Errorprone checks

[Eric Stevenson]

Planning to add some more Errorprone checks:

• Prefer AppCompatResources.getDrawable() over VectorDrawableCompat.create() (from https://groups.google.com/a/chromium.org/d/topic/chromium-dev/Z-g3RytRCU8/discussion)
• Prefer android.support.annotation.Nullable (over javax.annotation.Nullable and android.annotation.Nullable - from crbug.com/771683)
• Don't add redundant exception info (https://chromium.googlesource.com/chromium/src/+/master/styleguide/java/java.md#Exceptions)

Any objections? Are there other Errorprone checks that folks think would be useful?

Thanks,

Eric

[Nate Fischer]

Excited to enforce the correct @Nullable import. Can you do the same for @NonNull imports (or, ban them outright if we want to assume @NonNull by default)?

As an extra bonus, it would be great if our compiler could enforce @NonNull's semantics. I've heard of 3rd party plugins (uber/NullAway)--not sure if we're planning to use this, however.

Nate Fischer | Software Engineer | ntf...@google.com

--
You received this message because you are subscribed to the Google Groups "java" group.
To unsubscribe from this group and stop receiving emails from it, send an email to java+uns...@chromium.org.
To post to this group, send email to ja...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/java/CACO9cobFtP8CzGB%2BYdA9Wu7pog971GebWxKKy1h0F-e1MT2brg%40mail.gmail.com.

[Tommy Nyquist]

Adding the checks you mentioned for @Nullable and exceptions SGTM.

Regarding the Drawable check, any thoughts on this, twellington@ or huayinz@?

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/java/CACiwn8aEoSQyh%3DF8Eq9mzBm1sM8LutiyTx5FWWXgy17MzYTVgg%40mail.gmail.com.

[Boris Sazonov]

Thanks for doing that, Eric! We definitely need more checks for our Java codebase.

+1 for AppCompatResources.getDrawable check. Related internal discussion. By the way, is it much harder to make it a custom Lint check? Ideally, this check should be one of Support Library Lint checks.

For Chrome-specific checks, it would make sense to check that Chrome widget classes don't directly extend ImageView/ImageButton/AppCompatImageView/AppCompatImageButton, subclassing ChromeImageView/ChromeImageButton instead (see https://crbug.com/894770).

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/java/CAA6XDjMcd_bT3BkjA_5pYk2ayJAhsHjWRhbVgd0Etyv_-3Uozg%40mail.gmail.com.

[Peter Wen]

We currently have quite a few custom errorprone checks, but as far as I am aware, no custom lint checks.

Unfortunately it seems like the custom lint API is not stable, so custom errorprone checks are preferred.

Thanks,

Peter

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/java/CAJJcB998pM%3DsN%2BkrP%3DNSHtLB%2BBSWoNDnE1H17xKp49hSoPbNqw%40mail.gmail.com.