def test_all_numeric_with_no_numeric_columns(self):
sio = StringIO()
PandasTools.WriteSDF(self.df, sio, allNumeric=True)
s = sio.getvalue()
self.assertFalse(">" in s, s)
self.assertNotIn("7\n\n", s) # double-check that the numeric tests don't pass by accident
self.assertNotIn("8\n\n", s)
def test_all_numeric_with_no_numeric_columns(self):
sio = StringIO()
PandasTools.WriteSDF(self.df, sio, allNumeric=True)
s = sio.getvalue()
self.assertFalse(">" in s, s)
self.assertNotIn("7\n\n", s) # double-check that the numeric tests don't pass by accident
self.assertNotIn("8\n\n", s)
def SaveXlsxFromFrame(frame, outFile, molCol='ROMol', size=(300, 300)):
"""
Saves pandas DataFrame as a xlsx file with embedded images.
It maps numpy data types to excel cell types:
int, float -> number
datetime -> datetime
object -> string (limited to 32k character - xlsx limitations)
Cells with compound images are a bit larger than images due to excel.
Column width weirdness explained (from xlsxwriter docs):
The width corresponds to the column width value that is specified in Excel.
It is approximately equal to the length of a string in the default font of Calibri 11.
Unfortunately, there is no way to specify "AutoFit" for a column in the Excel file format.
This feature is only available at runtime from within Excel.
"""
import xlsxwriter # don't want to make this a RDKit dependency
cols = list(frame.columns)
cols.remove(molCol)
dataTypes = dict(frame.dtypes)
workbook = xlsxwriter.Workbook(outFile) # New workbook
worksheet = workbook.add_worksheet() # New work sheet
worksheet.set_column('A:A', size[0] / 6.) # column width
# Write first row with column names
c2 = 1
for x in cols:
worksheet.write_string(0, c2, x)
c2 += 1
c = 1
for index, row in frame.iterrows():
image_data = StringIO()
img = Draw.MolToImage(row[molCol], size=size)
img.save(image_data, format='PNG')
worksheet.set_row(c, height=size[1]) # looks like height is not in px?
worksheet.insert_image(c, 0, "f", {'image_data': image_data})
c2 = 1
for x in cols:
if str(dataTypes[x]) == "object":
worksheet.write_string(
c, c2,
str(row[x])[:32000]) # string length is limited in xlsx
elif ('float' in str(dataTypes[x])) or ('int' in str(
dataTypes[x])):
if (row[x] != np.nan) or (row[x] != np.inf):
worksheet.write_number(c, c2, row[x])
elif 'datetime' in str(dataTypes[x]):
worksheet.write_datetime(c, c2, row[x])
c2 += 1
c += 1
workbook.close()
image_data.close()
def test_specify_numeric_column(self):
sio = StringIO()
df = self.df
df["len2"] = df["ID"].map(len)
PandasTools.WriteSDF(df, sio, properties=["len2"])
s = sio.getvalue()
self.assertEqual(s.count("<len2>"), 2)
self.assertIn("7\n\n", s)
self.assertIn("8\n\n", s)
def test_all_numeric_with_numeric_columns(self):
sio = StringIO()
df = self.df
df["len"] = df["ID"].map(len)
PandasTools.WriteSDF(df, sio, allNumeric=True)
s = sio.getvalue()
self.assertEqual(s.count("<len>"), 2)
self.assertIn("7\n\n", s)
self.assertIn("8\n\n", s)
def test_specify_numeric_column(self):
sio = StringIO()
df = self.df
df["len2"] = df["ID"].map(len)
PandasTools.WriteSDF(df, sio, properties=["len2"])
s = sio.getvalue()
self.assertEqual(s.count("<len2>"), 2)
self.assertIn("7\n\n", s)
self.assertIn("8\n\n", s)
def test_all_numeric_with_numeric_columns(self):
sio = StringIO()
df = self.df
df["len"] = df["ID"].map(len)
PandasTools.WriteSDF(df, sio, allNumeric=True)
s = sio.getvalue()
self.assertEqual(s.count("<len>"), 2)
self.assertIn("7\n\n", s)
self.assertIn("8\n\n", s)
def SaveXlsxFromFrame(frame, outFile, molCol='ROMol', size=(300,300)):
"""
Saves pandas DataFrame as a xlsx file with embedded images.
It maps numpy data types to excel cell types:
int, float -> number
datetime -> datetime
object -> string (limited to 32k character - xlsx limitations)
Cells with compound images are a bit larger than images due to excel.
Column width weirdness explained (from xlsxwriter docs):
The width corresponds to the column width value that is specified in Excel.
It is approximately equal to the length of a string in the default font of Calibri 11.
Unfortunately, there is no way to specify "AutoFit" for a column in the Excel file format.
This feature is only available at runtime from within Excel.
"""
import xlsxwriter # don't want to make this a RDKit dependency
cols = list(frame.columns)
cols.remove(molCol)
dataTypes = dict(frame.dtypes)
workbook = xlsxwriter.Workbook(outFile) # New workbook
worksheet = workbook.add_worksheet() # New work sheet
worksheet.set_column('A:A', size[0]/6.) # column width
# Write first row with column names
c2 = 1
for x in cols:
worksheet.write_string(0, c2, x)
c2 += 1
c = 1
for index, row in frame.iterrows():
image_data = StringIO()
img = Draw.MolToImage(row[molCol], size=size)
img.save(image_data, format='PNG')
worksheet.set_row(c, height=size[1]) # looks like height is not in px?
worksheet.insert_image(c, 0, "f", {'image_data': image_data})
c2 = 1
for x in cols:
if str(dataTypes[x]) == "object":
worksheet.write_string(c, c2, str(row[x])[:32000]) # string length is limited in xlsx
elif ('float' in str(dataTypes[x])) or ('int' in str(dataTypes[x])):
if (row[x] != np.nan) or (row[x] != np.inf):
worksheet.write_number(c, c2, row[x])
elif 'datetime' in str(dataTypes[x]):
worksheet.write_datetime(c, c2, row[x])
c2 += 1
c += 1
workbook.close()
image_data.close()
def _get_image(x):
"""displayhook function for PIL Images, rendered as PNG"""
import pandas as pd
sio = StringIO()
x.save(sio,format='PNG')
s = b64encode(sio.getvalue())
pd.set_option('display.max_columns',len(s)+1000)
pd.set_option('display.max_rows',len(s)+1000)
if len(s)+100 > pd.get_option("display.max_colwidth"):
pd.set_option("display.max_colwidth",len(s)+1000)
return s
def test2(self):
fName = os.path.join(RDConfig.RDDataDir, 'NCI', 'first_200.props.sdf')
suppl = Chem.SDMolSupplier(fName)
io = StringIO()
try:
Convert(suppl, io, keyCol='AMW', stopAfter=5)
except Exception:
import traceback
traceback.print_exc()
self.fail('conversion failed')
txt = io.getvalue()
lines = [line for line in txt.split('\n') if line.strip() != '']
self.assertTrue(len(lines) == 6, 'bad num lines: %d' % len(lines))
line0 = lines[0].split(',')
self.assertEqual(len(line0), 20)
self.assertTrue(line0[0] == 'AMW')
self.assertTrue(line0[1] == 'SMILES')
def test2(self):
fName = os.path.join(RDConfig.RDDataDir, 'NCI', 'first_200.props.sdf')
suppl = Chem.SDMolSupplier(fName)
io = StringIO()
try:
Convert(suppl, io, keyCol='AMW', stopAfter=5)
except Exception:
import traceback
traceback.print_exc()
self.fail('conversion failed')
txt = io.getvalue()
lines = [line for line in txt.split('\n') if line.strip() != '']
self.assertTrue(len(lines) == 6, 'bad num lines: %d' % len(lines))
line0 = lines[0].split(',')
self.assertEqual(len(line0), 20)
self.assertTrue(line0[0] == 'AMW')
self.assertTrue(line0[1] == 'SMILES')
def test1(self):
fName = os.path.join(RDConfig.RDDataDir, 'NCI', 'first_200.props.sdf')
suppl = Chem.SDMolSupplier(fName)
io = StringIO()
try:
Convert(suppl, io)
except Exception:
import traceback
traceback.print_exc()
self.fail('conversion failed')
txt = io.getvalue()
lines = txt.split('\n')
if not lines[-1]:
del lines[-1]
self.assertTrue(len(lines) == 201, 'bad num lines: %d' % len(lines))
line0 = lines[0].split(',')
self.assertEqual(len(line0), 20)
self.assertTrue(line0[0] == 'SMILES')
def test1(self):
fName = os.path.join(RDConfig.RDDataDir, 'NCI', 'first_200.props.sdf')
suppl = Chem.SDMolSupplier(fName)
io = StringIO()
try:
Convert(suppl, io)
except Exception:
import traceback
traceback.print_exc()
self.fail('conversion failed')
txt = io.getvalue()
lines = txt.split('\n')
if not lines[-1]:
del lines[-1]
self.assertTrue(len(lines) == 201, 'bad num lines: %d' % len(lines))
line0 = lines[0].split(',')
self.assertEqual(len(line0), 20)
self.assertTrue(line0[0] == 'SMILES')
def test2(self):
import os
from rdkit.six.moves import cStringIO as StringIO
fName = os.path.join(RDConfig.RDDataDir,'NCI','first_200.props.sdf')
suppl = Chem.SDMolSupplier(fName)
io = StringIO()
try:
Convert(suppl,io,keyCol='AMW',stopAfter=5)
except:
import traceback
traceback.print_exc()
self.fail('conversion failed')
txt = io.getvalue()
lines = txt.split('\n')
if not lines[-1]:
del lines[-1]
self.assertTrue(len(lines)==6,'bad num lines: %d'%len(lines))
line0 = lines[0].split(',')
self.assertEqual(len(line0),20)
self.assertTrue(line0[0]=='AMW')
self.assertTrue(line0[1]=='SMILES')
def test2(self):
import os
from rdkit.six.moves import cStringIO as StringIO #@UnresolvedImport #pylint: disable=F0401
fName = os.path.join(RDConfig.RDDataDir, 'NCI', 'first_200.props.sdf')
suppl = Chem.SDMolSupplier(fName)
io = StringIO()
try:
Convert(suppl, io, keyCol='AMW', stopAfter=5)
except Exception:
import traceback
traceback.print_exc()
self.fail('conversion failed')
txt = io.getvalue()
lines = txt.split('\n')
if not lines[-1]:
del lines[-1]
self.assertTrue(len(lines) == 6, 'bad num lines: %d' % len(lines))
line0 = lines[0].split(',')
self.assertEqual(len(line0), 20)
self.assertTrue(line0[0] == 'AMW')
self.assertTrue(line0[1] == 'SMILES')
def test1(self):
import os
from rdkit.six.moves import cStringIO as StringIO # @UnresolvedImport #pylint: disable=F0401
fName = os.path.join(RDConfig.RDDataDir, "NCI", "first_200.props.sdf")
suppl = Chem.SDMolSupplier(fName)
io = StringIO()
try:
Convert(suppl, io)
except Exception:
import traceback
traceback.print_exc()
self.fail("conversion failed")
txt = io.getvalue()
lines = txt.split("\n")
if not lines[-1]:
del lines[-1]
self.assertTrue(len(lines) == 201, "bad num lines: %d" % len(lines))
line0 = lines[0].split(",")
self.assertEqual(len(line0), 20)
self.assertTrue(line0[0] == "SMILES")
def BuildFuncGroupHierarchy(fileNm=None,data=None,force=False):
global groupDefns,hierarchy,lastData,lastFilename
if not force and hierarchy and (not data or data==lastData) and \
(not fileNm or fileNm==lastFilename):
return hierarchy[:]
lastData=data
splitter = re.compile('\t+')
from rdkit import Chem
if not fileNm and not data:
fileNm = os.path.join(RDConfig.RDDataDir,'Functional_Group_Hierarchy.txt')
if fileNm:
inF = open(fileNm,'r')
lastFilename = fileNm
elif data:
inF = StringIO(data)
else:
raise ValueError("need data or filename")
groupDefns={}
res = []
lineNo=0
for line in inF.readlines():
lineNo+=1
line=line.strip()
line = line.split('//')[0]
if not line:
continue
splitL = splitter.split(line)
if len(splitL)<3:
raise FuncGroupFileParseError("Input line %d (%s) is not long enough."%(lineNo,repr(line)))
label = splitL[0].strip()
if label in groupDefns:
raise FuncGroupFileParseError("Duplicate label on line %d."%lineNo)
labelHierarchy = label.split('.')
if len(labelHierarchy)>1:
for i in range(len(labelHierarchy)-1):
tmp = '.'.join(labelHierarchy[:i+1])
if not tmp in groupDefns:
raise FuncGroupFileParseError("Hierarchy member %s (line %d) not found."%(tmp,lineNo))
parent = groupDefns['.'.join(labelHierarchy[:-1])]
else:
parent = None
smarts = splitL[1]
try:
patt = Chem.MolFromSmarts(smarts)
except:
import traceback
traceback.print_exc()
patt = None
if not patt:
raise FuncGroupFileParseError('Smarts "%s" (line %d) could not be parsed.'%(smarts,lineNo))
name = splitL[2].strip()
rxnSmarts=''
if len(splitL)>3:
rxnSmarts=splitL[3]
node = FGHierarchyNode(name,patt,smarts=smarts,label=label,parent=parent,rxnSmarts=rxnSmarts)
if parent:
parent.children.append(node)
else:
res.append(node)
groupDefns[label] = node
hierarchy=res[:]
return res
def BuildFuncGroupHierarchy(fileNm=None,data=None,force=False):
global groupDefns,hierarchy,lastData,lastFilename
if not force and hierarchy and (not data or data==lastData) and \
(not fileNm or fileNm==lastFilename):
return hierarchy[:]
lastData=data
splitter = re.compile('\t+')
from rdkit import Chem
if not fileNm and not data:
fileNm = os.path.join(RDConfig.RDDataDir,'Functional_Group_Hierarchy.txt')
if fileNm:
inF = open(fileNm,'r')
lastFilename = fileNm
elif data:
inF = StringIO(data)
else:
raise ValueError("need data or filename")
groupDefns={}
res = []
lineNo=0
for line in inF.readlines():
lineNo+=1
line=line.strip()
line = line.split('//')[0]
if not line:
continue
splitL = splitter.split(line)
if len(splitL)<3:
raise FuncGroupFileParseError("Input line %d (%s) is not long enough."%(lineNo,repr(line)))
label = splitL[0].strip()
if label in groupDefns:
raise FuncGroupFileParseError("Duplicate label on line %d."%lineNo)
labelHierarchy = label.split('.')
if len(labelHierarchy)>1:
for i in range(len(labelHierarchy)-1):
tmp = '.'.join(labelHierarchy[:i+1])
if not tmp in groupDefns:
raise FuncGroupFileParseError("Hierarchy member %s (line %d) not found."%(tmp,lineNo))
parent = groupDefns['.'.join(labelHierarchy[:-1])]
else:
parent = None
smarts = splitL[1]
patt = Chem.MolFromSmarts(smarts)
if not patt:
raise FuncGroupFileParseError('Smarts "%s" (line %d) could not be parsed.'%(smarts,lineNo))
name = splitL[2].strip()
rxnSmarts=''
if len(splitL)>3:
rxnSmarts=splitL[3]
node = FGHierarchyNode(name,patt,smarts=smarts,label=label,parent=parent,rxnSmarts=rxnSmarts)
if parent:
parent.children.append(node)
else:
res.append(node)
groupDefns[label] = node
hierarchy=res[:]
return res
def test_default_write_does_not_include_tags(self):
sio = StringIO()
PandasTools.WriteSDF(self.df, sio)
s = sio.getvalue()
self.assertNotIn(s, "prop2")
def test_identifier_from_a_column(self):
sio = StringIO()
PandasTools.WriteSDF(self.df, sio, idName="prop2")
s = sio.getvalue()
first_line = s.split("\n", 1)[0]
self.assertEqual(first_line, "qwe")
def test_identifier_from_a_column(self):
sio = StringIO()
PandasTools.WriteSDF(self.df, sio, idName="prop2")
s = sio.getvalue()
first_line = s.split("\n", 1)[0]
self.assertEqual(first_line, "qwe")
def test_default_write_does_not_include_tags(self):
sio = StringIO()
PandasTools.WriteSDF(self.df, sio)
s = sio.getvalue()
self.assertNotIn(s, "prop2")