Support editing existing comments instead of overwriting

[deecewan]

I think it'd be nice (maybe it happens on danger.rb?) for problems that are resolved to be struck out? It makes it hard to track progress and reasons for certain commits.

In this, I'd also like to see some logic to check if the comment if actually a danger comment before overwriting. We use the same bot on a bunch of repos, and it comments about various things in various places. It took me a while to realise that it was still working, but danger was overwriting the comment instead of creating a new one.

[orta]

We do support strikeouts in Danger.rb - I'm hesitant to add it in, as it adds a lot of complexity but ended up rarely used when I asked people if they used it.

In rb there is a --danger_id flag that you can pass, and then it would only update/edit/delete the comment with the same ID. This allowed for multiple Danger's to comment on the same thread. Would that fix it?

[deecewan]

yeah, as i was writing the strikeout thing, I don't really think it adds much to be honest.

Editing comments would be 100% great, although I'm not sure how I'd know which comment to flag ahead of time?

[orta]

Danger JS is already editing comments, and deleting them as they become empty.

Markdown allows for arbitrary text to be inserted inside HTML elements ;)

So I usually put data like that inside the table element

[deecewan]

i guess the thing is how does it know which comment to edit?

I had a case where the bot would comment with the IP of a server, and then danger would run and overwrite the server IP. After that it just edited it's own comment.

[orta]

So the danger-id thing is used for when you have danger being ran multiple times on the same commit ( maybe from different build machines etc )

in RB, the flag --dangerID=123 would mean that inside the markdown there would be a data-danger-id="123" then later as Danger is looking for the right comment to edit, it checks that there is a corresponding dangerID and the one that was ran during the argument. If it doesn't find one, then it creates a new comment instead

[deecewan]

right. I think what i'm proposing is more 'only edit a comment that Danger has created, otherwise add a new one'. Maybe we can use a flag for that to retain existing behavior if people want it.

[orta]

This feels like the current behavior to me though? can you write out a flow of what would happen after these changes

…

On Mar 15, 2017 10:55 AM, "David Buchan-Swanson" ***@***.***> wrote: right. I think what i'm proposing is more 'only edit a comment that Danger has created, otherwise add a new one'. Maybe we can use a flag for that to retain existing behavior if people want it. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#166 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAC_jhJ-e_lHFoa0nZQKUXNB6jg8KSmfks5rl0UDgaJpZM4Mcd9x> .

[orta]

Related #196

[deecewan]

yes! this is pretty much exactly what I was meaning, i think.

[pranshuchittora]

Hi, is there any workaround/hotfix for the same. This feature is really required in our monorepo setup with multiple danger files.
Thank you : )