docs(web/guides): clarify box CLI vs wheels CLI in v4 active docs#2542
Conversation
Add Aside to BoxLang section of cfml-engines guide explaining that `wheels start` boots the bundled Lucee Express and CommandBox is required for BoxLang server management. Update vm-deployment tip to remove confusing CI/Docker reference and explicitly distinguish CommandBox server management from the `wheels` dev CLI. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
There was a problem hiding this comment.
Wheels Bot — Reviewer A
TL;DR: Docs-only PR that adds a BoxLang <Aside> to cfml-engines.mdx and updates the CommandBox tip in vm-deployment.mdx. Both changes are directionally correct and the core goal is achieved. One minor completeness gap in the BoxLang aside makes this a comment rather than an approval.
Docs
cfml-engines.mdx — BoxLang aside is less comprehensive than the Adobe CF aside
The PR body says the new BoxLang <Aside> is "mirroring the existing note in the Adobe CF section", but the two notes are not equivalent. The Adobe CF note (lines 70–72) lists four affected commands:
The `wheels start`, `wheels stop`, `wheels reload`, and `wheels test` commands all assume
the Lucee Express bundled with the CLI. On Adobe CF you'll use `box server start/stop/restart`
instead, and reload via `?reload=true&password=<your-password>` directly.The new BoxLang note (lines 84–86 after the diff) only mentions wheels start:
`wheels start` boots the Lucee Express bundled with the CLI — it does not start BoxLang.A BoxLang developer who reads the new aside will correctly understand that wheels start is a no-op for them, but they may still be confused about:
wheels stop— same Lucee-only assumption, won't stop their BoxLang serverwheels reload— won't work; they need?reload=true&password=<your-password>like Adobe CF userswheels test— hits the bundled Lucee's test endpoint, not the BoxLang server
Suggested fix: expand the BoxLang aside to match the Adobe CF aside's coverage:
<Aside type="note">
The `wheels start`, `wheels stop`, `wheels reload`, and `wheels test` commands all assume
the Lucee Express bundled with the CLI — they do not interact with BoxLang. BoxLang developers
manage their server through CommandBox (`box server start/stop/restart`) and reload via
`?reload=true&password=<your-password>` directly. Generators (`wheels generate ...`) are
pure file-system operations and work the same regardless of engine.
</Aside>vm-deployment.mdx — removal of CI Docker reference is correct
The dropped sentence — "It's the same mechanism the Wheels test images use (see tools/docker/lucee7/server.json)" — referenced a file whose "profile": "development" and host: localhost make it actively misleading as production guidance. Removing it and replacing it with a wheels start scope note is the right call.
Commits
Single commit: docs(web/guides): clarify box CLI vs wheels CLI in v4 active docs — scope web/guides is in the allowlist, type docs is valid, subject is under 100 chars, not ALL-CAPS. Clean.
Tests
Docs-only change. No test coverage needed.
Wheels Bot - Reviewer B (round 1)A's review is accurate and actionable. The finding holds up against the actual file: the Adobe CF aside lists four commands (wheels start, wheels stop, wheels reload, wheels test) plus the reload workaround URL, while the new BoxLang aside lists only wheels start. A's suggested fix aligns the two asides correctly. Converging on changes needed. Sycophancy Minor: A writes 'the core goal is achieved' despite the PR body explicitly stating the BoxLang aside 'mirrors' the Adobe CF section, which it does not. This framing slightly softens a real gap. Not consequential enough to block convergence. False positives None detected. I verified the Adobe CF aside content against the actual file at the PR head SHA. A's quote is accurate word-for-word. Missed issues None. The CHANGELOG entry is in the right section, the commit message is valid (docs(web/guides), under 100 chars, not ALL-CAPS), and the vm-deployment change correctly removes the misleading CI Docker reference. Verdict alignment A chose COMMENTED rather than REQUEST_CHANGES. Given the PR body explicitly promises to mirror the Adobe CF section and does not fully deliver on that, REQUEST_CHANGES would be marginally more precise. However, COMMENTED with a concrete, specific fix is functionally equivalent. Convergence Aligned on changes needed. A's finding is valid and the fix is specific: expand the BoxLang aside to add wheels stop, wheels reload with the reload URL path, and wheels test. No disputes remain. Apply A's suggested fix. |
…ines Address Reviewer A/B consensus findings (round 1): - Expand the BoxLang <Aside> in v4-0-0-snapshot/start-here/cfml-engines.mdx to match the Adobe CF aside's coverage. Now lists all four affected commands (wheels start, stop, reload, test) and includes the ?reload=true&password=... URL workaround, matching the Adobe CF section that the PR body promised to mirror.
Wheels Bot — Address Review (round 1)Applied consensus findings from Reviewer A and Reviewer B's convergence (round 1):
No findings were skipped — A and B aligned on a single, specific change. Tests: docs-only change, no test run needed. The new commit will trigger a fresh Reviewer A run on the updated SHA. Convergence loop continues until reviewers align on |
There was a problem hiding this comment.
Wheels Bot — Reviewer A
TL;DR: Docs-only PR that clarifies BoxLang and CommandBox server management in two v4 active-docs files. The second commit directly addresses the only finding from my previous review round — the BoxLang <Aside> now fully mirrors the Adobe CF aside. The diff is clean; no issues found. Verdict: approve.
Correctness
The new commit (9911f2e1c) expands the BoxLang aside from a one-sentence stub to the full four-command coverage the PR body promised. Side-by-side comparison of the two asides after the fix:
| Section | Commands listed | Reload URL | Generators note |
|---|---|---|---|
| Adobe CF (line 71) | start, stop, reload, test | yes | yes |
| BoxLang (line 85) | start, stop, reload, test | yes | yes |
The BoxLang variant uses "they do not interact with BoxLang" as a bridging clause where the Adobe CF version uses "On Adobe CF you'll use … instead" — both are accurate and the engine-specific phrasing is appropriate rather than a gap.
The vm-deployment.mdx tip removal (CI Docker test-image reference) and scope note addition are both correct. The hardcoded lucee@7.0.1+100 version in that tip is pre-existing, not introduced here.
Docs
CHANGELOG [Unreleased] entry is present and accurate.
Commits
013ccd322 docs(web/guides): clarify box CLI vs wheels CLI in v4 active docs— typedocs, scopeweb/guides, clean.9911f2e1c docs(web/guides): align BoxLang aside with Adobe CF aside in cfml-engines— typedocs, scopeweb/guides, clean.
Both conform to commitlint.config.js.
Tests
Docs-only change. No test coverage needed.
Wheels Bot — Reviewer B (round 1)A's review checks out. The side-by-side table accurately reflects the actual file content — both asides now carry identical coverage (four commands, reload URL, generators note). No sycophancy, no false positives, no missed issues. Converging on approve. SycophancyNone detected. A provides a concrete side-by-side comparison table and verifies the engine-specific phrasing difference is intentional rather than a gap, rather than simply saying LGTM. False positivesNone detected. I read the Adobe CF aside and the new BoxLang aside from the head SHA directly. A's table is accurate:
Missed issuesNone detected. Both commits conform to Verdict alignmentA's verdict (approve) is consistent with finding no issues after the address-review commit resolved the prior round's only finding. ConvergenceAligned on approve. The BoxLang aside now fully mirrors the Adobe CF aside as the PR body promised. No outstanding disputes or missed issues remain. The PR is review-clean for this SHA. |
This PR addresses the two active-docs files that contain
boxCLI references needing clarification, as identified in #2541. Thebox server startcommands in these files are intentional (CommandBox is required for Adobe CF and BoxLang, and is a valid alternative Lucee installer on VMs), but the surrounding prose lacked context explaining whywheels startis not used in those paths.Changes:
cfml-engines.mdx— Added an<Aside>note after the BoxLang code block, mirroring the existing note in the Adobe CF section. Clarifies thatwheels startboots the bundled Lucee Express and that BoxLang developers usebox server start/stop/restartfor server management.vm-deployment.mdx— Updated the CommandBox tip to remove a confusing reference to CI Docker test images as production guidance, and added an explicit note thatwheels startis a local development convenience, not a production Lucee installation.Fixes #2541
Scope note
This PR covers the Priority 1 changes only — the 2 files in
v4-0-0-snapshot/(the active docs tree). The Priority 2 pass across ~23 files inv3-0-0/falls outside the ~5-file threshold for an automated docs PR and warrants a separate human-led pass. Thev3-0-0/files are versioned documentation for a prior major release; many of thosebox wheels …/box server …references may never have had a directwheelsCLI equivalent in v3 context, and a human should evaluate each file before replacing commands.Screenshots needed
None — these changes are prose and Aside note additions only.