def compile(self):
"""compile phase (calculation from MAST DiffusionCoefficient.py)"""
from pymatgen.symmetry.analyzer import SpacegroupAnalyzer
from pymatgen.matproj.rest import MPRester
indirs = glob.glob(os.path.join(self.rootdir, "*CuCu*")) + [
fn for fn in glob.glob(os.path.join(self.rootdir, "*Cu*"))
if not fnmatch.fnmatch(fn, "*CuCu*")
]
mp_id, E0, nw = None, None, 5 # five-frequency model
df = None
for idx,indir in enumerate(indirs):
element = os.path.basename(indir).split('_')[2][2:]
struct = self.find_entry_for_directory(
os.path.join(indir, 'perfect_stat*'), oszicar=False
).structure
reduced = SpacegroupAnalyzer(struct, symprec=1e-2).get_primitive_standard_structure()
if idx == 0:
ENDPOINT = "https://www.materialsproject.org/rest"
with MPRester(endpoint=ENDPOINT) as m:
matches = m.find_structure(reduced)
if len(matches) == 1: mp_id = matches[0]
else: raise ValueError(
"found {} matching structure(s) in MP and hence cannot "
"assign structure in {}.".format(len(matches), indir)
)
a = reduced.lattice.abc[0] * math.sqrt(2) * 10**(-8)
enebarr = np.array([ self.get_barrier(indir, i) for i in range(nw) ], dtype=float)
if idx == 0: E0 = min(enebarr[~np.isnan(enebarr)])
else: enebarr[0] = E0 # TODO: is this correct?
v = np.array([ self.get_attempt_frequency(indir, i) for i in range(nw) ])
v[0], HVf, kB, f0 = 1.0, 0.4847, 8.6173324e-5, 0.7815 # TODO set v[0] to 1.0? HVf dynamic how?
t, tempstep, tempend = 0.0, 0.1, 2.0 # default temperature range
x, y = [], []
while t < tempend + tempstep:
v *= np.exp(-enebarr/kB/1e3*t)
alpha = v[4]/v[0]
F_num = 10*np.power(alpha,4) + 180.5*np.power(alpha,3)
F_num += 927*np.power(alpha,2) + 1341*alpha
F_denom = 2*np.power(alpha,4) + 40.2*np.power(alpha,3)
F_denom += 254*np.power(alpha,2) + 597*alpha + 435
FX = 1-(1.0/7.0)*(F_num/F_denom)
f2 = 1+3.5*FX*(v[3]/v[1])
f2 /= 1+(v[2]/v[1]) + 3.5*FX*(v[3]/v[1])
cV = np.exp(-HVf/kB/1e3*t) if t > 0. else 1.
D = f0*cV*a**2*v[0] * f2/f0 * v[4]/v[0] * v[2]/v[3]
x.append(t)
y.append(D)
t += tempstep
if df is None: df = DataFrame(np.array(x), columns=['1/T'])
df.loc[:,element] = Series(np.array(y), index=df.index)
# generate (physical) MPFile
mpfile = MPFile()
mpfile.add_data_table(mp_id, df, 'data')
#print mpfile
mpfile.write_file(os.path.join(self.rootdir, 'output.mpf'))
def get_contributions(self, limit=20):
docs = self.query_contributions(
projection={'_id': 1, 'identifier': 1, 'content': 1}, limit=limit) # use URL for all data
if not docs:
raise Exception('No contributions found for CarrierTransport Explorer!')
data = []
columns = ['##'.join(['general', sk]) for sk in ['mp-id', 'cid', 'formula']]
keys, subkeys = ['<mₑᶜᵒⁿᵈ>', '<S>', '<σ>', '<S²σ>'], ['n', 'p']
columns += ['##'.join([k, sk]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['extra_data']['pretty_formula']]
row += [
contrib['data'].get(k[1:-1], {}).get(sk, {}).get('<ε>', 'n.a. mₑ')
for k in keys for sk in subkeys
]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, doping):
dopings = ['n', 'p']
if doping not in dopings:
raise Exception('doping has to be n or p!')
docs = self.query_contributions(projection={
'_id': 1,
'mp_cat_id': 1,
'content': 1
})
if not docs:
raise Exception('No contributions found for Boltztrap Explorer!')
data = []
columns = [
'##'.join(['general', sk]) for sk in ['mp-id', 'cid', 'formula']
]
keys, subkeys = [u'mₑᶜᵒⁿᵈ', u"Seebeck"], [u"e₁", u"e₂", u"e₃"]
columns += ['##'.join([k, sk]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['pretty_formula']]
row += [
contrib[k].get(doping, {}).get(sk, '') for k in keys
for sk in subkeys
]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for MpWorkshop2017 Explorer!')
data = []
columns = ['mp-id', 'contribution']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
for k in contrib.keys():
if k not in columns:
columns.append(k)
row = [mp_id, cid_url]
for col in columns[2:]:
row.append(contrib.get(col, ''))
data.append([mp_id, row])
# enforce equal row lengths
ncols = len(columns)
for entry in data:
n = len(entry[1])
if n != ncols:
entry[1] += [''] * (ncols - n)
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
data = []
columns = [
'mp-id', 'contribution', 'formula', 'CIF', 'dISO', 'etaQ', 'QCC',
'B'
]
docs = self.query_contributions(projection={
'_id': 1,
'mp_cat_id': 1,
'content': 1
})
if not docs:
raise Exception('No contributions found for Dibbs Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['formula']]
cif_url = ''
structures = mpfile.sdata.get(mp_id)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif',
structures.keys()[0]
])
row.append(cif_url)
row += [contrib['data'][col] for col in columns[-4:]]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
data = []
columns = ['mp-id', 'contribution', 'formula', 'CIF', 'dISO', 'etaQ', 'QCC', 'B']
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
)
if not docs:
raise Exception('No contributions found for Dibbs Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['formula']]
cif_url = ''
structures = mpfile.sdata.get(mp_id)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif', structures.keys()[0]
])
row.append(cif_url)
row += [contrib['data'][col] for col in columns[-4:]]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
data = []
columns = ['mp-id', 'contribution', 'efermi', 'ehull', 'bandgap']
docs = self.query_contributions(criteria=self.tam_perovskites_query,
projection={
'_id': 1,
'mp_cat_id': 1,
'content.efermi': 1,
'content.ehull': 1,
'content.bandgap': 1
})
if not docs:
raise Exception(
'No contributions found for TamPerovskites Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id']
])
row = [
mp_id, cid_url, contrib['efermi'], contrib['ehull'],
contrib['bandgap']
]
data.append((mp_id, row))
return DataFrame.from_items(data, orient='index', columns=columns)
def get_contributions(self):
docs = self.query_contributions(projection={
'_id': 1,
'mp_cat_id': 1,
'content': 1
})
if not docs:
raise Exception('No contributions found for SLAC MoSe2 Explorer!')
import pandas as pd
doc = docs[0] # there should be only one for MoSe2
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
response = {}
response['graphs'] = [
plot for key, plot in mpfile.gdata[mp_id].items() if 'pump' in key
]
tdata = mpfile.tdata[mp_id]
name = tdata.keys()[-1]
table = tdata[name]
table = table.apply(pd.to_numeric)
table.dropna(how='any', inplace=True)
response['traces'] = []
for col in table.columns[1:]:
response['traces'].append({
'x': table[table.columns[0]].values,
'y': table[col].values,
'name': col
})
return response
def get_contributions(self):
projection = {'_id': 1, 'identifier': 1, 'content.data': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception(
'No contributions found for RedoxThermoCsp Explorer!')
data, columns = [], ['identifier', 'contribution']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = self.get_cid_url(doc)
row = [identifier, cid_url]
scope = []
for key, value in contrib.iterate():
level, key = key
level_reduction = bool(level < len(scope))
if level_reduction:
del scope[level:]
if value is None:
scope.append(key)
else:
col = '##'.join(scope + [key]).replace('_', ' ')
if col not in columns:
columns.append(col)
row.append(value)
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, phase=None):
data = []
phase_query_key = {'$exists': 1} if phase is None else phase
columns = ['mp-id', 'contribution', 'formula']
if phase is None:
columns.append('phase')
columns += ['dH (formation)', 'dH (hydration)', 'GS?', 'CIF']
for doc in self.query_contributions(criteria={
'project': {
'$in': ['LBNL', 'MIT']
},
'content.info.Phase':
phase_query_key
},
projection={
'_id': 1,
'mp_cat_id': 1,
'content': 1
}):
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
info = mpfile.hdata[mp_id]['info']
row = [mp_id, get_short_object_id(doc['_id']), info['Formula']]
if phase is None:
row.append(info['Phase'])
row += [info['dHf'], info['dHh'], info['GS'], 'TODO']
# TODO URLs for mp_id and cid
data.append((mp_id, row))
return DataFrame.from_items(data, orient='index', columns=columns)
def get_provenance(self):
for doc in self.query_contributions(criteria={
'project': {
'$in': ['LBNL', 'MIT']
},
'content.Authors': {
'$exists': 1
},
'content.Title': {
'$exists': 1
},
'content.Reference': {
'$exists': 1
}
},
projection={
'_id': 1,
'mp_cat_id': 1,
'content.Authors': 1,
'content.Reference': 1,
'content.Title': 1
}):
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
return mpfile.hdata[mp_id]
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
)
if not docs:
raise Exception('No contributions found for SLAC MoSe2 Explorer!')
import pandas as pd
doc = docs[0] # there should be only one for MoSe2
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
response = {}
response['graphs'] = [
plot for key, plot in mpfile.gdata[mp_id].items() if 'pump' in key
]
tdata = mpfile.tdata[mp_id]
name = tdata.keys()[-1]
table = tdata[name]
table = table.apply(pd.to_numeric)
table.dropna(how='any', inplace=True)
response['traces'] = []
for col in table.columns[1:]:
response['traces'].append({
'x': table[table.columns[0]].values,
'y': table[col].values,
'name': col
})
return response
def get_contributions(self):
projection = {'_id': 1, 'mp_cat_id': 1, 'content': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for ALS Beamline Explorer!')
data = []
columns = ['formula', 'cid']
keys = RecursiveDict([
('composition', ['Co', 'Cu', 'Ce']),
#('position', ['x', 'y']),
('XAS', ['min', 'max']),
('XMCD', ['min', 'max'])
])
columns += ['##'.join([k, sk]) for k, subkeys in keys.items() for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = self.get_cid_url(doc)
row = [identifier, cid_url]
row += [contrib[k][sk] for k, subkeys in keys.items() for sk in subkeys]
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, limit=20):
docs = self.query_contributions(projection={
'_id': 1,
'identifier': 1,
'content': 1
},
limit=limit) # use URL for all data
if not docs:
raise Exception(
'No contributions found for CarrierTransport Explorer!')
data = []
columns = [
'##'.join(['general', sk]) for sk in ['mp-id', 'cid', 'formula']
]
keys, subkeys = ['<mₑᶜᵒⁿᵈ>', '<S>', '<σ>', '<S²σ>'], ['n', 'p']
columns += ['##'.join([k, sk]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['extra_data']['pretty_formula']]
row += [
contrib['data'].get(k[1:-1], {}).get(sk,
{}).get('<ε>', 'n.a. mₑ')
for k in keys for sk in subkeys
]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_detail_data(self,doping):
'''
function to get doping and temperature related to the max values
and the three eigenvalues of the effective mass tensor
'''
dopings = ['n', 'p']
if doping not in dopings:
raise Exception('doping has to be n or p!')
docs = self.query_contributions(projection={'_id': 1, 'identifier': 1, 'content': 1})
if not docs:
raise Exception('No contributions found for CarrierTransport Explorer!')
data = []
columns = ['mp-id', 'cid', 'formula', u'mₑᶜᵒⁿᵈ', u"Sₘₐₓ", u"σₘₐₓ", u"κₑ₋ₘᵢₙ"]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
details = {k:[contrib[k][doping]['temperature'],contrib[k][doping]['doping']] for k in columns[4:]}
data.append((mp_id, details))
return data
class PerovskitesDiffusionRester(MPContribsRester):
"""PerovskitesDiffusion-specific convenience functions to interact with MPContribs REST interface"""
mpfile = MPFile.from_file(os.path.join(
os.path.dirname(__file__), '..', 'mpfile_init.txt'
))
query = {'content.doi': mpfile.hdata.general['doi']}
provenance_keys = [k for k in mpfile.hdata.general.keys() if k != 'google_sheet']
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1, 'content.abbreviations': 1}
)
if not docs:
raise Exception('No contributions found for PerovskitesDiffusion Explorer!')
data, columns = [], None
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
if columns is None:
columns = ['mp-id', 'contribution'] + contrib.keys()
row = [mp_id, cid_url] + contrib.values()
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_abbreviations(self):
return self.get_global_hierarchical_data(['abbreviations']).get('abbreviations')
def get_contributions(self):
projection = {'_id': 1, 'identifier': 1, 'content': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for ALS Beamline Explorer!')
data = []
columns = ['formula', 'cid']
keys = RecursiveDict([
('composition', ['Co', 'Cu', 'Ce']),
#('position', ['x', 'y']),
('XAS', ['min', 'max']),
('XMCD', ['min', 'max'])
])
columns += ['##'.join([k, sk]) for k, subkeys in keys.items() for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = self.get_cid_url(doc)
row = [identifier, cid_url]
row += [contrib[k][sk] for k, subkeys in keys.items() for sk in subkeys]
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
projection = {'_id': 1, 'identifier': 1, 'content.data': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for RedoxThermoCsp Explorer!')
data, columns = [], ['identifier', 'contribution']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = self.get_cid_url(doc)
row = [identifier, cid_url]
scope = []
for key, value in contrib.iterate():
level, key = key
level_reduction = bool(level < len(scope))
if level_reduction:
del scope[level:]
if value is None:
scope.append(key)
else:
col = '##'.join(scope + [key]).replace('_', ' ')
if col not in columns:
columns.append(col)
row.append(value)
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_ionic_radii(self):
data = []
columns = ['mp-id', 'cid', 'species', 'charge', u'rᵢₒₙ', 'HS/LS', 'CN']
docs = self.query_contributions(
criteria={'content.title': 'Ionic Radii'},
projection={'_id': 1, 'identifier': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for RedoxThermoCsp Ionic Radii!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials', doc['_id']
])
nrows = sum(1 for v in contrib.values() if isinstance(v, dict))
rows = [[identifier, cid_url] for i in range(nrows)]
for col in columns[2:]:
for irow, row in enumerate(rows):
val = contrib.get(col)
if val is None:
val = contrib[str(irow)].get(col, '-')
row.append(val)
for row in rows:
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for SWF Explorer!')
data = []
columns = ['formula', 'contribution']
ncols = 9
for doc in docs:
mpfile = MPFile.from_contribution(doc)
formula = mpfile.ids[0]
contrib = mpfile.hdata[formula].get('data')
if contrib is None:
continue
cid_url = self.get_cid_url(doc)
for k in contrib.keys():
if k not in columns:
columns.append(k)
row = [formula, cid_url]
for col in columns[2:]:
row.append(contrib.get(col, ''))
n = len(row)
if n < ncols:
row += [''] * (ncols - n)
data.append((formula, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
data = []
columns = ['identifier', 'contribution', 'composition', 'CIF']
docs = self.query_contributions(
criteria={'content.title': {'$ne': 'Ionic Radii'}},
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
)
if not docs:
raise Exception('No contributions found for DlrVieten Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]
cid_url = self.get_cid_url(doc)
row = [identifier, cid_url, contrib['composition']]
cif_url = ''
structures = mpfile.sdata.get(identifier)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif', structures.keys()[0]
])
row.append(cif_url)
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_ionic_radii(self):
data = []
columns = ['mp-id', 'cid', 'species', 'charge', u'rᵢₒₙ', 'HS/LS', 'CN']
docs = self.query_contributions(
criteria={'content.title': 'Ionic Radii'},
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for DlrVieten Ionic Radii!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials', doc['_id']
])
nrows = sum(1 for v in contrib.values() if isinstance(v, dict))
rows = [[identifier, cid_url] for i in range(nrows)]
for col in columns[2:]:
for irow, row in enumerate(rows):
val = contrib.get(col)
if val is None:
val = contrib[str(irow)].get(col, '-')
row.append(val)
for row in rows:
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_uwsi2_contributions(self):
data = []
for doc in self.query_contributions(criteria={
'project': {
'$in': ['LBNL', 'UW-Madison']
},
'content.figshare_id': {
'$exists': 1
}
},
projection={
'_id': 1,
'mp_cat_id': 1,
'content': 1
}):
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
table = mpfile.tdata[mp_id]['data_D0_Q']
anums = [self.z[el] for el in table['element']]
table.insert(0, 'Z', Series(anums, index=table.index))
table.sort_values('Z', inplace=True)
table.reset_index(drop=True, inplace=True)
hdata = mpfile.hdata[mp_id]
data.append({
'mp_id': mp_id,
'cid': doc['_id'],
'short_cid': get_short_object_id(doc['_id']),
'formula': hdata['formula'],
'table': table
})
return data
class MpWorkshop2017Rester(MPContribsRester):
"""MpWorkshop2017-specific convenience functions to interact with MPContribs REST interface"""
mpfile = MPFile.from_file(os.path.join(
os.path.dirname(__file__), '..', 'mpfile_init.txt'
))
query = {'content.source': mpfile.hdata.general['source']}
provenance_keys = [k for k in mpfile.hdata.general.keys() if k != 'google_sheet']
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for MpWorkshop2017 Explorer!')
data = []
columns = ['mp-id', 'contribution']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
for k in contrib.keys():
if k not in columns:
columns.append(k)
row = [mp_id, cid_url]
for col in columns[2:]:
row.append(contrib.get(col, ''))
data.append([mp_id, row])
# enforce equal row lengths
ncols = len(columns)
for entry in data:
n = len(entry[1])
if n != ncols:
entry[1] += [''] * (ncols - n)
return Table.from_items(data, orient='index', columns=columns)
def get_graphs(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
)
if not docs:
raise Exception('No contributions found for MpWorkshop2017 Explorer!')
graphs = {}
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
graphs[mp_id] = mpfile.gdata[mp_id]
return graphs
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
)
if not docs:
raise Exception('No contributions found for JarvisDft Explorer!')
data, data_jarvis = [], []
general_columns = ['mp-id', 'cid', 'formula']
keys, subkeys = ['NUS', 'JARVIS'], ['id', 'Eₓ', 'CIF']
columns = general_columns + ['##'.join([k, sk]) for k in keys for sk in subkeys]
columns_jarvis = general_columns + ['id', 'E', 'ΔE|optB88vdW', 'ΔE|mbj', 'CIF']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
structures = mpfile.sdata.get(mp_id)
cif_urls = {}
for k in keys:
cif_urls[k] = ''
name = '{}_{}'.format(contrib['formula'], k)
if structures.get(name) is not None:
cif_urls[k] = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif', name
])
row = [mp_id, cid_url, contrib['formula']]
for k in keys:
for sk in subkeys:
if sk == subkeys[-1]:
row.append(cif_urls[k])
else:
cell = contrib.get(k, {sk: ''})[sk]
row.append(cell)
data.append((mp_id, row))
row_jarvis = [mp_id, cid_url, contrib['formula']]
for k in columns_jarvis[len(general_columns):]:
if k == columns_jarvis[-1]:
row_jarvis.append(cif_urls[keys[1]])
else:
row_jarvis.append(contrib.get(keys[1], {k: ''}).get(k, ''))
if row_jarvis[3]:
data_jarvis.append((mp_id, row_jarvis))
return [
Table.from_items(data, orient='index', columns=columns),
Table.from_items(data_jarvis, orient='index', columns=columns_jarvis)
]
def compile(self):
"""compile phase (calculation from MAST DiffusionCoefficient.py)"""
from pymatgen.symmetry.analyzer import SpacegroupAnalyzer
from pymatgen.matproj.rest import MPRester
indirs = glob.glob(os.path.join(self.rootdir, "*CuCu*")) + [
fn for fn in glob.glob(os.path.join(self.rootdir, "*Cu*"))
if not fnmatch.fnmatch(fn, "*CuCu*")
]
mp_id, E0, nw = None, None, 5 # five-frequency model
df = None
for idx, indir in enumerate(indirs):
element = os.path.basename(indir).split('_')[2][2:]
struct = self.find_entry_for_directory(os.path.join(
indir, 'perfect_stat*'),
oszicar=False).structure
reduced = SpacegroupAnalyzer(
struct, symprec=1e-2).get_primitive_standard_structure()
if idx == 0:
ENDPOINT = "https://www.materialsproject.org/rest"
with MPRester(endpoint=ENDPOINT) as m:
matches = m.find_structure(reduced)
if len(matches) == 1: mp_id = matches[0]
else:
raise ValueError(
"found {} matching structure(s) in MP and hence cannot "
"assign structure in {}.".format(
len(matches), indir))
a = reduced.lattice.abc[0] * math.sqrt(2) * 10**(-8)
enebarr = np.array([self.get_barrier(indir, i) for i in range(nw)],
dtype=float)
if idx == 0: E0 = min(enebarr[~np.isnan(enebarr)])
else: enebarr[0] = E0 # TODO: is this correct?
v = np.array(
[self.get_attempt_frequency(indir, i) for i in range(nw)])
v[0], HVf, kB, f0 = 1.0, 0.4847, 8.6173324e-5, 0.7815 # TODO set v[0] to 1.0? HVf dynamic how?
t, tempstep, tempend = 0.0, 0.1, 2.0 # default temperature range
x, y = [], []
while t < tempend + tempstep:
v *= np.exp(-enebarr / kB / 1e3 * t)
alpha = v[4] / v[0]
F_num = 10 * np.power(alpha, 4) + 180.5 * np.power(alpha, 3)
F_num += 927 * np.power(alpha, 2) + 1341 * alpha
F_denom = 2 * np.power(alpha, 4) + 40.2 * np.power(alpha, 3)
F_denom += 254 * np.power(alpha, 2) + 597 * alpha + 435
FX = 1 - (1.0 / 7.0) * (F_num / F_denom)
f2 = 1 + 3.5 * FX * (v[3] / v[1])
f2 /= 1 + (v[2] / v[1]) + 3.5 * FX * (v[3] / v[1])
cV = np.exp(-HVf / kB / 1e3 * t) if t > 0. else 1.
D = f0 * cV * a**2 * v[0] * f2 / f0 * v[4] / v[0] * v[2] / v[3]
x.append(t)
y.append(D)
t += tempstep
if df is None: df = DataFrame(np.array(x), columns=['1/T'])
df.loc[:, element] = Series(np.array(y), index=df.index)
# generate (physical) MPFile
mpfile = MPFile()
mpfile.add_data_table(mp_id, df, 'data')
#print mpfile
mpfile.write_file(os.path.join(self.rootdir, 'output.mpf'))
def get_provenance(self):
provenance_keys = ['authors', 'abbreviations']
projection = {'_id': 1, 'mp_cat_id': 1}
for key in provenance_keys:
projection['content.' + key] = 1
docs = self.query_contributions(criteria=self.tam_perovskites_query,
projection=projection)
if not docs:
raise Exception('No contributions found for DTU Explorer!')
mpfile = MPFile.from_contribution(docs[0])
mp_id = mpfile.ids[0]
return mpfile.hdata[mp_id]
def table(request, db_type=None, mdb=None):
try:
page = int(request.GET.get('page', '1'))
crit = {'project': 'carrier_transport'}
search = request.GET.get('q')
if search is not None:
crit.update({'content.extra_data.pretty_formula': {'$regex': search}})
proj = {'content.data': 1, 'content.extra_data': 1, 'identifier': 1}
total_count = mdb.contrib_ad.query_contributions(crit).count()
last_id = None
if page is not None and page > 1:
limit = (page-1) * 20 # TODO page_size from where?
ids = [d['_id'] for d in mdb.contrib_ad.query_contributions(crit, limit=limit)]
last_id = ids[-1]
sort = {'key': request.GET.get('sort'), 'order': request.GET.get('order')}
docs, last_id = mdb.contrib_ad.query_paginate(crit, projection=proj, last_id=last_id, sort=sort)
if not docs:
raise Exception('No contributions found for CarrierTransport Explorer!')
items = []
columns = ['##'.join(['general', sk]) for sk in ['mp-id', 'cid', 'formula']]
keys, subkeys = {'<mₑᶜᵒⁿᵈ>': '[mₑ]', '<S>': '[μV/K]', '<σ>': '[(Ωms)⁻¹]', '<S²σ>': '[μW/(cmK²s)]'}, ['n', 'p']
columns += ['##'.join([k, ' '.join([sk, keys[k]])]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = '/'.join([
request.build_absolute_uri(request.path).rsplit('/', 3)[0],
'explorer', 'materials' , unicode(doc['_id'])
])
mp_id_url = 'https://materialsproject.org/materials/{}'.format(mp_id)
row = [mp_id_url, cid_url, contrib['extra_data']['pretty_formula']]
row += [
contrib['data'].get(k[1:-1], {}).get(sk, {}).get('<ε>', '-').split()[0]
for k in keys for sk in subkeys
]
items.append(dict((k, v) for k, v in zip(columns, row)))
per_page = len(items)
total_pages = total_count/per_page
if total_pages%per_page:
total_pages += 1
response = {
'total_count': total_count, 'total_pages': total_pages, 'page': page,
'last_page': total_pages, 'per_page': per_page, 'last_id': last_id, 'items': items
}
except Exception as ex:
raise ValueError('"REST Error: "{}"'.format(str(ex)))
return response
def get_graphs(self):
docs = self.query_contributions(
criteria={'project': {'$in': ['LBNL']}, 'content.formula': 'MoSe2'},
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
)
if not docs:
raise Exception('No contributions found for SLAC MoSe2 Explorer!')
doc = docs[0] # there should be only one for MoSe2
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
graphs = mpfile.gdata[mp_id]
return graphs
def get_graphs(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
)
if not docs:
raise Exception('No contributions found for MpWorkshop2017 Explorer!')
graphs = {}
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
graphs[mp_id] = mpfile.gdata[mp_id]
return graphs
def get_provenance(self):
provenance_keys = [
'title', 'url', 'explanation', 'references', 'authors',
'contributor'
]
projection = {'_id': 1, 'mp_cat_id': 1}
for key in provenance_keys:
projection['content.' + key] = 1
docs = self.query_contributions(criteria=self.dtu_query,
projection=projection)
if not docs:
raise Exception('No contributions found for DTU Explorer!')
mpfile = MPFile.from_contribution(docs[0])
mp_id = mpfile.ids[0]
return mpfile.hdata[mp_id]
class SwfRester(MPContribsRester):
"""SWF-specific convenience functions to interact with MPContribs REST interface"""
mpfile = MPFile.from_file(
os.path.join(os.path.dirname(__file__), '..', 'mpfile_init.txt'))
query = {'content.doi': mpfile.hdata.general['doi']}
provenance_keys = [
k for k in mpfile.hdata.general.keys() if k != 'google_sheet'
]
def get_contributions(self):
docs = self.query_contributions(projection={
'_id': 1,
'mp_cat_id': 1,
'content.data': 1
})
if not docs:
raise Exception('No contributions found for SWF Explorer!')
data = []
columns = ['formula', 'contribution']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
formula = mpfile.ids[0]
contrib = mpfile.hdata[formula].get('data')
if contrib is None:
continue
cid_url = self.get_cid_url(doc)
for k in contrib.keys():
if k not in columns:
columns.append(k)
row = [formula, cid_url]
for col in columns[2:]:
row.append(contrib.get(col, ''))
data.append([formula, row])
# enforce equal row lengths
ncols = len(columns)
for entry in data:
n = len(entry[1])
if n != ncols:
entry[1] += [''] * (ncols - n)
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, typ):
types = ['2d', '3d']
if typ not in types:
raise Exception('typ has to be 2d or 3d!')
docs = self.query_contributions(
criteria={'content.data.{}'.format(typ): {
'$exists': 1
}},
projection={
'_id': 1,
'mp_cat_id': 1,
'content': 1
})
if not docs:
raise Exception('No contributions found for JarvisDft Explorer!')
data = []
columns = [
'mp-id', 'cid', 'CIF', 'final_energy', 'optB88vDW_bandgap',
'mbj_bandgap', 'bulk_modulus', 'shear_modulus', 'jid'
]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
hdata = mpfile.hdata[mp_id]
contrib = hdata['data'][typ]
cid_url = self.get_cid_url(doc)
cif_url = ''
structures = mpfile.sdata.get(mp_id)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif',
structures.keys()[0]
])
row = [mp_id, cid_url, cif_url
] + [contrib[k] for k in columns[3:-1]]
row.append(hdata['details_url'].format(contrib['jid']))
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, bandgap_range=None):
projection = {
'_id': 1,
'identifier': 1,
'content.formula': 1,
'content.ICSD': 1,
'content.data': 1
}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for DTU Explorer!')
data = []
columns = ['mp-id', 'cid', 'formula', 'ICSD', 'C']
keys, subkeys = ['ΔE-KS', 'ΔE-QP'], ['indirect', 'direct']
columns += ['##'.join([k, sk]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [
mp_id, cid_url, contrib['formula'], contrib['ICSD'],
contrib['data']['C']
]
row += [contrib['data'][k][sk] for k in keys for sk in subkeys]
if bandgap_range:
in_filter = True
for k, v in bandgap_range.iteritems():
ks = k.split('_')
val = contrib['data'][ks[0]][
ks[1]] if len(ks) > 1 else contrib['data'][k]
dec = float(val.split()[0])
if dec < v[0] or dec > v[1]:
in_filter = False
break
if in_filter:
data.append((mp_id, row))
else:
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_provenance(self):
docs = self.query_contributions(
criteria={
'project': {'$in': ['LBNL']},
'content.authors': {'$exists': 1},
'content.description': {'$exists': 1},
'content.reference': {'$exists': 1}
},
projection={
'_id': 1, 'mp_cat_id': 1, 'content.authors': 1,
'content.reference': 1, 'content.description': 1
}
)
if not docs:
raise Exception('No contributions found for SLAC MoSe2 Explorer!')
mpfile = MPFile.from_contribution(docs[0])
mp_id = mpfile.ids[0]
return mpfile.hdata[mp_id]
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1, 'content.abbreviations': 1}
)
if not docs:
raise Exception('No contributions found for PerovskitesDiffusion Explorer!')
data, columns = [], None
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
if columns is None:
columns = ['mp-id', 'contribution'] + contrib.keys()
row = [mp_id, cid_url] + contrib.values()
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, phase=None):
data = []
phase_query_key = {'$exists': 1} if phase is None else phase
columns = ['mp-id', 'contribution', 'formula']
if phase is None:
columns.append('phase')
columns += ['dH (formation)', 'dH (hydration)', 'GS?', 'CIF']
docs = self.query_contributions(
criteria={
'content.doi': '10.1021/jacs.6b11301',
'content.data.Phase': phase_query_key
},
projection={
'_id': 1, 'mp_cat_id': 1, 'content.data': 1,
'content.{}'.format(mp_level01_titles[3]): 1
}
)
if not docs:
raise Exception('No contributions found for MnO2 Phase Selection Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials', doc['_id']
])
row = [mp_id, cid_url, contrib['Formula']]
if phase is None:
row.append(contrib['Phase'])
row += [contrib['dHf'], contrib['dHh'], contrib['GS']]
cif_url = ''
structures = mpfile.sdata.get(mp_id)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif', structures.keys()[0]
])
row.append(cif_url)
data.append((mp_id, row))
return DataFrame.from_items(data, orient='index', columns=columns)
def get_all_spectra(self, typ):
types = ['XAS', 'XMCD']
if typ not in types:
raise Exception('{} not in {}'.format(typ, types))
projection = {'_id': 1, 'identifier': 1, 'content.Co': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for ALS Beamline Explorer!')
table = Table()
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
df = mpfile.tdata[identifier]['Co']
if 'Energy' not in table.columns:
table['Energy'] = df['Energy']
table[identifier] = df[typ]
return Plot({'x': 'Energy', 'table': typ, 'showlegend': False}, table)
def get_provenance(self):
docs = self.query_contributions(
criteria={
'content.doi': '10.1021/jacs.6b11301',
'content.authors': {'$exists': 1},
'content.title': {'$exists': 1},
'content.reference': {'$exists': 1}
},
projection={
'_id': 1, 'mp_cat_id': 1, 'content.authors': 1,
'content.reference': 1, 'content.title': 1
}
)
if not docs:
raise Exception('No contributions found for MnO2 Phase Selection Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
return mpfile.hdata[mp_id]
def get_all_spectra(self, typ):
types = ['XAS', 'XMCD']
if typ not in types:
raise Exception('{} not in {}'.format(typ, types))
projection = {'_id': 1, 'mp_cat_id': 1, 'content.Co': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for ALS Beamline Explorer!')
table = Table()
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
df = mpfile.tdata[identifier]['Co']
if 'Energy' not in table.columns:
table['Energy'] = df['Energy']
table[identifier] = df[typ]
return Plot({'x': 'Energy', 'table': typ, 'showlegend': False}, table)
def get_contributions(self):
data = []
for doc in self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
):
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
table = mpfile.tdata[mp_id][mp_level01_titles[1]+'_D0_Q']
anums = [self.z[el] for el in table['element']]
table.insert(0, 'Z', Series(anums, index=table.index))
table.sort_values('Z', inplace=True)
table.reset_index(drop=True, inplace=True)
hdata = mpfile.hdata[mp_id]
data.append({
'mp_id': mp_id, 'cid': doc['_id'],
'short_cid': get_short_object_id(doc['_id']),
'formula': hdata['formula'],
'table': table
})
return data
def get_contributions(self):
data = []
columns = [
'mp-id', 'contribution', 'kohn-sham_bandgap(indirect)',
'kohn-sham_bandgap(direct)', 'derivative_discontinuity',
'quasi-particle_bandgap(indirect)',
'quasi-particle_bandgap(direct)'
]
docs = self.query_contributions(
criteria=self.dtu_query,
projection={
'_id': 1,
'mp_cat_id': 1,
'content.kohn-sham_bandgap.indirect': 1,
'content.kohn-sham_bandgap.direct': 1,
'content.derivative_discontinuity': 1,
'content.quasi-particle_bandgap.indirect': 1,
'content.quasi-particle_bandgap.direct': 1
})
if not docs:
raise Exception('No contributions found for DTU Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id']
])
row = [
mp_id, cid_url, contrib['kohn-sham_bandgap']['indirect'],
contrib['kohn-sham_bandgap']['direct'],
contrib['derivative_discontinuity'],
contrib['quasi-particle_bandgap']['indirect'],
contrib['quasi-particle_bandgap']['direct']
]
data.append((mp_id, row))
return DataFrame.from_items(data, orient='index', columns=columns)
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for PCFC Explorer!')
data = []
columns = ['mp-id', 'cid']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mpid = mpfile.ids[0]
contrib = mpfile.hdata[mpid]['data']
cid_url = self.get_cid_url(doc)
row = [mpid, cid_url]
if len(columns) == 2:
columns += [k for k in contrib.keys()]
for col in columns[2:]:
row.append(contrib.get(col, ''))
data.append((mpid, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, phase=None):
data = []
phase_query_key = {'$exists': 1} if phase is None else phase
columns = ['mp-id', 'contribution', 'formula']
if phase is None:
columns.append('phase')
columns += ['ΔH', 'ΔHₕ', 'GS?', 'CIF']
docs = self.query_contributions(
criteria={'content.data.Phase': phase_query_key},
projection={
'_id': 1, 'mp_cat_id': 1, 'content.data': 1,
'content.{}'.format(mp_level01_titles[3]): 1
}
)
if not docs:
raise Exception('No contributions found for MnO2 Phase Selection Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['Formula'].replace(' ', '')]
if phase is None:
row.append(contrib['Phase'])
row += [contrib['ΔH'], contrib['ΔHₕ'], contrib['GS']]
cif_url = ''
structures = mpfile.sdata.get(mp_id)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif', structures.keys()[0]
])
row.append(cif_url)
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, bandgap_range=None):
projection = {
'_id': 1, 'mp_cat_id': 1,
'content.formula': 1, 'content.ICSD': 1, 'content.data': 1
}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for DTU Explorer!')
data = []
columns = ['mp-id', 'cid', 'formula', 'ICSD', 'C']
keys, subkeys = ['ΔE-KS', 'ΔE-QP'], ['indirect', 'direct']
columns += ['##'.join([k, sk]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['formula'], contrib['ICSD'], contrib['data']['C']]
row += [contrib['data'][k][sk] for k in keys for sk in subkeys]
if bandgap_range:
in_filter = True
for k, v in bandgap_range.iteritems():
ks = k.split('_')
val = contrib['data'][ks[0]][ks[1]] if len(ks) > 1 else contrib['data'][k]
dec = float(val.split()[0])
if dec < v[0] or dec > v[1]:
in_filter = False
break
if in_filter:
data.append((mp_id, row))
else:
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def test_get_string():
mpfile = MPFile.from_file('test_files/test_archieml.txt')
mpfile_test = MPFile.from_string(mpfile.get_string())
assert_equal(mpfile.document, mpfile_test.document)
def test_get_string():
test_file = os.path.join(os.path.dirname(__file__), 'test_archieml.txt')
mpfile = MPFile.from_file(test_file)
mpfile_test = MPFile.from_string(mpfile.get_string())
assert mpfile.document == mpfile_test.document
def run(mpfile, nmax=None, dup_check_test_site=True):
existing_mpids = {}
for b in [False, True]:
with PerovskitesDiffusionRester(test_site=b) as mpr:
for doc in mpr.query_contributions(
projection={'content.data.directory': 1, 'mp_cat_id': 1}
):
key = '_'.join([doc['mp_cat_id'], doc['content']['data']['directory']])
existing_mpids[key] = doc['_id']
if not dup_check_test_site:
break
general = mpfile.document[mp_level01_titles[0]]
google_sheet = general.pop('google_sheet') + '/export?format=xlsx'
contcars_filepath = general.pop('contcars_filepath')
contcars = tarfile.open(contcars_filepath)
df = read_excel(google_sheet)
keys = df.iloc[[0]].to_dict(orient='records')[0]
abbreviations = RecursiveDict()
count, skipped, update = 0, 0, 0
for index, row in df[1:].iterrows():
mpid = None
data = RecursiveDict()
mpfile_single = MPFile()
for col, value in row.iteritems():
if col == 'level_0' or col == 'index':
continue
key = keys[col]
if isinstance(key, string_types):
key = key.strip()
if not key in abbreviations:
abbreviations[key] = col
else:
key = col.strip().lower()
if key == 'pmgmatchid':
mpid = value.strip()
if mpid == 'None':
mpid = None
name = '_'.join(data['directory'].split('/')[1:])
contcar_path = 'bulk_CONTCARs/{}_CONTCAR'.format(
data['directory'].replace('/', '_')
)
contcar = contcars.extractfile(contcar_path)
mpid_match = mpfile_single.add_structure(
contcar.read(), fmt='poscar',
name=name, identifier=mpid
)
if not mp_id_pattern.match(mpid_match):
print 'skipping', name
continue
mpid = mpid_match
else:
data[key] = value
if mpid is None:
continue
mpid_mod = '_'.join([mpid, data['directory']])
if nmax is not None and mpid_mod in existing_mpids:
print 'skipping', mpid_mod
skipped += 1
continue # skip duplicates
mpfile_single.add_hierarchical_data({'data': data}, identifier=mpid)
if mpid_mod in existing_mpids:
cid = existing_mpids[mpid_mod]
mpfile_single.insert_id(mpid, cid)
update += 1
mpfile.concat(mpfile_single)
if nmax is not None and count >= nmax-1:
break
count += 1
mpfile.add_hierarchical_data({'abbreviations': abbreviations})
print len(mpfile.ids), 'mp-ids to submit.'
if nmax is None and update > 0:
print update, 'mp-ids to update.'
if nmax is not None and skipped > 0:
print skipped, 'duplicates to skip.'
import argparse, os
from mpcontribs.io.archieml.mpfile import MPFile
from pre_submission import *
parser = argparse.ArgumentParser(
description="""generate MPFile from directory of related XAS measurements"""
)
parser.add_argument(
'-i', '--input_mpfile', type=str, metavar="PATH", help="""path to input
MPFile with shared MetaData and processing instructions for each
composition""", default="input.txt"
)
parser.add_argument(
'-o', '--output_mpfile', type=str, metavar="FILENAME", help="""name of
output MPFile with shared MetaData and processing results for each
composition (will be created in same directory as `input_mpfile`)""",
default="output.txt"
)
args = parser.parse_args()
mpfile = MPFile.from_file(args.input_mpfile)
run(mpfile)
work_dir = os.path.dirname(os.path.realpath(args.input_mpfile))
output_mpfile = os.path.join(work_dir, args.output_mpfile)
mpfile.write_file(output_mpfile)
import sys, json, os
from pympler import asizeof
if __name__ == '__main__':
if not os.path.exists('martin_lab.json'):
from mpcontribs.users.martin_lab.pre_submission import run
from mpcontribs.io.archieml.mpfile import MPFile
from mpcontribs.rest.adapter import ContributionMongoAdapter
from mpcontribs.builder import MPContributionsBuilder, export_notebook
mpfile = MPFile.from_file('MPContribs/mpcontribs/users/martin_lab/mpfile_init.txt')
run(mpfile)
cma = ContributionMongoAdapter()
for mpfile_single in mpfile.split():
contributor = 'Patrick Huck <*****@*****.**>'
doc = cma.submit_contribution(mpfile_single, contributor)
cid = doc['_id']
print doc.keys()
mcb = MPContributionsBuilder(doc)
build_doc = mcb.build(contributor, cid)
nb = build_doc[-1]
print nb.keys()
with open('martin_lab.json', 'w') as f:
json.dump(nb, f)
print 'DONE'
with open('martin_lab.json', 'r') as f:
nb = json.load(f)
for idx, cell in enumerate(nb['cells']):
if idx: # skip first cell
