def test_init_from_and_to_stream(self):
mols = makeStereoExamples() * 10
holder = rdSubstructLibrary.CachedSmilesMolHolder()
# one day I'll fix this, but we need to write text but read binary
# grrr.... something about the python_streambuf handler.
slib = rdSubstructLibrary.SubstructLibrary(holder, None)
for mol in mols:
holder.AddSmiles(Chem.MolToSmiles(mol, isomericSmiles=True))
if rdSubstructLibrary.SubstructLibraryCanSerialize():
fd, path = tempfile.mkstemp()
with open(path, 'w') as file:
slib.ToStream(file)
with open(path, 'rb') as file:
slib2 = rdSubstructLibrary.SubstructLibrary()
slib2.InitFromStream(file)
self.assertEqual(len(slib), len(slib2))
from io import StringIO, BytesIO
s = StringIO()
slib.ToStream(s)
sb = BytesIO(s.getvalue().encode("ascii"))
self.assertTrue(len(sb.getvalue()) > 0)
slib3 = rdSubstructLibrary.SubstructLibrary()
slib3.InitFromStream(sb)
self.assertEqual(len(slib), len(slib2))
def test0SubstructLibrary(self):
for fpholderCls in [None, rdSubstructLibrary.PatternHolder]:
for holder in [rdSubstructLibrary.MolHolder(), rdSubstructLibrary.CachedMolHolder(),
rdSubstructLibrary.CachedSmilesMolHolder()]:
if fpholderCls: fpholder = fpholderCls()
else: fpholder = None
slib_ = rdSubstructLibrary.SubstructLibrary(holder, fpholder)
for i in range(100):
m = Chem.MolFromSmiles("c1ccccc1")
self.assertEqual(slib_.AddMol(m), i)
libs = [slib_]
if rdSubstructLibrary.SubstructLibraryCanSerialize():
serialized1 = pickle.loads(pickle.dumps(slib_))
serialized2 = rdSubstructLibrary.SubstructLibrary(slib_.Serialize())
libs.append(serialized1)
libs.append(serialized2)
for slib in libs:
res = slib.GetMatches(m)
t2 = time.time()
self.assertTrue(len(res) == 100)
res = slib.GetMatches(m)
self.assertEqual(len(res), 100)
self.assertTrue(set(res) == set(list(range(100))))
res = slib.GetMatches(m, maxResults=100);
self.assertEqual(len(res), 100)
self.assertEqual(len(slib.GetMatches(m, startIdx=0, endIdx=100)), 100)
self.assertTrue(slib.HasMatch(m))
self.assertEqual(slib.CountMatches(m), 100)
def testOptions(self):
mols = makeStereoExamples() * 10
for holderCls in [
rdSubstructLibrary.MolHolder,
rdSubstructLibrary.CachedMolHolder,
rdSubstructLibrary.CachedSmilesMolHolder,
rdSubstructLibrary.CachedTrustedSmilesMolHolder,
]:
holder = holderCls()
slib_ = rdSubstructLibrary.SubstructLibrary(holder, None)
for mol in mols:
slib_.AddMol(mol)
libs = [slib_]
if rdSubstructLibrary.SubstructLibraryCanSerialize():
serialized1 = pickle.loads(pickle.dumps(slib_))
serialized2 = rdSubstructLibrary.SubstructLibrary(
slib_.Serialize())
libs.append(serialized1)
libs.append(serialized2)
for slib in libs:
core = Chem.MolFromSmarts("C-1-C-C-O-C(-*)(-*)1")
res = slib.GetMatches(core)
self.assertEqual(
len(res),
len([
x for x in mols
if x.HasSubstructMatch(core, useChirality=True)
]))
core = Chem.MolFromSmarts("C-1-C-C-O-C(-[O])(-[N])1")
core.SetProp("core", "core")
res = slib.GetMatches(core, useChirality=False)
self.assertEqual(
len(res),
len([
x for x in mols
if x.HasSubstructMatch(core, useChirality=False)
]))
core = Chem.MolFromSmarts("C-1-C-C-O-[C@@](-[O])(-[N])1")
res = slib.GetMatches(core, useChirality=False)
self.assertEqual(
len(res),
len([
x for x in mols
if x.HasSubstructMatch(core, useChirality=False)
]))
core = Chem.MolFromSmarts("C-1-C-C-O-[C@@](-[O])(-[N])1")
res = slib.GetMatches(core)
self.assertEqual(
len(res),
len([
x for x in mols
if x.HasSubstructMatch(core, useChirality=True)
]))
def test1SubstructLibrary(self):
for keyholderCls in [None, rdSubstructLibrary.KeyFromPropHolder]:
for fpholderCls in [None, rdSubstructLibrary.PatternHolder]:
for holder in [
rdSubstructLibrary.MolHolder(),
rdSubstructLibrary.CachedMolHolder(),
rdSubstructLibrary.CachedSmilesMolHolder()
]:
if fpholderCls: fpholder = fpholderCls()
else: fpholder = None
if keyholderCls:
keyholder = keyholderCls()
self.assertEqual(keyholder.GetPropName(), "_Name")
else:
keyholder = None
slib_ = rdSubstructLibrary.SubstructLibrary(
holder, fpholder, keyholder)
mols = []
for i in range(100):
m = Chem.MolFromSmiles("c1ccccc1")
m.SetProp("_Name", str(i * 2))
self.assertEqual(slib_.AddMol(m), i * 2)
mols.append(m)
m2 = Chem.MolFromSmiles("CCCC")
m2.SetProp("_Name", str(i * 2 + 1))
self.assertEqual(slib_.AddMol(m2), i * 2 + 1)
mols.append(m2)
libs = [slib_]
if rdSubstructLibrary.SubstructLibraryCanSerialize():
serialized1 = pickle.loads(pickle.dumps(slib_))
serialized2 = rdSubstructLibrary.SubstructLibrary(
slib_.Serialize())
libs.append(serialized1)
libs.append(serialized2)
for slib in libs:
res = slib.GetMatches(m)
self.assertEqual(len(res), 100)
self.assertEqual(set(res), set(list(range(0, 200, 2))))
if keyholderCls:
self.assertEqual(
[str(idx) for idx in res],
[str(idx) for idx in range(0, 200, 2)])
res = slib.GetMatches(m2)
self.assertEqual(len(res), 100)
self.assertTrue(
set(res) == set(list(range(1, 200, 2))))
if keyholderCls:
self.assertEqual(
[str(idx) for idx in res],
[str(idx) for idx in range(1, 200, 2)])
res = slib.GetMatches(m)
self.assertEqual(len(res), 100)
res = slib.GetMatches(m, maxResults=100)
self.assertEqual(len(res), 100)
self.assertEqual(
len(slib.GetMatches(m, startIdx=0, endIdx=50 * 2)),
50)
self.assertEqual(
len(
slib.GetMatches(m2,
startIdx=1,
endIdx=50 * 2 + 1)), 50)
self.assertTrue(slib.HasMatch(m))
self.assertTrue(slib.HasMatch(m2))
self.assertEqual(slib.CountMatches(m), 100)
self.assertEqual(slib.CountMatches(m2), 100)