CKAN Extension of custom tags using a vocabulary

# This Helper function creates the region vocabulary from a text file
def createRegionsVocab():
    user = toolkit.get_action('get_site_user')({'ignore_auth': True}, {})
    context = {'user': user['name']}

    try:
        data = {'id': 'ILRI_prjregions'}
        toolkit.get_action('vocabulary_show')(context, data)
        print "ILRI_prjregions vocabulary already exists, skipping."
    except toolkit.ObjectNotFound:
        print "Creating vocab 'ILRI_prjregions'"
        data = {'name': 'ILRI_prjregions'}
        vocab = toolkit.get_action('vocabulary_create')(context, data)
        for tag in getArrayFromFile("/opt/ckan/lib/default/src/ckanext-ilriapi/ckanext/sourcetxt/ilri-regions.txt"):
            logging.info("Adding tag {0} to vocab 'ILRI_prjregions'".format(tag))
            data = {'name': tag, 'vocabulary_id': vocab['id']}
            toolkit.get_action('tag_create')(context, data)
    return True

#This is almost the same as CKAN but instead of passing key we pass the new tag
def convertToTags(vocab,newtag,data,error,context):

    new_tags = newtag

    if not new_tags:
        return
    if isinstance(new_tags, basestring):
        new_tags = [new_tags]

    # get current number of tags
    n = 0
    for k in data.keys():
        if k[0] == 'tags':
            n = max(n, k[1] + 1)

    v = model.Vocabulary.get(vocab)
    if not v:
        raise df.Invalid(_('Tag vocabulary "%s" does not exist') % vocab)
    context['vocabulary'] = v

    for tag in new_tags:
        validators.tag_in_vocabulary_validator(tag, context)

    for num, tag in enumerate(new_tags):
        data[('tags', num + n, 'name')] = tag
        data[('tags', num + n, 'vocabulary_id')] = v.id

#Separate a string of tags into individual tags using coma and then adds each tag using convertToTags
#Used in datasets to store Countries, Regions and Species as vocabulary tags
def comasToTags(key,data,error,context):
    #print key[0]
    tag_string = data.get(key)
    #print tag_string
    tags = tag_string.split(",")
    vocab = "None"
    if key[0] == "ILRI_prjregions":
        vocab = "ILRI_prjregions"

    if key[0] == "ILRI_prjcountries":
        vocab = "ILRI_prjcountries"

    if key[0] == "ILRI_prjspecies":
        vocab = "ILRI_prjspecies"

    if vocab == "None":
        raise df.Invalid(_('Tag vocabulary for key "%s" does not exist') % key[0])

    for tag in tags:
        convertToTags(vocab,tag,data,error,context)

# Combines a series of tags into a string of tags separated by separator
# Used in dataset to show the tags with the proper javascript
def tagsToString(tags,separator):
    tag_string = ""
    for tag in tags:
        tag_string = tag_string + tag + separator
    pprint.pprint(tag_string)
    tag_string = tag_string[:len(tag_string)-1]
    pprint.pprint(tag_string)

    return tag_string


def _add_custom_metadata_to_schema(self, schema):

        #Project Level metadata
        schema.update({'ILRI_prjregions': [comasToTags]}) #Here is the custom handling of tags


def show_package_schema(self):
        schema = super(ILRIMetadataPlugin, self).show_package_schema()

        schema['tags']['__extras'].append(toolkit.get_converter('free_tags_only'))

        #Project Level metadata
        schema.update({'ILRI_prjregions': [toolkit.get_converter('convert_from_tags')("ILRI_prjregions"),toolkit.get_validator('ignore_missing')]})