port(codename):
aliases = DictTool.filter(
lambda k, v: v, {
"en": h_codename2aliases_en.get(codename),
"ko": h_codename2aliases_ko.get(codename),
})
category = h_codename2category.get(codename)
port = DictTool.filter(
lambda k, v: bool(v), {
Tradegoodtype.Field.CODENAME: codename,
Tradegoodtype.Field.CATEGORY: category,
Tradegoodtype.Field.ALIASES: aliases,
})
return DictTool.filter(lambda k, v: v, port)
return lmap(codename2port, codename_list)
class CultureGooglesheets:
@classmethod
def spreadsheetId(cls):
return "1s_EBQGNu0DlPedOXQNcfmE_LDk4wRq5QgJ9TsdBCCDE"
@classmethod
@cached(cache=TTLCache(maxsize=2, ttl=60 * 10))
# @FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def dict_sheetname2data_ll(cls, ):
sheetname_list = [
NameskoSheet.NAME, NamesenSheet.NAME, PrefersSheet.NAME
]
return GooglesheetsTool.sheet_ranges2dict_range2data_ll(
HenriqueGoogleapi.credentials(),
cls.spreadsheetId(),
sheetname_list,
)
@classmethod
def sheetname2data_ll(cls, sheetname):
return cls.dict_sheetname2data_ll()[sheetname]
@classmethod
def culture_list_all(cls):
logger = HenriqueLogger.func_level2logger(cls.culture_list_all,
logging.DEBUG)
h_codename2aliases_en = NamesenSheet.dict_codename2aliases()
h_codename2aliases_ko = NameskoSheet.dict_codename2aliases()
h_codename2prefers = PrefersSheet.dict_codename2prefers()
codename_list = luniq(
chain(
h_codename2aliases_en.keys(),
h_codename2aliases_ko.keys(),
h_codename2prefers.keys(),
))
def codename2culture(codename):
aliases = DictTool.filter(
lambda k, v: v, {
"en": h_codename2aliases_en.get(codename),
"ko": h_codename2aliases_ko.get(codename),
})
culture = {
Culture.Field.CODENAME: codename,
Culture.Field.ALIASES: aliases,
Culture.Field.PREFERS: h_codename2prefers.get(codename) or [],
}
return DictTool.filter(lambda k, v: v, culture)
list_all = lmap(codename2culture, codename_list)
# logger.debug({"list_all":list_all})
return list_all
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@cached(cache=TTLCache(maxsize=2, ttl=60 * 10))
# @FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def dict_codename2culture(cls):
culture_list = cls.culture_list_all()
assert_is_not_none(culture_list)
h_codename2culture = merge_dicts([{
Culture.culture2codename(culture):
culture
} for culture in culture_list])
return h_codename2culture
class TradegoodtypeEntity:
@classmethod
def entity_type(cls):
return ClassTool.class2fullpath(cls)
@classmethod
def text2norm(cls, text):
return str2lower(text)
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
def _dict_lang2matcher(cls, ):
return {
lang: cls.lang2matcher(lang)
for lang in HenriqueLocale.langs()
}
@classmethod
@cached(cache=TTLCache(maxsize=HenriqueLocale.lang_count(),
ttl=HenriqueEntity.Cache.DEFAULT_TTL))
# @FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=HenriqueLocale.lang_count()))
def lang2matcher(cls, lang):
tgt_list = Tradegoodtype.list_all()
langs_recognizable = HenriqueLocale.lang2langs_recognizable(lang)
def tgt2aliases(tgt):
for _lang in langs_recognizable:
yield from Tradegoodtype.tradegoodtype_lang2aliases(tgt, _lang)
h_value2aliases = merge_dicts([{
Tradegoodtype.tradegoodtype2codename(tgt):
list(tgt2aliases(tgt))
} for tgt in tgt_list],
vwrite=vwrite_no_duplicate_key)
config = {
GazetteerMatcher.Config.Key.NORMALIZER: cls.text2norm,
# GazetteerMatcher.Config.Key.TEXTS2PATTERN: HenriqueEntity.texts2rstr_word_with_cardinal_suffix,
}
matcher = GazetteerMatcher(h_value2aliases, config)
return matcher
@classmethod
@CacheTool.cache2hashable(
cache=cached(
TTLCache(maxsize=HenriqueEntity.Cache.DEFAULT_SIZE,
ttl=HenriqueEntity.Cache.DEFAULT_TTL), ),
f_pair=CacheTool.JSON.func_pair(),
)
def text2entity_list(cls, text_in, config=None):
locale = HenriqueEntity.Config.config2locale(
config) or HenriqueLocale.DEFAULT
lang = LocaleTool.locale2lang(locale) or LocaleTool.locale2lang(
HenriqueLocale.DEFAULT)
matcher = cls.lang2matcher(lang)
span_value_list = list(matcher.text2span_value_iter(text_in))
entity_list = [{
FoxylibEntity.Field.SPAN:
span,
FoxylibEntity.Field.TEXT:
StringTool.str_span2substr(text_in, span),
FoxylibEntity.Field.VALUE:
value,
FoxylibEntity.Field.TYPE:
cls.entity_type(),
} for span, value in span_value_list]
return entity_list
class TimedeltaElement:
class Field:
QUANTITY = "quantity"
UNIT = "unit"
SPAN = "span"
@classmethod
def element2quantity(cls, element):
return element[cls.Field.QUANTITY]
@classmethod
def element2unit(cls, element):
return element[cls.Field.UNIT]
@classmethod
def element2span(cls, element):
return element[cls.Field.SPAN]
@classmethod
@WARMER.add(cond=not HenriqueEnv.is_skip_warmup())
@FunctionTool.wrapper2wraps_applied(lru_cache(maxsize=2))
def pattern_number(cls):
rstr_leftbound = RegexTool.rstr2left_bounded(
r"\d{1,2}", RegexTool.left_wordbounds())
rstr_bound_right_list = lchain(
RegexTool.right_wordbounds(),
lchain(*TimedeltaEntityUnit.gazetteer_all().values()),
)
rstr_bound = RegexTool.rstr2right_bounded(rstr_leftbound,
rstr_bound_right_list)
return re.compile(rstr_bound, re.I)
@classmethod
@FunctionTool.wrapper2wraps_applied(
lru_cache(maxsize=HenriqueEntity.Cache.DEFAULT_SIZE))
def text2element_list(cls, text_in, lang):
logger = HenriqueLogger.func_level2logger(cls.text2element_list,
logging.DEBUG)
langs = HenriqueLocale.lang2langs_recognizable(lang)
logger.debug({"langs": langs})
match_list_number = list(cls.pattern_number().finditer(text_in))
span_list_number = lmap(lambda m: m.span(), match_list_number)
matcher = TimedeltaEntityUnit.langs2matcher(langs)
span_value_list_timedelta_unit = list(
matcher.text2span_value_iter(text_in))
spans_list = [
span_list_number,
lmap(ig(0), span_value_list_timedelta_unit),
]
gap2is_valid = partial(StringTool.str_span2match_blank_or_nullstr,
text_in)
indextuple_list = ContextfreeTool.spans_list2reducible_indextuple_list(
spans_list, gap2is_valid)
def indextuple2element(indextuple):
i, j = indextuple
quantity = int(match_list_number[i].group())
unit = span_value_list_timedelta_unit[j][1]
span = (
span_list_number[i][0],
span_value_list_timedelta_unit[j][0][1],
)
element = {
cls.Field.QUANTITY: quantity,
cls.Field.UNIT: unit,
cls.Field.SPAN: span,
}
return element
element_list = lmap(indextuple2element, indextuple_list)
return element_list
@classmethod
def element2relativedelta(cls, element):
logger = HenriqueLogger.func_level2logger(cls.element2relativedelta,
logging.DEBUG)
unit = cls.element2unit(element)
quantity = cls.element2quantity(element)
kwargs = {TimedeltaEntityUnit.unit2plural(unit): quantity}
# logger.debug({"kwargs":kwargs})
return relativedelta(**kwargs)