Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    Jake Ginnivan
    @JakeGinnivan
    @/all have added shouldly/shouldly#308 - would appreciate any help anyone can give
    Suraj Gupta
    @SurajGupta
    @JakeGinnivan - I simply did the diff between FluentAssertions and Shoudly. On one hand you can say that the utility of each was justified when they were added to FluentAssertions. But ultimately it's up to the community to decide. For example, I think "has method" is probably not that useful. Maybe a bunch of reflection based assertions fall into this camp. In contrast, I think ShouldBePositive and ShouldBeNegative are super expressive and have high utility. Why don't you go ahead and just close out all the proposals that you can't justify and I'll do a sweep of those and comment on any that I'm particularly passionate about.
    Suraj Gupta
    @SurajGupta
    @JakeGinnivan - in fact, so far I only disagree with a small set of bugs that you've closed. please see comments in those bugs. My bar is readability.
    Jake Ginnivan
    @JakeGinnivan
    @SurajGupta I just did an initial close of ones I initially don't think add that much value, more than happy to reopen after a bit more discussion. Adding additional methods crowds the API so we need to think through every extension we add
    Jake Ginnivan
    @JakeGinnivan
    @SurajGupta other issues worth reading: #286, #284 and #290 - would love more in depth breakdown of them all including error messages and examples of usage (just for the ones which are tagged for discussion, the jump in ones make sense)
    #289 rather
    Suraj Gupta
    @SurajGupta
    @JakeGinnivan - totally agreed, need to keep bar high. I think I've posted to all bugs that required input (and also see that @ceregador responded on a few). Let me know!
    Gert Nelissen
    @gertnelissen
    Hi, I am using version 2.6.0 (nugetted today) and notice that ShouldSatisfyAllConditions doesn't process all Actions. It shows only the first encountered failure. The output is "the following errors were found: ..." but it will only show "Error 1". The docs say that ShouldSatisfyAllConditions should process all Actions and report them all at once.
    Oded Welgreen
    @odedw
    Hey! I'd like to contribute, should I just pick a 'Jump in' issue?
    Joseph Woodward
    @JosephWoodward
    That's great hear! Yes, pick a Jump In issue and away you go! If you have any questions then feel free to add a comment to the issue and we'll be able to help.
    Oded Welgreen
    @odedw
    Ok great thanks!
    Jake Ginnivan
    @JakeGinnivan
    @/all if anyone pulled in the last 24 hours, I just rewrote history on master (I just noticed one of my PR's included a 50mb TestResults MDF :().
    This message was deleted
    Jake Ginnivan
    @JakeGinnivan
    @/all 2.7.0-beta.1 released. Enjoy all. Please test :)
    Dave Newman
    @whatupdave
    wow, 6 years on this thing's still going! :D
    Matthias Koch
    @matkoch
    Hey
    what do you do to generate a text like "contestant.Points should be 1337 but was 0" from "contestant.Points.ShouldBe(1337)" ?
    I was thinking of rewriting, but I can't see anything in the nuspec that gives a hint about that
    Jake Ginnivan
    @JakeGinnivan
    You shouldn't need to do anything
    just make sure you are compiled as debug, and the source is still available
    Matthias Koch
    @matkoch
    ehmm... that's not even close an answer to my question :D
    Jake Ginnivan
    @JakeGinnivan
    I don't understand the question then?
    @matkoch did you install Shouldly through nuget?
    ShouldBe is an extension method, so you need to import the shouldly namespace
    then it is just an assertion within your test method
    Phil Scott
    @enkafan
    @matkoch are you asking how shouldly itself knows how to generate the text?
    Matthias Koch
    @matkoch
    I was asking for the specific technique, how you generate a text that equals to the actual statement.
    Jake Ginnivan
    @JakeGinnivan
    Ahh, sorry. I thought it was about usage of the library
    We use the stack trace to get the line number of the caller
    Matthias Koch
    @matkoch
    When I spoke about rewriting, I was relating to IL rewriting, which seemed like the most reliable way.
    Yes I got it already =)
    Jake Ginnivan
    @JakeGinnivan
    Ok cool
    Matthias Koch
    @matkoch
    well, performance is not so important for test projects
    but you know .... :)
    Jake Ginnivan
    @JakeGinnivan
    Yeah, shouldly has the opinion better error messages are more useful than saving a few hundred ms total
    We use another technique internally which would be quicker
    for Should.Throw() - we support expressions, then we format the expression tree
    Matthias Koch
    @matkoch
    I guess you guys are interested in testing, so I would like to tell you about my test framework, if you don't mind :) It's orthogonal to yours indeed. Could be combined.
    Jake Ginnivan
    @JakeGinnivan
    Sure
    Matthias Koch
    @matkoch
    Jake Ginnivan
    @JakeGinnivan
    Have not seen it
    nice
    Matthias Koch
    @matkoch
    I'm not putting so much effort into promotion unfortunately...
    But I'm actively developing it for 2 years now
    almost three :))
    Jake Ginnivan
    @JakeGinnivan
    Nice, I know how that feels. I have been building GitVersion for about that same time too
    Have you seen BDDfy? Also another project based on it https://github.com/mwhelan/Specify - Looks like we share similar concepts, except yours is also a test running whereas BDDfy/specify fit within existing test libraries
    https://github.com/JakeGinnivan/ExpressionToString - is the expression tree formatter i talked about
    Matthias Koch
    @matkoch
    ahh... i don't like this language noise :DD
    no test discovery in visual studio or resharper?