infection/Lobby

Maks Rafalko

@maks-rafalko

@computamike thanks for letting me know. Glad you found an issue !

Mike Hingley

@computamike

Hi - Question about false positives - so I have the following mutant output :

    ____      ____          __  _
   /  _/___  / __/__  _____/ /_(_)___  ____ 
   / // __ \/ /_/ _ \/ ___/ __/ / __ \/ __ \
 _/ // / / / __/  __/ /__/ /_/ / /_/ / / / /
/___/_/ /_/_/  \___/\___/\__/_/\____/_/ /_/

Running initial test suite...

Phpunit version: 6.2.4

   61 [============================] 3 mins

Generate mutants...

Processing source code files: 4/4
Creating mutated files and processes: 4/4
.: killed, M: escaped, S: uncovered, E: fatal error, T: timed out

SMMM                                                 (4 / 4)

4 mutations were generated:
       0 mutants were killed
       1 mutants were not covered by tests
       3 covered mutants were not detected
       0 errors were encountered
       0 time outs were encountered

Metrics:
         Mutation Score Indicator (MSI): 0%
         Mutation Code Coverage: 75%
         Covered Code MSI: 0%

Please note that some mutants will inevitably be harmless (i.e. false positives).

so I took one of the mutants that i found in my infection log :

3) /Users/mikehingley/Code/octopus/pacific/wordpress/web/app/plugins/pacific-octopus/pacific-octopus.php:131    [M] FalseValue
exec /usr/local/Cellar/php71/7.1.14_25/bin/php -c /private/var/folders/6h/jk4qf_555dd4561g67lw8z600000gp/T/infectionq5df5x /Users/mikehingley/Code/octopus/pacific/wordpress/vendor/phpunit/phpunit/phpunit --configuration /var/folders/6h/jk4qf_555dd4561g67lw8z600000gp/T/infection/phpunitConfiguration.882e7e50be1628ccb7cd7e543af8b36a.infection.xml --stop-on-failure 
--- Original
+++ New
@@ @@
             if (is_wp_error($response)) {
-                return false;
+                return true;
             }
             return $response->ID;
         }
     }
 }
 /**
PHPUnit 6.2.4 by Sebastian Bergmann and contributors.

Runtime:       PHP 7.1.14
Configuration: /private/var/folders/6h/jk4qf_555dd4561g67lw8z600000gp/T/infection/phpunitConfiguration.882e7e50be1628ccb7cd7e543af8b36a.infection.xml

......                                                              6 / 6 (100%)

Time: 1.63 seconds, Memory: 4.00MB

OK (6 tests, 6 assertions)

Which I think is a false positive (somehow) because when I made that mutation to my code, and ran the tests for that file it failed a test, as one of the assertions does check that the return value is false - so I'm confused by the test would pass. I'll investigate further.

Mike Hingley

@computamike

Ok - I have done more research - here is what I have found :

I cleared out the /private/var/folders//6h/jk4qf_555dd4561g67lw8z600000gp/T/infection folder - so that I could be sure that my results were accurate.
Re-ran the mutation - same result.
The infection folder was re-created - so that was good.

Found the infection log - and found the line that was run for the 3rd mutant - which I ran.

exec /usr/local/Cellar/php71/7.1.14_25/bin/php -c /private/var/folders/6h/jk4qf_555dd4561g67lw8z600000gp/T/infectionb4u3ig /Users/mikehingley/Code/octopus/pacific/wordpress/vendor/phpunit/phpunit/phpunit --configuration /var/folders/6h/jk4qf_555dd4561g67lw8z600000gp/T/infection/phpunitConfiguration.882e7e50be1628ccb7cd7e543af8b36a.infection.xml &

I ran it with & to allow it to run in the background, so that I wouldn’t have to restart a new terminal.
I also didn’t include the stop-on-failure option - I could but I didn’t copy the line properly.

Mike Hingley

@computamike

The interception doesn’t seem to be wokring - running the test and ading output to the original unit present the out put from that unit - therefore instead of running the mutant, this test is actually running the original unit. This is why the tests pass, because the original code is being called.

Mike Hingley

@computamike

I’m not sure - but i am currently working on the idea / premise that the issue is related to unit testing tags : * @runInSeparateProcess

 * @preserveGlobalState disabled

Mike Hingley

@computamike

There doesn’t seem to be any coverage around infection/src/StreamWrapper/IncludeInterceptor.php - I’ll write a test project to confirm my theory and try and write a patch… I’m not a php guy so wish me luck...

Maks Rafalko

@maks-rafalko

Thanks @computamike for the info. You can create a repo where tje issue is reproduced and create an issue on githib, we will look into it

Looks really weird

Mike Hingley

@computamike

Thanks Maks - I’ve started trying to create a really simple version of this - to see if I can create a minimum test example. - I’ll let you know when I have something to demo

Maks Rafalko

@maks-rafalko

:+1:

Mike Hingley

@computamike

Hi Maks - I have a git hub repo set up - but I’m struggling a little - I have some basic tests that in theory should allow true/false mutants to exist. I have 100% coverage, but should still have mutatnts here. Strangely infection doesn’t generate any mutants -

➜  infection-test git:(master) vendor/bin/infection --coverage=coverage
You are running Infection with xdebug enabled.
    ____      ____          __  _
   /  _/___  / __/__  _____/ /_(_)___  ____ 
   / // __ \/ /_/ _ \/ ___/ __/ / __ \/ __ \
 _/ // / / / __/  __/ /__/ /_/ / /_/ / / / /
/___/_/ /_/_/  \___/\___/\__/_/\____/_/ /_/

Running initial test suite...

PHPUnit version: 7.0.2

    7 [============================] < 1 sec

Generate mutants...

Processing source code files: 1/1
Creating mutated files and processes:    0/0
.: killed, M: escaped, S: uncovered, E: fatal error, T: timed out



0 mutations were generated:
       0 mutants were killed
       0 mutants were not covered by tests
       0 covered mutants were not detected
       0 errors were encountered
       0 time outs were encountered

Metrics:
         Mutation Score Indicator (MSI): 0%
         Mutation Code Coverage: 0%
         Covered Code MSI: 0%

Please note that some mutants will inevitably be harmless (i.e. false positives).

GitHub repo can be found here : https://github.com/computamike/infection-test

Maks Rafalko

@maks-rafalko

Thanj you @computamike , I will have a look

Gert de Pagter

@BackEndTea

Hi @computamike I'm not sure if anyone got back to you already, but due to auto-loading reasons, infection can't cover functions, only class methods.

Gert de Pagter

@BackEndTea

From what i saw in your code snippets, you are working with wordpress. I must admit that i haven't worked with wordpress before, but from what i have seen it works a lot with functions, rather than classes and methods, which may be where the issue lies.

Grummfy

@Grummfy

anyone on line?

Maks Rafalko

@maks-rafalko

yes

Grummfy

@Grummfy

hi, just to speak here about the atoum integration instead of all on the issue, I think it will be faster, if you ar eok

Maks Rafalko

@maks-rafalko

no problem

Grummfy

@Grummfy

regarding what I see now, I think it could interesting to perhaps refactor a bit (later) the way the framework are integrated. Perhaps to be seen more as plugins (if possible). Because a lot of code seems redundant . But perhaps I see optimization where there is not ;)

otherwise, regarding the implementation of my integration of atoum. is there any specific point that require attention. I see some comment about being sure it's isolated from the testing framework to avoid dependency if possible. but is there some other points to be carefull?

Grummfy

@Grummfy

one other question, if I see some refactoring that make integration easier (I speak about small one), do you prefer to have a pull request about it on the side instead of all together?

Maks Rafalko

@maks-rafalko

is there any specific point that require attention

you can take a look at the comments inside Codeception integration. Many of them are still not addressed. For example - there is no need to update public API of existing interfaces, adding $mutant parameter where it is not needed (https://github.com/infection/infection/pull/143#discussion_r165826810)

do you prefer to have a pull request about it on the side instead of all together?

I would prefer to have it separated to reduce complexity of the code review

Grummfy

@Grummfy

oki, thanks.
I read the pr about codeception a lot are common and still valid, so it was helpfull ;

Maks Rafalko

@maks-rafalko

but is there some other points to be carefull?

I have no other points at the moment. Just make sure to generate needed coverage reports to make it fast

btw, the Slack channel is much more popular for Infection

Grummfy

@Grummfy

regarding the header of file with the license, Did I put the same as the rest of the app or do you prefer I put my own like in the codeception PR

I miss the slack channel ;)

Maks Rafalko

@maks-rafalko

https://symfony-devs.slack.com/messages/C78M7A4BY

regarding license - we have a php-cs-fixer rule for it

just run make cs-fix

Grummfy

@Grummfy

yep, cs rule and other stuff like that I was thinking to run it after ;)

regarding slack, is there any autoinvitation stuff or did I need to send you my email address?

Maks Rafalko

@maks-rafalko

also, please read https://github.com/infection/infection/blob/master/.github/CONTRIBUTING.md just in case

Grummfy

@Grummfy

thanks, well I miss nothing ;)

Maks Rafalko

@maks-rafalko

regarding slack, is there any autoinvitation stuff or did I need to send you my email address?

https://symfony.com/slack-invite - our channel is inside symfony-dev account

or whatever it's named ;)

Grummfy

@Grummfy

thanks

Maks Rafalko

@maks-rafalko

you are welcome

Stanislav Krasnoyarov

@alvin777

Hi there! Does infection still claim php 7.0 support despite using Symphony 4 which requires PHP 7.1.3+ to run?

Owen Voke

@pxgamer

Looks like it's using Symfony ^3.2 or ^4.0 so it should support PHP 7.0

Gert de Pagter

@BackEndTea

Hey @alvin777 Infection uses symfony 3.2, but supports 4.0 as well. If your project requires php 7.0 you could consider adding

"config": {
  "platform": {
    "php": "7.0.8"
  }
}

This to your composer.json, which forces dependencies to be compatible with that php version

Stanislav Krasnoyarov

@alvin777

Thanks, that seemed to help!

Edvin Malinovskis

@nCrazed

Is it possible to disable xsd check for phpunit? the remote xsd storage appears to be broken again with no ETA on a fix which means that we have to either wait indefinitely or disable mutation tests in our CI.

Maks Rafalko

@maks-rafalko

@nCrazed you can always use the local version from vendor folder: see https://github.com/infection/infection/blob/b969a0e1e7c7c488f98e29fbaf64f05ca9d6e25b/tests/TestFramework/PhpUnit/Config/XmlConfigurationHelperTest.php#L306

<phpunit xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:noNamespaceSchemaLocation="./vendor/phpunit/phpunit/phpunit.xsd"

Edvin Malinovskis

@nCrazed

@borNfreee perfect, thanks

Joe Watkins

@krakjoe

I hear you have a problem with phpdbg ... tell me all about it ...

Bernhard Breytenbach

@Xethron

haha. I just came here to ask about phpdbg :/

I can't seem to set the memory limit. I believe infection spins off another thread for the phpunit tests? Is it possible that the parameters aren't carried across?

I tried the following command which didn't work: phpdbg -qrr -dmemory_limit=1G vendor/bin/infection --min-msi=95 --min-covered-msi=98 --threads=4

Where communities thrive

People

Repo info

Activity