cisco-sas / kitty / 232 / 1
91%
master: 90%

LAST BUILD BRANCH: patch-3
DEFAULT BRANCH: master
Ran
Files 47
Run time 3s
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 86.744% (+0.09%) from 86.658%
232.1

push

travis-ci

BinyaminSharet 
[Fuzzer] New test listing options

This commits changes the way the fuzzer decides which test should be
executed. Instead of just holding start/end indexes and letting the
model provide sequential mutations, it now uses a separate "Test list"
object to handle the next test that should run.

Currently, there are two classes that might decide about the next test:
1. StartEndList - using the legacy strategy of just running tests based
on start/end index. It will be removed eventually and unified with the
second class, RangesList.
2. RangesList - this list can manage sorted non-sequential tests. For
example, you can run a single test by passing its number as a string to
the constructor ('182'), run sequential tests ('200-210'), run all tests
up to a specific test ('-301'), run all tests from a specific tests
('1700-') or combine all those options ('-10,13,15,17-19,21-').

Two main changes were required (except of the obvious ones):
1. The current index refers to the current index in the test list and
not the mutation index. When a session is complete, the current index is
set to None. While it introduced many issues at first, it actually made
the code clearer, providing a distinct way to signal the end of the
session.
2. In order to give the test list control over what test is executed,
the caching mechanism that was implemented in the GraphModel was
removed.

This commit partially closes issue #26 - "Add kitty-option: validate -
to rerun all the failed/error tests".
While there is no option to just re-run the tests, one can now use the
new fuzzer API: `set_test_list` or the command line argument '-t
--test-list' to run a specific list of tests.

3344 of 3855 relevant lines covered (86.74%)

0.87 hits per line

Source Files on job 232.1