backup/pywb
1
0
Fork 0
mirror of https://github.com/webrecorder/pywb.git synced 2025-03-15 00:03:28 +01:00
Code Issues Projects Releases Activity

tests: add 'bad.cdx' for testing cdx lines with missing original for revisit,

Browse Source 
missing/non-existant warc
This commit is contained in:
Ilya Kreymer 2014-06-25 12:32:57 -07:00
parent 913a1e9f31
commit fb07775d38
3 changed files with 23 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
pywb/cdx/test/test_cdxops.py
View File

@ -54,7 +54,7 @@ com,example)/?example=1 20140103030321 http://example.com?example=1 text/html 20
com,example)/?example=1 20140103030341 http://example.com?example=1 warc/revisit - B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36A - - 553 1864 example.warc.gz com,example)/?example=1 20140103030341 http://example.com?example=1 warc/revisit - B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36A - - 553 1864 example.warc.gz
# Filter exact invert # Filter exact invert
>>> cdx_ops_test(url = 'http://example.com', sources = [test_cdx_dir], matchType = 'prefix', filter = ['!=urlkey:com,example)/?example=1', '!=urlkey:com,example)/?example=2']) >>> cdx_ops_test(url = 'http://example.com', sources = [test_cdx_dir], matchType = 'prefix', filter = ['!=urlkey:com,example)/?example=1', '!=urlkey:com,example)/?example=2', '!=urlkey:com,example)/?example=3'])
com,example)/ 20130729195151 http://test@example.com/ warc/revisit - B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36A - - 591 355 example-url-agnostic-revisit.warc.gz com,example)/ 20130729195151 http://test@example.com/ warc/revisit - B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36A - - 591 355 example-url-agnostic-revisit.warc.gz
com,example)/ 20140127171200 http://example.com text/html 200 B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36A - - 1046 334 dupes.warc.gz com,example)/ 20140127171200 http://example.com text/html 200 B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36A - - 1046 334 dupes.warc.gz
com,example)/ 20140127171251 http://example.com warc/revisit - B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36A - - 553 11875 dupes.warc.gz com,example)/ 20140127171251 http://example.com warc/revisit - B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36A - - 553 11875 dupes.warc.gz

5
sample_archive/cdx/bad.cdx Normal file
View File

@ -0,0 +1,5 @@
CDX N b a m s k r M S V g
com,example)/?example=2 20140703030321 http://example.com?example=2 text/html 200 B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36A - - 1987 0 non-existent.warc
com,example)/?example=2 20140603030351 http://example.com?example=2 warc/revisit - B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36B - - 504 2701 example-extra.warc
com,example)/?example=3 20140703030321 http://example.com?example=3 text/html 200 B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36A - - 1987 0 non-existent.warc
com,example)/?example=3 20140603030351 http://example.com?example=3 warc/revisit - B2LTWWPUOYAH7UIPQ7ZUPQ4VMBSVC36B - - 504 2701 example-extra.warc

17
tests/test_integration.py
View File

@ -240,6 +240,23 @@ class TestWb:
resp = self.testapp.get(uri, headers = [('Referer', referrer), ('Host', host)], status = 302) resp = self.testapp.get(uri, headers = [('Referer', referrer), ('Host', host)], status = 302)
assert resp.status_int == 302 assert resp.status_int == 302
def test_not_existant_warc_other_capture(self):
resp = self.testapp.get('/pywb/20140703030321mp_/http://example.com?example=2')
assert resp.status_int == 302
assert resp.headers['Location'].endswith('/pywb/20140603030341mp_/http://example.com?example=2')
def test_missing_revisit_other_capture(self):
resp = self.testapp.get('/pywb/20140603030351mp_/http://example.com?example=2')
assert resp.status_int == 302
assert resp.headers['Location'].endswith('/pywb/20140603030341mp_/http://example.com?example=2')
def test_not_existant_warc_no_other(self):
resp = self.testapp.get('/pywb/20140703030321mp_/http://example.com?example=3', status = 503)
assert resp.status_int == 503
def test_missing_revisit_no_other(self):
resp = self.testapp.get('/pywb/20140603030351mp_/http://example.com?example=3', status = 503)
assert resp.status_int == 503
def test_post_1(self): def test_post_1(self):
resp = self.testapp.post('/pywb/mp_/httpbin.org/post', {'foo': 'bar', 'test': 'abc'}) resp = self.testapp.post('/pywb/mp_/httpbin.org/post', {'foo': 'bar', 'test': 'abc'})