def get_nearest_obj(self,ra_in,dec_in,size=2):
a=self.BC.assets
#find a bounding box to speed up search (size depended on search regon)
if (ra_in<60) or (ra_in>308.5):
if (dec_in+size)<-1:
size=abs(dec_in+1)
elif (dec_in-size)>1:
size=abs(dec_in-1)
else:
if (dec_in+size)<-3:
size=abs(dec_in+3)
elif (dec_in-size)>76:
size=abs(dec_in-76)
box=geo_bounding_box(ra_in,dec_in,size=size)
#this distance is correct up to multiplications factors (order is correct)
dis=func.asin(func.sqrt(func.power(func.sin(0.5*func.radians(dec_in-a.dec)),2) + func.cos(func.radians(dec_in))*func.cos(func.radians(a.dec))*func.power(func.sin(.5*func.radians(ra_in-a.ra)),2)))
if isinstance(box[0],list):
#the search wraps around 360
result=self.session.query(a.name,a.id,a.ra,a.dec,a.location).filter(between(a.dec,box[1][0],box[1][1])).filter((between(a.ra,box[0][0][0],box[0][0][1]))|(between(a.ra,box[0][1][0],box[0][1][1]))).order_by(dis.asc()).first()
else:
result=self.session.query(a.name,a.id,a.ra,a.dec,a.location).filter(between(a.ra,box[0][0],box[0][1])).filter(between(a.dec,box[1][0],box[1][1])).order_by(dis.asc()).first()
#result=self.session.query(a.name,a.id,a.ra,a.dec,a.location).order_by(dis.asc()).first()
if result is None:
result=self.get_nearest_obj(ra_in,dec_in,size=size+20)
return result
def average_composition(queryset, **kwargs):
"""
The average oxide composition of a queryset, calculated in
one of several ways (either wt%, molar%, or normalized wt% abundances).
Returns the same result as the partial functions above, but is a bit
more descriptive.
:param type: Options
- weight Oxide weight % (default)
- normalized_weight Oxide weight % (normalized to 100% abundances)
- molar Oxide molar %
:param uncertainties: Whether to include uncertainties (default FALSE)
"""
mapping = dict(weight=ProbeDatum.weight_percent,
normalized_weight=ProbeDatum.weight_percent * 100 /
ProbeMeasurement.oxide_total,
molar=ProbeDatum.molar_percent)
type = kwargs.pop("type", "weight")
try:
quantity = mapping[type]
except KeyError:
raise ArgumentError("Parameter `type` must be either"
"`weight`,`molar`, or `normalized_weight`")
try:
q = queryset.with_entities(ProbeMeasurement.id)
filt = ProbeDatum.measurement_id.in_(q)
except AttributeError:
filt = ProbeDatum.measurement_id == queryset.id
average = func.avg(quantity)
qvars = [ProbeDatum._oxide, average.label('avg')]
uncertainties = kwargs.pop("uncertainties", False)
if uncertainties:
e = ProbeDatum.error
sum_ = func.sum(func.pow(e / 100 * quantity, 2))
std = func.sqrt(sum_) / func.count("*")
qvars.append(std.label("analytic_std"))
std2 = func.stddev(quantity)
qvars.append(std2.label("sample_std"))
data = db.session.query(ProbeDatum)\
.join(ProbeMeasurement)\
.filter(filt)\
.with_entities(*tuple(qvars))\
.group_by(ProbeDatum._oxide)\
.all()
if uncertainties:
try:
return {o: ufloat(n, a + s) for o, n, a, s in data}
except TypeError:
return {o: ufloat(n, a) for o, n, a, s in data}
else:
return {o: n for o, n in data}
conn = engine.connect()
c = [
## date/time functions ##
#func.timeofday(), # for postgresql
func.localtime(),
func.current_timestamp(),
#func.date_part("month", func.now()), # for postgresql
func.now(),
## mathematical functions ##
func.pow(4, 2),
func.sqrt(441),
func.pi(),
func.floor(func.pi()),
func.ceil(func.pi()),
## string functions ##
func.lower("ABC"),
func.upper("abc"),
func.length("abc"),
func.trim(" ab c "),
#func.chr(65), # for postgresql
]
s = select(c)
rs = conn.execute(s)
print(rs.keys())
def get_nearest_obj(self,ra_in,dec_in,size=5):
a=self.BC.assets
#this distance is correct up to multiplications factors (order is correct)
dis=func.asin(func.sqrt(func.power(func.sin(0.5*func.radians(dec_in-a.dec)),2) + func.cos(func.radians(dec_in))*func.cos(func.radians(a.dec))*func.power(func.sin(.5*func.radians(ra_in-a.ra)),2)))
result=self.session.query(a.name,a.id,a.ra,a.dec,a.location).order_by(dis.asc()).first()
return result