wiki:CSVExport

Version 3 (modified by hans, 12 years ago) (diff)

--

We have to follow the standards used by the portal for our Content export.

These are the types that we will be gathering the information for: MultimediaPage, PILOTMaterial, PresentationMaterial? and LeMillReference?. Our Tools and Methods are not included (that is negotiable).

This is the structure for our csv file (I've taken a CSV file put together by Martin and used that as an example):

  • Identifier - This one should get the UID of our objects, as this is the only unique identifier we have.
  • Metadata language - This one will get the language code for the language of the resource.
  • Title - Title of the object.
  • Description - not all of out objects have these, so for every type there will be a different solution.
    • MultimediaPage - we should take the first paragraph (I suppose we will have to strip the HTML tags, so some better solution might be offered).
    • PILOTMaterial - we should take Short Description or leave empty.
    • PresentationMaterial? - we should take the Caption of the first slide or leave empty.
    • LeMillReference? - we should take the Description or leave empty.
  • LO language - Language of the object.
  • LRE Keyword - We should take the contents of the Subject area field and translate these into some codes that are recognized by the portal. !!! We should get the codes from somewhere.
  • Keyword - We can provide the Tags of the resource separated by commas.
  • URL - This one will get the absolute_url for the object.
  • Learning resource type - We should get the meta_type of the object and then translate that into the standard format of the portal.
  • Age range - this one should be based upon our system of grades !!!(detailed info can be found below)
  • Description of rights - this one should contain link to a license page (should be the same for any created resource), for reference we should just check the license name that is there
  • Author - name of the author or authors (I would just put all the Full names of the authors separated by commas), !!! this field might need to contain some standard information ???

Some other fields have also been there, but I do not have any idea on how should we get information for these: Language of translation, Translated title, Translated description, Translated keywords.


How to transform grades to age:

  • pre-school - it should get a value of U - 6 (Undefined - when we do not know the limit)
  • 1st grade - 6 - 7
  • 2nd grade - 7 - 8
  • 3rd grade - 8 - 9
  • And so on until the last one
  • 12th grade - 18 - 19
  • higher education - 18 - U
  • adult education - 18 - U
  • teachers - 21 - 65
  • special education - 6 - 19 (will only be included in case of nothing else selected)

The result will be a combination of existing values, but we can only give one range (it's a standard by the system).

For example: if we have 1st grade and 12th grade selected, then we will get a range 6 - 19e

The system is something like:

  • One element selected - we will take the range by that element
  • Two or more selected - we take the first value of the first element as the first value for the result and we take the second value of the last element and use that as the second value for the result (as shown in example above: 6 from first element and 19 from second element)
  • The only exception at the moment is special education. We should only consider take that into consideration in case it is the only element.

ANY SUGGESTIONS OR COMMENTS ARE MORE THAN WELCOME


Hans:

In general I agree with Pjotr's proposal, because we were discussing it together. I have the following comments:

  • We should include only published learning resources (Multimedia Page, Presentation, PILOT).
  • According to LRE Application Profile 3.0 Identifier is composed of 2 sub-elements: ‘1.1.1 Catalog’ and ‘1.1.2 Entry’. The catalog should be a globally unique identifier, whereas the entry is a resource identifier that uniquely identifies the resource within this named catalog. I understand that in case of LeMill we should use lemill for ‘1.1.1 Catalog’ and UID for ‘1.1.2 Entry’... something like lemill:UID.
  • I am not sure how to map author to LRE Application Profile. LRE Application Profile element ‘2.3 Life Cycle.Contribute’ is intended to describe who has contributed to the LO. The ‘2.3 Life Cycle.Contribute’ is a container element. If this element is used, in elements ‘2.3.1 Life Cycle.Contribute.Role’ and ‘2.3.2 Life Cycle.Contribute.Entity’ there have to be values, i.e. they become mandatory elements. I guess we have to export ‘2.3.1 Life Cycle.Contribute.Role’ with value "author" and ‘2.3.2 Life Cycle.Contribute.Entity’ with author's name for each author.