def test_cache_is_reused_once_expired(self):
get_client(job_board=JOB_BOARD_CHOICES[0]) # NB the real one
self.assertEqual(len(responses.calls), 1)
self.cache.clear()
get_client(job_board=JOB_BOARD_CHOICES[0]) # NB the real one
self.assertEqual(len(responses.calls), 2)
def test_creating_normal_client_calls_url(self):
with open(wsdl_file_path, "r") as f:
xml = f.read()
responses.add(
responses.GET,
settings.TALENTLINK_API_WSDL,
xml,
status=200,
content_type="text/xml",
)
get_client(job_board=JOB_BOARD_CHOICES[0])
self.assertEqual(len(responses.calls), 1)
def import_attachments_for_job(job, client=None):
doc_imported = 0
if not client:
client = get_client(job_board=job.homepage.job_board)
# This will return list of attachments with
# 'content', 'description', 'fileName', 'id', 'mimeType'
attachments_response = client.service.getAttachments(job.talentlink_id)
for attachment in attachments_response:
if attachment["id"] and attachment["fileName"]:
doc, created = CustomDocument.objects.get_or_create(
talentlink_attachment_id=attachment["id"])
if created:
doc.title = (attachment["description"]
or attachment["fileName"].split(".")[0])
doc.file = ContentFile(
attachment["content"],
name=attachment["fileName"],
)
doc.save()
doc_imported += 1
job.attachments.add(doc)
job.save()
return doc_imported
def get_client(self, job_board=JOB_BOARD_CHOICES[0]):
wsdl_file_path = pathlib.Path(__file__).parent / "wsdl.xml"
wsdl_file_url = f"file://{wsdl_file_path}"
return get_client(wsdl=wsdl_file_url, job_board=job_board)
def test_cache_is_used(self):
get_client(job_board=JOB_BOARD_CHOICES[0]) # NB the real one
self.assertEqual(len(responses.calls), 1)
get_client(job_board=JOB_BOARD_CHOICES[0]) # NB the real one
self.assertEqual(len(responses.calls), 1)
def test_cache_is_set(self):
get_client(job_board=JOB_BOARD_CHOICES[0]) # NB the real one
self.assertNotEqual(self.cache.get(settings.TALENTLINK_API_WSDL), None)
def handle(self, *args, **options):
# We import jobs for each Recruitment Homepage, including draft ones
# so it is possible to import jobs before going live.
if not RecruitmentHomePage.objects.exists():
msg = f"Please create a RecruitmentHomePage page before running the import."
self.stdout.write(self.style.ERROR(msg))
for homepage in RecruitmentHomePage.objects.all():
self.stdout.write(
f"Starting import for recruitment site with homepage id {homepage.id}:"
)
job_board = homepage.job_board
try:
client = get_client(job_board=job_board)
except AttributeError as exception:
msg = exception.args[0] + "\n"
msg += f"Error fetching jobs for '{homepage.job_board}' job board. Skipping import for this site."
self.stdout.write(self.style.ERROR(msg))
continue
page = 1
results = True
num_updated = 0
num_created = 0
doc_imported = 0
image_imported = 0
errors = []
import_timestamp = now()
while results:
self.stdout.write(
f"Fetching page {page} from job board '{job_board}'")
response = client.service.getAdvertisementsByPage(
pageNumber=page, showJobLocation=True)
results = response["advertisements"]
page += 1
if results:
self.stdout.write(
f"{len(results['advertisement'])} advertisements")
for ad in response["advertisements"]["advertisement"]:
try:
job = TalentLinkJob.objects.get(
talentlink_id=ad["id"])
created = False
except TalentLinkJob.DoesNotExist:
job = TalentLinkJob(talentlink_id=ad["id"],
homepage=homepage)
created = True
try:
job = update_job_from_ad(
job,
ad,
homepage=homepage,
defaults={"last_imported": import_timestamp},
import_categories=options["import_categories"],
)
except Exception as e:
msg = (
f"Error occurred while processing job {ad['id']} {ad['jobNumber']}:\n"
+ str(e))
errors.append(msg)
else:
if created:
num_created += 1
else:
num_updated += 1
# Fetch attachments via a different call
try:
doc_imported += import_attachments_for_job(
job, client)
except Exception as e:
msg = (
"Error occurred while importing attachments "
+
f"for job {ad['id']} {ad['jobNumber']}:\n"
+ str(e))
errors.append(msg)
# Fetch logo image via a different call
image_found = False
try:
# This will return a assignedImageDto object with
# 'id', 'url', 'position'
logo_response = client.service.getAdvertisementImages(
job.talentlink_id)
for image in logo_response:
if image["position"] == "Logo":
talentlink_image_id = image["id"]
# Only update image if it is changed or new
try:
logo_image = CustomImage.objects.get(
talentlink_image_id=
talentlink_image_id)
except CustomImage.DoesNotExist:
logo_image = import_image_from_url(
title=job.title,
url=image["url"],
filename="logo " +
str(job.talentlink_id),
talentlink_image_id=
talentlink_image_id,
collection_name="jobs_logo",
)
image_imported += 1
finally:
job.logo = logo_image
job.save()
image_found = True
break
except Exception as e:
msg = (
f"Error occurred while importing logo image for job {ad['id']} {ad['jobNumber']}:\n"
+ str(e))
errors.append(msg)
# Remove logo from existing job if it is gone from this import
if (not created) and (
not image_found) and job.logo:
# Also delete logo image if not used anywhere else too
if (TalentLinkJob.objects.filter(
logo=job.logo).count() == 1):
job.logo.delete()
job.logo = None
job.save()
# Check for outdated jobs
num_deleted = 0
try:
num_deleted = delete_jobs(imported_before=import_timestamp,
homepage=homepage)
except Exception as e:
msg = f"Error occurred while deleting jobs:\n" + str(e)
errors.append(msg)
self.stdout.write("No more results")
self.stdout.write(f"{num_updated} existing jobs updated")
self.stdout.write(f"{num_created} new jobs created")
self.stdout.write(f"{doc_imported} new documents imported")
self.stdout.write(f"{image_imported} new images imported")
self.stdout.write(f"{num_deleted} jobs deleted")
self.stdout.write(self.style.ERROR(f"{len(errors)} errors"))
for msg in errors:
self.stdout.write(self.style.ERROR(msg))