You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
148 lines
4.1 KiB
148 lines
4.1 KiB
# vim:fileencoding=utf-8
|
|
|
|
import json
|
|
import logging
|
|
import threading
|
|
from io import open
|
|
|
|
logger = logging.getLogger("pycountry.db")
|
|
|
|
|
|
class Data:
|
|
def __init__(self, **fields):
|
|
self._fields = fields
|
|
|
|
def __getattr__(self, key):
|
|
if key not in self._fields:
|
|
raise AttributeError
|
|
return self._fields[key]
|
|
|
|
def __setattr__(self, key, value):
|
|
if key != "_fields":
|
|
self._fields[key] = value
|
|
super(Data, self).__setattr__(key, value)
|
|
|
|
def __repr__(self):
|
|
cls_name = self.__class__.__name__
|
|
fields = ", ".join("%s=%r" % i for i in sorted(self._fields.items()))
|
|
return "%s(%s)" % (cls_name, fields)
|
|
|
|
def __dir__(self):
|
|
return dir(self.__class__) + list(self._fields)
|
|
|
|
|
|
def lazy_load(f):
|
|
def load_if_needed(self, *args, **kw):
|
|
if not self._is_loaded:
|
|
with self._load_lock:
|
|
self._load()
|
|
return f(self, *args, **kw)
|
|
|
|
return load_if_needed
|
|
|
|
|
|
class Database:
|
|
|
|
data_class_base = Data
|
|
data_class_name = None
|
|
root_key = None
|
|
no_index = []
|
|
|
|
def __init__(self, filename):
|
|
self.filename = filename
|
|
self._is_loaded = False
|
|
self._load_lock = threading.Lock()
|
|
|
|
def _load(self):
|
|
if self._is_loaded:
|
|
# Help keeping the _load_if_needed code easier
|
|
# to read.
|
|
return
|
|
self.objects = []
|
|
self.index_names = set()
|
|
self.indices = {}
|
|
|
|
self.data_class = type(
|
|
self.data_class_name, (self.data_class_base,), {}
|
|
)
|
|
|
|
with open(self.filename, "r", encoding="utf-8") as f:
|
|
tree = json.load(f)
|
|
|
|
for entry in tree[self.root_key]:
|
|
obj = self.data_class(**entry)
|
|
self.objects.append(obj)
|
|
# Inject into index.
|
|
for key, value in entry.items():
|
|
# Lookups and searches are case insensitive. Normalize
|
|
# here.
|
|
value = value.lower()
|
|
if key in self.no_index:
|
|
continue
|
|
index = self.indices.setdefault(key, {})
|
|
if value in index:
|
|
logger.debug(
|
|
"%s %r already taken in index %r and will be "
|
|
"ignored. This is an error in the databases."
|
|
% (self.data_class_name, value, key)
|
|
)
|
|
index[value] = obj
|
|
|
|
self._is_loaded = True
|
|
|
|
# Public API
|
|
|
|
@lazy_load
|
|
def __iter__(self):
|
|
return iter(self.objects)
|
|
|
|
@lazy_load
|
|
def __len__(self):
|
|
return len(self.objects)
|
|
|
|
@lazy_load
|
|
def get(self, **kw):
|
|
kw.setdefault("default", None)
|
|
default = kw.pop("default")
|
|
if len(kw) != 1:
|
|
raise TypeError("Only one criteria may be given")
|
|
field, value = kw.popitem()
|
|
if not isinstance(value, str):
|
|
raise LookupError()
|
|
# Normalize for case-insensitivity
|
|
value = value.lower()
|
|
index = self.indices[field]
|
|
try:
|
|
return index[value]
|
|
except KeyError:
|
|
# Pythonic APIs implementing get() shouldn't raise KeyErrors.
|
|
# Those are a bit unexpected and they should rather support
|
|
# returning `None` by default and allow customization.
|
|
return default
|
|
|
|
@lazy_load
|
|
def lookup(self, value):
|
|
if not isinstance(value, str):
|
|
raise LookupError()
|
|
|
|
# Normalize for case-insensitivity
|
|
value = value.lower()
|
|
|
|
# Use indexes first
|
|
for key in self.indices:
|
|
try:
|
|
return self.indices[key][value]
|
|
except LookupError:
|
|
pass
|
|
|
|
# Use non-indexed values now. Avoid going through indexed values.
|
|
for candidate in self:
|
|
for k in self.no_index:
|
|
v = candidate._fields.get(k)
|
|
if v is None:
|
|
continue
|
|
if v.lower() == value:
|
|
return candidate
|
|
|
|
raise LookupError("Could not find a record for %r" % value)
|