def update(self, data):
props = (
"title",
"summary",
"author",
"crawler",
"source_url",
"file_name",
"mime_type",
"headers",
"date",
"authored_at",
"modified_at",
"published_at",
"retrieved_at",
"languages",
"countries",
"keywords",
)
countries = ensure_list(data.get("countries", []))
data["countries"] = [registry.country.clean(val) for val in countries]
languages = ensure_list(data.get("languages", []))
data["languages"] = [registry.language.clean(val) for val in languages]
for prop in props:
text = data.get(prop, self.meta.get(prop))
if isinstance(text, list):
self.meta[prop] = [sanitize_text(txt) for txt in text]
else:
self.meta[prop] = sanitize_text(text)
if self.meta.get(prop) is None:
self.meta.pop(prop, None)
flag_modified(self, "meta")
def __init__(self, args, authz, limit=None, max_limit=MAX_PAGE):
if not isinstance(args, MultiDict):
args = OrderedMultiDict(args)
self.args = args
self.authz = authz
self.offset = max(0, self.getint("offset", 0))
if limit is None:
limit = min(max_limit, max(0, self.getint("limit", 20)))
self.limit = limit
self.text = sanitize_text(self.get("q"))
self.prefix = sanitize_text(self.get("prefix"))
# Disable or enable query caching
self.cache = self.getbool("cache", settings.CACHE)
def __init__(self, model, data, key_prefix=None):
data = dict(data)
properties = ensure_dict(data.pop('properties', {}))
self.schema = model.get(data.pop('schema', None))
if self.schema is None:
raise InvalidData(gettext('No schema for entity.'))
self.id = sanitize_text(data.pop('id', None))
self.key_prefix = sanitize_text(key_prefix)
self.context = data
self._properties = {}
self._size = 0
if is_mapping(properties):
for key, value in properties.items():
self.add(key, value, cleaned=True, quiet=True)
def getlist(self, name, default=None):
values = []
for value in self.args.getlist(name):
value = sanitize_text(value, encoding="utf-8")
if value is not None:
values.append(value)
return values or (default or [])
def clean(self, text, format=None, **kwargs):
"""The classic: date parsing, every which way."""
# handle date/datetime before converting to text.
date = self._clean_datetime(text)
if date is not None:
return date
text = sanitize_text(text)
if text is None:
return
if format is not None:
# parse with a specified format
try:
obj = datetime.strptime(text, format)
text = obj.date().isoformat()
if self.MONTH_FORMATS.search(format) is None:
text = text[:4]
elif self.DAY_FORMATS.search(format) is None:
text = text[:7]
return text
except Exception:
return None
return self._clean_text(text)
def clean(self, text: Any, **kwargs) -> Optional[str]:
"""Create a clean version of a value of the type, suitable for storage
in an entity proxy."""
text = sanitize_text(text)
if text is None:
return None
return self.clean_text(text, **kwargs)
def __init__(self, model, data, key_prefix=None, cleaned=True):
data = dict(data)
properties = data.pop("properties", {})
if not cleaned:
properties = ensure_dict(properties)
self.schema = model.get(data.pop("schema", None))
if self.schema is None:
raise InvalidData(gettext("No schema for entity."))
self.key_prefix = key_prefix
self.id = data.pop("id", None)
if not cleaned:
self.id = sanitize_text(self.id)
self.context = data
self._properties = {}
self._size = 0
for key, value in properties.items():
if key not in self.schema.properties:
continue
if not cleaned:
self.add(key, value, cleaned=cleaned, quiet=True)
else:
values = set(value)
self._properties[key] = values
self._size += sum([len(v) for v in values])
def items(self):
for (key, value) in self.args.items(multi=True):
if key in ("offset", "limit", "next_limit"):
continue
value = sanitize_text(value, encoding="utf-8")
if value is not None:
yield key, value
def items(self):
for (key, value) in self.args.items(multi=True):
if key == 'offset':
continue
value = sanitize_text(value, encoding='utf-8')
if value is not None:
yield key, value
def read_csv(cls, fh: Iterable[str]) -> Generator[Record, None, None]:
for row in DictReader(fh, skipinitialspace=True):
data: Record = {}
for ref, ref_value in row.items():
value = sanitize_text(ref_value)
if value is not None:
data[ref] = value
yield data
def pick(self, values):
values = [sanitize_text(v) for v in ensure_list(values)]
values = [v for v in values if v is not None]
if not len(values):
return None
if 1 == len(values):
return values[0]
return setmedian(values)
def records(self):
"""Iterate through the table applying filters on-the-go."""
for url in self.urls:
for row in self.read_csv_url(url):
data = {}
for ref in self.query.refs:
data[ref] = sanitize_text(row.get(ref))
if self.check_filters(row):
yield data
def ingest(self, file_path, entity):
entity.schema = model.get('PlainText')
text = self.read_file_decoded(entity, file_path)
text = sanitize_text(text)
entity.set('bodyText', text)
try:
for card in vobject.readComponents(text):
self.ingest_card(entity, card)
except (ParseError, UnicodeDecodeError) as err:
raise ProcessingException('Cannot parse vcard: %s' % err) from err
def update(self, data):
props = ('title', 'summary', 'author', 'crawler', 'source_url',
'file_name', 'mime_type', 'headers', 'date', 'authored_at',
'modified_at', 'published_at', 'retrieved_at', 'languages',
'countries', 'keywords')
for prop in props:
text = data.get(prop, self.meta.get(prop))
self.meta[prop] = sanitize_text(text)
flag_modified(self, 'meta')
def validate(self, email, **kwargs):
"""Check to see if this is a valid email address."""
# TODO: adopt email.utils.parseaddr
email = sanitize_text(email)
if email is None:
return False
if not self.EMAIL_REGEX.match(email):
return False
_, domain = email.rsplit('@', 1)
return self.domains.validate(domain, **kwargs)
def validate(self, value: str) -> bool:
"""Check to see if this is a valid email address."""
# TODO: adopt email.utils.parseaddr
email = sanitize_text(value)
if email is None or not self.REGEX.match(email):
return False
_, domain = email.rsplit("@", 1)
if len(domain) < 4 or "." not in domain:
return False
return True
def clean(
self,
raw: Any,
fuzzy: bool = False,
format: Optional[str] = None,
proxy: Optional["EntityProxy"] = None,
) -> Optional[str]:
if not isinstance(raw, str):
return self.pack(raw)
else:
return sanitize_text(raw)
def validate(self, obj, **kwargs):
"""Check if a thing is a valid domain name."""
text = sanitize_text(obj)
if text is None:
return False
if '.' not in text:
return False
if '@' in text or ':' in text:
return False
if len(text) < 4:
return False
return True
def clean(
self,
raw: Any,
fuzzy: bool = False,
format: Optional[str] = None,
proxy: Optional["EntityProxy"] = None,
) -> Optional[str]:
"""Create a clean version of a value of the type, suitable for storage
in an entity proxy."""
text = sanitize_text(raw)
if text is None:
return None
return self.clean_text(text, fuzzy=fuzzy, format=format, proxy=proxy)
def make_sheet(self, title: str, headers: List[str]) -> Worksheet:
sheet = self.workbook.create_sheet(title=title)
sheet.freeze_panes = "A2"
sheet.sheet_properties.filterMode = True
cells = []
for header in headers:
header_ = sanitize_text(header)
cell = WriteOnlyCell(sheet, value=header_)
cell.font = self.HEADER_FONT
cell.fill = self.HEADER_FILL
cells.append(cell)
sheet.append(cells)
return sheet
def update(self, data):
props = ('title', 'summary', 'author', 'crawler', 'source_url',
'file_name', 'mime_type', 'headers', 'date', 'authored_at',
'modified_at', 'published_at', 'retrieved_at', 'languages',
'countries', 'keywords')
data['countries'] = ensure_list(data.get('countries', []))
data['countries'] = [
registry.country.clean(val) for val in data['countries']
] # noqa
data['languages'] = ensure_list(data.get('languages', []))
data['languages'] = [
registry.language.clean(val) for val in data['languages']
] # noqa
for prop in props:
text = data.get(prop, self.meta.get(prop))
if isinstance(text, list):
self.meta[prop] = [sanitize_text(txt) for txt in text]
else:
self.meta[prop] = sanitize_text(text)
if self.meta.get(prop) is None:
self.meta.pop(prop, None)
flag_modified(self, 'meta')
def records(self):
"""Compose the actual query and return an iterator of ``Record``."""
mapping = [(r, self.get_column(r).name) for r in self.query.refs]
q = self.compose_query()
log.info("Query: %s", q)
rp = self.engine.execute(q)
while True:
rows = rp.fetchmany(size=DATA_PAGE)
if not len(rows):
break
for row in rows:
data = {}
for ref, name in mapping:
data[ref] = sanitize_text(row[name])
yield data
def __init__(
self,
model: "Model",
data: Dict[str, Any],
key_prefix: Optional[str] = None,
cleaned: bool = True,
):
data = dict(data or {})
properties = data.pop("properties", {})
if not cleaned:
properties = ensure_dict(properties)
#: The schema definition for this entity, which implies the properties
#: That can be set on it.
schema = model.get(data.pop("schema", None))
if schema is None:
raise InvalidData(gettext("No schema for entity."))
self.schema = schema
#: When using :meth:`~make_id` to generate a natural key for this entity,
#: the prefix will be added to the ID as a salt to make it easier to keep
#: IDs unique across datasets. This is somewhat redundant following the
#: introduction of :class:`~followthemoney.namespace.Namespace`.
self.key_prefix = key_prefix
#: A unique identifier for this entity, usually a hashed natural key,
#: a UUID, or a very simple slug. Can be signed using a
#: :class:`~followthemoney.namespace.Namespace`.
self.id = data.pop("id", None)
if not cleaned:
self.id = sanitize_text(self.id)
#: If the input dictionary for the entity proxy contains fields other
#: than ``id``, ``schema`` or ``properties``, they will be kept in here
#: and re-added upon serialization.
self.context = data
self._properties: Dict[str, Set[str]] = {}
self._size = 0
for key, value in properties.items():
if key not in self.schema.properties:
continue
if not cleaned:
self.add(key, value, cleaned=cleaned, quiet=True)
else:
values = set(value)
self._properties[key] = values
self._size += sum([len(v) for v in values])
def save(
cls,
collection,
parent=None,
foreign_id=None,
content_hash=None,
meta=None,
role_id=None,
):
"""Try and find a document by various criteria."""
foreign_id = sanitize_text(foreign_id)
q = cls.all()
q = q.filter(Document.collection_id == collection.id)
if parent is not None:
q = q.filter(Document.parent_id == parent.id)
if foreign_id is not None:
q = q.filter(Document.foreign_id == foreign_id)
elif content_hash is not None:
q = q.filter(Document.content_hash == content_hash)
else:
raise ValueError("No unique criterion for document.")
document = q.first()
if document is None:
document = cls()
document.schema = cls.SCHEMA
document.collection_id = collection.id
document.role_id = role_id
if parent is not None:
document.parent_id = parent.id
if foreign_id is not None:
document.foreign_id = foreign_id
document.content_hash = content_hash
if content_hash is None:
document.schema = cls.SCHEMA_FOLDER
if meta is not None:
document.update(meta)
db.session.add(document)
return document
def emit_row_dicts(self, table, rows, headers=None):
csv_path = self.make_work_file(table.id)
row_count = 0
with open(csv_path, 'w', encoding='utf-8') as fp:
csv_writer = csv.writer(fp, dialect='unix')
for row in rows:
if headers is None:
headers = list(row.keys())
values = [sanitize_text(row.get(h)) for h in headers]
csv_writer.writerow(values)
self.manager.emit_text_fragment(table, values, row_count)
row_count += 1
if row_count > 0 and row_count % 1000 == 0:
log.info("Table emit [%s]: %s...", table, row_count)
if row_count > 0:
csv_hash = self.manager.store(csv_path, mime_type=CSV)
table.set('csvHash', csv_hash)
table.set('rowCount', row_count + 1)
table.set('columns', registry.json.pack(headers))
def emit_row_dicts(self, table, rows, headers=None):
csv_path = self.make_work_file(table.id)
row_count = 0
with open(csv_path, "w", encoding=self.DEFAULT_ENCODING) as fp:
csv_writer = csv.writer(fp, dialect="unix")
for row in rows:
if headers is None:
headers = list(row.keys())
values = [sanitize_text(row.get(h)) for h in headers]
length = sum((len(v) for v in values if v is not None))
if length == 0:
continue
csv_writer.writerow(values)
self.manager.emit_text_fragment(table, values, row_count)
row_count += 1
if row_count > 0 and row_count % 1000 == 0:
log.info("Table emit [%s]: %s...", table, row_count)
if row_count > 0:
csv_hash = self.manager.store(csv_path, mime_type=CSV)
table.set("csvHash", csv_hash)
table.set("rowCount", row_count + 1)
table.set("columns", registry.json.pack(headers))
def __init__(self, query, data, prop):
self.query = query
data = deepcopy(data)
self.data = data
self.prop = prop
self.name = prop.name
self.type = prop.type
self.refs = keys_values(data, "column", "columns")
self.literals = keys_values(data, "literal", "literals")
self.join = data.pop("join", None)
self.split = data.pop("split", None)
self.entity = data.pop("entity", None)
self.required = data.pop("required", False)
self.template = sanitize_text(data.pop("template", None))
self.replacements = {}
if self.template is not None:
# this is hacky, trying to generate refs from template
for ref in self.FORMAT_PATTERN.findall(self.template):
self.refs.append(ref)
self.replacements["{{%s}}" % ref] = ref
def __init__(self, query: "QueryMapping", data: Dict[str, Any],
prop: Property) -> None:
self.query = query
data = deepcopy(data)
self.prop = prop
self.refs = cast(List[str], keys_values(data, "column", "columns"))
self.join = cast(Optional[str], data.pop("join", None))
self.split = cast(Optional[str], data.pop("split", None))
self.entity = stringify(data.pop("entity", None))
self.format = stringify(data.pop("format", None))
self.fuzzy = as_bool(data.pop("fuzzy", False))
self.required = as_bool(data.pop("required", False))
self.literals = cast(List[str], keys_values(data, "literal",
"literals"))
self.template = sanitize_text(data.pop("template", None))
self.replacements: Dict[str, str] = {}
if self.template is not None:
# this is hacky, trying to generate refs from template
for ref in self.FORMAT_PATTERN.findall(self.template):
self.refs.append(ref)
self.replacements["{{%s}}" % ref] = ref
def ingest_component(self, entity, idx, comp):
if comp.name == "VCALENDAR":
entity.add("generator", comp.get("PRODID"))
if comp.name == "VEVENT":
event = self.manager.make_entity("Event")
self.manager.apply_context(event, entity)
uid = sanitize_text(comp.get("UID"))
if uid is not None:
event.make_id(uid)
else:
event.make_id(entity.id, idx)
event.add("proof", entity)
event.add("name", comp.get("SUMMARY"))
event.add("description", comp.get("DESCRIPTION"))
event.add("location", comp.get("LOCATION"))
event.add("sourceUrl", comp.get("URL"))
event.add("startDate", cal_date(comp.get("DTSTART")))
event.add("endDate", cal_date(comp.get("DTEND")))
event.add("date", cal_date(comp.get("CREATED")))
event.add("modifiedAt", cal_date(comp.get("LAST-MODIFIED")))
event.add("organizer", self.address_entity(comp.get("ORGANIZER")))
for attendee in ensure_list(comp.get("ATTENDEE")):
event.add("involved", self.address_entity(attendee))
self.manager.emit_entity(event, fragment=idx)
def validate(self, text, **kwargs):
text = sanitize_text(text)
try:
return iban.validate(text)
except ValidationError:
return False
