From: Alex Sadleir Date: Sun, 14 Apr 2013 06:48:31 +0000 Subject: enforce valid package url X-Git-Url: https://maxious.lambdacomplex.org/git/?p=disclosr.git&a=commitdiff&h=4bfca5ebcf1771fdd497046e81de89596f6f640d --- enforce valid package url Former-commit-id: 26accf46eff2958223ffae9a1234d4291379bce4 --- --- a/documents/datagov-export.py +++ b/documents/datagov-export.py @@ -3,6 +3,7 @@ from ckanclient import CkanApiError import re import html2text # aaronsw :( +import ckanapi class LoaderError(Exception): @@ -10,10 +11,12 @@ # Instantiate the CKAN client. #ckan = ckanclient.CkanClient(base_location='http://localhost:5000/api', api_key='b47b24cd-591d-40c1-8677-d73101d56d1b') +api_key = 'aeab80a6-0a58-4026-96a4-c07e57f9c434' ckan = ckanclient.CkanClient(base_location='http://data.disclosurelo.gs/api', - api_key='482a9dd2-a976-4adf-ac77-d71d92a98a52') -#couch = couchdb.Server('http://127.0.0.1:5984/') -couch = couchdb.Server('http://192.168.1.113:5984/') + api_key=api_key) +ckandirect = ckanapi.RemoteCKAN('http://data.disclosurelo.gs', api_key=api_key) +couch = couchdb.Server('http://127.0.0.1:5984/') +#couch = couchdb.Server('http://192.168.1.113:5984/') # http://code.activestate.com/recipes/578019-bytes-to-human-human-to-bytes-converter/ SYMBOLS = { @@ -88,9 +91,11 @@ name = re.sub('__', '_', name).lower() return name - +#todo "{'name': [u'Url must be purely lowercase alphanumeric (ascii) characters and these symbols: -_']}" +# http://data.gov.au/dataset/australian-domestic-regional-and-international-airline-activity-%E2%80%93-time-series/ def name_munge(input_name): - return munge(input_name.replace(' ', '').replace('.', '_').replace('&', 'and')) + return munge(input_name.replace(' ', '').replace('.', '_').replace('&', 'and')) + #[:100] #return input_name.replace(' ', '').replace('.', '_').replace('&', 'and') @@ -115,11 +120,12 @@ docsdb = couch['disclosr-documents'] if __name__ == "__main__": + orgs_list = [] for doc in docsdb.view('app/datasets'): print doc.id - if doc.value['url'] != "http://data.gov.au/data/": + if doc.value['url'] != "http://data.gov.au/data/" and doc.value['agencyID'] != "qld": # Collect the package metadata. - pkg_name = doc.value['url'].replace("http://data.gov.au/dataset/",'').replace('/',''); _ + pkg_name = filter( lambda x: x in '0123456789abcdefghijklmnopqrstuvwxyz-_',doc.value['url'].replace("http://data.gov.au/dataset/",'').replace('/','')[:100]); tags = [] if doc.value['agencyID'] == "AGIMO": if len(doc.value['metadata']["Keywords / Tags"]) > 0: @@ -133,7 +139,7 @@ else: tags = tags + [doc.value['metadata']['data.gov.au Category']] tags = [re.sub('[^a-zA-Z0-9-_.]', '', tag.replace('&', 'and')).lower() for tag in tags if tag] - print tags + #print tags package_entity = { 'name': pkg_name, 'title': doc.value['metadata']['DCTERMS.Title'], @@ -148,15 +154,16 @@ package_entity = doc.value['metadata'] try: - print package_entity + #print package_entity ckan.package_register_post(package_entity) except CkanApiError, e: - if ckan.last_status == 409: + if ckan.last_message == "{\"name\": [\"That URL is already in use.\"]}": print "package already exists" else: + print ckan.last_message raise LoaderError('Unexpected status %s checking for package under \'%s\': %r' % ( ckan.last_status, pkg_name, e.args)) - + pkg = ckan.package_entity_get(pkg_name) #add to group @@ -185,12 +192,52 @@ } print group_entity ckan.group_register_post(group_entity) + elif ckan.last_status == 409: + print "group already exists" else: raise LoaderError('Unexpected status %s adding to group under \'%s\': %r' % ( ckan.last_status, pkg_name, e.args)) + + # #add to or create organization using direct API + # org_name = name_munge(doc.value['metadata']["Agency"][:100]) + # if org_name not in orgs_list: + # orgs_list = ckandirect.action.organization_list()['result'] + # print orgs_list + # if org_name not in orgs_list: + # try: + # print "org not found, creating "+org_name + # ckandirect.action.organization_create(name = org_name, title= doc.value['metadata']["Agency"], + # description= doc.value['metadata']["Agency"]) + # orgs_list.append(org_name) + # except ckanapi.ValidationError, e: + # print e + # raise LoaderError('Unexpected status') + # + # try: + # print "linking package to "+org_name + # + # #brute force + # org = ckandirect.action.organization_show(id=org_name) + # #print org["result"]["packages"] + # #print pkg + # org["result"]["packages"].append({"id":pkg['id'], "name": pkg_name}) + # #print org["result"] + # ckandirect.call_action(action="organization_update", data_dict=org["result"]) + # + # # if only this worked + # print ckandirect.action.package_owner_org_update(id=pkg['id'],organization_id=org["result"]["id"]) + # + # except ckanapi.ValidationError, e: + # print e + # raise LoaderError('Unexpected status') + # except ckanapi.CKANAPIError, e: + # print "lol" + + + if 'Download' in doc.value['metadata'].keys(): try: - pkg = ckan.package_entity_get(pkg_name) + resources = pkg.get('resources', []) if len(resources) < len(doc.value['metadata']['Download']): for resource in doc.value['metadata']['Download']: --- a/documents/genericScrapers.py +++ b/documents/genericScrapers.py @@ -202,7 +202,7 @@ def getDate(self, content, entry, doc): strdate = ''.join(content.stripped_strings).strip() (a, b, c) = strdate.partition("(") - strdate = self.remove_control_chars(a.replace("Octber", "October").replace("Janrurary", "January").replace("1012","2012")) + strdate = self.remove_control_chars(a.replace("Octber", "October").replace("1012","2012")replace("Janrurary", "January").replace("1012","2012")) print strdate try: edate = parse(strdate, dayfirst=True, fuzzy=True).strftime("%Y-%m-%d") --- a/documents/scrapers/0049d35216493c545ef5f7f000e6b252.py +++ b/documents/scrapers/0049d35216493c545ef5f7f000e6b252.py @@ -42,7 +42,6 @@ 'data': {'request': '', 'session': '', 'more': ''} } - - amonpy.exception(data) + #amonpy.exception(data) pass --- a/documents/scrapers/6fa04af95fbe7de96daa2c7560e0aad3.py +++ b/documents/scrapers/6fa04af95fbe7de96daa2c7560e0aad3.py @@ -6,8 +6,6 @@ #http://www.doughellmann.com/PyMOTW/abc/ class ScraperImplementation(genericScrapers.GenericOAICDisclogScraper): - def getTable(self,soup): - return soup.find(id = "content_div_50269").table def getColumns(self,columns): (id, date, title, description, notes) = columns return (id, date, title, description, notes) --- /dev/null +++ b/documents/scrapers/b0ca7fddcd1c965787daea47f2d32e0a.py @@ -1,1 +1,17 @@ +import sys,os +sys.path.insert(0, os.path.join(os.path.dirname(__file__) or '.', '../')) +import genericScrapers +import scrape +from bs4 import BeautifulSoup +#http://www.doughellmann.com/PyMOTW/abc/ +class ScraperImplementation(genericScrapers.GenericOAICDisclogScraper): + def getColumns(self,columns): + (id, date, title, description, notes) = columns + return (id, date, title, description, notes) + +if __name__ == '__main__': + print 'Subclass:', issubclass(ScraperImplementation, genericScrapers.GenericOAICDisclogScraper) + print 'Instance:', isinstance(ScraperImplementation(), genericScrapers.GenericOAICDisclogScraper) + ScraperImplementation().doScrape() +