Overview - Ghost

7 days30 days

Latest CI Pipeline Executions

Fuzzy

Succeeded
feat/member-metafields-api
45054d10 Merge 5f6ee0b40b170e12b311bf79e4760dd263548161 into 2478bf700c2e434da51475b8c7902e20443f4edf
by Rob Lester
Failed
30399
5f6ee0b4 Removed the legacy custom_fields CSV column fallback The importer kept reading the pre-rename custom_fields.* columns so an old export would re-import without remapping. No such export exists outside development, because the feature has never been on in production, and the mapping step already lets a publisher point any column of any file at any field. The fallback goes, along with its tests and the last fixtures written in the old vocabulary.
by Rob Lester
Failed
30399
e8891c21 Changed the member custom fields API to namespaced metafields
by Rob Lester
Succeeded
30399
a73aed81 Removed the legacy custom_fields CSV column fallback The importer kept reading the pre-rename custom_fields.* columns so an old export would re-import without remapping. No such export exists outside development, because the feature has never been on in production, and the mapping step already lets a publisher point any column of any file at any field. The fallback goes, along with its tests and the last fixtures written in the old vocabulary.
by Rob Lester
Failed
30399
5a2d74b3 Removed the legacy custom_fields CSV column fallback The importer kept reading the pre-rename custom_fields.* columns so an old export would re-import without remapping. No such export exists outside development, because the feature has never been on in production, and the mapping step already lets a publisher point any column of any file at any field. The fallback goes, along with its tests and the last fixtures written in the old vocabulary.
by Rob Lester
Succeeded
30399
028ec9ce Removed the legacy custom_fields CSV column fallback The importer kept reading the pre-rename custom_fields.* columns so an old export would re-import without remapping. No such export exists outside development, because the feature has never been on in production, and the mapping step already lets a publisher point any column of any file at any field. The fallback goes, along with its tests and the last fixtures written in the old vocabulary.
by Rob Lester
Failed
30399
285dc82f Removed the legacy custom_fields CSV column fallback The importer kept reading the pre-rename custom_fields.* columns so an old export would re-import without remapping. No such export exists outside development, because the feature has never been on in production, and the mapping step already lets a publisher point any column of any file at any field. The fallback goes, along with its tests and the last fixtures written in the old vocabulary.
by Rob Lester
Failed
30399
98cba547 Removed the legacy custom_fields CSV column fallback The importer kept reading the pre-rename custom_fields.* columns so an old export would re-import without remapping. No such export exists outside development, because the feature has never been on in production, and the mapping step already lets a publisher point any column of any file at any field. The fallback goes, along with its tests and the last fixtures written in the old vocabulary.
by Rob Lester
Succeeded
30399
65faa373 Changed the filter picker to build field ids from the definition's namespace The picker was the last place in Admin that spelled a metafield address by hand, writing the publisher's namespace into a template literal twice. Every other surface derives addresses through the addressing module, so a field list that one day carries another namespace would have worked everywhere but the picker. Routing both entries through metafieldFieldId makes the namespace flow from the definition itself, and typing the helper's return after the metafields prefix lets the picker accept it without a cast. The acceptance fake for definitions was untyped, so its fixtures could omit the namespace the picker now reads; typing it against MemberCustomField makes the compiler hold every fixture to the shape the real API serves.
by Rob Lester
Failed
30399
98d41298 Changed metafield namespaces to flow as data above the storage layer Namespaces were half-introduced as a compile-time registry, which made every future namespace a code event when they are going to arrive as data, with an app that installs. The registry is gone: an unknown namespace is now the same non-event as an unknown key. A browse of it is an empty collection, a filter on it matches nobody, and a write into it is refused as an unknown field by its full address, so a namespace gaining its first field starts working with no change above the storage layer. Values travel keyed by field identity, every admin filter id, column key, CSV column and value lookup derives from the namespace each field carries, and the member payload nests by whatever namespaces the backend serializes. The one hard-coded namespace left is at the query boundaries, where storage that predates namespace storage implicitly holds the publisher's fields: the definitions codec and scoping, the values lookup and read grouping, the filter transformer's condition mapping, and the checkout bindings store. Once a leaves view carries namespace and key columns, each of those boundaries collapses into a column condition. A side effect worth naming: a __proto__ key sent as a field name is now refused loudly as an unknown field instead of being silently dropped, which matches how every other unrecognised name is treated.
by Rob Lester
Succeeded
feat/member-metafields-api
6b6f3252 Merge 98d41298b3c87453caa85ef4d87d5c1cccddf879 into 85fc64ee3250dc306e58979267b11a62f9d35f98
by Rob Lester
Failed
30399
98d41298 Changed metafield namespaces to flow as data above the storage layer Namespaces were half-introduced as a compile-time registry, which made every future namespace a code event when they are going to arrive as data, with an app that installs. The registry is gone: an unknown namespace is now the same non-event as an unknown key. A browse of it is an empty collection, a filter on it matches nobody, and a write into it is refused as an unknown field by its full address, so a namespace gaining its first field starts working with no change above the storage layer. Values travel keyed by field identity, every admin filter id, column key, CSV column and value lookup derives from the namespace each field carries, and the member payload nests by whatever namespaces the backend serializes. The one hard-coded namespace left is at the query boundaries, where storage that predates namespace storage implicitly holds the publisher's fields: the definitions codec and scoping, the values lookup and read grouping, the filter transformer's condition mapping, and the checkout bindings store. Once a leaves view carries namespace and key columns, each of those boundaries collapses into a column condition. A side effect worth naming: a __proto__ key sent as a field name is now refused loudly as an unknown field instead of being silently dropped, which matches how every other unrecognised name is treated.
by Rob Lester
Failed
30399
0c3035ec Changed metafield namespaces to flow as data above the storage layer Namespaces were half-introduced as a compile-time registry, which made every future namespace a code event when they are going to arrive as data, with an app that installs. The registry is gone: an unknown namespace is now the same non-event as an unknown key. A browse of it is an empty collection, a filter on it matches nobody, and a write into it is refused as an unknown field by its full address, so a namespace gaining its first field starts working with no change above the storage layer. Values travel keyed by field identity, every admin filter id, column key, CSV column and value lookup derives from the namespace each field carries, and the member payload nests by whatever namespaces the backend serializes. The one hard-coded namespace left is at the query boundaries, where storage that predates namespace storage implicitly holds the publisher's fields: the definitions codec and scoping, the values lookup and read grouping, the filter transformer's condition mapping, and the checkout bindings store. Once a leaves view carries namespace and key columns, each of those boundaries collapses into a column condition. A side effect worth naming: a __proto__ key sent as a field name is now refused loudly as an unknown field instead of being silently dropped, which matches how every other unrecognised name is treated.
by Rob Lester
Failed
30399
a94232e6 Fixed stale member snapshots that expected no creation attribution Three member-create snapshots recorded attribution as null, a state inherited from however the worker was ordered when they were recorded, while an admin-created member carries a Created manually attribution and the same file's own inline matchers expect it. The mismatch made those tests throw at the create step, and the one inside the destroy test aborted before the destroy ran, leaking its member into the shared database and knocking over the member counts and newsletter filter snapshots downstream. The three entries now record the attribution the API returns, which also lets the destroy complete and the downstream snapshots hold.
by Rob Lester
Failed
30399
ec30af7b Fixed tests left speaking the old custom fields vocabulary The metafields rename converted every surface it knew about, and CI plus a rebase over main found the stragglers: the body-schema package's own tests still pinned the old member payload key, the front-end member test pinned the account payload's exact key set without the new one, tests that merged in from main for the checkout collected-fields fix read values through the old payload shape, a bread service unit stub predated the wire unwrap, and the import mapping fixtures offered old column names. The member import suite also never deleted the members it created, which leaked one extra member into every file sharing its worker and shifted counts and snapshots downstream, so its cleanup now covers the members table too.
by Rob Lester
Failed
30399
bb964fd2 Changed metafield namespaces to flow as data above the storage layer Namespaces were half-introduced as a compile-time registry, which made every future namespace a code event when they are going to arrive as data, with an app that installs. The registry is gone: an unknown namespace is now the same non-event as an unknown key. A browse of it is an empty collection, a filter on it matches nobody, and a write into it is refused as an unknown field by its full address, so a namespace gaining its first field starts working with no change above the storage layer. Values travel keyed by field identity, every admin filter id, column key, CSV column and value lookup derives from the namespace each field carries, and the member payload nests by whatever namespaces the backend serializes. The one hard-coded namespace left is at the query boundaries, where storage that predates namespace storage implicitly holds the publisher's fields: the definitions codec and scoping, the values lookup and read grouping, the filter transformer's condition mapping, and the checkout bindings store. Once a leaves view carries namespace and key columns, each of those boundaries collapses into a column condition. A side effect worth naming: a __proto__ key sent as a field name is now refused loudly as an unknown field instead of being silently dropped, which matches how every other unrecognised name is treated.
by Rob Lester
Failed
30399
084b20dd Added member metafields to the members API account surface A member now reads and writes their own custom field values through the account endpoints: the member payload carries metafields nested by namespace, an account update plans and validates values before anything is applied and writes them with the member recorded as the writer, and a read-only definitions slice at /api/member/metafields serves the identity, label and type the account area renders inputs from. Every active field is present and writable; which fields each surface serializes is a per-definition policy, so a future access model narrows the data rather than the shape. With the feature flag off the payload key is absent, sent values are dropped with the rest of the body still applying, and the definitions slice 404s. The test that pinned this surface as staff-only flips to pinning the projection, including the stored provenance saying the member wrote the value.
by Rob Lester
Failed
30399
bab07d59 Added member metafields to the members API account surface A member now reads and writes their own custom field values through the account endpoints: the member payload carries metafields nested by namespace, an account update plans and validates values before anything is applied and writes them with the member recorded as the writer, and a read-only definitions slice at /api/member/metafields serves the identity, label and type the account area renders inputs from. Every active field is present and writable; which fields each surface serializes is a per-definition policy, so a future access model narrows the data rather than the shape. With the feature flag off the payload key is absent, sent values are dropped with the rest of the body still applying, and the definitions slice 404s. The test that pinned this surface as staff-only flips to pinning the projection, including the stored provenance saying the member wrote the value.
by Rob Lester