Hey, is it possible to exclude a rule based on the receiver Name/type?

Hi folks, after updating to 1.20.0, I get this error when building:

Execution failed for task ':app:detekt'.
> Run failed with 1 invalid config property.
  	- Property 'formatting' is misspelled or does not exist.

If I take out the

formatting

section in the config (which is only setting

active: false)

, it works fine. Looking at the 1.20.0 release notes, there's a mention about something changing in how the formatting rules are configured, but there is no example, migration guide, nothing — only a link to a couple PRs with no documentation. I don't use the

formatting

plugin, so am I ok with just removing the corresponding section from the config? And how would it work if I actually wanted to use the plugin? I'm not super clear on this... do I need a separate config file? And if so, how do I associate it with the plugin?

I’m trying to run both the compiler plugin and the main gradle plugin at the same time. When I do this, the main gradle plugin cannot find it’s

versions.properties

in resources anymore, and I’m struggling to figure out why. Any ideas that could get me unstuck?

Happy to contribute to the compiler plugin. I think one of the important features is autocorrect. But I can’t imagine changing the source files during compile is allowed (or even a good idea?). Currently my solution is to load both plugins and rename the gradle extensions in the compiler plugin so they do not conflict. Then run the detekt task from

io.gitlab.arturbosch.detekt

with auto correct before the build. Should be fine because I don’t think any type resolution rules have auto correct? I have a very rough POC doing this now. Any feedback on this approach?

Is it possible to configure detekt to allow characters outside of the

a-z

range in e.g. function names? I see that `parameterPattern: '[a-z][A-Za-z0-9]*'`is used, but I would like to use the Norwegian characters

æ

,

ø

and

å

in function names, too. Any help would be highly appreciated!

fun vilkårsØnske() {}

how often is autocorrect implemented? I was surprised to see that

VarCouldBeVal

didn’t implement it for what seems like a trivial-ish change. In some quick testing, I don’t really see any other autocorrects kicking in when updating from 1.18 to 1.20

hi everyone, is there a feature in Detekt like importing custom rule from community to our project? for example, I want to use all rules that widely use by Airbnb or Stream instead of the default one on my project. one of my friend told me that this feature is available in ESLInt, and its quite interesting actually to have it on Detekt too when digging to this, I come to conclusion that replacing the

config.yml

could be a way to do it, cmiiw thanks and have a great day

Is there any reason to expect

detektBaselineMain

and

detektBaseline

to return different results? We're seeing that on our codebase. Specifically, in the sense that rules we've turned off are actually turned off in

detektBaseline

but do not seem to be turned off in

detektBaselineMain

Hi all! I’ve a doubt on using detekt (:android: ). I’ve a rule that ?ve added a test for it, when running the test the rule complains sucesfully, but later on running

./gradlew deteket

on the real app code doesn’t throw any smell… code in 🧵

We have released a set of Detekt rules which help us find common errors while working on Compose code. https://github.com/appKODE/detekt-rules-compose

hm, would it make sense to have a rule for defining exception as object? I think stacktrace would not be correct, if I do following:

object TitanicSinking : RuntimeException()

// somewhere else
throw TitanicSinking

Detekt Ij plugin seems to be unable to support mpp projects. Is this a known issue or just me?

Hi, How can I include checking of xml files in detekt? I would like to apply to some rules on them. Can someone help me with this please?

Hello guys! I think I might have found a bug in Detekt. https://github.com/LeoColman/Petals/commit/7c16d82a999f3f65b30f2c19bd411f6d3db1c8de#diff-6bbb3aa0c72ccab5cb844fd4b[…]8ab4c6110c06add1d36615718f64R62 In this commit we implemented an workaround to the following issue:

AndroidView({ LineChart(it) }, Modifier) {
    val a = "abc"
    val b = "def
}

This codes makes detekt fail and autocorrect it to

AndroidView({ LineChart(it) }, Modifier) {
val a = "abc"
val b = "def
}

A workaround for this case is

AndroidView(::LineChart, Modifier) {
    val a = "abc"
}

Which is considered correct by

./gradlew detekt

Is this a known/intended behaviour? Should I open an issue on this?

Hello all, How can i set the *detekt.yam*l path, using detekt.configure {} ?

Hello there, we are running detekt as a plugin in our gradle-build, and also as a GitHub Action. Both are using the same configuration file. At least we hope they do 😅 Nonetheless, the local build fails with

DataSource.kt:23:1: Unexpected indentation (20) (should be 16) [Indentation]

Whereas the GHA-variant runs ok. Any suggestions to why? We are using the 1.20.0 version of the gradle plugin, and also the 1.20.0 version of the GHA. Any help would be highly appreciated 🙂

Hi there! After changing some code in

detekt

repo, i can’t see the new result in terminals. Is there something missing that i need to do before i see the new result? Thanks!

does anybody have a working prototype/example with a rule module and a consumer project module?

I’ve got an empty override method that detekt 1.20.0 is complaining about (

EmptyFunctionBlock

), but I found this bug which was fixed, which should allow overridden methods to have empty blocks. It was a while ago, though, has something changed since this?

Should the intellij plugin “sync” with the ruleset file in the project? I’ve tweaked some rules, but still see warnings in intellij. Closing and reopening the project didn’t fix it either.

Hey, is it possible to ignore TrailingComma for unit tests? As it adds it to all Spek tests because of ({})

I recently reinstalled Android Studio version

2021.1.1 Patch 2

which then installed the latest version of the detekt plugin (

1.20.1

). I'm using version

1.19.0

of detekt. After that I'm getting several false positives for

Missing newline after "("

and

Missing newline before ")"

warnings. They all report as being on at line 1 column 1. Running detekt from gradle does not report any issues. Not sure the plugin is incompatible with the version of Android Studio being used, or if I need to upgrade to a new version of detekt.

Would there be any interest in an update to the XML report merger to filter out duplicate smell entries? We merge our detekt reports from multiple android flavors, and it creates duplicate smells in the merged report. Those duplicate smells are sent to our Sonar. Good value add for everyone or too specialized / not complexity we want in the detekt codebase?

Has anyone run into any issues with Detekt 1.21.0-RC2 and Kotlin 1.7 when using type resolution - https://github.com/detekt/detekt/issues/5021

I’m trying to integrate

detekt

in my project. In the top level project’s

build.gradle.kts

file I have:

apply(plugin = "io.gitlab.arturbosch.detekt")

buildscript {
    repositories.applyDefault()
    dependencies {
        (...)
        classpath("io.gitlab.arturbosch.detekt:detekt-gradle-plugin:1.20.0")
    }
}

it synchronizes just fine, but when I try to configure it in the

detekt

block I get

Unresolved reference: detekt

error. Is there anything I’m missing? I had very similar integration in another project and it worked as expected. The only difference was that I was using groovy instead of kotlin in gradle files there.

Hi there! I can’t find any jetbrain’s documentation about their API’s that’ll help me understand the project more. Is there any official or not official documentation, blog or talk anything helpful? Thanks in advance!

Hey guys! Do you usually display Detekt results in the README some way? I was thinking that it would be an interesting shield to have in projects, a code-quality based on detekt findings. I couldn't find anything that could help me with this in

<http://shields.io|shields.io>

. Do you have any ideas?

After enabling Markdown, I have seen that it is being generated in a subfolder, in which html is included too, is it correct? Previously in detekt 1.20 I was getting the html at same level than the sarif and xml files.

Would detekt ever expose something like registerDetektTask ? It would save a lot of headache when registering custom detekt tasks for things like intermediate KMP source sets.