1
65%
master: 65%

Ran 21 Apr 2018 04:01AM UTC

Files 29

Run time 2s

Badge

Embed ▾

Committed 21 Apr 2018 03:57AM UTC coverage: 60.994% (+0.05%) from 60.947%

Job # 772.1

Build Type

push

travis-ci

Committed by

mfateev

Commit Message

Un-pin zap, bump to a newer version to pull in zaptest logger (#438)

* Un-pin zap, bump to a newer version to pull in zaptest logger

Mostly motivated by the addition of a testing.T.Logf-compatible logger
provided by zap, which allows quite a lot cleaner test-logs, especially
when they are run in parallel.

But generally: there's no reason for zap to be pinned to a *specific*
version like it was. Ranges are far nicer for users of libraries.

* Remove testify.Suite from backoff tests

While trying to change these >= / < checks to something like
`assert.InDelta`, and generally trying to trigger edge cases, I
found that these tests aren't really sufficient:

```
cadence$ go test -race -count 30 ./internal/common/backoff/
--- FAIL: TestDefaultPublishRetryPolicy (0.00s)
        Error Trace:    retrypolicy_test.go:156
	Error:      	Should be true
	Test:       	TestDefaultPublishRetryPolicy
	Messages:   	NextBackoff too low: actual: 1.035903185s, expected: 1.3s
FAIL
FAIL	go.uber.org/cadence/internal/common/backoff	0.684s
```

If you run the tests enough, you eventually get values that
exceed the 0.8x limit implied by the tests.  I haven't sat down
and run through the math to figure out real bounds, but they're
currently misleading, it'd be good to get that fixed.

Run Details

5129 of 8409 relevant lines covered (60.99%)

94.95 hits per line

uber-go / cadence-client / 772 / 1
65%
master: 65%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 772.1

uber-go / cadence-client / 772 / 1 65% master: 65%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 772.1

uber-go / cadence-client / 772 / 1
65%
master: 65%

README BADGES
x