-
Notifications
You must be signed in to change notification settings - Fork 61
Conversation
ready to start message in /start response
fix config and warning
uses default json first unless 0 then uses 30d
remove console log and adjust warning
Config validation error
hotfix: issue creator multiplier should be optional
* feat: extract time duration with ms * feat: use regex
feat: incentive for PullRequest reviewers
* fix: use bot penalty uses bot penalty created at * fix: defaults set to false * fix: use reopen time vs using penalty comment * fix: future living implemented futures fix
* Fix: renamed default config file * Fix: resolved changes * chore: update ubiquibot-config-default.ts Co-authored-by: アレクサンダー.eth <[email protected]> * Update ubiquibot-config-default.ts as const / as read only isn't working * Fix: added default config to src --------- Co-authored-by: 0xCodercrane <[email protected]> Co-authored-by: アレクサンダー.eth <[email protected]>
* feat(action): parsing the permitUrl to an URL object * refactor(action) : using permitBaseUrl from config * feat: use regex * fix(action): - removing unused import statement --------- Co-authored-by: 0xcodercrane <[email protected]>
* fix: close pr issues and stop running handlers twice * chore: remove unused * chore: remove unused * chore: check org and repo name * chore: refactor gitLinkedPrParser * chore: make pr parser generic * Update src/helpers/parser.ts Co-authored-by: whilefoo <[email protected]> * Update src/helpers/parser.ts Co-authored-by: whilefoo <[email protected]> * Update src/helpers/parser.ts Co-authored-by: whilefoo <[email protected]> * Update src/helpers/parser.ts Co-authored-by: whilefoo <[email protected]> * chore: whilefoo review refactor --------- Co-authored-by: whilefoo <[email protected]>
* feat: add repo specific wallet support * feat: repo specific wallet support * feat: repo specific wallet support * feat: add repo specific wallet support --------- Co-authored-by: whilefoo <[email protected]>
* chore: move incentives to main handler * chore: setting up creator reward * chore: setting up commenter reward * chore: remove type string * fix: syntax errors * fix: make sure assignee cannot get comment rewards * fix: undefined for payout command * fix: add types to fix lodash error * fix: lint error on const * chore: refactor reward generation * fix: remove unused * chore: edit payout function * chore: edit payout function * chore: fix error from latest merge * chore: remove assetion and make camelcase * chore: remove all assertions * chore: fix reviews * chore: merge commenter and reviewer * chore: make interface for rewards user * chore: fix regexp error * fix: new merge issues
* feat(deps): added openai added openAI * feat: added ask command Simple prompt engineering, one response per user per question. * fix: update prompt better initial prompt * fix: update system prompt update prompt * feat: simple memory original context indexable across repos, issues and pull requests * fix: comments comments etc * fix: streamlined context waiting for more input * chore: pulling context pulls from sources using the hashtag in issue body * chore: tokenization total token usage now displayed * chore: diff comparison feed gpt the spec and diff for it to analyze * fix: ci and spec check will only spec check once now fixed tsc issues * fix: merge mistake sort import * fix: key from config allow it to be set in org and repo config * fix: set ask to false reset default * fix(deps): langchain unused dep * fix: apikey undefined no longer undefined * fix: using token limit set to 8k in config 4k by default * chore: improve ask command improved pr-review and context for both ask and pr-review commands * fix: gpt helpers moved base ask and context call into helpers * chore: use error diff replace text error with diff * fix: remove unused requested changes * fix: remove unused missed the shared default * fix: schema new types * fix: nullable and error start with a bang and clear api key error * fix: duplicated --------- Co-authored-by: 0xcodercrane <[email protected]>
* feat(camelCase): changing config file case from kebab case to camel case * refactor(config): reverting file name to ubiquibot-config.yml * feat(camelCase): - changing ubiquibot-config-default.json to cameCase * feat(private): - changing to camelCase in private * feat(config.ts): - updating WideConfigSchema and MergedConfigSchema * feat(update-base): - updating price-multiplier to camelCase counterpart priceMultiplier * fix(config): - removing quotes * fix(quotes): - removing left out quotes * fix(ts-errors): - fixing ts errors * fix: build --------- Co-authored-by: 0xCodercrane <[email protected]>
✅ Deploy Preview for ubiquibot-staging ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
✅ Deploy Preview for ubiquibot-production ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
`; | ||
|
||
export const gptContextTemplate = ` | ||
You are the UbiquityAI, designed to review and analyze pull requests. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think the command is not registered. I couldn't find it earlier.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@Keyrxng rfc
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
My mistake fellas, forgot to replace it following shuffling the order of the contexts. In gpt.ts inside /review, the template is supplied first before anything else #L202. It's not used in /review as we are only passing issue/pull body as context now but it should be passed in to chatHistory for ask.
Without it GPT responds in a story telling way rather than a json format but achieves the same thing, but JSON does provide more data for the dependant call.
Let me know what happens next, whether I'm to fix it or what bit wet behind the ears with this
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You can file a new issue and reference this comment, and start working on it to get paid. If its too small of a task it might be more worth to just open the pull request for it without filing an issue. Either way is fine.
I should be able to price it before you get it done.
I asked @whilefoo to address all these changes but if only you really understand the solution it makes sense for you to work on it!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
No no new issue or anything needed, it's just adding one convo object into chatHistory but I appreciate that. I'll have time after 6pm tonight to catch it, don't trust doing it on my phone lmao too many fat thumbs waiting to happen
result.push(pr); | ||
} | ||
} | ||
return result; | ||
}; | ||
|
||
// Strips out all links from the body of an issue or pull request and fetches the conversational context from each linked issue or pull request | ||
export const getAllLinkedIssuesAndPullsInBody = async (issueNumber: number) => { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think we have a redundant function, something like getAllLinkedPulls
I saw in the beginning of this review
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not sure what's the context of this function, but this functions retrieves all URLs to lead to PRs and issues in the body of the issue which is different from gitLinkedPrParser
which retrieves only linked PRs on the specific issue.
@ubiquity/software-development I left a review RFC Assuming that all of my opinions are agreeable @whilefoo would you be able to implement my requests? |
ubiquity/ubiquibot#795 camel cased the configs
ubiquity/ubiquibot#795 camel cased the configs
ubiquity/ubiquibot#795 camel cased the configs
Resolves #
Quality Assurance: