0 Fork 0

Files

T

WangShen007 54b949b392 chore(brand): rename the product to ReplyPier

The upstream project name is gone from code, assets, docs and build
output: brand strings, domains (replypier.site and its app/api/docs/help
subdomains), env prefixes, image and network names, the widget's public
JS API (window.ReplyPier, replypierId/BaseUrl/Config/Token), CSS tokens
(--rp-*), file and directory names, and the WordPress plugin package.

- widget loader is now webclient/replypier.min.js; both bundles rebuilt
  from source (backend/assets/widget.{js,css}, frontend/public/webclient)
- NOTICE, llms.txt, CONTRIBUTING, HELP_CENTER_INFRA and config.dev.js
  rewritten rather than string-swapped; the demo config no longer ships
  the previous vendor's Firebase/Razorpay/Google keys
- new migration replypier_ai_model_001 adds REPLYPIER to the managed
  AIModelType and rewrites pre-rename rows
- local compose project, database, role and volume renamed; LOCAL_DEPLOY
  documents carrying the old local data over

2026-10-26 00:51:03 +08:00

2.1 KiB

Raw Blame History

name, description

name	description
write-e2e-test	Write or change a ReplyPier Playwright end-to-end test (e2e/). Use when adding e2e coverage for a feature, fixing a failing e2e spec, or adding a page object, fixture or fake-LLM script.

Writing a ReplyPier e2e test

The suite lives in e2e/. Read e2e/AUTHORING.md first — it is the source of truth for layout and rules. The essentials:

Specs speak product language. specs/<feature>/<name>.spec.ts, imports only from @e2e, no selectors (page.locator) and no raw HTTP — lint fails otherwise. Title ends with tags: @<feature>, plus @smoke only for the few journeys that must run on every PR.
Selectors go in src/pages/ (a screen) or src/components/ (shared UI). Prefer getByRole/getByLabel; if the element has no accessible name, add an aria-label to the Vue component rather than reaching for CSS. CSS needs a one-line "why" comment.
Seed through src/api/, assert through the UI. Need a new endpoint? Add a function to the matching src/api/*.ts module.
Script the AI with llm.script(agent, s => …) — reply, handOver, createTicket, callTool(...).thenReply(...). Replies are structured ChatResponse JSON; handover and tickets are boolean fields, not tool calls.
Fixtures give each test its own agent + widget. test.use({ agentSettings: { transferToHuman: true } }) for handover. inbox is Sam Agent's dashboard in a second browser. Never change org-wide settings outside a *.org.spec.ts.
Prove it is stable before finishing: cd e2e && npx playwright test specs/<feature> --repeat-each=10 && npm run lint && npm run typecheck (start the stack once with ./scripts/stack.sh up; in a Claude worktree pass E2E_PYTHON and E2E_FRONTEND_NODE_MODULES pointing at the main checkout, and run with the sandbox disabled — Postgres, ports and the node_modules symlink need it).
Update e2e/FEATURES.md — the row for the journey, with the spec path and ✅.

Enterprise specs live in the private enterprise_frontend repo under e2e/, import only from @e2e, and must never be added to this public repo.

2.1 KiB Raw Blame History

Writing a ReplyPier e2e test

2.1 KiB

Raw Blame History