Introduction
Qualifiers
Maths
CDataFile
CI2XmlFile - XML Files
Lists
Use the console mode to try these. I expect that plugins will mostly initialise data structures from DEF and DATA files so you will not need to do this basic initialisation stuff often. Please ask if you need explanation or other examples or feel free to edit in more examples.
from CCP4Data import *
i = CInt(5)
print type(i),i >>>>> <class 'CCP4Data.CInt'> 5
i.set(7)
print type(i),i >>>>> <class 'CCP4Data.CInt'> 7
i = 7
print type(i),i >>>>> <type 'int'> 7 (Beware!!!)
# A 'complex' class with two items - several equivalent ways to initialise..
r = CFloatRange(start=5,end=10)
r = CFloatRange(value={'start':5,'end':10})
r = CFloatRange(name='foo',value={'start':5,'end':10})
print type(r),r >>>>> <class 'CCP4Data.CFloatRange'> {'start': 5.0, 'end': 10.0}
# addressing the individual items
r = CFloatRange()
r.start= 5.0
r.end = 10.0
print type(r.start),r.start >>>>> <class 'CCP4Data.CFloat'> 5.0
# When you need getattr (or setattr)
for item in ['start','end']:
sum = sum + getattr(r,item)
i = CInt(min=10)
i.set(5) >>>>> 'below minimum' exception
r = CFloatRange(start_min=0.0,end_max=10.0,compare=1) # Note setting qualifiers for items
r.start = 5.0
r.end = 2.0 >>>>> 'End of range less than start' exception
r.end = 12.0 >>>>> 'above maximum' exception
r.get() >>>>> { 'start' : 5.0, 'end' : None }
# List all qualifers
r.qualifiers() >>>>> {'compare': 1}
r.start.qualifiers() >>>>> {'charWidth': None, 'min': 0.0, 'default': None, 'max': None, 'allowUndefined': True, 'onlyEnumerators': False, 'enumerators': [], 'menuText': []}
# List qualifiers but not those that have default values
r.start.qualifiers(default=False) >>>> {'min': 0.0}
# Get a specific qualifier
r.start.qualifier('min') >>>>> 0.0
NOTE: The CData classes do a lot of work validating after ever value update. So probably not wise to use them for serious calculations.
s = r.start + 4 print r.start,type(s),s >>>> 5.09.0 r.start.add(2.5) print r >>>>> {'start': 7.5, 'end': None} r.start>3 >>>>> True
Initialising and retrieving a file name. Applies to subclasses such as CPdbDataFile and CMtzDataFile
There is a project 'my_project' with directory /Users/lizp/Desktop/myproject.
N.B. CDataFile.__str__() is best way to return a string with full file name. e.g. str(f)
from CCP4File import *
f = CDataFile('/Users/lizp/Desktop/whatsit.wht')
print f >>>>> /Users/lizp/Desktop/whatsit.wht
str(f) >>>>> /Users/lizp/Desktop/whatsit.wht
f.get() >>>>> {'project': None, 'baseName': 'whatsit.wht', 'relPath': '/Users/lizp/Desktop'}
f = CDataFile(project='my_project',relPath='job_2',baseName='whatsit.wht')
str(f) >>>>> /Users/lizp/Desktop/myproject/job_2/whatsit.wht
f = CDataFile('/Users/lizp/Desktop/myproject/job_2/whatsit.wht')
f.get() >>>>> {'project': 'my_project', 'baseName': 'whatsit.wht', 'relPath': 'job_2'}
f.relPath = 'job_66'
f.get() >>>>> {'project': 'my_project', 'baseName': 'whatsit.wht', 'relPath': 'job_66'}
Note that CDataFile.dbFileId is an integer id for the file in the database and is vital for the core i2 code that saves the file provenance to the database. If you are copying filenames in pipelines (for example setting the input to one wrapper from the output of another) please copy the whole of the CDataFile object:
secondPlugin.container.inputData.XYZIN = firstPlugin.container.outputData.XYZOUT
Here, in a pipeline, firstPlugin and secondPlugin are two consecutive wrappers and XYZOUT and XYZIN are instances of CPdbDataFile that are the output and input of the two plugins. This is explained in the pipelines section.
Reading an XML file. Note that the 'CCP4I2' directory alias can be used as 'project'
The body of the file can be return as an etree which can be addressed using the functions from lxml
from CCP4File import *
f = CI2XmlDataFile(project='CCP4I2',relPath='test/data',baseName='pdbset.def.xml')
f.header.get() >>>>> {'function': 'DEF', 'comment': None, 'creationTime': None,
'userId': 'lizp', 'ccp4iVersion': '0.1', 'jobId': None,
'project': None, 'pluginName': 'pdbset',
'pluginVersion': None, 'pluginTitle': 'PDBSet'}
root = f.getBodyEtree()
# Use lxml functionality to address the body of the xml file..
for child in root: print child.get('id') >>>>>>>>> inputData
outputData
controlParameters
Creating an XML file
from CCP4File import *
from lxml import etree
f = CI2XmlDataFile ('/Users/lizp/Desktop/whatsit.xml')
f.header.setCurrent() # Sets the current time and user in header
f.header.function = 'DEF'
tree = etree.Element('mystuff')
tree.append(etree.Element('something'))
tree.append(etree.Element('something_else'))
f.saveFile(bodyEtree=tree)
The list base class CList is intended to behave like a Python list but has the restriction that all items in the list must be of the same type and that type is a subclass of CData. The CList has qualifiers listMinLength and listMaxLength which do what they say and listCompare which is an integer - if it is defined with value 1/-1 then each item in the list must be greater/less than preceeding item - for this to work the list item class must have a __cmp__() method (see Python docs). In the example below a list of integers with minimum length 1 and minimum integer value 1 is created.
from CCP4Data import *
l = CList(subItem= { 'class' : CInt, 'qualifiers' : { 'min' : 1 , 'default' : 1 }}, listMinLength=1)
print l >>>> [1]
l.append(6)
print l >>>> [1,6]
l[1] = -1 >>>> below minimum error
l[1] = l[1] + 2
print l >>>>> [1,8]
del l[1]
del l[0] >>>>> list below minimum length error
Below is an example of a CList subclass from the CCP4ModelData - this has list items CResidueRange that has components: chainId and first and last residue.
from CCP4ModelData import *
l = CResidueRangeList()
l.append({'chainId':'A','firstRes':'23','lastRes':'45'})
print l,type(l[0]) >>>> [{'firstRes': '23', 'chainId': 'A', 'lastRes': '45'}] ,
print l[0].chainId >>>> A