feat: migrate enterprise customer business logic to the BFF layer

[brobro10000]

Utilizes the BFF layer for metadata related to the enterprise customer metadata with the following serializer response update for the BFF layer within enterprise access (PR).

This PR also acts as a full cutover for the BFF layer in terms of removing any waffle or MFE config based flagging for the BFF API call. This PR refactors logic such that the determination for a BFF API call is solely dependent on the user's current route and deprecates all BFF related feature flags.

For all changes

• Ensure adequate tests are in place (or reviewed existing tests cover changes)
• Ensure English strings are marked for translation. See documentation for more details.

Only if submitting a visual change

• Ensure to attach screenshots
• Ensure to have UX team confirm screenshots

[codecov]

Codecov Report

Attention: Patch coverage is 94.59459% with 2 lines in your changes missing coverage. Please review.

Project coverage is 89.57%. Comparing base (64030b7) to head (0a70519).
Report is 12 commits behind head on master.

Files with missing lines	Patch %	Lines
.../components/app/data/hooks/useEnterpriseLearner.js	90.90%	1 Missing ⚠️
src/components/app/routes/loaders/rootLoader.ts	91.66%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1263      +/-   ##
==========================================
- Coverage   89.59%   89.57%   -0.02%     
==========================================
  Files         405      406       +1     
  Lines        8824     8809      -15     
  Branches     2064     2112      +48     
==========================================
- Hits         7906     7891      -15     
+ Misses        883      882       -1     
- Partials       35       36       +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[adamstankiewicz]

[curious] Why was this return type removed? Without it, TypeScript won't know this function returns a promise resolving Types.EnterpriseCustomer without it (currently shows Promise<any>).

It looks like removing this would be OK iff the allLinkedEnterpriseCustomerUsers was typed. As is, it looks like this property is associated with the type EnterpriseCustomerUser[], however it doesn't seem like the type EnterpriseCustomerUser exists yet?

For example, defining the EnterpriseCustomerUser type above EnterpriseLearnerData makes it so TypeScript can properly infer the type returned for foundEnterpriseCustomerForSlug.

export interface EnterpriseCustomerUser {
  enterpriseCustomer: EnterpriseCustomer;
}

export interface EnterpriseLearnerData {
  enterpriseCustomer: Types.EnterpriseCustomer | null;
  activeEnterpriseCustomer: Types.EnterpriseCustomer | null;
  activeEnterpriseCustomerUserRoleAssignments: any[];
  allLinkedEnterpriseCustomerUsers: EnterpriseCustomerUser[];
  enterpriseCustomerUserRoleAssignments: any[];
  staffEnterpriseCustomer: Types.EnterpriseCustomer | null;
  enterpriseFeatures: Types.EnterpriseFeatures;
  shouldUpdateActiveEnterpriseCustomerUser: boolean;
}

[adamstankiewicz]

[curious] Why is this type using any? Including it largely defeats the purpose of including Types.EnterpriseLearnerData, since it forces enterpriseLearnerData to be any instead of the actual type, EnterpriseLearnerData.

[adamstankiewicz]

Bit of a code smell to use any. Should these types by Types.EnterpriseCustomer and Types.EnterpriseCustomerUser[], respectively?

[adamstankiewicz]

[clarification/consideration] Keep in mind that properties now required in EnterpriseLearnerData must be returned/derived from the BFF-enabled cases, as well. Does activeEnterpriseCustomer, activeEnterpriseCustomerUserRoleAssignments, and enterpriseCustomerUserRoleAssignments need to be included in the BFF API responses via openedx/enterprise-access#624?

Or is it expected/OK that, in the BFF-enabled cases, these data are missing? If expected, should the additional fields be made optional?

[adamstankiewicz]

I believe the enterprise-learner API is still called on the dashboard route, coming from dashboardLoader calling extractEnterpriseCustomer. Perhaps a similar implementation as here, where enterpriseLearnerData is conditional depending on matchedBFFQuery, is needed within extractEnterpriseCustomer, too?

There might be an opportunity to make this logic more DRY across loaders, by abstracting the conditional enterpriseLearnerData into a helper function.

[adamstankiewicz]

[example] Sample helper function:

export async function getEnterpriseLearnerQueryData({
  requestUrl,
  queryClient,
  enterpriseSlug,
  authenticatedUser,
}: ExtractEnterpriseCustomerArgs) {
  // Retrieve linked enterprise customers for the current user from query cache
  // or fetch from the server if not available.
  const matchedBFFQuery = resolveBFFQuery(requestUrl.pathname);
  if (matchedBFFQuery) {
    const bffResponse = await queryClient.ensureQueryData(
      matchedBFFQuery({ enterpriseSlug }),
    ) as Types.BaseBFFResponse;
    const enterpriseLearnerData: Types.EnterpriseLearnerData = {
      enterpriseCustomer: bffResponse.enterpriseCustomer,
      enterpriseCustomerUserRoleAssignments: bffResponse.enterpriseCustomerUserRoleAssignments,
      activeEnterpriseCustomer: bffResponse.activeEnterpriseCustomer,
      activeEnterpriseCustomerUserRoleAssignments: bffResponse.activeEnterpriseCustomerUserRoleAssignments,
      allLinkedEnterpriseCustomerUsers: bffResponse.allLinkedEnterpriseCustomerUsers,
      staffEnterpriseCustomer: bffResponse.staffEnterpriseCustomer,
      enterpriseFeatures: bffResponse.enterpriseFeatures,
      shouldUpdateActiveEnterpriseCustomerUser: bffResponse.shouldUpdateActiveEnterpriseCustomerUser,
    };
    return enterpriseLearnerData;
  }
  const enterpriseLearnerData = await queryClient.ensureQueryData(
    queryEnterpriseLearner(authenticatedUser.username, enterpriseSlug),
  );
  return enterpriseLearnerData;
}

/**
 * Extracts the appropriate enterprise ID for the current user and enterprise slug.
 */
async function extractEnterpriseCustomer({
  queryClient,
  authenticatedUser,
  enterpriseSlug,
  requestUrl,
} : ExtractEnterpriseCustomerArgs) {
  const enterpriseLearnerData = await getEnterpriseLearnerQueryData({
    requestUrl,
    queryClient,
    enterpriseSlug,
    authenticatedUser,
  });

  const {
    activeEnterpriseCustomer,
    allLinkedEnterpriseCustomerUsers,
    staffEnterpriseCustomer,
  } = enterpriseLearnerData;

  // ...
}