source: main/waeup.sirp/trunk/src/waeup/sirp/students/batching.py @ 6843

"""Batch processing components for academics objects.

Batch processors eat CSV files to add, update or remove large numbers
of certain kinds of objects at once.

Here we define the processors for academics specific objects like
faculties, departments and the like.
"""
import grok
from zope.interface import Interface
from zope.schema import getFields
from zope.component import queryUtility
from zope.catalog.interfaces import ICatalog
from waeup.sirp.interfaces import IBatchProcessor, FatalCSVError
from waeup.sirp.students.interfaces import (
    IStudent, IStudentStudyCourse, IStudentStudyCourseImport)
from waeup.sirp.utils.batching import BatchProcessor

class StudentProcessor(BatchProcessor):
    """A batch processor for IStudent objects.
    """
    grok.implements(IBatchProcessor)
    grok.provides(IBatchProcessor)
    grok.context(Interface)
    util_name = 'studentimporter'
    grok.name(util_name)

    name = u'Student Importer'
    iface = IStudent

    location_fields = ['student_id',]
    factory_name = 'waeup.Student'

    mode = None

    @property
    def req(self):
        result = dict(
            create = self.required_fields,
            update = self.location_fields,
            remove = self.location_fields,
        )
        return result

    def parentsExist(self, row, site):
        return 'students' in site.keys()

    # The entry never exists in create mode.
    # To do: Use reg_number and matric_number for locating as well.
    def entryExists(self, row, site):
        if row.has_key('student_id'):
            return row['student_id'] in site['students'].keys()
        return False

    def getParent(self, row, site):
        return site['students']

    def getEntry(self, row, site):
        if not self.entryExists(row, site):
            return None
        parent = self.getParent(row, site)
        return parent.get(row['student_id'])

    def addEntry(self, obj, row, site):
        parent = self.getParent(row, site)
        parent.addStudent(obj)
        return

    def delEntry(self, row, site):
        parent = self.getParent(row, site)
        del parent[row['student_id']]
        pass

class StudentStudyCourseProcessor(BatchProcessor):
    """A batch processor for IStudentStudyCourse objects.
    """
    grok.implements(IBatchProcessor)
    grok.provides(IBatchProcessor)
    grok.context(Interface)
    util_name = 'studycourseupdater'
    grok.name(util_name)

    name = u'StudentStudyCourse Importer (update only)'
    iface = IStudentStudyCourseImport
    factory_name = 'waeup.StudentStudyCourse'

    mode = None

    @property
    def available_fields(self):
        result = []
        return sorted(list(set(
            ['student_id','reg_number','matric_number'] + getFields(
                self.iface).keys())))

    def checkHeaders(self, headerfields, mode='ignore'):
        if not 'reg_number' in headerfields and not 'student_id' in headerfields and not 'matric_number' in headerfields:
            raise FatalCSVError(
                "Need at least columns student_id or reg_number or matric_number for import!")
        # Check for fields to be ignored...
        not_ignored_fields = [x for x in headerfields
                              if not x.startswith('--')]
        if len(set(not_ignored_fields)) < len(not_ignored_fields):
            raise FatalCSVError(
                "Double headers: each column name may only appear once.")
        return True

    def parentsExist(self, row, site):
        if not 'students' in site.keys() and row['student_id']:
            return False
        if 'student_id' in row.keys():
            if row['student_id'] in site['students']:
                student = site['students'][row['student_id']]
                return student
        elif 'reg_number' in row.keys() and row['reg_number']:
            reg_number = row['reg_number']
            cat = queryUtility(ICatalog, name='students_catalog')
            results = list(
                cat.searchResults(reg_number=(reg_number, reg_number)))
            if results:
                return results[0]
        elif 'matric_number' in row.keys() and row['matric_number']:
            #import pdb; pdb.set_trace()
            matric_number = row['matric_number']
            cat = queryUtility(ICatalog, name='students_catalog')
            results = list(
                cat.searchResults(matric_number=(matric_number, matric_number)))
            if results:
                return results[0]
        return False

    def entryExists(self, row, site):
        student = self.parentsExist(row, site)
        if not student:
            return False
        if 'studycourse' in student:
            return student
        return False

    def getEntry(self, row, site):
        student = self.entryExists(row, site)
        if not student:
            return None
        return student.get('studycourse')