diff --git a/setup.py b/setup.py
index b19e5f14..a98aeb88 100644
--- a/setup.py
+++ b/setup.py
@@ -18,6 +18,7 @@
 install_requires = [
     'autosemver~=0.2',
     'inspire-schemas~=42.0',
+    'inspire-dojson~=41.0',
     'Scrapy>=1.1.0',
     # TODO: unpin once they support wheel building again
     'scrapyd==1.1.0',
diff --git a/tests/functional/arxiv/test_arxiv.py b/tests/functional/arxiv/test_arxiv.py
index a9677b89..0f58b17d 100644
--- a/tests/functional/arxiv/test_arxiv.py
+++ b/tests/functional/arxiv/test_arxiv.py
@@ -72,6 +72,7 @@ def test_arxiv(set_up_local_environment, expected_results):
         app=celery_app,
         monitor_timeout=5,
         monitor_iter_limit=100,
+        events_limit=1,
         crawler_instance=crawler,
         project=set_up_local_environment.get('CRAWLER_PROJECT'),
         spider='arXiv',
diff --git a/tests/functional/wsp/test_wsp.py b/tests/functional/wsp/test_wsp.py
index 70996466..a0411b8e 100644
--- a/tests/functional/wsp/test_wsp.py
+++ b/tests/functional/wsp/test_wsp.py
@@ -13,7 +13,6 @@
 
 import pytest
 import os
-import shutil
 
 from time import sleep
 
@@ -21,6 +20,7 @@
 from hepcrawl.testlib.fixtures import (
     get_test_suite_path,
     expected_json_results_from_file,
+    clean_dir,
 )
 from hepcrawl.testlib.tasks import app as celery_app
 from hepcrawl.testlib.utils import get_crawler_instance
@@ -90,10 +90,6 @@ def remove_generated_files(package_location):
             os.unlink(os.path.join(package_location, file_name))
 
 
-def clean_dir(path='/tmp/WSP/'):
-    shutil.rmtree(path, ignore_errors=True)
-
-
 @pytest.mark.parametrize(
     'expected_results',
     [
@@ -114,6 +110,7 @@ def test_wsp_ftp(set_up_ftp_environment, expected_results):
         app=celery_app,
         monitor_timeout=5,
         monitor_iter_limit=100,
+        events_limit=1,
         crawler_instance=crawler,
         project=set_up_ftp_environment.get('CRAWLER_PROJECT'),
         spider='WSP',
@@ -147,6 +144,7 @@ def test_wsp_local_package_path(set_up_local_environment, expected_results):
         app=celery_app,
         monitor_timeout=5,
         monitor_iter_limit=100,
+        events_limit=1,
         crawler_instance=crawler,
         project=set_up_local_environment.get('CRAWLER_PROJECT'),
         spider='WSP',
diff --git a/tests/unit/test_alpha.py b/tests/unit/test_alpha.py
index eef140b1..96bf9af1 100644
--- a/tests/unit/test_alpha.py
+++ b/tests/unit/test_alpha.py
@@ -20,13 +20,15 @@
 def results():
     """Return results generator from the Alpha spider."""
     spider = alpha_spider.AlphaSpider()
-    records = list(
+    parsed_items = list(
         spider.parse(
             fake_response_from_file('alpha/test_1.htm')
         )
     )
 
+    records = [parsed_item.record for parsed_item in parsed_items]
     assert records
+
     return records
 
 
diff --git a/tests/unit/test_aps.py b/tests/unit/test_aps.py
index eb53269d..3bb3698c 100644
--- a/tests/unit/test_aps.py
+++ b/tests/unit/test_aps.py
@@ -21,7 +21,7 @@ def results():
     from scrapy.http import TextResponse
 
     spider = aps_spider.APSSpider()
-    records = list(
+    parsed_items = list(
         spider.parse(
             fake_response_from_file(
                 'aps/aps_single_response.json',
@@ -30,6 +30,8 @@ def results():
         )
     )
 
+    records = [parsed_item.record for parsed_item in parsed_items]
+
     assert records
     return records
 
diff --git a/tests/unit/test_arxiv_all.py b/tests/unit/test_arxiv_all.py
index bd75e5a4..1f4155c9 100644
--- a/tests/unit/test_arxiv_all.py
+++ b/tests/unit/test_arxiv_all.py
@@ -11,7 +11,8 @@
 
 import pytest
 
-from scrapy.crawler import  Crawler
+from scrapy.crawler import Crawler
+from scrapy.http import TextResponse
 
 from hepcrawl.pipelines import InspireCeleryPushPipeline
 from hepcrawl.spiders import arxiv_spider
@@ -25,36 +26,16 @@ def spider():
     return spider
 
 
-@pytest.fixture
-def one_result(spider):
-    """Return results generator from the arxiv spider. Tricky fields, one
-    record.
-    """
-    from scrapy.http import TextResponse
-
-    records = list(
-        spider.parse(
-            fake_response_from_file(
-                'arxiv/sample_arxiv_record0.xml',
-                response_type=TextResponse,
-            )
-        )
-    )
-
-    assert records
-    pipeline = InspireCeleryPushPipeline()
-    pipeline.open_spider(spider)
-    return [pipeline.process_item(record, spider) for record in records]
-
-
 @pytest.fixture
 def many_results(spider):
     """Return results generator from the arxiv spider. Tricky fields, many
     records.
     """
-    from scrapy.http import TextResponse
+    def _get_processed_record(item, spider):
+        record = pipeline.process_item(item, spider)
+        return record
 
-    records = list(
+    parsed_items = list(
         spider.parse(
             fake_response_from_file(
                 'arxiv/sample_arxiv_record.xml',
@@ -63,10 +44,10 @@ def many_results(spider):
         )
     )
 
-    assert records
     pipeline = InspireCeleryPushPipeline()
     pipeline.open_spider(spider)
-    return [pipeline.process_item(record, spider) for record in records]
+
+    return [_get_processed_record(parsed_item, spider) for parsed_item in parsed_items]
 
 
 def test_page_nr(many_results):
diff --git a/tests/unit/test_arxiv_single.py b/tests/unit/test_arxiv_single.py
index a6ed66d6..329a2a49 100644
--- a/tests/unit/test_arxiv_single.py
+++ b/tests/unit/test_arxiv_single.py
@@ -24,10 +24,15 @@
 def results():
     """Return results generator from the arxiv spider. All fields, one record.
     """
+    def _get_processed_item(item, spider):
+        record = pipeline.process_item(item, spider)
+        validate(record, 'hep')
+        assert record
+        return record
 
     crawler = Crawler(spidercls=arxiv_spider.ArxivSpider)
     spider = arxiv_spider.ArxivSpider.from_crawler(crawler)
-    records = list(
+    parsed_items = list(
         spider.parse(
             fake_response_from_file(
                 'arxiv/sample_arxiv_record0.xml',
@@ -36,16 +41,10 @@ def results():
         )
     )
 
-    assert records
     pipeline = InspireCeleryPushPipeline()
     pipeline.open_spider(spider)
-    processed_records = []
-    for record in records:
-        processed_record = pipeline.process_item(record, spider)
-        validate(processed_record, 'hep')
-        processed_records.append(processed_record)
 
-    return processed_records
+    return [_get_processed_item(parsed_item, spider) for parsed_item in parsed_items]
 
 
 
diff --git a/tests/unit/test_base.py b/tests/unit/test_base.py
index cc6ef093..b8ec5b8a 100644
--- a/tests/unit/test_base.py
+++ b/tests/unit/test_base.py
@@ -38,9 +38,12 @@ def record():
     nodes = selector.xpath('.//%s' % spider.itertag)
     response.meta["record"] = nodes[0].extract()
     response.meta["urls"] = ["http://hdl.handle.net/1885/10005"]
-    parsed_record = spider.build_item(response)
-    assert parsed_record
-    return parsed_record
+
+    parsed_item = spider.build_item(response)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 @pytest.fixture
@@ -169,7 +172,12 @@ def splash():
                 'Content-Type': 'text/html',
             },
         )
-        return spider.scrape_for_pdf(splash_response)
+
+        parsed_item = spider.scrape_for_pdf(splash_response)
+        assert parsed_item
+        assert parsed_item.record
+
+        return parsed_item.record
 
 
 def test_splash(splash):
@@ -201,7 +209,12 @@ def parsed_node():
     response = fake_response_from_string(text=body)
     node = get_node(spider, 'OAI-PMH:record', text=body)
     response.meta["record"] = node[0].extract()
-    return spider.parse_node(response, node[0])
+
+    parsed_item = spider.parse_node(response, node[0])
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_parsed_node(parsed_node):
diff --git a/tests/unit/test_brown.py b/tests/unit/test_brown.py
index 0b42b4df..8d0f20de 100644
--- a/tests/unit/test_brown.py
+++ b/tests/unit/test_brown.py
@@ -41,10 +41,12 @@ def record():
 
         splash_response = fake_response_from_file('brown/test_splash.html')
         splash_response.meta["jsonrecord"] = jsonrecord
-        parsed_record = spider.scrape_splash(splash_response)
 
-        assert parsed_record
-        return parsed_record
+        parsed_item = spider.scrape_splash(splash_response)
+        assert parsed_item
+        assert parsed_item.record
+
+        return parsed_item.record
 
 
 @pytest.fixture
@@ -200,7 +202,11 @@ def parsed_node_no_splash():
     jsonrecord = jsonresponse["items"]["docs"][0]
     response.meta["jsonrecord"] = jsonrecord
 
-    return spider.parse(response).next()
+    parsed_item = spider.parse(response).next()
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_no_splash(parsed_node_no_splash):
diff --git a/tests/unit/test_dnb.py b/tests/unit/test_dnb.py
index b00aff3d..5aa05a64 100644
--- a/tests/unit/test_dnb.py
+++ b/tests/unit/test_dnb.py
@@ -72,7 +72,12 @@ def record(scrape_pos_page_body):
             body=scrape_pos_page_body,
             **{'encoding': 'utf-8'}
         )
-        return request.callback(response)
+
+        parsed_item = request.callback(response)
+        assert parsed_item
+        assert parsed_item.record
+
+        return parsed_item.record
 
 
 def test_title(record):
@@ -241,7 +246,12 @@ def parse_without_splash():
                 'Content-Type': 'application/pdf;charset=base64',
             }
         )
-        return spider.parse_node(response, nodes[0])
+
+        parsed_item = spider.parse_node(response, nodes[0])
+        assert parsed_item
+        assert parsed_item.record
+
+        return parsed_item.record
 
 
 def test_parse_without_splash(parse_without_splash):
diff --git a/tests/unit/test_edp.py b/tests/unit/test_edp.py
index cc7885bd..115abda6 100644
--- a/tests/unit/test_edp.py
+++ b/tests/unit/test_edp.py
@@ -40,6 +40,7 @@ def scrape_pos_page_body():
         )
     )
 
+
 @pytest.fixture
 def targzfile():
     """Path to test tar.gz file with JATS XML file."""
@@ -50,6 +51,7 @@ def targzfile():
         'test_gz.tar.gz'
     )
 
+
 @pytest.fixture
 def package_jats(targzfile):
     """Extract tar.gz package with JATS XML file."""
@@ -75,7 +77,12 @@ def record_jats(package_jats, scrape_pos_page_body):
         body=scrape_pos_page_body,
         **{'encoding': 'utf-8'}
     )
-    return request.callback(response)
+
+    parsed_item = request.callback(response)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 @pytest.fixture
@@ -107,7 +114,11 @@ def record_rich(package_rich):
     fake_resp.meta["rich"] = True
     node = get_node(spider, "//EDPSArticle", fake_resp)[0]
 
-    return spider.parse_node(fake_resp, node)
+    parsed_item = spider.parse_node(fake_resp, node)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_title(record_jats):
@@ -145,6 +156,7 @@ def test_abstract(record_jats):
     assert 'abstract' in record_jats
     assert record_jats['abstract'] == abstract
 
+
 def test_date_published(record_jats):
     """Test extracting date_published."""
     date_published = "2015-01-01"
@@ -179,6 +191,7 @@ def test_doi(record_jats):
     assert 'dois' in record_jats
     assert record_jats['dois'][0]['value'] == doi
 
+
 def test_publication_info(record_jats):
     """Test extracting publication info."""
     assert 'journal_title' in record_jats
@@ -206,7 +219,6 @@ def test_keywords(record_jats):
         assert keyw["value"] in keywords
 
 
-
 def test_authors(record_jats):
     """Test authors."""
     authors = ["Arasoglu, Ali", "Ozdemir, Omer Faruk"]
@@ -326,7 +338,6 @@ def test_authors_rich(record_rich):
         assert astr[index]["affiliations"][0]["value"] == affiliations[index]
 
 
-
 def test_tarfile(tarbzfile, tmpdir):
     """Test untarring a tar.bz package with a test XML file.
 
@@ -343,7 +354,6 @@ def test_tarfile(tarbzfile, tmpdir):
     assert "aas/xml_rich/2000/01" not in xml_files_flat[0]
 
 
-
 def test_handle_package_ftp(tarbzfile):
     """Test getting the target folder name for xml files."""
     spider = edp_spider.EDPSpider()
@@ -351,7 +361,8 @@ def test_handle_package_ftp(tarbzfile):
     request = spider.handle_package_ftp(response).next()
 
     assert isinstance(request, Request)
-    assert request.meta["package_path"] == tarbzfile
+    assert request.meta["source_folder"] == tarbzfile
+
 
 def test_no_dois_jats():
     """Test parsing when no DOI in record. JATS format."""
@@ -370,7 +381,11 @@ def test_no_dois_jats():
     """
     response = fake_response_from_string(body)
     node = get_node(spider, "//article", response)[0]
-    record = spider.parse_node(response, node)
+
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+    record = parsed_item.record
 
     assert "dois" not in record
     assert "additional_files" not in record
@@ -390,7 +405,11 @@ def test_no_dois_rich():
     response = fake_response_from_string(body)
     response.meta["rich"] = True
     node = get_node(spider, "//EDPSArticle", response)[0]
-    record = spider.parse_node(response, node)
+
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+    record = parsed_item.record
 
     assert "dois" not in record
     assert "additional_files" not in record
@@ -416,7 +435,11 @@ def test_addendum_jats():
     """
     response = fake_response_from_string(body)
     node = get_node(spider, "//article", response)[0]
-    record = spider.parse_node(response, node)
+
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+    record = parsed_item.record
 
     assert "related_article_doi" in record
     assert record["related_article_doi"][0][
@@ -439,7 +462,11 @@ def test_author_with_email():
     """
     response = fake_response_from_string(body)
     node = get_node(spider, "//article", response)[0]
-    record = spider.parse_node(response, node)
+
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+    record = parsed_item.record
 
     assert 'email' in record['authors'][0]
     assert record['authors'][0]['email'] == "Fname.Sname@university.org"
@@ -472,7 +499,11 @@ def test_aff_with_email():
     """
     response = fake_response_from_string(body)
     node = get_node(spider, "//article", response)[0]
-    record = spider.parse_node(response, node)
+
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+    record = parsed_item.record
 
     affiliation = "Department of Physics, Western Michigan University, Kalamazoo, MI 49008, USA"
     assert 'affiliations' in record['authors'][0]
@@ -481,8 +512,6 @@ def test_aff_with_email():
     assert record['authors'][0]['email'] is None
 
 
-
-
 def test_no_valid_article():
     """Test parsing when filtering out non-interesting article types."""
     spider = edp_spider.EDPSpider()
@@ -506,7 +535,11 @@ def test_collections_review():
     """
     response = fake_response_from_string(body)
     node = get_node(spider, "//article", response)[0]
-    record = spider.parse_node(response, node)
+
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+    record = parsed_item.record
 
     assert "collections" in record
     assert record["collections"] == [{'primary': 'HEP'}, {'primary': 'Review'}]
@@ -533,7 +566,12 @@ def record_references_only():
     """
     response = fake_response_from_string(body)
     node = get_node(spider, "//article", response)[0]
-    return spider.parse_node(response, node)
+
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_references(record_references_only):
diff --git a/tests/unit/test_elsevier.py b/tests/unit/test_elsevier.py
index ca023122..3d5fb3f5 100644
--- a/tests/unit/test_elsevier.py
+++ b/tests/unit/test_elsevier.py
@@ -41,9 +41,12 @@ def record():
         response.meta["xml_url"] = 'elsevier/sample_consyn_record.xml'
         tag = '//%s' % spider.itertag
         nodes = get_node(spider, tag, response)
-        parsed_record = spider.parse_node(response, nodes)
-        assert parsed_record
-        return parsed_record
+
+        parsed_item = spider.parse_node(response, nodes)
+        assert parsed_item
+        assert parsed_item.record
+
+        return parsed_item.record
 
 
 @pytest.fixture(scope="module")
@@ -97,7 +100,12 @@ def parsed_node():
         response.meta["xml_url"] = 'elsevier/sample_consyn_record.xml'
         parse_response = spider.parse_node(response, node)
         parse_response.status = 404
-        return spider.scrape_sciencedirect(parse_response)
+
+        parsed_item = spider.scrape_sciencedirect(parse_response)
+        assert parsed_item
+        assert parsed_item.record
+
+        return parsed_item.record
 
 
 def test_collection(parsed_node):
@@ -164,7 +172,11 @@ def cover_display_date():
 
     node = get_node(spider, '/doc', text=body)
     response = fake_response_from_string(body)
-    return spider.parse_node(response, node)
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_cover_display_date(cover_display_date):
@@ -187,7 +199,11 @@ def cover_display_date_y_m():
     </doc>"""
     node = get_node(spider, '/doc', text=body)
     response = fake_response_from_string(body)
-    return spider.parse_node(response, node)
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_cover_display_date_y_m(cover_display_date_y_m):
@@ -210,7 +226,11 @@ def cover_display_date_y():
     </doc>"""
     node = get_node(spider, '/doc', text=body)
     response = fake_response_from_string(body)
-    return spider.parse_node(response, node)
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_cover_display_date_y(cover_display_date_y):
@@ -1579,11 +1599,11 @@ def test_handle_package(handled_package):
     for astro, nima in zip(astropart, nima):
         assert nima
         assert astro
-        assert astro.meta["package_path"] == "tests/unit/responses/elsevier/fake_astropart.zip"
+        assert astro.meta["source_folder"] == "tests/unit/responses/elsevier/fake_astropart.zip"
         url_to_match = u'file:///tmp/elsevier_fake_astropart_*/0927-6505/aip/S0927650515001656/S0927650515001656.xml'
         assert astro.meta["xml_url"] == fnmatch.filter([astro.meta["xml_url"]], url_to_match)[0]
 
-        assert nima.meta["package_path"] == "tests/unit/responses/elsevier/fake_nima.zip"
+        assert nima.meta["source_folder"] == "tests/unit/responses/elsevier/fake_nima.zip"
         url_to_match = u'file:///tmp/elsevier_fake_nima_*/0168-9002/S0168900215X00398/S0168900215015636/S0168900215015636.xml'
         assert nima.meta["xml_url"] == fnmatch.filter([nima.meta["xml_url"]], url_to_match)[0]
 
@@ -1644,7 +1664,12 @@ def sciencedirect():
         ])
     response.meta["info"] = {}
     response.meta["node"] = get_node(spider, '/head', text=body)
-    return spider.scrape_sciencedirect(response)
+
+    parsed_item = spider.scrape_sciencedirect(response)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_sciencedirect(sciencedirect):
diff --git a/tests/unit/test_hindawi.py b/tests/unit/test_hindawi.py
index 37e5e183..3af8ba3a 100644
--- a/tests/unit/test_hindawi.py
+++ b/tests/unit/test_hindawi.py
@@ -26,9 +26,11 @@ def record():
     response = fake_response_from_file("hindawi/test_1.xml")
     nodes = get_node(spider, "//marc:record", response)
 
-    parsed_record = spider.parse_node(response, nodes[0])
-    assert parsed_record
-    return parsed_record
+    parsed_item = spider.parse_node(response, nodes[0])
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_title(record):
diff --git a/tests/unit/test_infn.py b/tests/unit/test_infn.py
index 0c60799a..c15ef727 100644
--- a/tests/unit/test_infn.py
+++ b/tests/unit/test_infn.py
@@ -28,9 +28,12 @@ def record():
     """Return scraping results from the INFN spider."""
     spider = infn_spider.InfnSpider()
     response = fake_response_from_file('infn/test_splash.html')
-    parsed_record = spider.scrape_splash(response)
-    assert parsed_record
-    return parsed_record
+
+    parsed_item = spider.scrape_splash(response)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_title(record):
@@ -121,6 +124,7 @@ def test_non_thesis():
 
     assert record is None
 
+
 def test_parse_node():
     """Test parse_node function. This should be a scrapy Request object.
 
@@ -148,6 +152,8 @@ def test_parse_node_nolink():
     response = fake_response_from_file('infn/test_1_nolink.html')
     selector = Selector(response, type='html')
     node = selector.xpath('//%s' % spider.itertag)[0]
-    record = spider.parse_node(response, node).next()
+    parsed_item = spider.parse_node(response, node).next()
+    assert parsed_item
+    assert parsed_item.record
 
-    assert isinstance(record, hepcrawl.items.HEPRecord)
+    assert isinstance(parsed_item.record, hepcrawl.items.HEPRecord)
diff --git a/tests/unit/test_iop.py b/tests/unit/test_iop.py
index b776adfa..bb01766c 100644
--- a/tests/unit/test_iop.py
+++ b/tests/unit/test_iop.py
@@ -38,9 +38,12 @@ def record():
     response = fake_response_from_file('iop/xml/test_standard.xml')
     node = get_node(spider, "Article", response)
     spider.pdf_files = TEST_PDF_DIR
-    parsed_record = spider.parse_node(response, node)
-    assert parsed_record
-    return parsed_record
+
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_abstract(record):
@@ -182,10 +185,12 @@ def erratum_open_access_record():
         'iop',
         'pdf',
     )
-    parsed_record = spider.parse_node(response, node)
-    assert parsed_record
-    return parsed_record
 
+    parsed_item = spider.parse_node(response, node)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_files_erratum_open_access_record(erratum_open_access_record):
diff --git a/tests/unit/test_magic.py b/tests/unit/test_magic.py
index eeb574fe..f3c0f355 100644
--- a/tests/unit/test_magic.py
+++ b/tests/unit/test_magic.py
@@ -23,6 +23,7 @@
     get_node,
 )
 
+
 @pytest.fixture
 def record():
     """Return results from the MAGIC spider. First parse node, then scrape,
@@ -39,9 +40,11 @@ def record():
     splash_response.meta["date"] = parsed_node.meta["date"]
     splash_response.meta["urls"] = parsed_node.meta["urls"]
 
-    parsed_record = spider.scrape_for_pdf(splash_response).next()
-    assert parsed_record
-    return parsed_record
+    parsed_item = spider.scrape_for_pdf(splash_response).next()
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_abstract(record):
@@ -102,7 +105,6 @@ def test_abstract(record):
     assert record["abstract"] == abstract
 
 
-
 def test_title(record):
     """Test extracting title."""
     title = "Limits to the violation of Lorentz invariance using the emission of the CRAB pulsar at TeV energies, discovered with archival data from the MAGIC telescopes"
@@ -139,6 +141,7 @@ def test_url(record):
     assert 'urls' in record
     assert record['urls'][0]['value'] == url
 
+
 def test_pdf_link(record):
     """Test pdf link(s)"""
     files = "http://stlab.adobe.com/wiki/images/d/d3/Test.pdf"
@@ -164,7 +167,10 @@ def test_no_author_no_date_no_url():
     """
     response = fake_response_from_string(body)
     node = get_node(spider, spider.itertag, text=body)
-    record = spider.parse_node(response, node).next()
+    parsed_item = spider.parse_node(response, node).next()
+    assert parsed_item
+    assert parsed_item.record
+    record = parsed_item.record
 
     assert isinstance(record, hepcrawl.items.HEPRecord)
     assert "date" not in record
@@ -184,7 +190,10 @@ def test_no_aff():
     </html>
     """
     response = fake_response_from_string(body)
-    record = spider.scrape_for_pdf(response).next()
+    parsed_item = spider.scrape_for_pdf(response).next()
+    assert parsed_item
+    assert parsed_item.record
+    record = parsed_item.record
 
     assert isinstance(record, hepcrawl.items.HEPRecord)
     assert "date" not in record
@@ -216,7 +225,10 @@ def test_no_spash_page():
     response.status = 404
     response.meta["title"] = parsed_node.meta["title"]
     response.meta["urls"] = parsed_node.meta["urls"]
-    record = spider.scrape_for_pdf(response).next()
+    parsed_item = spider.scrape_for_pdf(response).next()
+    assert parsed_item
+    assert parsed_item.record
+    record = parsed_item.record
 
     assert isinstance(record, hepcrawl.items.HEPRecord)
     assert "urls" in record
diff --git a/tests/unit/test_mit.py b/tests/unit/test_mit.py
index 0253d91f..8a185cef 100644
--- a/tests/unit/test_mit.py
+++ b/tests/unit/test_mit.py
@@ -25,9 +25,12 @@ def record():
     """Return scraping results from the MIT spider."""
     spider = mit_spider.MITSpider()
     response = fake_response_from_file('mit/test_splash.html')
-    parsed_record = spider.build_item(response)
-    assert parsed_record
-    return parsed_record
+
+    parsed_item = spider.build_item(response)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 @pytest.fixture
@@ -37,7 +40,11 @@ def parsed_node():
     response = fake_response_from_file('mit/test_list.html')
     tag = spider.itertag
     node = get_node(spider, tag, response, rtype="html")
-    return spider.parse_node(response, node).next()
+
+    parsed_item = spider.parse_node(response, node).next()
+    assert parsed_item
+
+    return parsed_item
 
 
 def test_url(parsed_node):
@@ -159,7 +166,12 @@ def supervisors():
     <html>
     """
     response = fake_response_from_string(body)
-    return spider.build_item(response)
+
+    parsed_item = spider.build_item(response)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_two_supervisors(supervisors):
diff --git a/tests/unit/test_phenix.py b/tests/unit/test_phenix.py
index 75384350..c272683f 100644
--- a/tests/unit/test_phenix.py
+++ b/tests/unit/test_phenix.py
@@ -29,9 +29,13 @@ def record():
     response = fake_response_from_file('phenix/test_1.html')
     selector = Selector(response, type='html')
     nodes = selector.xpath('//%s' % spider.itertag)
-    parsed_record = spider.parse_node(response, nodes[0])
-    assert parsed_record
-    return parsed_record
+
+    parsed_item = spider.parse_node(response, nodes[0])
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
+
 
 @pytest.fixture
 def non_thesis():
@@ -49,10 +53,12 @@ def non_thesis():
     node = get_node(spider, '//li', text=body)
     return spider.parse_node(response, node)
 
+
 def test_non_thesis(non_thesis):
     """Test MSc thesis skipping."""
     assert non_thesis is None
 
+
 def test_title(record):
     """Test extracting title."""
     title = "MEASUREMENT OF THE DOUBLE HELICITY ASYMMETRY IN INCLUSIVE $\pi^{0}$ PRODUCTION IN POLARIZED PROTON-PROTON COLLISIONS AT $\sqrt{s}$ = 510 GeV"
@@ -82,6 +88,7 @@ def test_authors(record):
             aff['value'] for aff in record['authors'][index]['affiliations']
         ]
 
+
 def test_pdf_link(record):
     """Test pdf link(s)"""
     files = "http://www.phenix.bnl.gov/phenix/WWW/talk/archive/theses/2015/Guragain_Hari-DISSERTATION.pdf"
diff --git a/tests/unit/test_phil.py b/tests/unit/test_phil.py
index e99064b2..6db536ef 100644
--- a/tests/unit/test_phil.py
+++ b/tests/unit/test_phil.py
@@ -33,9 +33,12 @@ def record():
         "http://philpapers.org/go.pl?id=BROBB&proxyId=none&u=http%3A%2F%2Fanalysis.oxfordjournals.org%2Fcontent%2F66%2F3%2F194.full.pdf%2Bhtml%3Fframe%3Dsidebar",
         "http://philpapers.org/go.pl?id=BROBB&proxyId=none&u=http%3A%2F%2Fbrogaardb.googlepages.com%2Ftensedrelationsoffprint.pdf"
     ]
-    parsed_record = spider.build_item(response)
-    assert parsed_record
-    return parsed_record
+
+    parsed_item = spider.build_item(response)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 @pytest.fixture
@@ -48,7 +51,12 @@ def journal():
     response = fake_response_from_file('phil/test_journal.json')
     jsonrecord = json.loads(response.body_as_unicode())
     response.meta["jsonrecord"] = jsonrecord[0]
-    return spider.build_item(response)
+
+    parsed_item = spider.build_item(response)
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 @pytest.fixture
@@ -223,7 +231,11 @@ def splash():
         ]
     }
 
-        return spider.scrape_for_pdf(response)
+        parsed_item = spider.scrape_for_pdf(response)
+        assert parsed_item
+        assert parsed_item.record
+
+        return parsed_item.record
 
 
 def test_scrape(splash):
diff --git a/tests/unit/test_pos.py b/tests/unit/test_pos.py
index 20c872f4..bea29b34 100644
--- a/tests/unit/test_pos.py
+++ b/tests/unit/test_pos.py
@@ -51,8 +51,11 @@ def record(scrape_pos_page_body):
     assert response
     pipeline = InspireCeleryPushPipeline()
     pipeline.open_spider(spider)
-    record = request.callback(response)
-    return pipeline.process_item(record, spider)
+    parsed_item = request.callback(response)
+    parsed_record = pipeline.process_item(parsed_item, spider)
+    assert parsed_record
+
+    return parsed_record
 
 
 def test_titles(record):
diff --git a/tests/unit/test_t2k.py b/tests/unit/test_t2k.py
index 283a02e5..d9395aa2 100644
--- a/tests/unit/test_t2k.py
+++ b/tests/unit/test_t2k.py
@@ -36,9 +36,11 @@ def record():
     splash_response.meta["urls"] = parsed_node.meta["urls"]
     splash_response.meta["authors"] = parsed_node.meta["authors"]
 
-    parsed_record = spider.scrape_for_pdf(splash_response).next()
-    assert parsed_record
-    return parsed_record
+    parsed_item = spider.scrape_for_pdf(splash_response).next()
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_abstact(record):
@@ -125,9 +127,11 @@ def non_url():
     selector = Selector(response, type='html')
     nodes = selector.xpath('//%s' % spider.itertag)
 
-    parsed_record = spider.parse_node(response, nodes[0]).next()
-    assert parsed_record
-    return parsed_record
+    parsed_item = spider.parse_node(response, nodes[0]).next()
+    assert parsed_item
+    assert parsed_item.record
+
+    return parsed_item.record
 
 
 def test_non_url(non_url):
diff --git a/tests/unit/test_world_scientific.py b/tests/unit/test_world_scientific.py
index 36438ab4..291d00d0 100644
--- a/tests/unit/test_world_scientific.py
+++ b/tests/unit/test_world_scientific.py
@@ -48,8 +48,11 @@ def get_records(response_file_name):
 
 
 def get_one_record(response_file_name):
-    results = get_records(response_file_name)
-    return results.next()
+    records = get_records(response_file_name)
+    record = records.next()
+    assert record
+
+    return record
 
 
 def override_generated_fields(record):