Merge branch 'bnewbold-xml-html-ingest' into 'master'

HTML webcapture ingest (and XML file ingest)

See merge request webgroup/fatcat!88

1 files changed, 66 insertions, 2 deletions

diff --git a/python/tests/import_ingest.py b/python/tests/import_ingest.py
index 4a46232a..92539f1a 100644
--- a/python/tests/import_ingest.py
+++ b/python/tests/import_ingest.py
@@ -2,7 +2,7 @@
 import json
 import pytest
 
-from fatcat_tools.importers import IngestFileResultImporter, JsonLinePusher
+from fatcat_tools.importers import IngestFileResultImporter, IngestWebResultImporter, JsonLinePusher
 from fixtures import *
 
 
@@ -10,6 +10,10 @@ from fixtures import *
 def ingest_importer(api):
     yield IngestFileResultImporter(api)
 
+@pytest.fixture(scope="function")
+def ingest_web_importer(api):
+    yield IngestWebResultImporter(api)
+
 # TODO: use API to check that entities actually created...
 def test_ingest_importer_basic(ingest_importer):
     with open('tests/files/example_ingest.json', 'r') as f:
@@ -41,6 +45,60 @@ def test_ingest_importer(ingest_importer):
     assert counts['exists'] == 1
     assert counts['skip'] == 1
 
+def test_ingest_importer_xml(ingest_importer):
+    last_index = ingest_importer.api.get_changelog(limit=1)[0].index
+    with open('tests/files/example_ingest_xml.json', 'r') as f:
+        ingest_importer.bezerk_mode = True
+        counts = JsonLinePusher(ingest_importer, f).run()
+    print(counts)
+    assert counts['insert'] == 1
+    assert counts['exists'] == 0
+    assert counts['skip'] == 0
+
+    # fetch most recent editgroup
+    change = ingest_importer.api.get_changelog_entry(index=last_index+1)
+    eg = change.editgroup
+    assert eg.description
+    assert "crawled from web" in eg.description.lower()
+    assert eg.extra['git_rev']
+    assert "fatcat_tools.IngestFileResultImporter" in eg.extra['agent']
+
+    # re-import should skip
+    with open('tests/files/example_ingest_xml.json', 'r') as f:
+        ingest_importer.reset()
+        ingest_importer.bezerk_mode = False
+        counts = JsonLinePusher(ingest_importer, f).run()
+    assert counts['insert'] == 0
+    assert counts['exists'] == 1
+    assert counts['skip'] == 0
+
+def test_ingest_importer_web(ingest_web_importer):
+    last_index = ingest_web_importer.api.get_changelog(limit=1)[0].index
+    with open('tests/files/example_ingest_html.json', 'r') as f:
+        ingest_web_importer.bezerk_mode = True
+        counts = JsonLinePusher(ingest_web_importer, f).run()
+    print(counts)
+    assert counts['insert'] == 1
+    assert counts['exists'] == 0
+    assert counts['skip'] == 0
+
+    # fetch most recent editgroup
+    change = ingest_web_importer.api.get_changelog_entry(index=last_index+1)
+    eg = change.editgroup
+    assert eg.description
+    assert "crawled from web" in eg.description.lower()
+    assert eg.extra['git_rev']
+    assert "fatcat_tools.IngestWebResultImporter" in eg.extra['agent']
+
+    # re-import should skip
+    with open('tests/files/example_ingest_html.json', 'r') as f:
+        ingest_web_importer.reset()
+        ingest_web_importer.bezerk_mode = False
+        counts = JsonLinePusher(ingest_web_importer, f).run()
+    assert counts['insert'] == 0
+    assert counts['exists'] == 1
+    assert counts['skip'] == 0
+
 def test_ingest_importer_stage(ingest_importer, api):
     """
     Tests that ingest importer correctly handles release stage matching
@@ -57,7 +115,7 @@ def test_ingest_importer_stage(ingest_importer, api):
     with open('tests/files/example_ingest.json', 'r') as f:
         raw = json.loads(f.readline())
     for row in test_table:
-        print(row)
+        #print(row)
 
         # set dummy record stage
         eg = quick_eg(api)
@@ -94,6 +152,12 @@ def test_ingest_dict_parse(ingest_importer):
 def test_ingest_dict_parse_old(ingest_importer):
     with open('tests/files/example_ingest.old.json', 'r') as f:
         raw = json.loads(f.readline())
+
+        # ancient ingest requests had no type; skip them
+        f = ingest_importer.parse_record(raw)
+        assert f == None
+        raw['request']['ingest_type'] = 'pdf'
+
         f = ingest_importer.parse_record(raw)
         assert f.sha1 == "00242a192acc258bdfdb151943419437f440c313"
         assert f.md5 == "f4de91152c7ab9fdc2a128f962faebff"