# -*- coding: utf-8 -*-
"""
obspy.core.event.base - Classes for handling event metadata
===========================================================
This module provides a class hierarchy to consistently handle event metadata.
This class hierarchy is closely modelled after the de-facto standard format
`QuakeML <https://quake.ethz.ch/quakeml/>`_.
.. figure:: /_images/Event.png
.. note::
For handling additional information not covered by the QuakeML standard and
how to output it to QuakeML see the :ref:`ObsPy Tutorial <quakeml-extra>`.
:copyright:
The ObsPy Development Team (devs@obspy.org)
:license:
GNU Lesser General Public License, Version 3
(https://www.gnu.org/copyleft/lesser.html)
"""
from __future__ import (absolute_import, division, print_function,
unicode_literals)
from future.builtins import * # NOQA
from future.utils import native_str
import collections
import copy
import inspect
import re
import warnings
import weakref
from copy import deepcopy
from uuid import uuid4
import numpy as np
from obspy.core.event.header import DataUsedWaveType, ATTRIBUTE_HAS_ERRORS
from obspy.core.utcdatetime import UTCDateTime
from obspy.core.util import AttribDict
[docs]class QuantityError(AttribDict):
"""
Uncertainty information for a physical quantity.
:type uncertainty: float
:param uncertainty: Uncertainty as the absolute value of symmetric
deviation from the main value.
:type lower_uncertainty: float
:param lower_uncertainty: Uncertainty as the absolute value of deviation
from the main value towards smaller values.
:type upper_uncertainty: float
:param upper_uncertainty: Uncertainty as the absolute value of deviation
from the main value towards larger values.
:type confidence_level: float
:param confidence_level: Confidence level of the uncertainty, given in
percent (0-100).
"""
defaults = {"uncertainty": None, "lower_uncertainty": None,
"upper_uncertainty": None, "confidence_level": None}
warn_on_non_default_key = True
[docs] def __init__(self, uncertainty=None, lower_uncertainty=None,
upper_uncertainty=None, confidence_level=None):
super(QuantityError, self).__init__()
self.uncertainty = uncertainty
self.lower_uncertainty = lower_uncertainty
self.upper_uncertainty = upper_uncertainty
self.confidence_level = confidence_level
[docs] def __bool__(self):
"""
Boolean testing for QuantityError.
QuantityError evaluates ``True`` if any of the default fields is not
``None``. Setting non default fields raises also an UserWarning which
is the reason we have to skip those lines in the doctest below.
>>> err = QuantityError()
>>> bool(err)
False
>>> err.custom_field = "spam" # doctest: +SKIP
>>> bool(err)
False
>>> err.uncertainty = 0.05
>>> bool(err)
True
>>> del err.custom_field # doctest: +SKIP
>>> bool(err)
True
"""
return any([getattr(self, key) is not None for key in self.defaults])
# Python 2 compatibility
__nonzero__ = __bool__
[docs]def _bool(value):
"""
A custom bool() implementation that returns
True for any value (including zero) of int and float,
and for (empty) strings.
"""
if value == 0 or isinstance(value, (str, native_str)):
return True
return bool(value)
[docs]def _event_type_class_factory(class_name, class_attributes=[],
class_contains=[]):
"""
Class factory to unify the creation of all the types needed for the event
handling in ObsPy.
The types oftentimes share attributes and setting them manually every time
is cumbersome, error-prone and hard to do consistently. The classes created
with this method will inherit from :class:`~obspy.core.util.AttribDict`.
Usage to create a new class type:
The created class will assure that any given (key, type) attribute pairs
will always be of the given type and will attempt to convert any given
value to the correct type and raise an error otherwise. This happens to
values given during initialization as well as values set when the object
has already been created. A useful type is Enum if you want to restrict
the acceptable values.
>>> from obspy.core.util import Enum
>>> MyEnum = Enum(["a", "b", "c"])
>>> class_attributes = [ \
("resource_id", ResourceIdentifier), \
("creation_info", CreationInfo), \
("some_letters", MyEnum), \
("some_error_quantity", float, ATTRIBUTE_HAS_ERRORS), \
("description", str)]
Furthermore the class can contain lists of other objects. There is not much
to it so far. Giving the name of the created class is mandatory.
>>> class_contains = ["comments"]
>>> TestEventClass = _event_type_class_factory("TestEventClass", \
class_attributes=class_attributes, \
class_contains=class_contains)
>>> assert(TestEventClass.__name__ == "TestEventClass")
Now the new class type can be used.
>>> test_event = TestEventClass(resource_id="event/123456", \
creation_info={"author": "obspy.org", "version": "0.1"})
All given arguments will be converted to the right type upon setting them.
>>> test_event.resource_id
ResourceIdentifier(id="event/123456")
>>> print(test_event.creation_info)
CreationInfo(author='obspy.org', version='0.1')
All others will be set to None.
>>> assert(test_event.description is None)
>>> assert(test_event.some_letters is None)
If the resource_id attribute of the created class type is set, the object
the ResourceIdentifier refers to will be the class instance.
>>> assert(id(test_event) == \
id(test_event.resource_id.get_referred_object()))
They can be set later and will be converted to the appropriate type if
possible.
>>> test_event.description = 1
>>> assert(test_event.description == "1")
Trying to set with an inappropriate value will raise an error.
>>> test_event.some_letters = "d" # doctest:+ELLIPSIS
Traceback (most recent call last):
...
ValueError: Setting attribute "some_letters" failed. ...
If you pass ``ATTRIBUTE_HAS_ERRORS`` as the third tuple item for the
class_attributes, a error (type
:class:`~obspy.core.event.base.QuantityError`) will be be created that will
be named like the attribute with "_errors" appended.
>>> assert(hasattr(test_event, "some_error_quantity_errors"))
>>> test_event.some_error_quantity_errors # doctest: +ELLIPSIS
QuantityError(...)
"""
class AbstractEventType(AttribDict):
# Keep the class attributes in a class level list for a manual property
# implementation that works when inheriting from AttribDict.
_properties = []
for item in class_attributes:
_properties.append((item[0], item[1]))
if len(item) == 3 and item[2] == ATTRIBUTE_HAS_ERRORS:
_properties.append((item[0] + "_errors", QuantityError))
_property_keys = [_i[0] for _i in _properties]
_property_dict = {}
for key, value in _properties:
_property_dict[key] = value
_containers = class_contains
warn_on_non_default_key = True
defaults = dict.fromkeys(class_contains, [])
defaults.update(dict.fromkeys(_property_keys, None))
do_not_warn_on = ["extra"]
def __init__(self, *args, **kwargs):
# Make sure the args work as expected. Therefore any specified
# arg will overwrite a potential kwarg, e.g. arg at position 0 will
# overwrite kwargs class_attributes[0].
for _i, item in enumerate(args):
# Use the class_attributes list here because it is not yet
# polluted be the error quantities.
kwargs[class_attributes[_i][0]] = item
# Set all property values to None or the kwarg value.
for key, _ in self._properties:
value = kwargs.get(key, None)
# special handling for resource id
if key == "resource_id":
if kwargs.get("force_resource_id", False):
if value is None:
value = ResourceIdentifier()
setattr(self, key, value)
# Containers currently are simple lists.
for name in self._containers:
setattr(self, name, list(kwargs.get(name, [])))
# All errors are QuantityError. If they are not set yet, set them
# now.
for key, _ in self._properties:
if key.endswith("_errors") and getattr(self, key) is None:
setattr(self, key, QuantityError())
def clear(self):
super(AbstractEventType, self).clear()
self.__init__(force_resource_id=False)
def __str__(self, force_one_line=False):
"""
Fairly extensive in an attempt to cover several use cases. It is
always possible to change it in the child class.
"""
# Get the attribute and containers that are to be printed. Only not
# None attributes and non-error attributes are printed. The errors
# will appear behind the actual value.
# We use custom _bool() for testing getattr() since we want to
# print int and float values that are equal to zero and empty
# strings.
attributes = [_i for _i in self._property_keys if not
_i.endswith("_errors") and _bool(getattr(self, _i))]
containers = [_i for _i in self._containers if
_bool(getattr(self, _i))]
# Get the longest attribute/container name to print all of them
# nicely aligned.
max_length = max(max([len(_i) for _i in attributes])
if attributes else 0,
max([len(_i) for _i in containers])
if containers else 0) + 1
ret_str = self.__class__.__name__
# Case 1: Empty object.
if not attributes and not containers:
return ret_str + "()"
def get_value_repr(key):
value = getattr(self, key)
if isinstance(value, (str, native_str)):
value = native_str(value)
repr_str = value.__repr__()
# Print any associated errors.
error_key = key + "_errors"
if self.get(error_key, False):
err_items = sorted(getattr(self, error_key).items())
repr_str += " [%s]" % ', '.join(
sorted([str(k) + "=" + str(v) for k, v in err_items
if v is not None]))
return repr_str
# Case 2: Short representation for small objects. Will just print a
# single line.
if len(attributes) <= 3 and not containers or\
force_one_line:
att_strs = ["%s=%s" % (_i, get_value_repr(_i))
for _i in attributes if _bool(getattr(self, _i))]
ret_str += "(%s)" % ", ".join(att_strs)
return ret_str
# Case 3: Verbose string representation for large object.
if attributes:
format_str = "%" + str(max_length) + "s: %s"
att_strs = [format_str % (_i, get_value_repr(_i))
for _i in attributes if _bool(getattr(self, _i))]
ret_str += "\n\t" + "\n\t".join(att_strs)
# For the containers just print the number of elements in each.
if containers:
# Print delimiter only if there are attributes.
if attributes:
ret_str += '\n\t' + '---------'.rjust(max_length + 5)
element_str = "%" + str(max_length) + "s: %i Elements"
ret_str += "\n\t" + \
"\n\t".join(
[element_str % (_i, len(getattr(self, _i)))
for _i in containers])
return ret_str
def _repr_pretty_(self, p, cycle):
p.text(str(self))
def copy(self):
return copy.deepcopy(self)
def __repr__(self):
return self.__str__(force_one_line=True)
# called for bool on PY2
def __nonzero__(self):
return self.__bool__()
def __bool__(self):
# We use custom _bool() for testing getattr() since we want
# zero valued int and float and empty string attributes to be True.
if any([_bool(getattr(self, _i))
for _i in self._property_keys + self._containers]):
return True
return False
def __eq__(self, other):
"""
Two instances are considered equal if all attributes and all lists
are identical.
"""
# Looping should be quicker on average than a list comprehension
# because only the first non-equal attribute will already return.
for attrib in self._property_keys:
if not hasattr(other, attrib) or \
(getattr(self, attrib) != getattr(other, attrib)):
return False
for container in self._containers:
if not hasattr(other, container) or \
(getattr(self, container) != getattr(other, container)):
return False
return True
def __ne__(self, other):
return not self.__eq__(other)
def __setattr__(self, name, value):
"""
Custom property implementation that works if the class is
inheriting from AttribDict.
"""
# avoid type casting of 'extra' attribute, to make it possible to
# control ordering of extra tags by using an OrderedDict for
# 'extra'.
if name == 'extra':
dict.__setattr__(self, name, value)
return
# Pass to the parent method if not a custom property.
if name not in self._property_dict.keys():
AttribDict.__setattr__(self, name, value)
return
attrib_type = self._property_dict[name]
# If the value is None or already the correct type just set it.
if (value is not None) and (type(value) is not attrib_type):
# If it is a dict, and the attrib_type is no dict, than all
# values will be assumed to be keyword arguments.
if isinstance(value, dict):
new_value = attrib_type(**value)
else:
new_value = attrib_type(value)
if new_value is None:
msg = 'Setting attribute "%s" failed. ' % (name)
msg += 'Value "%s" could not be converted to type "%s"' % \
(str(value), str(attrib_type))
raise ValueError(msg)
value = new_value
# Make sure all floats are finite - otherwise this is most
# likely a user error.
if attrib_type is float and value is not None:
if not np.isfinite(value):
msg = "On %s object: Value '%s' for '%s' is " \
"not a finite floating point value." % (
type(self).__name__, str(value), name)
raise ValueError(msg)
AttribDict.__setattr__(self, name, value)
# if value is a resource id bind or unbind the resource_id
if isinstance(value, ResourceIdentifier):
if name == "resource_id": # bind the resource_id to self
self.resource_id.set_referred_object(self, warn=False)
else: # else unbind to allow event scoping later
value._object_id = None
class AbstractEventTypeWithResourceID(AbstractEventType):
def __init__(self, force_resource_id=True, *args, **kwargs):
kwargs["force_resource_id"] = force_resource_id
super(AbstractEventTypeWithResourceID, self).__init__(*args,
**kwargs)
def __deepcopy__(self, memodict=None):
"""
reset resource_id's object_id after deep copy to allow the
object specific behavior of get_referred_object
"""
memodict = memodict or {}
cls = self.__class__
result = cls.__new__(cls)
memodict[id(self)] = result
for k, v in self.__dict__.items():
setattr(result, k, deepcopy(v, memodict))
return result
if "resource_id" in [item[0] for item in class_attributes]:
base_class = AbstractEventTypeWithResourceID
else:
base_class = AbstractEventType
# Set the class type name.
setattr(base_class, "__name__", native_str(class_name))
return base_class
[docs]class ResourceIdentifier(object):
r"""
Unique identifier of any resource so it can be referred to.
In QuakeML many elements and types can have a unique id that other elements
use to refer to it. This is called a ResourceIdentifier and it is used for
the same purpose in the obspy.core.event classes.
In QuakeML it has to be of the following regex form::
(smi|quakeml):[\w\d][\w\d\-\.\*\(\)_~']{2,}/[\w\d\-\.\*\(\)_~']
[\w\d\-\.\*\(\)\+\?_~'=,;#/&]*
e.g.
* ``smi:sub.website.org/event/12345678``
* ``quakeml:google.org/pick/unique_pick_id``
smi stands for "seismological meta-information".
In this class it can be any hashable object, e.g. most immutable objects
like numbers and strings.
:type id: str, optional
:param id: A unique identifier of the element it refers to. It is
not verified, that it actually is unique. The user has to take care of
that. If no resource_id is given, uuid.uuid4() will be used to
create one which assures uniqueness within one Python run.
If no fixed id is provided, the ID will be built from prefix
and a random uuid hash. The random hash part can be regenerated by the
referred object automatically if it gets changed.
:type prefix: str, optional
:param prefix: An optional identifier that will be put in front of any
automatically created resource id. The prefix will only have an effect
if `id` is not specified (for a fixed ID string). Makes automatically
generated resource ids more reasonable. By default "smi:local" is used
which ensures a QuakeML conform resource identifier.
:type referred_object: Python object, optional
:param referred_object: The object this instance refers to. All instances
created with the same resource_id will be able to access the object as
long as at least one instance actually has a reference to it.
Additionally, ResourceIdentifier instances that have the same id but
different referred_objects will still return the referred object,
provided it doesn't get garbage collected. If the referred object no
longer exists any object with the same id will be returned.
.. rubric:: General Usage
>>> ResourceIdentifier('2012-04-11--385392')
ResourceIdentifier(id="2012-04-11--385392")
>>> # If 'id' is not specified it will be generated automatically.
>>> ResourceIdentifier() # doctest: +ELLIPSIS
ResourceIdentifier(id="smi:local/...")
>>> # Supplying a prefix will simply prefix the automatically generated ID
>>> ResourceIdentifier(prefix='event') # doctest: +ELLIPSIS
ResourceIdentifier(id="event/...")
ResourceIdentifiers can, and oftentimes should, carry a reference to the
object they refer to. This is a weak reference which means that if the
object get deleted or runs out of scope, e.g. gets garbage collected, the
reference will cease to exist.
>>> from obspy.core.event import Event
>>> event = Event()
>>> import sys
>>> ref_count = sys.getrefcount(event)
>>> res_id = ResourceIdentifier(referred_object=event)
>>> # The reference does not changed the reference count of the object.
>>> print(ref_count == sys.getrefcount(event))
True
>>> # It actually is the same object.
>>> print(event is res_id.get_referred_object())
True
>>> # Deleting it, or letting the garbage collector handle the object will
>>> # invalidate the reference.
>>> del event
>>> print(res_id.get_referred_object()) # doctest: +SKIP
None
The most powerful ability (and reason why one would want to use a resource
identifier class in the first place) is that once a ResourceIdentifier with
an attached referred object has been created, any other ResourceIdentifier
instances with the same ID can retrieve that object. This works
across all ResourceIdentifiers that have been instantiated within one
Python run.
This enables, e.g. the resource references between the different QuakeML
elements to work in a rather natural way.
>>> event_object = Event()
>>> obj_id = id(event_object)
>>> res_id = "obspy.org/event/test"
>>> ref_a = ResourceIdentifier(res_id)
>>> # The object is refers to cannot be found yet. Because no instance that
>>> # an attached object has been created so far.
>>> print(ref_a.get_referred_object())
None
>>> # This instance has an attached object.
>>> ref_b = ResourceIdentifier(res_id, referred_object=event_object)
>>> ref_c = ResourceIdentifier(res_id)
>>> # All ResourceIdentifiers will refer to the same object.
>>> assert(id(ref_a.get_referred_object()) == obj_id)
>>> assert(id(ref_b.get_referred_object()) == obj_id)
>>> assert(id(ref_c.get_referred_object()) == obj_id)
Resource identifiers are bound to an object once the get_referred_object
method has been called. The results is that get_referred_object will
always return the same object it did on the first call as long as the
object still exists. If the bound object gets garage collected a warning
will be issued and another object with the same resource_id will be
returned if one exists. If no other object has the same resource_id, an
additional warning will be issued and None returned.
>>> res_id = 'obspy.org/tests/test_resource_doc_example'
>>> obj_a = UTCDateTime(10)
>>> obj_b = UTCDateTime(10)
>>> ref_a = ResourceIdentifier(res_id, referred_object=obj_a)
>>> ref_b = ResourceIdentifier(res_id, referred_object=obj_b)
>>> assert ref_a.get_referred_object() == ref_b.get_referred_object()
>>> assert ref_a.get_referred_object() is not ref_b.get_referred_object()
>>> assert ref_a.get_referred_object() is obj_a
>>> assert ref_b.get_referred_object() is obj_b
>>> del obj_b # if obj_b gets garbage collected
>>> assert ref_b.get_referred_object() is obj_a # doctest: +SKIP
>>> del obj_a # now no object with res_id exists
>>> assert ref_b.get_referred_object() is None # doctest: +SKIP
The id can be converted to a valid QuakeML ResourceIdentifier by calling
the convert_id_to_quakeml_uri() method. The resulting id will be of the
form::
smi:authority_id/prefix/id
>>> res_id = ResourceIdentifier(prefix='origin')
>>> res_id.convert_id_to_quakeml_uri(authority_id="obspy.org")
>>> res_id # doctest: +ELLIPSIS
ResourceIdentifier(id="smi:obspy.org/origin/...")
>>> res_id = ResourceIdentifier(id='foo')
>>> res_id.convert_id_to_quakeml_uri()
>>> res_id
ResourceIdentifier(id="smi:local/foo")
>>> # A good way to create a QuakeML compatibly ResourceIdentifier from
>>> # scratch is
>>> res_id = ResourceIdentifier(prefix='pick')
>>> res_id.convert_id_to_quakeml_uri(authority_id='obspy.org')
>>> res_id # doctest: +ELLIPSIS
ResourceIdentifier(id="smi:obspy.org/pick/...")
>>> # If the given ID is already a valid QuakeML
>>> # ResourceIdentifier, nothing will happen.
>>> res_id = ResourceIdentifier('smi:test.org/subdir/id')
>>> res_id
ResourceIdentifier(id="smi:test.org/subdir/id")
>>> res_id.convert_id_to_quakeml_uri()
>>> res_id
ResourceIdentifier(id="smi:test.org/subdir/id")
ResourceIdentifiers are considered identical if the IDs are
the same.
>>> # Create two different resource identifiers.
>>> res_id_1 = ResourceIdentifier()
>>> res_id_2 = ResourceIdentifier()
>>> assert(res_id_1 != res_id_2)
>>> # Equalize the IDs. NEVER do this. This just an example.
>>> res_id_2.id = res_id_1.id = "smi:local/abcde"
>>> assert(res_id_1 == res_id_2)
ResourceIdentifier instances can be used as dictionary keys.
>>> dictionary = {}
>>> res_id = ResourceIdentifier(id="foo")
>>> dictionary[res_id] = "bar1"
>>> # The same ID can still be used as a key.
>>> dictionary["foo"] = "bar2"
>>> items = sorted(dictionary.items(), key=lambda kv: kv[1])
>>> for k, v in items: # doctest: +ELLIPSIS
... print(repr(k), v)
ResourceIdentifier(id="foo") bar1
...'foo' bar2
"""
# Class (not instance) attribute that keeps track of all resource
# identifier throughout one Python run. Will only store weak references
# and therefore does not interfere with the garbage collection.
# DO NOT CHANGE THIS FROM OUTSIDE THE CLASS.
__resource_id_weak_dict = {} # a nested dict for weak object references
# Use an additional dictionary to track all resource ids.
__resource_id_tracker = collections.defaultdict(int)
# yet another dictionary for keep track of resources id that are not bound
# keys are the id and values are a weak ref to the resource identifier
__unbound_resource_id = weakref.WeakValueDictionary()
[docs] def __init__(self, id=None, prefix="smi:local",
referred_object=None):
# Create a resource id if None is given and possibly use a prefix.
if id is None:
self.fixed = False
self._prefix = prefix
self._uuid = str(uuid4())
elif isinstance(id, ResourceIdentifier):
self.__dict__.update(id.__dict__)
return
else:
self.fixed = True
self.id = id
# Append the referred object in case one is given to the class level
# reference dictionary.
self._object_id = None # the object specific ID
if referred_object is not None:
self.set_referred_object(referred_object)
# Increment the counter for the current resource id.
ResourceIdentifier.__resource_id_tracker[self.id] += 1
[docs] def __del__(self):
if self.id not in ResourceIdentifier.__resource_id_tracker:
return
# Decrement the resource id counter.
ResourceIdentifier.__resource_id_tracker[self.id] -= 1
# If below or equal to zero, delete it and also delete it from the weak
# value dictionary.
if ResourceIdentifier.__resource_id_tracker[self.id] <= 0:
del ResourceIdentifier.__resource_id_tracker[self.id]
try:
del ResourceIdentifier.__resource_id_weak_dict[self.id]
except KeyError:
pass
[docs] @classmethod
def bind_resource_ids(cls):
"""
Bind the unbound ResourceIdentifier instances to referred objects.
Binds all of the unbound ResourceIdentifier instances to the most
recent object assigned to the resource_id. This ensures that all
resource identifiers will return the same object they are bound to
until the object goes out of scope.
"""
for rid_id in list(cls.__unbound_resource_id):
rid = cls.__unbound_resource_id.pop(rid_id, None)
if rid is not None: # if the resource id still exists
rid.get_referred_object() # will bind rid to referred object
[docs] def get_referred_object(self):
"""
Returns the object associated with the resource identifier.
This works as long as at least one ResourceIdentifier with the same
ID as this instance has an associate object.
Will return None if no object could be found.
"""
try:
rdic = ResourceIdentifier.__resource_id_weak_dict[self.id]
except KeyError:
return None
else:
if self._object_id in rdic and rdic[self._object_id]() is not None:
return rdic[self._object_id]()
else: # find last added obj that is not None
return self._get_similar_referred_object()
[docs] def _get_similar_referred_object(self):
"""
Find an object with the same resource_id that is not None and
return it. Also pop all keys that have None values in the
__resource_id_weak_dict
"""
rdic = ResourceIdentifier.__resource_id_weak_dict[self.id]
if self._object_id is not None:
msg = ("The object with identity of: %d no longer exists, "
"returning the most recently created object with a"
" resource id of: %s") % (self._object_id, self.id)
line_number = inspect.currentframe().f_back.f_lineno
warnings.warn_explicit(msg, UserWarning, __file__,
line_number)
self._object_id = None # reset object id
# find a obj that is not None starting at last in ordered dict
for key in list(reversed(rdic)):
obj = rdic[key]()
if obj is not None:
self._object_id = id(obj) # bind object ID
return obj
else: # remove references that are None
rdic.pop(key)
else: # if iter runs out all objects are none; pop rid, return None
ResourceIdentifier.__resource_id_weak_dict.pop(self.id)
msg = ('no object found with resource id %s, returning None'
% self.id)
line_number = inspect.currentframe().f_back.f_lineno
warnings.warn_explicit(msg, UserWarning, __file__, line_number)
return None
[docs] def set_referred_object(self, referred_object, warn=True):
"""
Binds a ResourceIdentifier instance to an object.
If it already a weak reference it will be used, otherwise one will be
created. If the object is None, None will be set.
Will also append self again to the global class level reference list
so everything stays consistent. Warning can be ignored by setting
the warn parameter to False.
"""
self._object_id = id(referred_object) # identity of object
rdic = ResourceIdentifier.__resource_id_weak_dict
# if the resource_id is in the rid_dict
if self.id in rdic and self._object_id not in rdic[self.id]:
last_obj = rdic[self.id][next(reversed(rdic[self.id]))]()
if warn and last_obj is not None and last_obj != referred_object:
msg = ('Warning, binding object to resource ID %s which '
'is not equal to the last object bound to this '
'resource_id') % self.id
line_number = inspect.currentframe().f_back.f_lineno
warnings.warn_explicit(msg, UserWarning, __file__,
line_number)
rdic[self.id][self._object_id] = weakref.ref(referred_object)
else:
rdic[self.id] = collections.OrderedDict()
rdic[self.id][self._object_id] = weakref.ref(referred_object)
[docs] def convert_id_to_quakeml_uri(self, authority_id="local"):
"""
Converts the current ID to a valid QuakeML URI.
Only an invalid QuakeML ResourceIdentifier string it will be converted
to a valid one. Otherwise nothing will happen but after calling this
method the user can be sure that the ID is a valid QuakeML URI.
The resulting ID will be of the form
smi:authority_id/prefix/resource_id
:type authority_id: str, optional
:param authority_id: The base url of the resulting string. Defaults to
``"local"``.
"""
self.id = self.get_quakeml_uri(authority_id=authority_id)
[docs] def get_quakeml_uri(self, authority_id="local"):
"""
Returns the ID as a valid QuakeML URI if possible. Does not
change the ID itself.
>>> res_id = ResourceIdentifier("some_id")
>>> print(res_id.get_quakeml_uri())
smi:local/some_id
>>> # Did not change the actual resource id.
>>> print(res_id.id)
some_id
"""
id = self.id
if str(id).strip() == "":
id = str(uuid4())
regex = r"^(smi|quakeml):[\w\d][\w\d\-\.\*\(\)_~']{2,}/[\w\d\-\." + \
r"\*\(\)_~'][\w\d\-\.\*\(\)\+\?_~'=,;#/&]*$"
result = re.match(regex, str(id))
if result is not None:
return id
id = 'smi:%s/%s' % (authority_id, str(id))
# Check once again just to be sure no weird symbols are stored in the
# ID.
result = re.match(regex, id)
if result is None:
msg = (
"The id '%s' is not a valid QuakeML resource "
"identifier. ObsPy tried modifying it to '%s' but it is still "
"not valid. Please make sure all resource ids are either "
"valid or can be made valid by prefixing them with "
"'smi:<authority_id>/'. Valid ids are specified in the "
"QuakeML manual section 3.1 and in particular exclude colons "
"for the final part." % (self.id, id))
raise ValueError(msg)
return id
[docs] def copy(self):
"""
Returns a copy of the ResourceIdentifier.
>>> res_id = ResourceIdentifier()
>>> res_id_2 = res_id.copy()
>>> print(res_id is res_id_2)
False
>>> print(res_id == res_id_2)
True
"""
return deepcopy(self)
@property
def _object_id(self):
return self.__dict__['_object_id']
@_object_id.setter
def _object_id(self, value):
if value is None: # add instance to unbound dict
self.__class__.__unbound_resource_id[id(self)] = self
else: # binding to object, remove instance from unbound dict
self.__class__.__unbound_resource_id.pop(id(self), None)
self.__dict__['_object_id'] = value
@property
def id(self):
"""
Unique identifier of the current instance.
"""
if self.fixed:
return self.__dict__.get("id")
else:
id = self.prefix
if not id.endswith("/"):
id += "/"
id += self.uuid
return id
@id.deleter
def id(self):
msg = "The resource id cannot be deleted."
raise Exception(msg)
@id.setter
def id(self, value):
self.fixed = True
# XXX: no idea why I had to add bytes for PY2 here
if not isinstance(value, (str, bytes)):
msg = "attribute id needs to be a string."
raise TypeError(msg)
self.__dict__["id"] = value
@property
def prefix(self):
return self._prefix
@prefix.deleter
def prefix(self):
self._prefix = ""
@prefix.setter
def prefix(self, value):
if not isinstance(value, (str, native_str)):
msg = "prefix id needs to be a string."
raise TypeError(msg)
self._prefix = value
@property
def uuid(self):
return self._uuid
@uuid.deleter
def uuid(self):
"""
Deleting is uuid hash is forbidden and will not work.
"""
msg = "The uuid cannot be deleted."
raise Exception(msg)
@uuid.setter
def uuid(self, value): # @UnusedVariable
"""
Setting is uuid hash is forbidden and will not work.
"""
msg = "The uuid cannot be set manually."
raise Exception(msg)
@property
def resource_id(self):
return self.id
@resource_id.deleter
def resource_id(self):
del self.id
@resource_id.setter
def resource_id(self, value):
self.id = value
[docs] def __str__(self):
return self.id
[docs] def _repr_pretty_(self, p, cycle):
p.text(str(self))
[docs] def __repr__(self):
return 'ResourceIdentifier(id="%s")' % self.id
[docs] def __eq__(self, other):
if self.id == other:
return True
if not isinstance(other, ResourceIdentifier):
return False
if self.id == other.id:
return True
return False
[docs] def __ne__(self, other):
return not self.__eq__(other)
[docs] def __hash__(self):
"""
Uses the same hash as the resource id. This means that class instances
can be used in dictionaries and other hashed types.
Both the object and it's id can still be independently used as
dictionary keys.
"""
# "Salt" the hash with a string so the hash of the object and a
# string identical to the id can both be used as individual
# dictionary keys.
return hash("RESOURCE_ID") + self.id.__hash__()
[docs] def regenerate_uuid(self):
"""
Regenerates the uuid part of the ID. Does nothing for resource
identifiers with a user-set, fixed id.
"""
self._uuid = str(uuid4())
__CreationInfo = _event_type_class_factory(
"__CreationInfo",
class_attributes=[("agency_id", str),
("agency_uri", ResourceIdentifier),
("author", str),
("author_uri", ResourceIdentifier),
("creation_time", UTCDateTime),
("version", str)])
[docs]class CreationInfo(__CreationInfo):
"""
CreationInfo is used to describe creation metadata (author, version, and
creation time) of a resource.
:type agency_id: str, optional
:param agency_id: Designation of agency that published a resource.
:type agency_uri: :class:`~obspy.core.event.base.ResourceIdentifier`,
optional
:param agency_uri: Resource Identifier of the agency that published a
resource.
:type author: str, optional
:param author: Name describing the author of a resource.
:type author_uri: :class:`~obspy.core.event.base.ResourceIdentifier`,
optional
:param author_uri: Resource Identifier of the author of a resource.
:type creation_time: :class:`~obspy.core.utcdatetime.UTCDateTime`, optional
:param creation_time: Time of creation of a resource.
:type version: str, optional
:param version: Version string of a resource
>>> info = CreationInfo(author="obspy.org", version="0.0.1")
>>> print(info)
CreationInfo(author='obspy.org', version='0.0.1')
.. note::
For handling additional information not covered by the QuakeML
standard and how to output it to QuakeML see the
:ref:`ObsPy Tutorial <quakeml-extra>`.
"""
__TimeWindow = _event_type_class_factory(
"__TimeWindow",
class_attributes=[("begin", float),
("end", float),
("reference", UTCDateTime)])
[docs]class TimeWindow(__TimeWindow):
"""
Describes a time window for amplitude measurements, given by a central
point in time, and points in time before and after this central point. Both
points before and after may coincide with the central point.
:type begin: float
:param begin: Absolute value of duration of time interval before reference
point in time window. The value may be zero, but not negative. Unit: s
:type end: float
:param end: Absolute value of duration of time interval after reference
point in time window. The value may be zero, but not negative. Unit: s
:type reference: :class:`~obspy.core.utcdatetime.UTCDateTime`
:param reference: Reference point in time ("central" point).
.. note::
For handling additional information not covered by the QuakeML
standard and how to output it to QuakeML see the
:ref:`ObsPy Tutorial <quakeml-extra>`.
"""
__CompositeTime = _event_type_class_factory(
"__CompositeTime",
class_attributes=[("year", int, ATTRIBUTE_HAS_ERRORS),
("month", int, ATTRIBUTE_HAS_ERRORS),
("day", int, ATTRIBUTE_HAS_ERRORS),
("hour", int, ATTRIBUTE_HAS_ERRORS),
("minute", int, ATTRIBUTE_HAS_ERRORS),
("second", float, ATTRIBUTE_HAS_ERRORS)])
[docs]class CompositeTime(__CompositeTime):
"""
Focal times differ significantly in their precision. While focal times of
instrumentally located earthquakes are estimated precisely down to seconds,
historic events have only incomplete time descriptions. Sometimes, even
contradictory information about the rupture time exist. The CompositeTime
type allows for such complex descriptions. If the specification is given
with no greater accuracy than days (i.e., no time components are given),
the date refers to local time. However, if time components are given, they
have to refer to UTC.
:type year: int
:param year: Year or range of years of the event's focal time.
:type year_errors: :class:`~obspy.core.event.base.QuantityError`
:param year_errors: AttribDict containing error quantities.
:type month: int
:param month: Month or range of months of the event’s focal time.
:type month_errors: :class:`~obspy.core.event.base.QuantityError`
:param month_errors: AttribDict containing error quantities.
:type day: int
:param day: Day or range of days of the event’s focal time.
:type day_errors: :class:`~obspy.core.event.base.QuantityError`
:param day_errors: AttribDict containing error quantities.
:type hour: int
:param hour: Hour or range of hours of the event’s focal time.
:type hour_errors: :class:`~obspy.core.event.base.QuantityError`
:param hour_errors: AttribDict containing error quantities.
:type minute: int
:param minute: Minute or range of minutes of the event’s focal time.
:type minute_errors: :class:`~obspy.core.event.base.QuantityError`
:param minute_errors: AttribDict containing error quantities.
:type second: float
:param second: Second and fraction of seconds or range of seconds with
fraction of the event’s focal time.
:type second_errors: :class:`~obspy.core.event.base.QuantityError`
:param second_errors: AttribDict containing error quantities.
>>> print(CompositeTime(2011, 1, 1))
CompositeTime(year=2011, month=1, day=1)
>>> # Can also be instantiated with the uncertainties.
>>> print(CompositeTime(year=2011, year_errors={"uncertainty":1}))
CompositeTime(year=2011 [uncertainty=1])
.. note::
For handling additional information not covered by the QuakeML
standard and how to output it to QuakeML see the
:ref:`ObsPy Tutorial <quakeml-extra>`.
"""
__Comment = _event_type_class_factory(
"__Comment",
class_attributes=[("text", str),
("resource_id", ResourceIdentifier),
("creation_info", CreationInfo)])
__WaveformStreamID = _event_type_class_factory(
"__WaveformStreamID",
class_attributes=[("network_code", str),
("station_code", str),
("channel_code", str),
("location_code", str),
("resource_uri", ResourceIdentifier)])
__ConfidenceEllipsoid = _event_type_class_factory(
"__ConfidenceEllipsoid",
class_attributes=[("semi_major_axis_length", float),
("semi_minor_axis_length", float),
("semi_intermediate_axis_length", float),
("major_axis_plunge", float),
("major_axis_azimuth", float),
("major_axis_rotation", float)])
[docs]class ConfidenceEllipsoid(__ConfidenceEllipsoid):
"""
This class represents a description of the location uncertainty as a
confidence ellipsoid with arbitrary orientation in space. See the QuakeML
documentation for the full details
:param semi_major_axis_length: Largest uncertainty, corresponding to the
semi-major axis of the confidence ellipsoid. Unit: m
:param semi_minor_axis_length: Smallest uncertainty, corresponding to the
semi-minor axis of the confidence ellipsoid. Unit: m
:param semi_intermediate_axis_length: Uncertainty in direction orthogonal
to major and minor axes of the confidence ellipsoid. Unit: m
:param major_axis_plunge: Plunge angle of major axis of confidence
ellipsoid. Corresponds to Tait-Bryan angle φ. Unit: deg
:param major_axis_azimuth: Azimuth angle of major axis of confidence
ellipsoid. Corresponds to Tait-Bryan angle ψ. Unit: deg
:param major_axis_rotation: This angle describes a rotation about the
confidence ellipsoid’s major axis which is required to define the
direction of the ellipsoid’s minor axis. Corresponds to Tait-Bryan
angle θ.
Unit: deg
.. note::
For handling additional information not covered by the QuakeML
standard and how to output it to QuakeML see the
:ref:`ObsPy Tutorial <quakeml-extra>`.
"""
__DataUsed = _event_type_class_factory(
"__DataUsed",
class_attributes=[("wave_type", DataUsedWaveType),
("station_count", int),
("component_count", int),
("shortest_period", float),
("longest_period", float)])
[docs]class DataUsed(__DataUsed):
"""
The DataUsed class describes the type of data that has been used for a
moment-tensor inversion.
:type wave_type: str
:param wave_type: Type of waveform data.
See :class:`~obspy.core.event.header.DataUsedWaveType` for allowed
values.
:type station_count: int, optional
:param station_count: Number of stations that have contributed data of the
type given in wave_type.
:type component_count: int, optional
:param component_count: Number of data components of the type given in
wave_type.
:type shortest_period: float, optional
:param shortest_period: Shortest period present in data. Unit: s
:type longest_period: float, optional
:param longest_period: Longest period present in data. Unit: s
.. note::
For handling additional information not covered by the QuakeML
standard and how to output it to QuakeML see the
:ref:`ObsPy Tutorial <quakeml-extra>`.
"""
if __name__ == '__main__':
import doctest
doctest.testmod(exclude_empty=True)