source: main/waeup.sirp/trunk/src/waeup/sirp/students/batching.py @ 6850

"""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
import csv
import copy
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, IObjectConverter)
from waeup.sirp.students.interfaces import (
    IStudent, IStudentStudyCourse, IStudentStudyCourseImport,
    IStudentUpdateByRegNo, IStudentUpdateByMatricNo)
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 = []
    factory_name = 'waeup.Student'

    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='create'):
        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!")
        if mode == 'create':
            for field in self.required_fields:
                if not field in headerfields:
                    raise FatalCSVError(
                        "Need at least columns %s for import!" %
                        ', '.join(["'%s'" % x for x in self.required_fields]))
        # 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):
        return 'students' in site.keys()

    def getLocator(self, row):
        if 'student_id' in row.keys() and row['student_id']:
            return 'student_id'
        elif 'reg_number' in row.keys() and row['reg_number']:
            return 'reg_number'
        elif 'matric_number' in row.keys() and row['matric_number']:
            return 'matric_number'
        else:
            return None

    # The entry never exists in create mode.
    def entryExists(self, row, site):
        if not 'students' in site.keys():
            return None
        if self.getLocator(row) == 'student_id':
            if row['student_id'] in site['students']:
                student = site['students'][row['student_id']]
                return student
        elif self.getLocator(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 self.getLocator(row) == 'matric_number':
            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 None

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

    def getEntry(self, row, site):
        return self.entryExists(row, site)

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

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

    def getMapping(self, path, headerfields, mode):
        """Get a mapping from CSV file headerfields to actually used fieldnames.
        """
        result = dict()
        reader = csv.reader(open(path, 'rb'))
        raw_header = reader.next()
        for num, field in enumerate(headerfields):
            if field not in ['student_id', 'reg_number', 'matric_number'] and mode == 'remove':
                continue
            if field == u'--IGNORE--':
                # Skip ignored columns in failed and finished data files.
                continue
            result[raw_header[num]] = field
        return result

    def checkConversion(self, row, mode='create'):
        """Validates all values in row.
        """
        if mode in ['update', 'remove']:
            if self.getLocator(row) == 'reg_number':
                iface = IStudentUpdateByRegNo
            elif self.getLocator(row) == 'matric_number':
                iface = IStudentUpdateByMatricNo
        else:
            iface = self.iface
        converter = IObjectConverter(iface)
        errs, inv_errs, conv_dict =  converter.fromStringDict(
            row, self.factory_name)
        return errs, inv_errs, conv_dict

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'

    location_fields = []

    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():
            return None
        if 'student_id' in row.keys() and row['student_id']:
            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']
            #import pdb; pdb.set_trace()
            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']:
            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 None

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

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