fix(studio): validate entity names against the entity-store pattern

[marcusds]

Problem

Creating an inference provider named Sparl passed client-side validation and then failed on submit:

{"detail": "[{'type': 'string_pattern_mismatch', 'loc': ['body', 'name'],
  'msg': \"String should match pattern '^[a-z](?!.*--)[a-z0-9\\\\-@.+_]{1,62}(?<!-)$'\", ...}]"}

The forms validated against the SDK-generated regex ^[\w\-.]+$ (max 255) — that's what the service DTOs advertise in the OpenAPI spec. The entity store enforces a stricter RFC-1035-ish NAME_PATTERN downstream (packages/nmp_common/src/nmp/common/entities/constants.py on main; moving to nemo_platform_plugin/entity_naming.py in #978), so the spec is looser than reality and invalid names got through to a confusing 422 toast.

Three places had already worked around this independently, each with its own copy of the pattern or its own wording of the same rule.

Change

New web/packages/common/src/utils/entityName.ts — mirrors NAME_PATTERN and reports the specific rule a value breaks, with a repaired-name suggestion where one exists:

input	message
Sparl	Name must be lowercase. Try "sparl".
invalid name!	Name cannot contain spaces, "!". Use lowercase letters, numbers, and - _ . @ + only. Try "invalid-name".
1provider	Name must start with a lowercase letter. Try "provider".
my--provider	Name cannot contain consecutive hyphens. Try "my-provider".
myprovider-	Name cannot end with a hyphen. Try "myprovider".
64 chars	Name must be 63 characters or fewer (currently 64). Try "…".

Adopted in every form that names an entity:

• CreateInferenceProviderSidePanel — replaced the SDK regex; also pre-empts the duplicate-name 409 using the provider list it already fetches
• CreateSecretModal — dropped its duplicate SECRET_NAME_REGEXP (closes the TODO #4082 there)
• FilesetNewRoute / FilesetCreateModal — dropped two more hand-written wordings
• SubmitEvaluationModal — replaced a manual .test() inside a superRefine

filesetName.ts now delegates to the shared helper and keeps only toValidFilesetName (domain-specific 'fileset' fallback) and FILESET_NAME_MAX_LENGTH.

The regex itself is no longer written here at all — ENTITY_NAME_REGEXP is entitiesCreateEntityBodyNameRegExp from the generated zod, taken from the generic entity-store endpoint rather than one of the per-resource copies. That only works because of #978 (see below).

ENTITY_NAME_MAX_LENGTH stays local: the entity-store schema declares no maxLength, so there is nothing generic to import. entityName.test.ts pins it against the fileset and secret create schemas that do declare one, and asserts all four generated schemas — entity, fileset, secret, model provider — still share a single pattern. Drift between them fails CI. Before #978 the model-provider assertion would have failed against ^[\w\-.]+$.

What stays hand-written is the part codegen can't express: the per-rule error messages and the sanitizer's character classes.

Stacked on #978 — merge that first

Important

This PR targets astd-349-entity-name-pattern-dtos/mschwab, not main. #978 must merge first; GitHub will retarget this to main automatically when it does.

@steramae-nvidia asked whether the spec itself was the problem. It was, and #978 (ASTD-349) fixes it:

• CreateModelProviderRequest, CreateFilesetRequest, and PlatformSecretCreateRequest now declare NAME_PATTERN instead of ^[\w\-.]+$, with max_length 255 → 63 to match. The secrets DTO enforced its rule in a field_validator, so it published no pattern at all; that moved to pattern=.
• The four Python copies of NAME_PATTERN collapse into one nemo_platform_plugin/entity_naming.py, re-exported by nmp_common. nmp_common depends on nemo-platform-plugin, so the plugin is the only side both can import from.

Originally this PR shipped its own mirrored regex as a stopgap. Stacking on #978 removes that: the generated zod now carries the strict pattern, so there is no hand-written copy of NAME_PATTERN left anywhere in web/.

Testing

• packages/studio — 2657/2657 pass
• packages/common — 1359/1359 pass
• Added entityName.test.ts (per-rule messages, sanitizer, zod schema, generated-schema agreement) and 7 cases to the provider panel test
• tsc --noEmit, eslint, prettier clean on both packages

One existing assertion changed, and it's an upgrade: FilesetNewRoute typing tiny-gpt2-A previously matched the generic blob (which happened to contain "must start with a lowercase letter"); it now asserts Name must be lowercase. Try "tiny-gpt2-a".

Summary by CodeRabbit

• New Features
  • Applied consistent name validation across filesets, inference providers, secrets, and evaluation submissions, including shared guidance and automatic sanitization/fallback.
  • Prevent duplicate inference provider names.
  • Disable the “Add Provider” action while providers are still loading.
• Refactor
  • Centralized entity-name validation and help text, replacing feature-specific regex rules.
• Bug Fixes
  • Updated validation messaging to be more specific (e.g., lowercase/capitalization suggestions).
• Tests
  • Added unit tests for entity-name rules and updated existing validation test expectations.

[coderabbitai]

📝 Walkthrough

Walkthrough

Adds shared entity-name validation, sanitization, error messaging, and help text. Frontend forms, backend models, job-step validation, and OpenAPI schemas now use common lowercase-first naming rules with length and hyphen constraints.

Changes

Entity-name validation

Layer / File(s)	Summary
Entity-name contract and utilities web/packages/common/src/utils/entityName.ts, web/packages/common/src/utils/entityName.test.ts	Defines shared regex, length limits, sanitization, validation errors, Zod integration, and tests.
Fileset name utility migration web/packages/common/src/utils/filesetName.ts, web/packages/common/src/utils/filesetName.test.ts	Delegates fileset sanitization to shared utilities and validates results against the shared regexp.
Studio form validation web/packages/studio/src/components/FilesetCreateModal/*, web/packages/studio/src/routes/FilesetNewRoute/*, web/packages/studio/src/routes/SecretsListRoute/*, web/packages/studio/src/routes/agents/..., web/packages/studio/src/routes/InferenceProvidersListRoute/*	Replaces inline naming rules with shared validation, help text, duplicate-name checks, loading-state gating, and updated tests.
Backend and API naming contract packages/nemo_platform_plugin/src/nemo_platform_plugin/..., packages/nmp_common/src/..., openapi/*	Centralizes Python naming constants and applies shared patterns and 63-character limits to backend models, job steps, secrets, filesets, providers, and OpenAPI schemas.

Sequence Diagram(s)

sequenceDiagram
  participant User
  participant StudioForm
  participant EntityNameSchema
  participant ProvidersAPI
  User->>StudioForm: Enter entity name
  StudioForm->>EntityNameSchema: Validate name
  EntityNameSchema-->>StudioForm: Return rule-specific result
  StudioForm->>ProvidersAPI: Load existing provider names
  ProvidersAPI-->>StudioForm: Return provider list
  StudioForm-->>User: Show validation or enable submission

Loading

Possibly related PRs

• NVIDIA-NeMo/nemo-platform#839: Touches the shared fileset/entity-name validation flow used by this change.

Suggested reviewers: steramae-nvidia, mckornfield, aray12

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 57.14% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarizes the main change: Studio entity-name validation now matches the shared entity-store pattern.

✨ Finishing Touches 💡 1

📝 Generate docstrings 💡

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch provider-name-validation/mschwab

---

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

web/packages/common/src/utils/entityName.ts (1)

93-100: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low value

Add an explicit return type.

entityNameSchema is an exported API but relies on inferred Zod types. Declare the compatible schema return type for the pinned Zod version. As per coding guidelines, “Use explicit return types for public APIs and complex functions in TypeScript.”

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@web/packages/common/src/utils/entityName.ts` around lines 93 - 100, Update
the exported entityNameSchema function with an explicit return type compatible
with the pinned Zod version, while preserving its current z.string().superRefine
validation behavior and label default.

Source: Coding guidelines

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In
`@web/packages/studio/src/routes/InferenceProvidersListRoute/CreateInferenceProviderSidePanel/index.tsx`:
- Around line 126-142: Update the provider-name validation around formSchema and
the useForm submission flow to validate against a complete provider set, not an
empty or first-page-only result. Ensure all provider pages are loaded (or
perform an equivalent server-side name-availability check), and block submission
while that validation data is unavailable so duplicate names are detected before
the API request.

---

Nitpick comments:
In `@web/packages/common/src/utils/entityName.ts`:
- Around line 93-100: Update the exported entityNameSchema function with an
explicit return type compatible with the pinned Zod version, while preserving
its current z.string().superRefine validation behavior and label default.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: e608ddaf-6bb9-4efd-8aa0-9a639e72c545

📥 Commits

Reviewing files that changed from the base of the PR and between 579de6b and 6ab428b.

📒 Files selected for processing (13)

• web/packages/common/src/utils/entityName.test.ts
• web/packages/common/src/utils/entityName.ts
• web/packages/common/src/utils/filesetName.test.ts
• web/packages/common/src/utils/filesetName.ts
• web/packages/studio/src/components/FilesetCreateModal/constants.ts
• web/packages/studio/src/routes/FilesetNewRoute/constants.ts
• web/packages/studio/src/routes/FilesetNewRoute/index.test.tsx
• web/packages/studio/src/routes/FilesetNewRoute/types.ts
• web/packages/studio/src/routes/InferenceProvidersListRoute/CreateInferenceProviderSidePanel/index.test.tsx
• web/packages/studio/src/routes/InferenceProvidersListRoute/CreateInferenceProviderSidePanel/index.tsx
• web/packages/studio/src/routes/SecretsListRoute/CreateSecretModal/constants.ts
• web/packages/studio/src/routes/SecretsListRoute/CreateSecretModal/index.tsx
• web/packages/studio/src/routes/agents/AgentEvaluationsRoute/components/SubmitEvaluationModal.tsx

💤 Files with no reviewable changes (2)

• web/packages/studio/src/routes/SecretsListRoute/CreateSecretModal/constants.ts
• web/packages/studio/src/routes/FilesetNewRoute/constants.ts

[github-actions]

Suite	Lines Covered	Line Rate	Branch Rate
Unit Tests	28393/36233	78.4%	62.8%
Integration Tests	16981/34951	48.6%	21.1%

[marcusds]

@coderabbitai Added the explicit return type in 6153573 — entityNameSchema now declares z.ZodEffects<z.ZodString, string, string> (zod ^3.25.76).

[coderabbitai]

Tip

For best results, initiate chat on the files or code changes.

@marcusds Confirmed.

[coderabbitai]

🧹 Nitpick comments (2)

packages/nemo_platform_plugin/src/nemo_platform_plugin/entity_naming.py (1)

10-25: 🎯 Functional Correctness | 🔵 Trivial | ⚡ Quick win

Regex verified correct; no dedicated tests for this new shared contract.

Traced NAME_PATTERN manually: enforces lowercase start, 2-63 total chars (1 + {1,62}), rejects any --, rejects trailing -. Matches NAME_PATTERN_DESCRIPTION and NAME_MAX_LENGTH=63. No functional issue found.

This module is now the single source of truth consumed by filesets, secrets, job steps, entities, model providers, workspaces, and projects — but ships with no unit tests in this change. A regression here has wide blast radius.

Want me to draft a test_entity_naming.py covering boundary cases (min/max length, leading digit, consecutive hyphens, trailing hyphen, allowed special chars)?

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@packages/nemo_platform_plugin/src/nemo_platform_plugin/entity_naming.py`
around lines 10 - 25, Add dedicated unit tests for the shared NAME_PATTERN
contract in a new test_entity_naming.py, covering minimum and maximum valid
lengths, lowercase-start enforcement, rejection of leading digits, consecutive
hyphens, and trailing hyphens, plus acceptance of the temporarily supported @,
., +, and _ characters.

openapi/ga/individual/platform.openapi.yaml (1)

9411-9416: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win

Name-field descriptions omit allowed special characters. All three new/updated name fields use pattern ^[a-z](?!.*--)[a-z0-9\-@.+_]{1,62}(?<!-)$, which permits _ . @ +, but each description claims only "lowercase letters, digits, and hyphens" are allowed.

• openapi/ga/individual/platform.openapi.yaml#L9411-L9416: update CreateFilesetRequest.name description to mention _ . @ + are also allowed.
• openapi/ga/individual/platform.openapi.yaml#L9682-L9687: same fix for CreateModelProviderRequest.name.
• openapi/ga/individual/platform.openapi.yaml#L16458-L16463: same fix for PlatformSecretCreateRequest.name.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@openapi/ga/individual/platform.openapi.yaml` around lines 9411 - 9416, Update
the descriptions for CreateFilesetRequest.name in
openapi/ga/individual/platform.openapi.yaml lines 9411-9416,
CreateModelProviderRequest.name at lines 9682-9687, and
PlatformSecretCreateRequest.name at lines 16458-16463 to state that lowercase
letters, digits, hyphens, underscores, periods, at signs, and plus signs are
allowed, while preserving the existing length, starting-character,
consecutive-hyphen, and trailing-hyphen constraints.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Nitpick comments:
In `@openapi/ga/individual/platform.openapi.yaml`:
- Around line 9411-9416: Update the descriptions for CreateFilesetRequest.name
in openapi/ga/individual/platform.openapi.yaml lines 9411-9416,
CreateModelProviderRequest.name at lines 9682-9687, and
PlatformSecretCreateRequest.name at lines 16458-16463 to state that lowercase
letters, digits, hyphens, underscores, periods, at signs, and plus signs are
allowed, while preserving the existing length, starting-character,
consecutive-hyphen, and trailing-hyphen constraints.

In `@packages/nemo_platform_plugin/src/nemo_platform_plugin/entity_naming.py`:
- Around line 10-25: Add dedicated unit tests for the shared NAME_PATTERN
contract in a new test_entity_naming.py, covering minimum and maximum valid
lengths, lowercase-start enforcement, rejection of leading digits, consecutive
hyphens, and trailing hyphens, plus acceptance of the temporarily supported @,
., +, and _ characters.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 71e96903-1f76-4655-b5cc-8d9870f2ff54

📥 Commits

Reviewing files that changed from the base of the PR and between 6153573 and 772d1e4.

📒 Files selected for processing (8)

• openapi/ga/individual/platform.openapi.yaml
• openapi/ga/openapi.yaml
• openapi/openapi.yaml
• packages/nemo_platform_plugin/src/nemo_platform_plugin/entity_naming.py
• packages/nemo_platform_plugin/src/nemo_platform_plugin/files/types.py
• packages/nemo_platform_plugin/src/nemo_platform_plugin/jobs/spec.py
• packages/nemo_platform_plugin/src/nemo_platform_plugin/secrets/types.py
• packages/nmp_common/src/nmp/common/entities/constants.py