def load_economic_data():
global HPV_data
if HPV_data is not None:
return None
global _EPACDRDict, _ECHATonnageDict
import pandas as pd
import zipfile
folder = os_path_join(os.path.dirname(__file__), 'Law')
'''OECD are chemicals produced by and OECD members in > 1000 tonnes/year.'''
HPV_data = pd.read_csv(os.path.join(folder, 'HPV 2015 March 3.csv'),
sep='\t',
index_col=0)
# 13061-29-2 not valid and removed
_ECHATonnageDict = {}
with zipfile.ZipFile(os.path.join(folder,
'ECHA Tonnage Bands.csv.zip')) as z:
with z.open(z.namelist()[0]) as f:
for line in f.readlines():
# for some reason, the file must be decoded to UTF8 first
CAS, band = line.decode("utf-8").strip('\n').split('\t')
if CAS in _ECHATonnageDict:
if band in _ECHATonnageDict[CAS]:
pass
else:
_ECHATonnageDict[CAS].append(band)
else:
_ECHATonnageDict[CAS] = [band]
_EPACDRDict = {}
with open(os.path.join(folder,
'EPA 2012 Chemical Data Reporting.csv')) as f:
'''EPA summed reported chemical usages. In metric tonnes/year after conversion.
Many producers keep their date confidential.
This was originally in terms of lb/year, but rounded to the nearest kg.
'''
next(f)
for line in f:
values = line.rstrip().split('\t')
CAS, manufactured, imported, exported = to_num(values)
_EPACDRDict[CAS] = {
"Manufactured": manufactured / 1000.,
"Imported": imported / 1000.,
"Exported": exported / 1000.
}
def load_mixture_composition():
global mixture_composition_loaded, common_mixtures_by_synonym, common_mixtures
common_mixtures = {}
common_mixtures_by_synonym = {}
with open(os.path.join(folder, 'Mixtures Compositions.tsv')) as f:
"""Read in a dict of 90 or so mixutres, their components, and synonyms.
Small errors in mole fractions not adding to 1 are known. Errors in
adding mass fraction are less common, present at the 5th decimal. Mass
basis is assumed for all mixtures.
"""
next(f)
for line in f:
values = to_num(line.strip('\n').strip('\t').split('\t'))
name, source, N = values[0:3]
N = int(N)
CASs, names, ws, zs = values[3:3 + N], values[
3 + N:3 + 2 * N], values[3 + 2 * N:3 +
3 * N], values[3 + 3 * N:3 + 4 * N]
synonyms = values[3 + 4 * N:]
if synonyms:
synonyms = [i.lower() for i in synonyms]
synonyms.append(name.lower())
obj = CommonMixtureMetadata(name=name,
CASs=CASs,
N=N,
source=source,
names=names,
ws=ws,
zs=zs,
synonyms=synonyms)
common_mixtures[name] = obj
for syn in synonyms:
common_mixtures_by_synonym[syn] = obj
mixture_composition_loaded = True
self.Cpc = Cpc
self.Cpd = Cpd
self.Hfus = Hfus
self.Hvap = Hvap
self.mua = mua
self.mub = mub
def __repr__(self):
return '''JOBACK(i=%r, name=%r, Tc=%r, Pc=%r, Vc=%r, Tb=%r, Tm=%r, Hform=%r, Gform=%r,
Cpa=%r, Cpb=%r, Cpc=%r, Cpd=%r, Hfus=%r, Hvap=%r, mua=%r, mub=%r)''' % (
self.i, self.name, self.Tc, self.Pc, self.Vc, self.Tb, self.Tm,
self.Hform, self.Gform, self.Cpa, self.Cpb, self.Cpc, self.Cpd,
self.Hfus, self.Hvap, self.mua, self.mub)
for i, line in enumerate(joback_data_txt.split('\n')):
parsed = to_num(line.split('\t'))
j = JOBACK(i+1, *parsed)
joback_groups_str_dict[parsed[0]] = j
joback_groups_id_dict[i+1] = j
def smarts_fragment(catalog, rdkitmol=None, smi=None, deduplicate=True):
r'''Fragments a molecule into a set of unique groups and counts as
specified by the `catalog`. The molecule can either be an rdkit
molecule object, or a smiles string which will be parsed by rdkit.
Returns a dictionary of groups and their counts according to the
indexes of the catalog provided.
Parameters
----------
catalog : dict