/
RUBY-TESTING
65 lines (58 loc) · 3.11 KB
/
RUBY-TESTING
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
=== Testing rFeedParser ===
Simply run `ruby tests/rfeedparsertest.rb` to run all of the FeedParser tests.
Optionally, you can start up rfeedparserserver.rb and, in another shell, run rfeedparser.rb
against "http://localhost:8097/tests/path/to/testcase.xml" if you want
to try a test individually. You can, of course, run rfeedparser.rb with
the file path to the testcase, but there may be HTTP headers required for
a successful run that can only be given by running rfeedparser against
the HTTP server path. I'll probably merge feedparserserver.rb into
rfeedparsertest.rb soon. Note that there are many test failures that occur
solely because of problems with a platforms iconv installation. Inconsistency
in iconv is the leading cause of errors in rfp's test cases.
=== Last Count 20070321 ===
By my last count, rfeedparsertest.rb says that there are 45 assertions
that fail, and 4 that error out. I've included here a few tests that
"Failed, Sort Of". By that I mean, the behaviors the tests are meant to
check are correct, but the test fails because of some other superficial
or unrelated behavior.
=== Tests Failed, Sort Of ===
Problem:
Hpricot adds end tags when it sees an unclosed tag. This means that
certain tests that rely on feedparser.py's _HTMLSanitizer not closing
tags will fail. Many of the tests affected (actually, all the ones
affected, AFAICT) would otherwise passed.
Tests Affected:
* tests/wellformed/rss/item_description_not_a_doctype.xml (extraneous trailing </a>)
* tests/illformed/rss/item_description_not_a_doctype.xml (ditto)
==
Problem:
The Hpricot#scrub method I've written does not remove the dangerous
markup in the same way feedparser.py does, but the output is still safe.
Tests Affected:
* tests/wellformed/sanitize/entry_content_crazy.xml
* tests/wellformed/sanitize/entry_summary_crazy.xml
* tests/wellformed/sanitize/entry_title_crazy.xml
* tests/wellformed/sanitize/feed_copyright_crazy.xml
* tests/wellformed/sanitize/feed_info_crazy.xml
* tests/wellformed/sanitize/feed_subtitle_crazy.xml
* tests/wellformed/sanitize/feed_tagline_crazy.xml
* tests/wellformed/sanitize/feed_title_crazy.xml
* tests/wellformed/sanitize/item_content_encoded_crazy.xml
* tests/wellformed/sanitize/item_description_crazy.xml
* tests/wellformed/sanitize/item_fullitem_crazy.xml
* tests/illformed/sanitize/entry_content_crazy.xml
* tests/illformed/sanitize/entry_summary_crazy.xml
* tests/illformed/sanitize/entry_title_crazy.xml
* tests/illformed/sanitize/feed_copyright_crazy.xml
* tests/illformed/sanitize/feed_info_crazy.xml
* tests/illformed/sanitize/feed_subtitle_crazy.xml
* tests/illformed/sanitize/feed_tagline_crazy.xml
* tests/illformed/sanitize/feed_title_crazy.xml
* tests/illformed/sanitize/item_content_encoded_crazy.xml
* tests/illformed/sanitize/item_description_crazy.xml
* tests/illformed/sanitize/item_fullitem_crazy.xml
==
Problem:
My current system lacks a few encodings that rfeedparser and Iconv need.
This results in failures that will probably not occur on other machines.
Tests Affected: