Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Jan 31 22:31
    sharwell unlabeled #2391
  • Jan 31 22:31
    sharwell labeled #2391
  • Jan 31 22:31
    sharwell unlabeled #2865
  • Jan 31 22:31
    sharwell unlabeled #1934
  • Jan 31 22:31
    sharwell labeled #1934
  • Jan 31 22:31
    sharwell unlabeled #2864
  • Jan 31 22:31
    sharwell unlabeled #1923
  • Jan 31 22:31
    sharwell labeled #1923
  • Jan 31 22:30
    sharwell unlabeled #2867
  • Jan 31 22:30

    sharwell on master

    Permanently disable SA1644 (Doc… Merge pull request #2866 from s… (compare)

  • Jan 31 22:30
    sharwell closed #2866
  • Jan 31 22:30
    sharwell closed #164
  • Jan 31 22:30
    sharwell assigned #2866
  • Jan 31 22:29
    sharwell unlabeled #2866
  • Jan 31 22:29
    sharwell unlabeled #2866
  • Jan 31 22:29
    sharwell edited #2866
  • Jan 31 22:29
    sharwell unlabeled #164
  • Jan 31 22:29
    sharwell assigned #164
  • Jan 31 22:29
    sharwell labeled #164
  • Jan 31 22:29
    sharwell labeled #164
Glenn
@glennawatson
using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;

using Microsoft.Build.Framework;
using Microsoft.Build.Utilities;

using NuGet.Packaging.Core;
using NuGet.Versioning;

using Pharmacist.Core;
using Pharmacist.Core.NuGet;

using Splat;
I seen that in the original stylecop examples
Dirk Lemstra
@dlemstra
That is probably why this was not added here. But in most places StyleCopAnalyzer is pretty strict about adding extra newlines. And for me it also felt that it should also check the using block. And that made me wonder what the process is of getting a new rule added. Where is decided what should and shouldn't be added to StyleCopAnalyzer?
Glenn
@glennawatson
I'm not really the one who makes that decision, just pointing out existing uses in the wild for new lines in wild
Sam Harwell
@sharwell
You can propose a new rule on the repostiory
in this case it seems likely that it would just extend the configuration available for spacing between groups
Simon Novak
@snovak7
@sharwell It’s me again :) Anyway it now reads the file, thank you for pointing that out! what do I do with this one ? CSC : warning SA0001: XML comment analysis is disabled due to project configuration
that page contains guidance about the warning
Simon Novak
@snovak7
You’re a hero! :)
It works, but now I get bunch of other warnings
but those are bearable, I mean fixable :)
weird, SA1629, then I build one more time, and doesn’t complain anymore
And then when I clean and rebuild warnings popup
Sam Harwell
@sharwell
the second time the project was up to date so the compilre was skipped. when the compiler is skipped, it doesn't report warnings.
Simon Novak
@snovak7
I see, yes it makes sense
Joseph Musser
@jnm2
Sam Harwell
@sharwell
Yeah, I lean against 37382
Still WIP
Johan Larsson
@JohanLarsson
Sam did you see that we created an azure devops org?
Sam Harwell
@sharwell
Yes
Johan Larsson
@JohanLarsson
We will hand over ownership to you of course.
Sam Harwell
@sharwell
We can't use it for StyleCop Analyzers or Documentation Analyzers yet
They don't support our status page
Johan Larsson
@JohanLarsson
devops is nicer than appveyor as it has a concept of org
Sam Harwell
@sharwell
AppVeyor isn't perfect, but of the issues I have with it, ADO doesn't fix very many of them and adds its own
It's not a problem with ADO per se, but more a side effect of the fact that I've been using AppVeyor for probably 5 years now and it works very well for my workflow
Johan Larsson
@JohanLarsson
AppVeyor is really nice for many things, nice simple UX. And the notifications defaults are perfect.
We can use both for a period, not the most environmentally friendly perhaps.
Sam Harwell
@sharwell
There is no ETA for ADO supporting the current status page strategy so someone will have to figure out an alternative
AppVeyor created the feature at least partially for this project at my request
Johan Larsson
@JohanLarsson
:thumbsup:
Not nice to replace them then.
Sam Harwell
@sharwell
I'll still be paying them the same either way so not sure it matters lol
That's the magic line
Johan Larsson
@JohanLarsson
do you pay for open source?
Sam Harwell
@sharwell
The status page relies on a CORS-friendly permalink to a named build artifact of the most recent successful master branch build. It downloads that JSON file, and then uses client-side JavaScript to render the entire page.
Johan Larsson
@JohanLarsson
fancy
Sam Harwell
@sharwell
Yeah, it means the page is always up-to-date as soon as the build completes on merge, and we don't need any extra build infrastructure or hosting to make it happen.
The same concept could be used to automate the creation and update of analyzer rule documentation pages as well, but I haven't done that (we still maintain per-rule Markdown files).
Johan Larsson
@JohanLarsson
Markdown files in repository have the benefit of history.
I have tests in the other repos that checks that .md are in sync with the code. Not pretty but works ok.
Joseph Musser
@jnm2

The status page relies on a CORS-friendly permalink to a named build artifact of the most recent successful master branch build.

For the two projects that rely on free build artifact hosting, AppVeyor makes perfect sense. It's easy to add an AppVeyor build if other projects start doing that.

Sam Harwell
@sharwell

do you pay for open source?

I pay a little bit of money; more time. It's no problem, people give what they can and take what they need.

Glenn
@glennawatson
in terms of https://github.com/DotNetAnalyzers/StyleCopAnalyzers/blob/master/documentation/SA1202.md where would 'private protected' fit in the ordering?
once you implement
above private?
just asking since I'm working on my own API generator of sorts and want it to comply as much as possible with stylecop ordering rules
Sam Harwell
@sharwell
yeah above private seems right