def setCompressionOptions(opt):
"""
Adds suitable compression options for an object of type
pyopenms.PeakFileOptions
- compresses mass / time arrays with numpress linear
- compresses intensity with slof (log integer)
- compresses ion mobility with slof (log integer)
"""
cfg = pyopenms.NumpressConfig()
cfg.estimate_fixed_point = True
cfg.numpressErrorTolerance = -1.0 # skip check, faster
cfg.setCompression(b"linear")
cfg.linear_fp_mass_acc = -1
# set the desired RT accuracy in seconds
opt.setNumpressConfigurationMassTime(cfg)
cfg = pyopenms.NumpressConfig()
cfg.estimate_fixed_point = True
cfg.numpressErrorTolerance = -1.0 # skip check, faster
cfg.setCompression(b"slof")
opt.setNumpressConfigurationIntensity(cfg)
opt.setCompression(True) # zlib compression
# Now also try to compress float data arrays (this is not enabled in all
# versions of pyOpenMS).
try:
cfg = pyopenms.NumpressConfig()
cfg.estimate_fixed_point = True
cfg.numpressErrorTolerance = -1.0 # skip check, faster
cfg.setCompression(b"slof")
opt.setNumpressConfigurationFloatDataArray(cfg)
except Exception:
pass
def main():
if len(sys.argv) < 3:
raise RuntimeError("need arguments: tdf_directory output.mzML")
analysis_dir = sys.argv[1]
output_fname = sys.argv[2]
if sys.version_info.major == 2:
analysis_dir = unicode(analysis_dir)
td = timsdata.TimsData(analysis_dir)
conn = td.conn
# Get total frame count:
q = conn.execute("SELECT COUNT(*) FROM Frames")
row = q.fetchone()
N = row[0]
print("Analysis has {0} frames.".format(N))
# Store output
if output_fname.lower().endswith("mzml"):
consumer = pyopenms.PlainMSDataWritingConsumer(output_fname)
# Compress output
try:
opt = consumer.getOptions()
cfg = pyopenms.NumpressConfig()
cfg.estimate_fixed_point = True
cfg.numpressErrorTolerance = -1.0 # skip check, faster
cfg.setCompression("linear");
cfg.linear_fp_mass_acc = -1; # set the desired RT accuracy in seconds
opt.setNumpressConfigurationMassTime(cfg)
cfg = pyopenms.NumpressConfig()
cfg = pyopenms.NumpressConfig()
cfg.estimate_fixed_point = True
cfg.numpressErrorTolerance = -1.0 # skip check, faster
cfg.setCompression("slof");
opt.setNumpressConfigurationIntensity(cfg)
opt.setCompression(True) # zlib compression
consumer.setOptions(opt)
except Exception:
pass
if output_fname.lower().endswith("sqmass"):
consumer = pyopenms.MSDataSqlConsumer(output_fname)
for frame_id in range(N):
store_frame(frame_id+1, td, conn, consumer, compressFrame=True)
def test_encodeNP_SLOF(self):
"""
String out;
MSNumpressCoder::NumpressConfig config;
config.np_compression = MSNumpressCoder::SLOF;
config.estimate_fixed_point = true; // critical
bool zlib_compression = false;
MSNumpressCoder().encodeNP(in, out, zlib_compression, config);
TEST_EQUAL(out.size(), 24)
TEST_EQUAL(out, "QMVagAAAAAAZxX3ivPP8/w==")
"""
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.SLOF
config.estimate_fixed_point = True
out_ = pyopenms.String()
coder.encodeNP(self.testData, out_, False, config)
out = out_.c_str()
self.assertEqual(len(out), 24)
self.assertEqual(out, "QMVagAAAAAAZxX3ivPP8/w==")
def test_encodeNP_LINEAR(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.LINEAR
config.estimate_fixed_point = True
out_ = pyopenms.String()
coder.encodeNP(self.testData, out_, False, config)
out = out_.c_str()
self.assertEqual(len(out), 28)
self.assertEqual(out, "QWR64UAAAADo//8/0P//f1kSgA==")
def test_encodeNP_PIC(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.PIC
config.estimate_fixed_point = True
out_ = pyopenms.String()
coder.encodeNP(self.testData, out_, False, config)
out = out_.c_str()
self.assertEqual(len(out), 12)
self.assertEqual(out, "ZGaMXCFQkQ==")
def test_encodeNP_PIC(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.PIC
config.estimate_fixed_point = True
out_ = pyopenms.String()
coder.encodeNPRaw(self.testData, out_, config)
out = out_.c_str()
self.assertEqual(len(out), 7)
self.assertEqual(out, b'df\x8c\\!P\x91')
def test_decodeNP_LINEAR(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.LINEAR
config.estimate_fixed_point = True
inData = "QWR64UAAAADo//8/0P//f1kSgA=="
out = []
coder.decodeNP(inData, out, False, config)
self.assertEqual(len(out), 4)
for a, b in zip(self.testData, out):
self.assertAlmostEqual(a, b, places=7)
def test_decodeNP_PIC(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.PIC
config.estimate_fixed_point = True
inData = "ZGaMXCFQkQ=="
out = []
coder.decodeNP(inData, out, False, config)
self.assertEqual(len(out), 4)
for a, b in zip(self.testData, out):
self.assertAlmostEqual(a, b, places=2)
def test_decodeNP_SLOF(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.SLOF
config.estimate_fixed_point = True
inData = "QMVagAAAAAAZxX3ivPP8/w=="
out = []
coder.decodeNP(inData, out, False, config)
self.assertEqual(len(out), 4)
for a, b in zip(self.testData, out):
self.assertAlmostEqual(a, b, places=2)
def test_decodeNP_LINEAR(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.LINEAR
config.estimate_fixed_point = True
inData = b'Adz\xe1@\x00\x00\x00\xe8\xff\xff?\xd0\xff\xff\x7fY\x12\x80'
out = []
coder.decodeNPRaw(inData, out, config)
self.assertEqual(len(out), 4)
for a, b in zip(self.testData, out):
self.assertAlmostEqual(a, b, places=7)
def test_encodeNP_LINEAR(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.LINEAR
config.estimate_fixed_point = True
out_ = pyopenms.String()
coder.encodeNPRaw(self.testData, out_, config)
out = out_.c_str()
self.assertEqual(len(out), 19)
self.assertEqual(
out, b'Adz\xe1@\x00\x00\x00\xe8\xff\xff?\xd0\xff\xff\x7fY\x12\x80')
def test_decodeNP_PIC(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.PIC
config.estimate_fixed_point = True
inData = b'df\x8c\\!P\x91'
out = []
coder.decodeNPRaw(inData, out, config)
self.assertEqual(len(out), 4)
for a, b in zip(self.testData, out):
self.assertAlmostEqual(a, b, places=2)
def test_decodeNP_SLOF(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.SLOF
config.estimate_fixed_point = True
inData = b'@\xc5Z\x80\x00\x00\x00\x00\x19\xc5}\xe2\xbc\xf3\xfc\xff'
out = []
coder.decodeNPRaw(inData, out, config)
self.assertEqual(len(out), 4)
for a, b in zip(self.testData, out):
self.assertAlmostEqual(a, b, places=2)
def test_encodeNP_SLOF(self):
coder = pyopenms.MSNumpressCoder()
config = pyopenms.NumpressConfig()
config.np_compression = pyopenms.MSNumpressCoder.NumpressCompression.SLOF
config.estimate_fixed_point = True
out_ = pyopenms.String()
coder.encodeNPRaw(self.testData, out_, config)
out = out_.c_str()
self.assertEqual(len(out), 16)
self.assertEqual(
out, b'@\xc5Z\x80\x00\x00\x00\x00\x19\xc5}\xe2\xbc\xf3\xfc\xff')