def strip(self, inpPath, outPath, stripList=[]):
""" Strip categories from inpPath and write to outPath
"""
try:
myDataList = []
with open(inpPath, "r") as ifh:
pRd = PdbxReader(ifh)
pRd.read(myDataList)
#
myBlock = myDataList[0]
myName = myBlock.getName()
newContainer = DataContainer(myName)
for objName in myBlock.getObjNameList():
myObj = myBlock.getObj(objName)
if myObj.getName() not in stripList:
newContainer.append(myObj)
#
with open(outPath, "w") as ofh:
pWr = PdbxWriter(ofh)
pWr.setPreferSingleQuotes()
pWr.write([newContainer])
return True
except Exception as e:
logger.exception("Failing with %s" % str(e))
return False
def testSerialize(self):
try:
for storeStringsAsBytes in [True, False]:
tcL = []
ioPy = IoAdapter()
containerList = ioPy.readFile(self.__pathTextCif)
for container in containerList:
cName = container.getName()
tc = DataContainer(cName)
for catName in container.getObjNameList():
dObj = container.getObj(catName)
tObj = DataCategoryTyped(dObj, dictionaryApi=self.__dApi, copyInputData=True)
tc.append(tObj)
tcL.append(tc)
#
bcw = BinaryCifWriter(self.__dApi, storeStringsAsBytes=storeStringsAsBytes, applyTypes=False, useFloat64=True)
bcw.serialize(self.__testBcifOutput, tcL)
self.assertEqual(containerList[0], containerList[0])
self.assertEqual(tcL[0], tcL[0])
bcr = BinaryCifReader(storeStringsAsBytes=storeStringsAsBytes)
cL = bcr.deserialize(self.__testBcifOutput)
#
ioPy = IoAdapter()
ok = ioPy.writeFile(self.__testBcifTranslated, cL)
self.assertTrue(ok)
self.assertTrue(self.__same(tcL[0], cL[0]))
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def test_single_row(self, rw_data):
myDataList = []
curContainer = DataContainer("myblock")
aCat = DataCategory("pdbx_seqtool_mapping_ref")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("entity_id")
aCat.appendAttribute("auth_mon_id")
aCat.appendAttribute("auth_mon_num")
aCat.appendAttribute("pdb_chain_id")
aCat.appendAttribute("ref_mon_id")
aCat.appendAttribute("ref_mon_num")
aCat.appendAttribute("details")
aCat.append([1, 2, 3, 4, 5, 6, 7, 'data_my_big_data_file'])
aCat.append([1, 2, 3, 4, 5, 6, 7, 'loop_my_big_data_loop'])
aCat.append([1, 2, 3, 4, 5, 6, 7, 'save_my_big_data_saveframe'])
aCat.append([1, 2, 3, 4, 5, 6, 7, '_category.item'])
curContainer.append(aCat)
bCat = curContainer.getObj("pdbx_seqtool_mapping_ref")
print("----attribute list %r\n" % bCat.getAttributeList())
row = bCat.getRow(0)
print("----ROW %r\n" % row)
with open(str(rw_data['pathOutputFile2']), "w") as ofh:
myDataList.append(curContainer)
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
assert len(myDataList) == 1
def testWriteDataFile(self):
"""Test case - write data file"""
try:
#
myDataList = []
curContainer = DataContainer("myblock")
aCat = DataCategory("pdbx_seqtool_mapping_ref")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("entity_id")
aCat.appendAttribute("auth_mon_id")
aCat.appendAttribute("auth_mon_num")
aCat.appendAttribute("pdb_chain_id")
aCat.appendAttribute("ref_mon_id")
aCat.appendAttribute("ref_mon_num")
aCat.append([1, 2, 3, 4, 5, 6, 7])
aCat.append([1, 2, 3, 4, 5, 6, 7])
aCat.append([1, 2, 3, 4, 5, 6, 7])
aCat.append([1, 2, 3, 4, 5, 6, 7])
aCat.append([7, 6, 5, 4, 3, 2, 1])
# aCat.printIt()
curContainer.append(aCat)
# curContainer.printIt()
#
myDataList.append(curContainer)
with open(self.__pathOutputFile1, "w") as ofh:
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
self.assertEqual(len(myDataList), 1)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testUpdateDataFile(self):
"""Test case - update data file
"""
try:
# Create a initial data file --
#
myDataList = []
curContainer = DataContainer("myblock")
aCat = DataCategory("pdbx_seqtool_mapping_ref")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("entity_id")
aCat.appendAttribute("auth_mon_id")
aCat.appendAttribute("auth_mon_num")
aCat.appendAttribute("pdb_chain_id")
aCat.appendAttribute("ref_mon_id")
aCat.appendAttribute("ref_mon_num")
aCat.append([9, 2, 3, 4, 5, 6, 7])
aCat.append([10, 2, 3, 4, 5, 6, 7])
aCat.append([11, 2, 3, 4, 5, 6, 7])
aCat.append([12, 2, 3, 4, 5, 6, 7])
curContainer.append(aCat)
myDataList.append(curContainer)
ofh = open(self.__pathOutputFile1, "w")
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
ofh.close()
#
#
# Read and update the data -
#
myDataList = []
ifh = open(self.__pathOutputFile1, "r")
pRd = PdbxReader(ifh)
pRd.read(myDataList)
ifh.close()
#
myBlock = myDataList[0]
# myBlock.printIt()
myCat = myBlock.getObj("pdbx_seqtool_mapping_ref")
# myCat.printIt()
for iRow in range(0, myCat.getRowCount()):
myCat.setValue("some value", "ref_mon_id", iRow)
myCat.setValue(100, "ref_mon_num", iRow)
with open(self.__pathOutputFile2, "w") as ofh:
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
#
self.assertEqual(len(myDataList), 1)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def testRowListInitialization(self):
"""Test case - Row list initialization of a data category and data block"""
try:
#
fn = self.__pathOutputFile4
attributeNameList = [
"aOne", "aTwo", "aThree", "aFour", "aFive", "aSix", "aSeven",
"aEight", "aNine", "aTen"
]
rowList = [
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
]
nameCat = "myCategory"
#
#
curContainer = DataContainer("myblock")
aCat = DataCategory(nameCat, attributeNameList, rowList)
# aCat.printIt()
curContainer.append(aCat)
# curContainer.printIt()
#
myContainerList = []
myContainerList.append(curContainer)
ofh = open(fn, "w")
pdbxW = PdbxWriter(ofh)
pdbxW.write(myContainerList)
ofh.close()
myContainerList = []
ifh = open(fn, "r")
pRd = PdbxReader(ifh)
pRd.read(myContainerList)
ifh.close()
for container in myContainerList:
for objName in container.getObjNameList():
name, aList, rList = container.getObj(objName).get()
logger.debug("Recovered data category %s", name)
logger.debug("Attribute list %r", repr(aList))
logger.debug("Row list %r", repr(rList))
self.assertEqual(len(myContainerList), 1)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def __write_mmcif(self, pathout, coef, entry_id):
"""Writes out the specific map coefficients
"""
# Categories that will not be copied
_striplist = [
'audit', 'diffrn_radiation_wavelength', 'exptl_crystal',
'reflns_scale'
]
# refln attributes to keep
_keepattr = ['index_h', 'index_k', 'index_l', 'fom']
if coef == 'fo':
_keepattr.extend(['pdbx_DELFWT', 'pdbx_DELPHWT'])
else:
_keepattr.extend(['pdbx_FWT', 'pdbx_PHWT'])
# Datablockname
blkname = "{}{}".format(entry_id, coef)
new_cont = DataContainer(blkname)
# Only care about first block
blockin = self.__sf[0]
for objname in blockin.getObjNameList():
if objname in _striplist:
continue
myobj = blockin.getObj(objname)
# Make a copy of the original - as likely will need to modify
modobj = copy.deepcopy(myobj)
if objname == 'entry':
modobj.setValue(entry_id, 'id', 0)
if objname in ['cell', 'symmetry']:
modobj.setValue(entry_id, 'entry_id', 0)
if objname == 'refln':
# Remove all but what we want
# Make a copy to ensure not messed with during operation
for attr in list(modobj.getAttributeList()):
if attr not in _keepattr:
modobj.removeAttribute(attr)
new_cont.append(modobj)
# new_cont.printIt()
io = IoAdapterCore()
# Write out a single block
ret = io.writeFile(pathout, [new_cont])
return ret
def __processContent(self, cifFileObj):
"""Internal method to transfer parsed data from the wrapped input C++ CifFile object into
the list of Python DataContainer objects.
Args:
cifFileObj (wrapped CifFile object): Wrapped input C++ CifFile object
Returns:
list of DataContainer objects: List of Python DataContainer objects
"""
containerList = []
containerNameList = []
try:
# ----- Repackage the data content ----
#
containerList = []
containerNameList = []
containerNameList = list(
cifFileObj.GetBlockNames(containerNameList))
for containerName in containerNameList:
#
aContainer = DataContainer(containerName)
#
block = cifFileObj.GetBlock(containerName)
tableNameList = []
tableNameList = list(block.GetTableNames(tableNameList))
for tableName in tableNameList:
table = block.GetTable(tableName)
attributeNameList = list(table.GetColumnNames())
numRows = table.GetNumRows()
rowList = []
for iRow in range(0, numRows):
row = table.GetRow(iRow)
# row = table.GetRow(iRow).decode('unicode_escape').encode('utf-8')
# row = [p.encode('ascii', 'xmlcharrefreplace') for p in table.GetRow(iRow)]
rowList.append(list(row))
aCategory = DataCategory(
tableName,
attributeNameList,
rowList,
copyInputData=False,
raiseExceptions=self._raiseExceptions)
aContainer.append(aCategory)
containerList.append(aContainer)
except Exception as e:
msg = "Failing packaging with %s" % str(e)
self._logError(msg)
return containerList
def _createfile2(pathout):
my_data_list = []
cur_container = DataContainer("test")
acat = DataCategory("new")
acat.appendAttribute("item")
acat.append(('1',))
cur_container.append(acat)
acat = DataCategory("second_category")
acat.appendAttribute('row')
acat.appendAttribute('rowb')
acat.append(('1', '2'))
cur_container.append(acat)
acat = DataCategory("third")
acat.appendAttribute('id')
acat.appendAttribute('val')
acat.append(('1', 'a'))
acat.append(('2', 'b'))
acat.append(('3', 'c'))
cur_container.append(acat)
acat = DataCategory("exptl")
acat.appendAttribute('method')
acat.appendAttribute('entry_id')
acat.append(('NEW', 'something'))
cur_container.append(acat)
acat = DataCategory("struct")
acat.appendAttribute('new')
acat.appendAttribute('pdbx_descriptor')
acat.append(('Something to add', 'Override descriptor'))
cur_container.append(acat)
my_data_list.append(cur_container)
with open(pathout, "w") as ofh:
pdbxw = PdbxWriter(ofh)
pdbxw.setAlignmentFlag(flag=True)
pdbxw.write(my_data_list)
def _createfile2(pathout):
my_data_list = []
cur_container = DataContainer("test")
acat = DataCategory("new")
acat.appendAttribute("item")
acat.append(("1",))
cur_container.append(acat)
acat = DataCategory("second_category")
acat.appendAttribute("row")
acat.appendAttribute("rowb")
acat.append(("1", "2"))
cur_container.append(acat)
acat = DataCategory("third")
acat.appendAttribute("id")
acat.appendAttribute("val")
acat.append(("1", "a"))
acat.append(("2", "b"))
acat.append(("3", "c"))
cur_container.append(acat)
acat = DataCategory("exptl")
acat.appendAttribute("method")
acat.appendAttribute("entry_id")
acat.append(("NEW", "something"))
cur_container.append(acat)
acat = DataCategory("struct")
acat.appendAttribute("new")
acat.appendAttribute("pdbx_descriptor")
acat.append(("Something to add", "Override descriptor"))
cur_container.append(acat)
my_data_list.append(cur_container)
with open(pathout, "w") as ofh:
pdbxw = PdbxWriter(ofh)
pdbxw.setAlignmentFlag(flag=True)
pdbxw.write(my_data_list)
def writeDefaultDataTypeMap(self, outPath, dataTyping="ANY"):
"""Write data file containing application default dictionary to application data type mapping
data_rcsb_data_type_map
loop_
_pdbx_data_type_application_map.application_name
_pdbx_data_type_application_map.type_code
_pdbx_data_type_application_map.app_type_code
_pdbx_data_type_application_map.app_precision_default
_pdbx_data_type_application_map.app_width_default
# .... type mapping data ...
"""
try:
#
containerList = []
curContainer = DataContainer("rcsb_data_type_map")
aCat = DataCategory("pdbx_data_type_application_map")
aCat.appendAttribute("application_name")
aCat.appendAttribute("type_code")
aCat.appendAttribute("app_type_code")
aCat.appendAttribute("app_width_default")
aCat.appendAttribute("app_precision_default")
for (cifType, simpleType, defWidth, defPrecision) in zip(
DataTypeApplicationInfo.cifTypes,
DataTypeApplicationInfo.appTypes,
DataTypeApplicationInfo.defaultWidths,
DataTypeApplicationInfo.defaultPrecisions):
if self.__isNull(cifType):
continue
aCat.append(
[dataTyping, cifType, simpleType, defWidth, defPrecision])
curContainer.append(aCat)
containerList.append(curContainer)
#
mU = MarshalUtil(workPath=self.__workPath)
ok = mU.doExport(outPath,
containerList,
fmt="mmcif",
enforceAscii=True,
useCharRefs=True,
raiseExceptions=True)
return ok
except Exception as e:
logger.exception("Failing with %s", str(e))
return False
def __generateData(self):
"""Generates data for test. __testValues must be in sync"""
curContainer = DataContainer("myblock")
aCat = DataCategory("pdbx_test")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("details")
aCat.append([1, "data_my_big_data_file"])
aCat.append([2, "loop_my_big_data_loop"])
aCat.append([3, "save_my_big_data_saveframe"])
aCat.append([4, "_category.item"])
aCat.append([5, "Data_my_big_data_file"])
aCat.append([6, "Loop_my_big_data_loop"])
aCat.append([7, "Save_my_big_data_saveframe"])
aCat.append([8, "DatA_my_big_data_file"])
curContainer.append(aCat)
return curContainer
def testRowDictInitialization(self):
"""Test case - Row dictionary initialization of a data category and data block
"""
try:
#
rLen = 10
fn = self.__pathOutputFile5
attributeNameList = ["a", "b", "c", "d"]
rowList = [{"a": 1, "b": 2, "c": 3, "d": 4} for i in range(rLen)]
nameCat = "myCategory"
#
#
curContainer = DataContainer("myblock")
aCat = DataCategory(nameCat, attributeNameList, rowList)
aCat.append({"a": 1, "b": 2, "c": 3, "d": 4})
aCat.append({"a": 1, "b": 2, "c": 3, "d": 4})
aCat.extend(rowList)
curContainer.append(aCat)
aCat.renameAttributes({"a": "aa", "b": "bb", "c": "cc", "d": "dd"})
aCat.setName("renamedCategory")
#
#
myContainerList = []
myContainerList.append(curContainer)
ofh = open(fn, "w")
pdbxW = PdbxWriter(ofh)
pdbxW.write(myContainerList)
ofh.close()
myContainerList = []
ifh = open(fn, "r")
pRd = PdbxReader(ifh)
pRd.read(myContainerList)
ifh.close()
for container in myContainerList:
for objName in container.getObjNameList():
name, aList, rList = container.getObj(objName).get()
logger.debug("Recovered data category %s", name)
logger.debug("Attribute list %r", repr(aList))
logger.debug("Row list %r", repr(rList))
self.assertEqual(len(myContainerList), 1)
self.assertEqual(len(rList), 2 * rLen + 2)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def test_update_data_file(self, writer_paths):
myDataList = []
curContainer = DataContainer("myblock")
aCat = DataCategory("pdbx_seqtool_mapping_ref")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("entity_id")
aCat.appendAttribute("auth_mon_id")
aCat.appendAttribute("auth_mon_num")
aCat.appendAttribute("pdb_chain_id")
aCat.appendAttribute("ref_mon_id")
aCat.appendAttribute("ref_mon_num")
aCat.append((1, 2, 3, 4, 5, 6, 7))
aCat.append((1, 2, 3, 4, 5, 6, 7))
aCat.append((1, 2, 3, 4, 5, 6, 7))
aCat.append((1, 2, 3, 4, 5, 6, 7))
curContainer.append(aCat)
myDataList.append(curContainer)
with open(str(writer_paths['pathOutputFile1']), "w") as ofh:
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
#
# Read and update the data -
#
myDataList = []
with open(str(writer_paths['pathOutputFile1']), "r") as ifh:
pRd = PdbxReader(ifh)
pRd.read(myDataList)
#
myBlock = myDataList[0]
# myBlock.printIt()
myCat = myBlock.getObj('pdbx_seqtool_mapping_ref')
# myCat.printIt()
for iRow in range(0, myCat.getRowCount()):
myCat.setValue('some value', 'ref_mon_id', iRow)
myCat.setValue(100, 'ref_mon_num', iRow)
with open(str(writer_paths['pathOutputFile2']), "w") as ofh:
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
assert len(myDataList) == 1
def test_update_data_file(self, rw_data):
myDataList = []
curContainer = DataContainer("myblock")
aCat = DataCategory("pdbx_seqtool_mapping_ref")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("entity_id")
aCat.appendAttribute("auth_mon_id")
aCat.appendAttribute("auth_mon_num")
aCat.appendAttribute("pdb_chain_id")
aCat.appendAttribute("ref_mon_id")
aCat.appendAttribute("ref_mon_num")
aCat.append([9, 2, 3, 4, 5, 6, 7])
aCat.append([10, 2, 3, 4, 5, 6, 7])
aCat.append([11, 2, 3, 4, 5, 6, 7])
aCat.append([12, 2, 3, 4, 5, 6, 7])
curContainer.append(aCat)
myDataList.append(curContainer)
ofh = open(str(rw_data['pathOutputFile1']), "w")
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
ofh.close()
myDataList = []
ifh = open(str(rw_data['pathOutputFile1']), "r")
pRd = PdbxReader(ifh)
pRd.read(myDataList)
ifh.close()
myBlock = myDataList[0]
myCat = myBlock.getObj('pdbx_seqtool_mapping_ref')
for iRow in range(0, myCat.getRowCount()):
myCat.setValue('some value', 'ref_mon_id', iRow)
myCat.setValue(100, 'ref_mon_num', iRow)
with open(str(rw_data['pathOutputFile2']), "w") as ofh:
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
assert len(myDataList) == 1
def test_row_list_initialization(self, rw_data):
fn = rw_data['pathOutputFile4']
attributeNameList = ['aOne', 'aTwo', 'aThree', 'aFour', 'aFive', 'aSix', 'aSeven', 'aEight', 'aNine', 'aTen']
rowList = [[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
]
nameCat = 'myCategory'
curContainer = DataContainer("myblock")
aCat = DataCategory(nameCat, attributeNameList, rowList)
curContainer.append(aCat)
myContainerList = []
myContainerList.append(curContainer)
ofh = open(str(fn), "w")
pdbxW = PdbxWriter(ofh)
pdbxW.write(myContainerList)
ofh.close()
myContainerList = []
ifh = open(str(fn), "r")
pRd = PdbxReader(ifh)
pRd.read(myContainerList)
ifh.close()
for container in myContainerList:
for objName in container.getObjNameList():
name, aList, rList = container.getObj(objName).get()
print("Recovered data category %s\n" % name)
print("Attribute list %r\n" % repr(aList))
print("Row list %r\n" % repr(rList))
assert len(myContainerList) == 1
def _createfile1(pathout):
my_data_list = []
cur_container = DataContainer("myblock")
acat = DataCategory("pdbx_item_enumeration")
acat.appendAttribute("name")
acat.appendAttribute("value")
acat.appendAttribute("detail")
acat.append(('1', '2', '3'))
cur_container.append(acat)
acat = DataCategory("exptl")
acat.appendAttribute('absorpt_coefficient_mu')
acat.appendAttribute('entry_id')
acat.appendAttribute('method')
acat.appendAttribute('details')
acat.append(('?', 'D_12345', 'X-RAY DIFFRACTION', 'some details'))
cur_container.append(acat)
acat = DataCategory("struct")
acat.appendAttribute('title')
acat.appendAttribute('pdbx_descriptor')
acat.append(('Start title', 'Start Descriptor'))
cur_container.append(acat)
my_data_list.append(cur_container)
# Second block
cur_container = DataContainer("secondblock")
acat = DataCategory("pdbx_item_enumeration")
acat.appendAttribute("name")
acat.appendAttribute("value")
acat.appendAttribute("detail")
acat.append(('3', '2', '1'))
cur_container.append(acat)
my_data_list.append(cur_container)
with open(pathout, "w") as ofh:
pdbxw = PdbxWriter(ofh)
pdbxw.setAlignmentFlag(flag=True)
pdbxw.write(my_data_list)
def _createfile1(pathout):
my_data_list = []
cur_container = DataContainer("myblock")
acat = DataCategory("pdbx_item_enumeration")
acat.appendAttribute("name")
acat.appendAttribute("value")
acat.appendAttribute("detail")
acat.append(("1", "2", "3"))
cur_container.append(acat)
acat = DataCategory("exptl")
acat.appendAttribute("absorpt_coefficient_mu")
acat.appendAttribute("entry_id")
acat.appendAttribute("method")
acat.appendAttribute("details")
acat.append(("?", "D_12345", "X-RAY DIFFRACTION", "some details"))
cur_container.append(acat)
acat = DataCategory("struct")
acat.appendAttribute("title")
acat.appendAttribute("pdbx_descriptor")
acat.append(("Start title", "Start Descriptor"))
cur_container.append(acat)
my_data_list.append(cur_container)
# Second block
cur_container = DataContainer("secondblock")
acat = DataCategory("pdbx_item_enumeration")
acat.appendAttribute("name")
acat.appendAttribute("value")
acat.appendAttribute("detail")
acat.append(("3", "2", "1"))
cur_container.append(acat)
my_data_list.append(cur_container)
with open(pathout, "w") as ofh:
pdbxw = PdbxWriter(ofh)
pdbxw.setAlignmentFlag(flag=True)
pdbxw.write(my_data_list)
def testSingleRow(self):
"""Test case - read /write single row and null row in data file
"""
try:
#
myDataList = []
# ofh = open(self.__pathOutputFile1, "w")
curContainer = DataContainer("myblock")
aCat = DataCategory("pdbx_seqtool_mapping_ref")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("entity_id")
aCat.appendAttribute("auth_mon_id")
aCat.appendAttribute("auth_mon_num")
aCat.appendAttribute("pdb_chain_id")
aCat.appendAttribute("ref_mon_id")
aCat.appendAttribute("ref_mon_num")
aCat.appendAttribute("details")
aCat.append([1, 2, 3, 4, 5, 6, 7, "data_my_big_data_file"])
aCat.append([1, 2, 3, 4, 5, 6, 7, "loop_my_big_data_loop"])
aCat.append([1, 2, 3, 4, 5, 6, 7, "save_my_big_data_saveframe"])
aCat.append([1, 2, 3, 4, 5, 6, 7, "_category.item"])
# aCat.dumpIt()
curContainer.append(aCat)
#
bCat = curContainer.getObj("pdbx_seqtool_mapping_ref")
logger.debug("----attribute list %r", bCat.getAttributeList())
row = bCat.getRow(0)
logger.debug("----ROW %r", row)
#
with open(self.__pathOutputFile2, "w") as ofh:
myDataList.append(curContainer)
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
self.assertEqual(len(myDataList), 1)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def test_row_dict_initialization(self, rw_data):
rLen = 10
fn = rw_data['pathOutputFile5']
attributeNameList = ['a', 'b', 'c', 'd']
rowList = [{'a': 1, 'b': 2, 'c': 3, 'd': 4} for i in range(rLen)]
nameCat = 'myCategory'
#
#
curContainer = DataContainer("myblock")
aCat = DataCategory(nameCat, attributeNameList, rowList)
aCat.append({'a': 1, 'b': 2, 'c': 3, 'd': 4})
aCat.append({'a': 1, 'b': 2, 'c': 3, 'd': 4})
aCat.extend(rowList)
curContainer.append(aCat)
aCat.renameAttributes({'a': 'aa', 'b': 'bb', 'c': 'cc', 'd': 'dd'})
aCat.setName('renamedCategory')
#
#
myContainerList = []
myContainerList.append(curContainer)
ofh = open(str(fn), "w")
pdbxW = PdbxWriter(ofh)
pdbxW.write(myContainerList)
ofh.close()
myContainerList = []
ifh = open(str(fn), "r")
pRd = PdbxReader(ifh)
pRd.read(myContainerList)
ifh.close()
for container in myContainerList:
for objName in container.getObjNameList():
name, aList, rList = container.getObj(objName).get()
print("Recovered data category %s\n" % name)
print("Attribute list %r\n" % repr(aList))
print("Row list %r\n" % repr(rList))
assert len(myContainerList) == 1
assert len(rList) == 2 * rLen + 2
def test_write_data_file(self, rw_data):
myDataList = []
curContainer = DataContainer("myblock")
aCat = DataCategory("pdbx_seqtool_mapping_ref")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("entity_id")
aCat.appendAttribute("auth_mon_id")
aCat.appendAttribute("auth_mon_num")
aCat.appendAttribute("pdb_chain_id")
aCat.appendAttribute("ref_mon_id")
aCat.appendAttribute("ref_mon_num")
aCat.append([1, 2, 3, 4, 5, 6, 7])
aCat.append([1, 2, 3, 4, 5, 6, 7])
aCat.append([1, 2, 3, 4, 5, 6, 7])
aCat.append([1, 2, 3, 4, 5, 6, 7])
aCat.append([7, 6, 5, 4, 3, 2, 1])
curContainer.append(aCat)
myDataList.append(curContainer)
with open(str(rw_data['pathOutputFile1']), "w") as ofh:
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
assert len(myDataList) == 1
def test_write_data_file(self, writer_paths):
myDataList = []
curContainer = DataContainer("myblock")
aCat = DataCategory("pdbx_seqtool_mapping_ref")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("entity_id")
aCat.appendAttribute("auth_mon_id")
aCat.appendAttribute("auth_mon_num")
aCat.appendAttribute("pdb_chain_id")
aCat.appendAttribute("ref_mon_id")
aCat.appendAttribute("ref_mon_num")
aCat.append(
(1, 2, 3, 4, '55555555555555555555555555555555555555555555', 6, 7))
aCat.append((1, 2, 3, 4, '5555', 6, 7))
aCat.append((1, 2, 3, 4, '5555555555', 6, 7))
aCat.append((1, 2, 3, 4, '5', 6, 7))
curContainer.append(aCat)
myDataList.append(curContainer)
with open(str(writer_paths['pathOutputFile1']), "w") as ofh:
pdbxW = PdbxWriter(ofh)
pdbxW.setAlignmentFlag(flag=True)
pdbxW.write(myDataList)
assert len(myDataList) == 1
def __dictionaryPragma(self, dictName, dictDescription, version,
updateDate, comment):
""" Add CIF dictionary header details including name, version and history.
; Returns:
Data container (object) data container with dictionary history and version details
"""
#
dataH = DataContainer("pdbx_vrpt_ext.dic")
dc = DataCategory("datablock", attributeNameList=["id", "description"])
dc.append([dictName, dictDescription])
dataH.append(dc)
dc = DataCategory(
"dictionary",
attributeNameList=["title", "datablock_id", "version"])
dc.append([dictName, dictName, version])
dataH.append(dc)
dc = DataCategory("dictionary_history",
attributeNameList=["version", "update", "revision"])
dc.append([version, updateDate, comment])
dataH.append(dc)
return dataH
class mmCIFUtil:
"""Using pdbx mmCIF utility to parse mmCIF file"""
def __init__(self, verbose=False, log=sys.stderr, filePath=None): # pylint: disable=unused-argument
# self.__verbose = verbose
self.__lfh = log
self.__filePath = filePath
self.__dataList = []
self.__dataMap = {}
self.__container = None
self.__blockID = None
self.__read()
#
def __read(self):
if not self.__filePath:
return
#
try:
ifh = open(self.__filePath, "r")
pRd = PdbxReader(ifh)
pRd.read(self.__dataList)
ifh.close()
if self.__dataList:
self.__container = self.__dataList[0]
self.__blockID = self.__container.getName()
idx = 0
for container in self.__dataList:
self.__dataMap[container.getName()] = idx
idx += 1
#
#
except Exception as e:
self.__lfh.write("Read %s failed %s.\n" %
(self.__filePath, str(e)))
#
def GetBlockID(self):
"""Return first block ID"""
return self.__blockID
def GetValueAndItemByBlock(self, blockName, catName):
"""Get category values and item names"""
dList = []
iList = []
if blockName not in self.__dataMap:
return dList, iList
#
catObj = self.__dataList[self.__dataMap[blockName]].getObj(catName)
if not catObj:
return dList, iList
#
iList = catObj.getAttributeList()
rowList = catObj.getRowList()
for row in rowList:
tD = {}
for idxIt, itName in enumerate(iList):
if row[idxIt] != "?" and row[idxIt] != ".":
tD[itName] = row[idxIt]
#
if tD:
dList.append(tD)
#
#
return dList, iList
def GetValueAndItem(self, catName):
dList, iList = self.GetValueAndItemByBlock(self.__blockID, catName)
return dList, iList
def GetValue(self, catName):
"""Get category values based on category name 'catName'. The results are stored
in a list of dictionaries with item name as key
"""
dList, _iList = self.GetValueAndItemByBlock(self.__blockID, catName)
return dList
def GetSingleValue(self, catName, itemName):
"""Get the first value of item name 'itemName' from 'itemName' item in 'catName' category."""
text = ""
dlist = self.GetValue(catName)
if dlist:
if itemName in dlist[0]:
text = dlist[0][itemName]
return text
#
def UpdateSingleRowValue(self, catName, itemName, row, value):
"""Update value in single row"""
catObj = self.__container.getObj(catName)
if catObj is None:
return
#
catObj.setValue(value, itemName, row)
def UpdateMultipleRowsValue(self, catName, itemName, value):
"""Update value in multiple rows"""
catObj = self.__container.getObj(catName)
if catObj is None:
return
#
rowNo = catObj.getRowCount()
for row in range(0, rowNo):
catObj.setValue(value, itemName, row)
#
def AddBlock(self, blockID):
"""Add Data Block"""
self.__container = DataContainer(blockID)
self.__blockID = blockID
self.__dataMap[blockID] = len(self.__dataList)
self.__dataList.append(self.__container)
def AddCategory(self, categoryID, items):
"""Add Category"""
category = DataCategory(categoryID)
for item in items:
category.appendAttribute(item)
#
self.__container.append(category)
def RemoveCategory(self, categoryID):
return self.__container.remove(categoryID)
def InsertData(self, categoryID, dataList):
""""""
catObj = self.__container.getObj(categoryID)
if catObj is None:
return
#
for data in dataList:
catObj.append(data)
#
def WriteCif(self, outputFilePath=None):
"""Write out cif file"""
if not outputFilePath:
return
#
ofh = open(outputFilePath, "w")
pdbxW = PdbxWriter(ofh)
pdbxW.write(self.__dataList)
ofh.close()
def GetCategories(self):
return self.__container.getObjNameList()
def GetAttributes(self, category):
return self.__container.getObj(category).getAttributeList()
def category_as_dict(self, category, block=None):
if block is None:
block = self.__blockID
values, attributes = self.GetValueAndItemByBlock(block, category)
data = [[x[y] if y in x else None for y in attributes] for x in values]
return {category: {"Items": attributes, "Values": data}}
def block_as_dict(self, block=None):
if block is None:
block = self.__blockID
data = {}
for category in self.GetCategories():
data.update(self.category_as_dict(category, block=block))
return data
def __buildCif(self, rD, containerName="vrpt"):
""" Construct a mmCIF data category objects for the input
extracted data.
Args:
rD (dict): extracted data organized by category.
containerName (str) : data block name
Returns:
containers (list): data container list
"""
#
curContainer = DataContainer(containerName)
for elName in rD:
catName = elName
if (not rD[elName]) or (not self.__attribD[catName]) or (
catName in ["programs"]):
continue
hasOrdinal = "ordinal" in self.__attribD[catName]
rowList = rD[elName]
# Find the unique attribute content across the rowlist and the ordinal value
atS = set()
for ii, rowD in enumerate(rowList, 1):
if hasOrdinal:
rowD["ordinal"] = ii
if "icode" in rowD:
rowD["icode"] = str(rowD["icode"]).strip()
if "altcode" in rowD:
rowD["altcode"] = str(rowD["altcode"]).strip()
atS.update(rowD.keys())
attributeNameList = list(atS)
#
# Set a reasonable order for these attributes
#
sD = {ky: self.__atOrdD[ky] for ky in attributeNameList}
srtAtL = [
tup[0]
for tup in sorted(sD.items(), key=operator.itemgetter(1))
]
logger.debug("Category %s sorted attributes %r", catName, srtAtL)
aCat = DataCategory(catName, srtAtL, rowList)
curContainer.append(aCat)
#
# Adjust schema names -
#
atD = self.__dictionaryMap["attributes"]
for catName in curContainer.getObjNameList():
catObj = curContainer.getObj(catName)
atNameList = catObj.getAttributeList()
mapD = {}
mapCatName = self.__dictionaryMap["categories"][
catName] if catName in self.__dictionaryMap[
"categories"] else catName
for atName in atNameList:
mapD[atName] = atD[(catName, atName)]["at"] if (
catName, atName) in atD else atName
catObj.renameAttributes(mapD)
catObj.setName(mapCatName)
#
# Map provenance items from programs.properties -
#
catObj = curContainer.getObj("program")
if catObj and catObj.hasAttribute("properties"):
for iRow in range(catObj.getRowCount()):
pV = catObj.getValue("properties", iRow)
pVL = [v.strip() for v in pV.split(",")]
nL = [
self.__atMap[ky] if ky in self.__atMap else ky
for ky in pVL
]
catObj.setValue(",".join(nL), "properties", iRow)
# logger.info("Row %r properties %r" % (iRow, pV))
#
return [curContainer]
def __writeModel(self, targetId, targetObj, fitFD, fitXyzMapD,
fitAtomUnMappedL, matchD, modelId, modelPath):
"""Write the chemical component model for the input chemical component Id and associated atom mapping and
feature details --
ComponentAtomDetails = namedtuple("ComponentAtomDetails", "index atNo name aType x y z fCharge")
AlignAtomMap = namedtuple("AlignAtomMap", "refId refAtIdx refAtNo refAtName fitId fitAtIdx fitAtNo fitAtName")
AlignAtomUnMapped = namedtuple("AlignAtomUnMapped", "fitId fitAtIdx fitAtNo fitAtType fitAtName fitAtFormalCharge x y z fitNeighbors")
"""
try:
unMappedTypeD = defaultdict(int)
hAtomPrefix = "HEX"
variantType = self.__getBuildVariant(targetId)
#
if not self.__testUnMappedProtonation(fitAtomUnMappedL):
logger.info(
"Unmapped non-hydrogen atoms target %r model %r unMapped count (%d)",
targetId, modelId, len(fitAtomUnMappedL))
return False, variantType
# Get atom partners for the unmapped atoms
fitAtMapD = {}
for refAtName, fAtTup in fitXyzMapD.items():
fitAtMapD[fAtTup.atName] = refAtName
if fitAtomUnMappedL:
# Check if neighbors are all mapped
ok = True
for fitUnTup in fitAtomUnMappedL:
for nAtName in fitUnTup.fitNeighbors:
if nAtName not in fitAtMapD:
ok = False
logger.info(
"Missing mapped neighbor for %r target %r model %r",
nAtName, targetId, modelId)
break
if not ok:
return False, variantType
else:
logger.debug("%s match has unmapped protonation", modelId)
variantType = "tautomer_protomer"
#
#
kList = ["xyz", "SMILES", "SMILES_STEREO", "InChI", "InChIKey"]
for k in kList:
if k not in fitFD:
logger.error(
"Fit feature dictionary for %s missing key %s",
targetId, k)
return False, variantType
# ------------
dataContainer = DataContainer(modelId)
#
myContainer = targetObj
dbName = myContainer.getName()
if dbName.upper() != targetId.upper():
logger.info("mismatch datablock (%r) and targetId (%r)",
dbName, targetId)
cObj = None
if myContainer.exists("chem_comp"):
cObj = myContainer.getObj("chem_comp")
#
#
catName = "pdbx_chem_comp_model"
if not dataContainer.exists(catName):
dataContainer.append(
DataCategory(catName, attributeNameList=["id", "comp_id"]))
#
parentId = targetId.split("|")[0]
wObj = dataContainer.getObj(catName)
wObj.setValue(modelId, "id", 0)
wObj.setValue(parentId, "comp_id", 0)
#
# -------- ---------
catName = "pdbx_chem_comp_model_atom"
if not dataContainer.exists(catName):
dataContainer.append(
DataCategory(catName,
attributeNameList=[
"model_id", "atom_id", "type_symbol",
"charge", "model_Cartn_x",
"model_Cartn_y", "model_Cartn_z",
"ordinal_id"
]))
wObj = dataContainer.getObj(catName)
#
if myContainer.exists("chem_comp_atom"):
cObj = myContainer.getObj("chem_comp_atom")
#
# Only write the mapped atoms in case we are missing hydrogens in the mapping
#
jj = 0
for ii in range(cObj.getRowCount()):
atName = cObj.getValue("atom_id", ii)
atType = cObj.getValue("type_symbol", ii)
if atName not in fitXyzMapD:
unMappedTypeD[atType] += 1
continue
fitXyz = fitXyzMapD[atName]
#
# fCharge = cObj.getValue("charge", ii)
#
wObj.setValue(modelId, "model_id", jj)
wObj.setValue(atName, "atom_id", jj)
wObj.setValue(atType, "type_symbol", jj)
#
wObj.setValue(fitXyz.atFormalCharge, "charge", jj)
wObj.setValue("%.4f" % fitXyz.x, "model_Cartn_x", jj)
wObj.setValue("%.4f" % fitXyz.y, "model_Cartn_y", jj)
wObj.setValue("%.4f" % fitXyz.z, "model_Cartn_z", jj)
wObj.setValue(jj + 1, "ordinal_id", jj)
jj += 1
#
# Add the unmapped atoms ...
# AlignAtomUnMapped = namedtuple("AlignAtomUnMapped", "fitId fitAtIdx fitAtNo fitAtType fitAtName fitNeighbors")
ii = wObj.getRowCount()
for jj, uTup in enumerate(fitAtomUnMappedL):
refAtomName = hAtomPrefix + str(jj)
wObj.setValue(modelId, "model_id", ii)
wObj.setValue(refAtomName, "atom_id", ii)
wObj.setValue(uTup.fitAtType, "type_symbol", ii)
wObj.setValue(uTup.fitAtFormalCharge, "charge", ii)
wObj.setValue("%.4f" % uTup.x, "model_Cartn_x", ii)
wObj.setValue("%.4f" % uTup.y, "model_Cartn_y", ii)
wObj.setValue("%.4f" % uTup.z, "model_Cartn_z", ii)
wObj.setValue(ii + 1, "ordinal_id", ii)
# -------- ---------
catName = "pdbx_chem_comp_model_bond"
if not dataContainer.exists(catName):
dataContainer.append(
DataCategory(catName,
attributeNameList=[
"model_id", "atom_id_1", "atom_id_2",
"value_order", "ordinal_id"
]))
wObj = dataContainer.getObj(catName)
#
if myContainer.exists("chem_comp_bond"):
cObj = myContainer.getObj("chem_comp_bond")
#
jj = 0
for ii in range(cObj.getRowCount()):
at1 = cObj.getValue("atom_id_1", ii)
if at1 not in fitXyzMapD:
continue
at2 = cObj.getValue("atom_id_2", ii)
if at2 not in fitXyzMapD:
continue
bType = cObj.getValue("value_order", ii)
#
wObj.setValue(modelId, "model_id", jj)
wObj.setValue(at1, "atom_id_1", jj)
wObj.setValue(at2, "atom_id_2", jj)
wObj.setValue(bType, "value_order", jj)
wObj.setValue(jj + 1, "ordinal_id", jj)
jj += 1
#
ii = wObj.getRowCount()
for jj, uTup in enumerate(fitAtomUnMappedL):
at1 = hAtomPrefix + str(jj)
for nAt in uTup.fitNeighbors:
at2 = fitAtMapD[nAt]
wObj.setValue(modelId, "model_id", ii)
wObj.setValue(at1, "atom_id_1", ii)
wObj.setValue(at2, "atom_id_2", ii)
wObj.setValue("SING", "value_order", ii)
wObj.setValue(ii + 1, "ordinal_id", ii)
# -------- ---------
catName = "pdbx_chem_comp_model_descriptor"
if not dataContainer.exists(catName):
dataContainer.append(
DataCategory(
catName,
attributeNameList=["model_id", "type", "descriptor"]))
wObj = dataContainer.getObj(catName)
#
ii = 0
wObj.setValue(modelId, "model_id", ii)
wObj.setValue("SMILES", "type", ii)
wObj.setValue(fitFD["SMILES"], "descriptor", ii)
ii += 1
wObj.setValue(modelId, "model_id", ii)
wObj.setValue("SMILES_CANONICAL", "type", ii)
wObj.setValue(fitFD["SMILES_STEREO"], "descriptor", ii)
ii += 1
wObj.setValue(modelId, "model_id", ii)
wObj.setValue("InChI", "type", ii)
wObj.setValue(fitFD["InChI"], "descriptor", ii)
ii += 1
wObj.setValue(modelId, "model_id", ii)
wObj.setValue("InChIKey", "type", ii)
wObj.setValue(fitFD["InChIKey"], "descriptor", ii)
#
# -------- ---------
if matchD["queryId"] is not None:
catName = "pdbx_chem_comp_model_reference"
if not dataContainer.exists(catName):
dataContainer.append(
DataCategory(catName,
attributeNameList=[
"model_id", "db_name", "db_code"
]))
wObj = dataContainer.getObj(catName)
ii = 0
wObj.setValue(modelId, "model_id", ii)
wObj.setValue("COD", "db_name", ii)
wObj.setValue(matchD["queryId"], "db_code", ii)
#
featureD = {}
v = matchD["rValue"]
vS = str(v)
if v is not None and len(vS) > 0:
featureD["r_factor"] = "%.3f" % float(v)
#
v = matchD["diffrnTemp"]
vS = str(v)
# remove string artifacts from temperature string ...
if v is not None and len(vS) > 0:
tV = vS.upper()
try:
if tV.endswith("DEG.C"):
tV = tV.replace("AT", "")
tV = tV.replace("DEG.C", "")
tV = float(tV.strip())
tV = tV + 273.15
else:
tV = tV.replace("AT", "")
tV = tV.replace("K", "")
tV = float(tV.strip())
featureD["experiment_temperature"] = tV
except Exception as e:
logger.exception(
"Temperature conversion fails for %s (%r) with %s",
modelId, vS, tV)
#
v = matchD["publicationDOI"]
vS = str(v)
if v is not None and len(vS) > 0:
featureD["publication_doi"] = v
#
v = matchD["version"]
vS = str(v)
if v is not None and len(vS) > 0:
featureD["cod_version"] = v
#
if matchD["radiationSource"] and "neutron" in matchD[
"radiationSource"]:
featureD["neutron_radiation_experiment"] = True
if matchD["hasDisorder"] in ["Y"]:
featureD["has_disorder"] = True
#
if len(unMappedTypeD) == 1 and "H" in unMappedTypeD:
logger.info("model %r heavy_atoms_only", modelId)
featureD["heavy_atoms_only"] = True
else:
featureD["all_atoms_have_sites"] = True
# -------- ---------
catName = "pdbx_chem_comp_model_feature"
if not dataContainer.exists(catName):
dataContainer.append(
DataCategory(catName,
attributeNameList=[
"model_id", "feature_name",
"feature_value"
]))
wObj = dataContainer.getObj(catName)
#
fKeyList = [
"experiment_temperature", "publication_doi", "r_factor",
"csd_version"
]
ii = 0
for fKey in fKeyList:
if fKey in featureD:
wObj.setValue(modelId, "model_id", ii)
wObj.setValue(fKey, "feature_name", ii)
wObj.setValue(str(featureD[fKey]), "feature_value", ii)
ii += 1
#
boolKeyList = [
"has_disorder", "neutron_radiation_experiment",
"heavy_atoms_only", "all_atoms_have_sites"
]
for fKey in boolKeyList:
if fKey in featureD:
if featureD[fKey]:
wObj.setValue(modelId, "model_id", ii)
wObj.setValue(fKey, "feature_name", ii)
wObj.setValue("Y", "feature_value", ii)
ii += 1
#
if variantType:
wObj.setValue(modelId, "model_id", ii)
wObj.setValue(variantType + "_match", "feature_name", ii)
wObj.setValue("Y", "feature_value", ii)
ii += 1
# -------- ---------
catName = "pdbx_chem_comp_model_audit"
if not dataContainer.exists(catName):
dataContainer.append(
DataCategory(
catName,
attributeNameList=["model_id", "action_type", "date"]))
wObj = dataContainer.getObj(catName)
#
ii = 0
wObj.setValue(modelId, "model_id", ii)
wObj.setValue("Initial release", "action_type", ii)
wObj.setValue(self.__getToday(), "date", ii)
# wObj.setValue('RCSB', 'processing_site', ii)
# wObj.setValue('JDW', 'annotator', ii)
# wObj.setValue('?', 'details', ii)
#
mU = MarshalUtil(workPath=self.__cachePath)
ok = mU.doExport(modelPath, [dataContainer], fmt="mmcif")
return ok, variantType
except Exception as e:
logger.exception("Failing for %r with %s", targetId, str(e))
return False, ""
def testGenDDLm(self):
"""Generating alternative DDLm metadata format. (starting point)"""
try:
myIo = IoAdapterPy(self.__verbose, self.__lfh)
self.__containerList = myIo.readFile(
inputFilePath=self.__pathPdbxDictionary)
dApi = DictionaryApi(containerList=self.__containerList,
consolidate=True,
verbose=self.__verbose)
parentD = dApi.getParentDictionary()
#
oCList = []
dDef = DataContainer("mmcif_pdbx_ddlm_auto")
dc = DataCategory("dictionary")
dc.appendAttribute("title")
dc.appendAttribute("class")
dc.appendAttribute("version")
dc.appendAttribute("date")
dc.appendAttribute("ddl_conformance")
dc.appendAttribute("text")
dc.append([
"mmcif_pdbx_ddlm_auto", "Instance", "latest", "2018-03-09",
"ddlm best effort",
"Software converted PDBx dictionary using DDLm semantics"
])
dDef.append(dc)
oCList.append(dDef)
catIdx = dApi.getCategoryIndex()
for catName in sorted(catIdx.keys()):
attNameList = catIdx[catName]
# created definition container -
cDef = DefinitionContainer(catName)
oCList.append(cDef)
#
dc = DataCategory("definition")
dc.appendAttribute("id")
dc.appendAttribute("scope")
dc.appendAttribute("class")
dc.appendAttribute("update")
dc.append([catName, "Category", "Loop", "2018-03-09"])
cDef.append(dc)
val = dApi.getCategoryDescription(category=catName)
dc = DataCategory("description")
dc.appendAttribute("text")
dc.append([val])
cDef.append(dc)
#
dc = DataCategory("name")
dc.appendAttribute("category_id")
dc.appendAttribute("object_id")
valList = dApi.getCategoryGroupList(category=catName)
pcg = catName
for val in valList:
if val != "inclusive_group":
pcg = val
break
dc.append([catName, pcg])
cDef.append(dc)
valList = dApi.getCategoryKeyList(category=catName)
if not valList:
self.__lfh.write("Missing caegory key for category %s\n" %
catName)
else:
dc = DataCategory("category")
dc.appendAttribute("key_id")
kItemName = CifName.itemName(catName, "synthetic_key")
dc.append([kItemName])
cDef.append(dc)
iDef = DefinitionContainer(kItemName)
self.__makeKeyItem(catName, "synthetic_key", valList, iDef)
oCList.append(iDef)
for attName in attNameList:
itemName = CifName.itemName(catName, attName)
iDef = DefinitionContainer(itemName)
oCList.append(iDef)
#
dc = DataCategory("definition")
dc.appendAttribute("id")
dc.appendAttribute("scope")
dc.appendAttribute("class")
dc.appendAttribute("update")
dc.append([itemName, "Item", "Single", "2013-08-22"])
iDef.append(dc)
#
val = dApi.getDescription(category=catName,
attribute=attName)
dc = DataCategory("description")
dc.appendAttribute("text")
dc.append([val])
iDef.append(dc)
#
dc = DataCategory("name")
dc.appendAttribute("category_id")
dc.appendAttribute("object_id")
#
if itemName in parentD:
dc.appendAttribute("linked_item_id")
dc.append([catName, attName, parentD[itemName][0]])
else:
dc.append([catName, attName])
iDef.append(dc)
#
#
aliasList = dApi.getItemAliasList(category=catName,
attribute=attName)
if aliasList:
dc = DataCategory("alias")
dc.appendAttribute("definition_id")
for alias in aliasList:
dc.append([alias[0]])
iDef.append(dc)
enList = dApi.getEnumListAltWithDetail(category=catName,
attribute=attName)
tC = dApi.getTypeCode(category=catName, attribute=attName)
tcontainer = "Single"
purpose = "Describe"
source = "Recorded"
contents = "Text"
#
if tC is None:
self.__lfh.write("Missing data type attribute %s\n" %
attName)
elif tC in [
"code", "atcode", "name", "idname", "symop", "fax",
"phone", "email", "code30", "ec-type"
]:
purpose = "Encode"
contents = "Text"
source = "Assigned"
elif tC in ["ucode"]:
purpose = "Encode"
contents = "Code"
source = "Assigned"
elif tC in ["line", "uline", "text"]:
purpose = "Describe"
source = "Recorded"
contents = "Text"
elif tC in ["int"]:
purpose = "Number"
source = "Recorded"
contents = "Integer"
elif tC in ["int-range"]:
purpose = "Number"
source = "Recorded"
contents = "Range"
elif tC in ["float"]:
purpose = "Measurand"
source = "Recorded"
contents = "Real"
elif tC in ["float-range"]:
purpose = "Measurand"
source = "Recorded"
contents = "Range"
elif tC.startswith("yyyy"):
source = "Assigned"
contents = "Date"
purpose = "Describe"
if enList:
purpose = "State"
dc = DataCategory("type")
dc.appendAttribute("purpose")
dc.appendAttribute("source")
dc.appendAttribute("contents")
dc.appendAttribute("container")
dc.append([purpose, source, contents, tcontainer])
iDef.append(dc)
#
if enList:
dc = DataCategory("enumeration_set")
dc.appendAttribute("state")
dc.appendAttribute("detail")
for en in enList:
dc.append([en[0], en[1]])
iDef.append(dc)
dfv = dApi.getDefaultValue(category=catName,
attribute=attName)
bvList = dApi.getBoundaryList(category=catName,
attribute=attName)
if ((dfv is not None) and
(dfv not in ["?", "."])) or bvList:
row = []
dc = DataCategory("enumeration")
if dfv is not None:
dc.appendAttribute("default")
row.append(dfv)
if bvList:
dc.appendAttribute("range")
mminVp = -1000000
mmaxVp = 10000000
mminV = mmaxVp
mmaxV = mminVp
for bv in bvList:
minV = float(bv[0]) if bv[0] != "." else mminVp
maxV = float(bv[1]) if bv[1] != "." else mmaxVp
mminV = min(mminV, minV)
mmaxV = max(mmaxV, maxV)
if mminV == mminVp:
mminV = ""
if mmaxV == mmaxVp:
mmaxV = ""
row.append(str(mminV) + ":" + str(mmaxV))
dc.append(row)
iDef.append(dc)
myIo.writeFile(outputFilePath=os.path.join(
HERE, "test-output", "mmcif_pdbx_ddlm_auto.dic"),
containerList=oCList)
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()
def test_gen_ddlm(self, in_tmpdir, test_files):
myIo = IoAdapterPy()
containerList = myIo.readFile(
inputFilePath=str(test_files / 'mmcif_pdbx_v5_next.dic'))
dApi = DictionaryApi(containerList=containerList, consolidate=True)
parentD = dApi.getParentDictionary()
#
oCList = []
dDef = DataContainer('mmcif_pdbx_ddlm_auto')
dc = DataCategory("dictionary")
dc.appendAttribute("title")
dc.appendAttribute("class")
dc.appendAttribute("version")
dc.appendAttribute("date")
dc.appendAttribute("ddl_conformance")
dc.appendAttribute("text")
dc.append([
'mmcif_pdbx_ddlm_auto', 'Instance', 'latest', '2018-03-09',
'ddlm best effort',
'Software converted PDBx dictionary using DDLm semantics'
])
dDef.append(dc)
oCList.append(dDef)
catIdx = dApi.getCategoryIndex()
for catName in sorted(catIdx.keys()):
attNameList = catIdx[catName]
# created definition container -
cDef = DefinitionContainer(catName)
oCList.append(cDef)
#
dc = DataCategory("definition")
dc.appendAttribute("id")
dc.appendAttribute("scope")
dc.appendAttribute("class")
dc.appendAttribute("update")
dc.append([catName, "Category", "Loop", "2018-03-09"])
cDef.append(dc)
val = dApi.getCategoryDescription(category=catName)
dc = DataCategory("description")
dc.appendAttribute("text")
dc.append([val])
cDef.append(dc)
#
dc = DataCategory("name")
dc.appendAttribute("category_id")
dc.appendAttribute("object_id")
valList = dApi.getCategoryGroupList(category=catName)
pcg = catName
for val in valList:
if val != 'inclusive_group':
pcg = val
break
dc.append([catName, pcg])
cDef.append(dc)
valList = dApi.getCategoryKeyList(category=catName)
if len(valList) < 1:
print("Missing caegory key for category %s\n" % catName)
else:
dc = DataCategory("category")
dc.appendAttribute("key_id")
kItemName = CifName.itemName(catName, "synthetic_key")
dc.append([kItemName])
cDef.append(dc)
iDef = DefinitionContainer(kItemName)
self._makeKeyItem(catName, "synthetic_key", valList, iDef)
oCList.append(iDef)
for attName in attNameList:
itemName = CifName.itemName(catName, attName)
iDef = DefinitionContainer(itemName)
oCList.append(iDef)
#
dc = DataCategory("definition")
dc.appendAttribute("id")
dc.appendAttribute("scope")
dc.appendAttribute("class")
dc.appendAttribute("update")
dc.append([itemName, "Item", "Single", "2013-08-22"])
iDef.append(dc)
#
val = dApi.getDescription(category=catName, attribute=attName)
dc = DataCategory("description")
dc.appendAttribute("text")
dc.append([val])
iDef.append(dc)
#
dc = DataCategory("name")
dc.appendAttribute("category_id")
dc.appendAttribute("object_id")
#
if itemName in parentD:
dc.appendAttribute("linked_item_id")
dc.append([catName, attName, parentD[itemName][0]])
else:
dc.append([catName, attName])
iDef.append(dc)
#
#
aliasList = dApi.getItemAliasList(category=catName,
attribute=attName)
if len(aliasList) > 0:
dc = DataCategory("alias")
dc.appendAttribute("definition_id")
for alias in aliasList:
dc.append([alias[0]])
iDef.append(dc)
enList = dApi.getEnumListAltWithDetail(category=catName,
attribute=attName)
tC = dApi.getTypeCode(category=catName, attribute=attName)
tcontainer = 'Single'
purpose = 'Describe'
source = 'Recorded'
contents = 'Text'
#
if tC is None:
self.__lfh.write("Missing data type attribute %s\n" %
attName)
elif tC in [
'code', 'atcode', 'name', 'idname', 'symop', 'fax',
'phone', 'email', 'code30', 'ec-type'
]:
purpose = 'Encode'
contents = 'Text'
source = 'Assigned'
elif tC in ['ucode']:
purpose = 'Encode'
contents = 'Code'
source = 'Assigned'
elif tC in ['line', 'uline', 'text']:
purpose = 'Describe'
source = 'Recorded'
contents = 'Text'
elif tC in ['int']:
purpose = 'Number'
source = 'Recorded'
contents = 'Integer'
elif tC in ['int-range']:
purpose = 'Number'
source = 'Recorded'
contents = 'Range'
elif tC in ['float']:
purpose = 'Measurand'
source = 'Recorded'
contents = 'Real'
elif tC in ['float-range']:
purpose = 'Measurand'
source = 'Recorded'
contents = 'Range'
elif tC.startswith('yyyy'):
source = 'Assigned'
contents = 'Date'
purpose = 'Describe'
if len(enList) > 0:
purpose = 'State'
dc = DataCategory("type")
dc.appendAttribute("purpose")
dc.appendAttribute("source")
dc.appendAttribute("contents")
dc.appendAttribute("container")
dc.append([purpose, source, contents, tcontainer])
iDef.append(dc)
#
if (len(enList) > 0):
dc = DataCategory("enumeration_set")
dc.appendAttribute("state")
dc.appendAttribute("detail")
for en in enList:
dc.append([en[0], en[1]])
iDef.append(dc)
dfv = dApi.getDefaultValue(category=catName, attribute=attName)
bvList = dApi.getBoundaryList(category=catName,
attribute=attName)
if (((dfv is not None) and (dfv not in ['?', '.']))
or len(bvList) > 0):
row = []
dc = DataCategory("enumeration")
if dfv is not None:
dc.appendAttribute("default")
row.append(dfv)
if len(bvList) > 0:
dc.appendAttribute("range")
mminVp = -1000000
mmaxVp = 10000000
mminV = mmaxVp
mmaxV = mminVp
for bv in bvList:
minV = float(bv[0]) if bv[0] != '.' else mminVp
maxV = float(bv[1]) if bv[1] != '.' else mmaxVp
mminV = min(mminV, minV)
mmaxV = max(mmaxV, maxV)
if mminV == mminVp:
mminV = ''
if mmaxV == mmaxVp:
mmaxV = ''
row.append(str(mminV) + ":" + str(mmaxV))
dc.append(row)
iDef.append(dc)
myIo.writeFile(outputFilePath="mmcif_pdbx_ddlm_auto.dic",
containerList=oCList)
def __deserialize(self, fh, storeStringsAsBytes=False):
cL = []
try:
dec = BinaryCifDecoders(storeStringsAsBytes=storeStringsAsBytes)
bD = msgpack.unpack(fh)
#
logger.debug("bD.keys() %r", bD.keys())
logger.debug("bD['dataBlocks'] %s",
bD[self.__toBytes("dataBlocks")])
#
for dataBlock in bD[self.__toBytes("dataBlocks")]:
header = self.__fromBytes(
dataBlock[self.__toBytes("header")]) if self.__toBytes(
"header") in dataBlock else None
logger.debug("header %r", header)
logger.debug("dataBlock %r", dataBlock)
#
dc = DataContainer(header)
categoryList = dataBlock[self.__toBytes(
"categories")] if self.__toBytes(
"categories") in dataBlock else []
for category in categoryList:
catName = self.__fromBytes(
category[self.__toBytes("name")])[1:]
colList = category[self.__toBytes("columns")]
logger.debug("catName %r columns %r", catName, colList)
colD = OrderedDict()
atNameList = []
for col in colList:
logger.debug("col.keys() %r", col.keys())
atName = self.__fromBytes(col[self.__toBytes("name")])
atData = col[self.__toBytes("data")]
logger.debug("atData encoding (%d) data (%d)",
len(atData[self.__toBytes("encoding")]),
len(atData[self.__toBytes("data")]))
atMask = col[self.__toBytes("mask")]
logger.debug("catName %r atName %r", catName, atName)
logger.debug(" >atData.data %r",
atData[self.__toBytes("data")])
logger.debug(" >atData.encoding (%d) %r",
len(atData[self.__toBytes("encoding")]),
atData[self.__toBytes("encoding")])
logger.debug(" >mask %r", atMask)
tVal = dec.decode(
col[self.__toBytes("data")][self.__toBytes(
"data")], col[self.__toBytes("data")][
self.__toBytes("encoding")])
if col[self.__toBytes("mask")]:
mVal = dec.decode(
col[self.__toBytes("mask")][self.__toBytes(
"data")], col[self.__toBytes("mask")][
self.__toBytes("encoding")])
tVal = [
"?" if m == 2 else "." if m == 1 else d
for d, m in zip(tVal, mVal)
]
colD[atName] = tVal
atNameList.append(atName)
#
cObj = DataCategory(catName, attributeNameList=atNameList)
genL = [colGen for colGen in colD.values()]
for row in zip(*genL):
logger.debug("row %r", row)
cObj.append(row)
#
dc.append(cObj)
cL.append(dc)
except Exception as e:
logger.exception("Failing with %s", str(e))
return cL
class CIF(object):
"""
This class uses the mmcif library to create an mmCIF-like object
Each object has one container, with a container ID and a list of DataCategory objects
"""
DUMMY_CONTAINER_ID = "emd_0000"
def __init__(self, cif_name_name):
self.filename = cif_name_name
# self.__dataList needed for PDBxWriter
self.__dataList = []
self.__container_id = None
self.__container = None
self.__dataMap = {}
def write(self):
"""
Given a file name, a pdbx writer is used to write data stored in self.__dataList
:return written: a boolean; True when pdf writer is finished
"""
written = False
if self.filename:
ofh = open(self.filename, "w")
pdbx_writer = PdbxWriter(ofh)
pdbx_writer.write(self.__dataList)
ofh.close()
written = True
return written
def add_container(self, container_id):
"""
This method provides the basic functionality to set up a container
:param container_id: a string; an mmcif category e.g. 'emd_admin'
:return:
"""
added = False
self.__container_id = container_id
self.__container = DataContainer(container_id)
self.__dataMap[container_id] = len(self.__dataList)
self.__dataList.append(self.__container)
if self.__container is not None:
added = True
return added
def prepare_container(self, container_id):
"""
Creates a container is it doesn't exist using either provided value or the dummy value
:param container_id: a string; an mmcif category e.g. 'emd_admin'
:return:
"""
if not self.__container:
if container_id is None:
container_id = self.DUMMY_CONTAINER_ID
return self.add_container(container_id)
def add_category(self, category_id, items):
"""
This method creates a data category object, adds all items to it and appends it to the container
:param category_id: a string; an mmcif category e.g. 'emd_admin'
:param items: a list of strings; each element in the list is an item of mmcif category as defined by category_id
:return: a list of strings; each element represents a value for the corresponding element in data_items
"""
category = DataCategory(category_id)
for item in items:
category.appendAttribute(item)
self.__container.append(category)
#
# def update_single_row_value(self, category_id, item_name, row, value):
# """Update value in single row
# """
# catObj = self.__container.getObj(category_id)
# if catObj is None:
# return
# #
# catObj.setValue(value, item_name, row)
#
# def update_multiple_rows_value(self, category_id, item_name, value):
# """Update value in multiple rows
# """
# cat_obj = self.__container.getObj(category_id)
# if cat_obj is None:
# return
# #
# row_no = cat_obj.getRowCount()
# for row in range(0, row_no):
# cat_obj.setValue(value, item_name, row)
def insert_data(self, category_id, data_list):
"""
This method appends the data in data_list to the container labeled category_id
:param category_id: a string; an mmcif category e.g. 'emd_admin'
:param data_list:
:return:
"""
cat_obj = self.__container.getObj(category_id)
if cat_obj is None:
return
if any(isinstance(el, list) for el in data_list):
# print(data_list)
for data_ord in data_list[0]:
new_list = []
ord_index = data_list[0].index(data_ord)
new_list.append(ord_index)
new_list.append(data_list[1][ord_index])
# print(new_list)
cat_obj.append(new_list)
else:
cat_obj.append(data_list)
def insert_data_into_category(self, category_id, data_items, data_list):
"""
Helper method: calls two other methods, one to add a category and its items into a container and
another to insert the data for the category items
:param category_id: a string; an mmcif category e.g. 'emd_admin'
:param data_items: a list of strings; each element in the list is an item of mmcif category as defined by category_id
:param data_list: a list of strings; each element represents a value for the corresponding element in data_items
:return:
"""
# print('INSERT DATA INTO CATEGORY:', category_id, data_items, data_list)
self.add_category(category_id, data_items)
self.insert_data(category_id, data_list)