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 testClassifyByGroup(self):
"""Test case - organize dictionary items by classes: SAMPLE, MX, NMR, EM, STRUCTURE, and DB"""
try:
myIo = IoAdapter(raiseExceptions=True)
self.__containerList = myIo.readFile(
inputFilePath=self.__pathPdbxDictionary)
dApi = DictionaryApi(containerList=self.__containerList,
consolidate=True,
verbose=self.__verbose)
#
itemList = []
groupList = dApi.getCategoryGroups()
categoryList = dApi.getCategoryList()
for category in categoryList:
itemList.extend(dApi.getItemNameList(category))
itemList = sorted(set(itemList))
logger.info("Total category length %d", len(categoryList))
logger.info("Total definition length %d", len(itemList))
logger.info("group length %s", len(groupList))
logger.debug("groupList %r", groupList)
#
findUnlinked = False
if findUnlinked:
tSet = set(["pdbx_group", "inclusive_group"])
for category in categoryList:
gList = dApi.getCategoryGroupList(category)
gSet = set(gList)
if gSet == tSet:
logger.info("unqualified %s", category)
# logger.info("%s -> %r", category, gList)
if not gList:
logger.info("--- No category group assignment for %s",
category)
#
classD = {}
# Add category group members -
for groupName, className in self.__groupClassTupL:
categoryL = dApi.getCategoryGroupCategories(
groupName, followChildren=True)
for category in categoryL:
classD.setdefault(className, []).extend(
dApi.getItemNameList(category))
#
# Add unlinked categories
#
for category, className in self.__unlinkedCategoryClassTup:
classD.setdefault(className,
[]).extend(dApi.getItemNameList(category))
#
sumItem = 0
classItemD = {}
for className, itemL in classD.items():
numItem = len(set(itemL))
sumItem += numItem
logger.info("class %s items %d", className, len(set(itemL)))
for item in itemL:
classItemD[item] = True
#
logger.info("Sum classified items is %d", sumItem)
logger.info("classified items %d", len(classItemD))
#
logger.debug("classItemD.items() %r",
list(classItemD.items())[:10])
missingGroupL = []
jj = 0
for item in itemList:
if item not in classItemD:
jj += 1
category = CifName.categoryPart(item)
logger.info("%d item %r category %r", jj, item, category)
missingGroupL.extend(dApi.getCategoryGroupList(category))
#
logger.info("missing groups %r", sorted(set(missingGroupL)))
except Exception as e:
logger.exception("Failing with %s", str(e))
self.fail()