@ai-matrx/records 0.39.0 → 0.40.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +31 -27
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -1,5 +1,36 @@
|
|
|
1
1
|
# Changelog — @ai-matrx/records
|
|
2
2
|
|
|
3
|
+
## 0.40.0
|
|
4
|
+
|
|
5
|
+
**The parity floor this package types is asserted against the one the store ships.**
|
|
6
|
+
|
|
7
|
+
`checkbox` landed as the fourteenth parity type and `store.generated.ts` picked it
|
|
8
|
+
up — but nothing made that agreement a test. Everything that spelled the floor out
|
|
9
|
+
by hand went stale in the same hour: three of the store's own refusal hints, the
|
|
10
|
+
`FieldTypeWord` doc comment, `UseCaseField.parityType`'s doc comment and the
|
|
11
|
+
use-case README's list, all still saying thirteen. A person who asked for a
|
|
12
|
+
checkbox could be told by the store that checkboxes do not exist.
|
|
13
|
+
|
|
14
|
+
* `src/__tests__/real-doors.test.ts` gains three cases, in the one suite that runs
|
|
15
|
+
against the live store: `PARITY_FIELD_TYPES` is exactly
|
|
16
|
+
`custom.parity_field_types()` name for name (a diff NAMES the missing type, it
|
|
17
|
+
never prints a count); every parity type and every behaviour the store ships is
|
|
18
|
+
a legal `parityType` word in `RESOLVABLE_FIELD_TYPE_WORDS`; and `checkbox` is
|
|
19
|
+
present with behaviour `boolean`, with a compile-time assertion that the
|
|
20
|
+
`ParityFieldType` union still admits it. Shown RED: with the `checkbox` row
|
|
21
|
+
removed from `store.generated.ts` all three fail and the first one says
|
|
22
|
+
"expected 'attachment, currency, …' to be 'attachment, checkbox, …'".
|
|
23
|
+
* The four places that wrote the count or the list out by hand now point at the
|
|
24
|
+
derived set instead. No number is written down anywhere in this package.
|
|
25
|
+
* `store.generated.ts` regenerated against the live store (542 doors, 1023 entity
|
|
26
|
+
tokens); `pnpm check:store-registry` was RED on that drift before this release.
|
|
27
|
+
|
|
28
|
+
**Consumer action:** none.
|
|
29
|
+
|
|
30
|
+
**Why 0.40.0 and not 0.38.0.** This work landed on `main` a few minutes after
|
|
31
|
+
0.39.0 was tagged, so the published 0.39.0 does not contain it. The version number
|
|
32
|
+
moves rather than the tag.
|
|
33
|
+
|
|
3
34
|
## 0.39.0
|
|
4
35
|
|
|
5
36
|
**Two things a records screen could not know: that a table's shape had changed,
|
|
@@ -45,33 +76,6 @@ of the set, and nothing else. `src/__tests__/page-truth.test.ts` asserts the
|
|
|
45
76
|
refusal against the live store, so the day it softens into a silent shrink the
|
|
46
77
|
suite goes red and names the inference that has to change with it.
|
|
47
78
|
|
|
48
|
-
## 0.38.0
|
|
49
|
-
|
|
50
|
-
**The parity floor this package types is asserted against the one the store ships.**
|
|
51
|
-
|
|
52
|
-
`checkbox` landed as the fourteenth parity type and `store.generated.ts` picked it
|
|
53
|
-
up — but nothing made that agreement a test. Everything that spelled the floor out
|
|
54
|
-
by hand went stale in the same hour: three of the store's own refusal hints, the
|
|
55
|
-
`FieldTypeWord` doc comment, `UseCaseField.parityType`'s doc comment and the
|
|
56
|
-
use-case README's list, all still saying thirteen. A person who asked for a
|
|
57
|
-
checkbox could be told by the store that checkboxes do not exist.
|
|
58
|
-
|
|
59
|
-
* `src/__tests__/real-doors.test.ts` gains three cases, in the one suite that runs
|
|
60
|
-
against the live store: `PARITY_FIELD_TYPES` is exactly
|
|
61
|
-
`custom.parity_field_types()` name for name (a diff NAMES the missing type, it
|
|
62
|
-
never prints a count); every parity type and every behaviour the store ships is
|
|
63
|
-
a legal `parityType` word in `RESOLVABLE_FIELD_TYPE_WORDS`; and `checkbox` is
|
|
64
|
-
present with behaviour `boolean`, with a compile-time assertion that the
|
|
65
|
-
`ParityFieldType` union still admits it. Shown RED: with the `checkbox` row
|
|
66
|
-
removed from `store.generated.ts` all three fail and the first one says
|
|
67
|
-
"expected 'attachment, currency, …' to be 'attachment, checkbox, …'".
|
|
68
|
-
* The four places that wrote the count or the list out by hand now point at the
|
|
69
|
-
derived set instead. No number is written down anywhere in this package.
|
|
70
|
-
* `store.generated.ts` regenerated against the live store (542 doors, 1023 entity
|
|
71
|
-
tokens); `pnpm check:store-registry` was RED on that drift before this release.
|
|
72
|
-
|
|
73
|
-
**Consumer action:** none.
|
|
74
|
-
|
|
75
79
|
## 0.37.0
|
|
76
80
|
|
|
77
81
|
**Three doors the store already had and this client could not reach.**
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@ai-matrx/records",
|
|
3
|
-
"version": "0.
|
|
3
|
+
"version": "0.40.0",
|
|
4
4
|
"description": "The headless client of the AI Matrx record store: the Table / Field / Rule / Record / Relation / Home types, the closed vocabularies and refusal codes read from the store itself, the typed calls over the store's doors and nothing else, the merge-field declaration, a validation mirror that predicts only what the store's own triggers refuse, and the never-throw result envelope. Zero React, zero DOM in the root and /core.",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"main": "./dist/index.cjs",
|