Batch Processing via Browser¶

Preliminaries:

We define a function that looks up a form with several submit buttons for the one with a given value (this functionality is missing in zope.testbrowser):

>>> def lookup_submit_value(name, value, browser):
...   """Find a button with a certain value."""
...   for num in range(0, 100):
...     try:
...       button = browser.getControl(name=name, index=num)
...       if button.value.endswith(value):
...         return button
...     except IndexError:
...       break
...   return None

Create a site:

>>> from waeup.kofa.app import University
>>> getRootFolder()['app'] = University()
>>> from zope.component.hooks import setSite
>>> setSite(getRootFolder()['app'])

Create a datacenter storage path:

>>> import os
>>> import tempfile
>>> dc_path = tempfile.mkdtemp()

Log in:

>>> from zope.testbrowser.testing import Browser
>>> browser = Browser()
>>> browser.addHeader('Authorization', 'Basic mgr:mgrpw')
>>> browser.handleErrors = False

Set datacenter path and deselect moving old data:

>>> browser.open('http://localhost/app')
>>> browser.getLink('Data Center').click()
>>> browser.getLink('Edit settings').click()
>>> browser.getControl(name='newpath').value = dc_path
>>> browser.getControl(name='move').value = False
>>> browser.getControl(name='save').click()

Set non-usable datacenter path:

>>> browser.getLink('Edit settings').click()
>>> browser.getControl(name='newpath').value = '/'
>>> browser.getControl(name='save').click()
>>> 'Given storage path cannot be used.' in browser.contents
True
>>> browser.getControl('Back to Data Center').click()

Batch Processing Faculties¶

Go to datacenter page:

>>> browser.open('http://localhost/app/datacenter')

Prepare a CSV file for faculties (extended ascii values are accepted):

>>> open('faculties.csv', 'wb').write(
... """code,title,title_prefix
... FAC1,Faculty 1,faculty
... FAC2,Faculty 2,institute
... FAC3,Fäcülty 3,school
... """)

Upload the file:

>>> import cStringIO
>>> browser.getLink('Upload data').click()
>>> filecontents = cStringIO.StringIO(
...   open('faculties.csv', 'rb').read())
>>> filewidget = browser.getControl(name='uploadfile:file')
>>> filewidget.add_file(filecontents, 'text/plain', 'faculties.csv')
>>> browser.getControl(name='SUBMIT').click()

Step 1: start batch processing:

>>> browser.getLink('Process data').click()
>>> browser.getLink('Switch maintenance mode').click()
>>> button = lookup_submit_value(
...   'select', 'faculties_zope.mgr.csv', browser)
>>> button.click()

Step 2: select a processor and mode:

>>> importerselect = browser.getControl(name='importer')
>>> importerselect.displayOptions
['AccessCodeBatch Processor',
'AccessCode Processor',
'ApplicantOnlinePayment Processor',
'Applicant Processor',
'ApplicantRefereeReport Processor',
'ApplicantsContainer Processor',
'Bed Processor (update only)',
'CertificateCourse Processor',
'Certificate Processor',
'ConfigurationConainer Processor (update only)',
'Course Processor',
'CourseTicket Processor',
'Department Processor',
'Faculty Processor',
'Hostel Processor',
'Public HTML Document Processor',
'StudentOnlinePayment Processor',
'Public PDF Document Processor',
'Public REST Document Processor',
'SessionConfiguration Processor',
'Student Processor',
'StudentStudyCourse Processor',
'StudentStudyLevel Processor',
'User Processor',
'Verdict Processor (special processor, update only)']

>>> importerselect.getControl('Faculty Processor').selected = True

>>> modeselect = browser.getControl(name='mode')
>>> modeselect.options
['create', 'update', 'remove']

>>> modeselect.getControl(value='create').selected = True
>>> browser.getControl('Proceed to step 3').click()

Step 3: Fix headerlines

We get informed that there are no problems with the current header:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...
Header fields OK
...

The submit button is enabled:

>>> browser.getControl('Perform import').disabled
False

>>> browser.getControl('Perform import').click()

Step 4: See import results

The import was successful:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...Successfully processed 3 rows...
...Batch processing finished...
...File:...faculties_zope.mgr.csv...

We can grep the entries generated in logfile:

>>> browser.open('http://localhost/app/datacenter/logs')
>>> browser.getControl('Show', index=0).click()
>>> print browser.contents
<!DOCTYPE ...
...<h1 class="kofa-content-label">Logfile datacenter.log</h1>...

>>> browser.getControl(name='query').value = "zope.mgr"
>>> browser.getControl('Search').click()
>>> 'zope.mgr - processed' in browser.contents
True

Batch Processing Departments¶

>>> browser.open('http://localhost/app/datacenter')

Prepare a CSV file for departments:

>>> open('departments.csv', 'wb').write(
... """code,faculty_code,title,title_prefix
... DEP1,FAC1,Department 1,department
... DEP2,FAC2,Department 2,centre
... """)

Upload the file:

>>> import cStringIO
>>> browser.getLink('Upload data').click()
>>> filecontents = cStringIO.StringIO(
...   open('departments.csv', 'rb').read())
>>> filewidget = browser.getControl(name='uploadfile:file')
>>> filewidget.add_file(filecontents, 'text/plain', 'departments.csv')
>>> browser.getControl(name='SUBMIT').click()

Step 1: start batch processing:

>>> browser.getLink('Process data').click()
>>> browser.getLink('Switch maintenance mode').click()
>>> button = lookup_submit_value(
...   'select', 'departments_zope.mgr.csv', browser)
>>> button.click()

Step 2: select a processor and mode:

>>> importerselect = browser.getControl(name='importer')
>>> importerselect.getControl('Department Processor').selected = True
>>> modeselect = browser.getControl(name='mode')
>>> modeselect.getControl(value='create').selected = True
>>> browser.getControl('Proceed to step 3').click()

Step 3: Fix headerlines

We get informed that there are no problems with the current header:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...
Header fields OK
...

The submit button is enabled:

>>> browser.getControl('Perform import').disabled
False

>>> browser.getControl('Perform import').click()

Step 4: See import results

The import was successful:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...Successfully processed 2 rows...
...Batch processing finished...
...File:...departments_zope.mgr.csv...

Batch Processing Courses¶

>>> browser.open('http://localhost/app/datacenter')

Prepare a CSV file for courses:

>>> open('courses.csv', 'wb').write(
... """code,faculty_code,department_code,title,level,passmark,credits,semester
... CRS1,FAC1,DEP1,Course 1,100,40,2,1
... CRS2,FAC1,DEP1,Course 2,100,40,2,2
... """)

Upload the file:

>>> import cStringIO
>>> browser.getLink('Upload data').click()
>>> filecontents = cStringIO.StringIO(
...   open('courses.csv', 'rb').read())
>>> filewidget = browser.getControl(name='uploadfile:file')
>>> filewidget.add_file(filecontents, 'text/plain', 'courses.csv')
>>> browser.getControl(name='SUBMIT').click()

Step 1: start batch processing:

>>> browser.getLink('Process data').click()
>>> browser.getLink('Switch maintenance mode').click()
>>> button = lookup_submit_value(
...   'select', 'courses_zope.mgr.csv', browser)
>>> button.click()

Step 2: select a processor and mode:

>>> importerselect = browser.getControl(name='importer')
>>> importerselect.getControl('Course Processor', index=1).selected = True
>>> modeselect = browser.getControl(name='mode')
>>> modeselect.getControl(value='create').selected = True
>>> browser.getControl('Proceed to step 3').click()

Step 3: Fix headerlines

We get informed that there are no problems with the current header:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...
Header fields OK
...

The submit button is enabled:

>>> browser.getControl('Perform import').disabled
False

>>> browser.getControl('Perform import').click()

Step 4: See import results

The import was successful:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...Successfully processed 2 rows...
...Batch processing finished...
...File:...courses_zope.mgr.csv...

Batch Processing Certificates¶

>>> browser.open('http://localhost/app/datacenter')

Prepare a CSV file for certificates:

>>> open('certificates.csv', 'wb').write(
... """code,faculty_code,department_code,title,study_mode,start_level,end_level,application_category
... CERT1,FAC1,DEP1,Certificate 1,pg_ft,999,999,basic
... CERT2,FAC1,DEP1,Certificate 2,ug_ft,200,300,cest
... """)

Upload the file:

>>> import cStringIO
>>> browser.getLink('Upload data').click()
>>> filecontents = cStringIO.StringIO(
...   open('certificates.csv', 'rb').read())
>>> filewidget = browser.getControl(name='uploadfile:file')
>>> filewidget.add_file(filecontents, 'text/plain', 'certificates.csv')
>>> browser.getControl(name='SUBMIT').click()

Step 1: start batch processing:

>>> browser.getLink('Process data').click()
>>> browser.getLink('Switch maintenance mode').click()
>>> button = lookup_submit_value(
...   'select', 'certificates_zope.mgr.csv', browser)
>>> button.click()

Step 2: select a processor and mode:

>>> importerselect = browser.getControl(name='importer')
>>> importerselect.getControl('Certificate Processor').selected = True
>>> modeselect = browser.getControl(name='mode')
>>> modeselect.getControl(value='create').selected = True
>>> browser.getControl('Proceed to step 3').click()

Step 3: Fix headerlines

We get informed that there are no problems with the current header:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...
Header fields OK
...

The submit button is enabled:

>>> browser.getControl('Perform import').disabled
False

>>> browser.getControl('Perform import').click()

Step 4: See import results

The import was successful:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...Successfully processed 2 rows...
...Batch processing finished...
...File:...certificates_zope.mgr.csv...

Batch Processing Certificate Courses¶

>>> browser.open('http://localhost/app/datacenter')

Prepare a CSV file for certificate courses:

>>> open('mycertcourses.csv', 'wb').write(
... """course,faculty_code,department_code,certificate_code,level,mandatory
... CRS1,FAC1,DEP1,CERT1,100,True
... CRS2,FAC1,DEP1,CERT1,100,True
... """)

Upload the file:

>>> import cStringIO
>>> browser.getLink('Upload data').click()
>>> filecontents = cStringIO.StringIO(
...   open('mycertcourses.csv', 'rb').read())
>>> filewidget = browser.getControl(name='uploadfile:file')
>>> filewidget.add_file(filecontents, 'text/plain', 'mycertcourses.csv')
>>> browser.getControl(name='SUBMIT').click()

Step 1: start batch processing:

>>> browser.getLink('Process data').click()
>>> browser.getLink('Switch maintenance mode').click()
>>> button = lookup_submit_value(
...   'select', 'mycertcourses_zope.mgr.csv', browser)
>>> button.click()

Step 2: select a processor and mode:

>>> importerselect = browser.getControl(name='importer')
>>> importerselect.getControl('CertificateCourse Processor').selected = True
>>> modeselect = browser.getControl(name='mode')
>>> modeselect.getControl(value='create').selected = True
>>> browser.getControl('Proceed to step 3').click()

Step 3: Fix headerlines

We get informed that there are no problems with the current header:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...
Header fields OK
...

The submit button is enabled:

>>> browser.getControl('Perform import').disabled
False

>>> browser.getControl('Perform import').click()

Step 4: See import results

The import was successful:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...Successfully processed 2 rows...
...Batch processing finished...
...File:...mycertcourses_zope.mgr.csv...

Batch Processing Users¶

>>> browser.open('http://localhost/app/datacenter')

Prepare a CSV file for users:

>>> open('users.csv', 'wb').write(
... """name,title,public_name,email,phone,roles
... uli,Uli Fouquet,Chief Developer,uli@abc.de,+49-234-567,[]
... henrik,Henrik Bettermann,Admin,henrik@abc.de,+49-234-567,"['waeup.PortalManager', 'waeup.ImportManager']"
... anne,Anne Palina,,anne@abc.de,+49-234-567,"['waeup.Nonsense']"
... """)

Upload the file:

>>> import cStringIO
>>> browser.getLink('Upload data').click()
>>> filecontents = cStringIO.StringIO(
...   open('users.csv', 'rb').read())
>>> filewidget = browser.getControl(name='uploadfile:file')
>>> filewidget.add_file(filecontents, 'text/plain', 'users.csv')
>>> browser.getControl(name='SUBMIT').click()

Step 1: start batch processing:

>>> browser.getLink('Process data').click()
>>> browser.getLink('Switch maintenance mode').click()
>>> button = lookup_submit_value(
...   'select', 'users_zope.mgr.csv', browser)
>>> button.click()

Step 2: select a processor and mode:

>>> importerselect = browser.getControl(name='importer')
>>> importerselect.getControl('User Processor').selected = True
>>> modeselect = browser.getControl(name='mode')
>>> modeselect.getControl(value='create').selected = True
>>> browser.getControl('Proceed to step 3').click()

Step 3: Fix headerlines

We get informed that there are no problems with the current header:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...
Header fields OK
...

The submit button is enabled:

>>> browser.getControl('Perform import').disabled
False

>>> browser.getControl('Perform import').click()

Step 4: See import results

The import was successful:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...Successfully processed 2 rows...
...Batch processing finished...
...File:...users_zope.mgr.csv...

User henrik has got the global roles:

>>> henrik = getRootFolder()['app']['users']['henrik']
>>> henrik.roles
['waeup.PortalManager', 'waeup.AcademicsOfficer', 'waeup.ImportManager']

Pending Files¶

When an error occurs during an import, two files are generated: a CSV file with finished files and a CSV file with pending data. Both are stored in the appropriate subdirectories in datacenter. We try to create faculties, from which one already exists.

Go to datacenter page:

>>> browser.open('http://localhost/app/datacenter')

Prepare a CSV file for faculties:

>>> open('newfaculties.csv', 'wb').write(
... """code,title,title_prefix
... FAC1,Faculty 1,faculty
... FAC4,Faculty 4,school
... FAC 5,Faculty 5,faculty
... """)

Upload the file:

>>> import cStringIO
>>> browser.getLink('Upload data').click()
>>> filecontents = cStringIO.StringIO(
...   open('newfaculties.csv', 'rb').read())
>>> filewidget = browser.getControl(name='uploadfile:file')
>>> filewidget.add_file(filecontents, 'text/plain', 'newfaculties.csv')
>>> browser.getControl(name='SUBMIT').click()

Since we now have a user with waeup.ImportManager role, an email has been sent:

>>> print browser.contents
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
...
...All import managers have been notified by email...
...

Step 1: start batch processing:

>>> browser.getLink('Process data').click()
>>> browser.getLink('Switch maintenance mode').click()
>>> button = lookup_submit_value(
...   'select', 'newfaculties_zope.mgr.csv', browser)
>>> button.click()

Step 2: select a processor and mode:

>>> importerselect = browser.getControl(name='importer')
>>> importerselect.getControl('Faculty Processor').selected = True
>>> modeselect = browser.getControl(name='mode')
>>> modeselect.getControl(value='create').selected = True
>>> browser.getControl('Proceed to step 3').click()

Step 3: Fix headerlines

As there should be no problem with the headers, we can immediately perfom the import:

>>> browser.getControl('Perform import').click()

Two lines could not be imported:

>>> print browser.contents
<!DOCTYPE html PUBLIC...
...
...Processing of 2 rows failed...
...Successfully processed 1 rows...
...

Now there are two files as a result in datacenter storage’s root and finished dirs:

>>> pending_file = dc_path + '/newfaculties_zope.mgr.create.pending.csv'
>>> print open(pending_file).read()
title_prefix,code,title,--ERRORS--
faculty,FAC1,Faculty 1,This object already exists.
faculty,FAC 5,Faculty 5,code: Invalid input

>>> finished_file = dc_path + '/finished/newfaculties_zope.mgr.create.finished.csv'
>>> print open(finished_file).read()
title_prefix,code,title
school,FAC4,Faculty 4

The finished-file contains the dataset we could import, while the pending file contains the dataset that failed, appended by an error message.

Fixing the Pending File¶

We ‘edit’ the pending file (setting code to FAC5 and title appropriately, and removing the –ERROR– column) and finish the import this way:

>>> open(dc_path + '/newfaculties_zope.mgr.create.pending.csv', 'wb').write(
... """title_prefix,--IGNORE--,code,title
... faculty,,FAC5,Faculty 5
... """)

Step 1: start batch processing:

>>> browser.open('http://localhost/app/datacenter')
>>> browser.getLink('Process data').click()
>>> browser.getLink('Switch maintenance mode').click()
>>> button = lookup_submit_value(
...   'select', 'newfaculties_zope.mgr.create.pending.csv', browser)
>>> button.click()

Step 2: select a processor and mode:

>>> importerselect = browser.getControl(name='importer')
>>> importerselect.getControl('Faculty Processor').selected = True
>>> modeselect = browser.getControl(name='mode')
>>> modeselect.getControl(value='create').selected = True
>>> browser.getControl('Proceed to step 3').click()

Step 3/4: Fix headerlines and import:

As there should be no problem with the headers, we can immediately perfom the import:

>>> browser.getControl('Perform import').click()

This time everything should work:

>>> print browser.contents
<!DOCTYPE html PUBLIC...
...
...Successfully processed 1 rows...
...

Oh no, we forgot Anne Palina. Her user record was not imported because she has a non-existent role:

>>> sorted(os.listdir(dc_path))
['deleted', 'finished', 'graduated', 'logs', 'unfinished', 'users_zope.mgr.create.pending.csv']

>>> os.listdir(dc_path + '/unfinished')
['users_zope.mgr.csv']

>>> pending_file = dc_path + '/users_zope.mgr.create.pending.csv'
>>> print open(pending_file).read()
name,roles,title,public_name,phone,email,--ERRORS--
anne,['waeup.Nonsense'],Anne Palina,<IGNORE>,+49-234-567,anne@abc.de,roles: invalid role

There are many finished-files:

>>> sorted(os.listdir(dc_path + '/finished'))
['certificates_zope.mgr.create.finished.csv', ...,
'users_zope.mgr.create.finished.csv']

Processed (finished) Files¶

>>> browser.open('http://localhost/app/datacenter/processed')
>>> 'download?filename=finished/certificates_zope.mgr.create.finished.csv' in browser.contents
True

Log Files¶

>>> browser.open('http://localhost/app/datacenter/logs')
>>> 'datacenter.log' in browser.contents
True
>>> browser.getControl('Show', index=0).click()
>>> browser.getControl('Back', index=0).click()
>>> browser.getControl('Back to Data Center').click()
>>> 'Storage path:' in browser.contents
True

Clean up:

>>> import shutil
>>> shutil.rmtree(dc_path)

Table Of Contents

Previous topic

Next topic

This Page

Batch Processing via Browser¶

Batch Processing Faculties¶

Batch Processing Departments¶

Batch Processing Courses¶

Batch Processing Certificates¶

Batch Processing Certificate Courses¶

Batch Processing Users¶

Pending Files¶

Fixing the Pending File¶

Processed (finished) Files¶

Log Files¶