uniweb 0.51.0 → 0.51.1

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/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "uniweb",
3
- "version": "0.51.0",
3
+ "version": "0.51.1",
4
4
  "description": "Create structured Vite + React sites with content/code separation",
5
5
  "type": "module",
6
6
  "bin": {
@@ -43,13 +43,13 @@
43
43
  "tar": "^7.0.0",
44
44
  "@uniweb/core": "^0.26.1",
45
45
  "@uniweb/kit": "^0.18.4",
46
- "@uniweb/semantic-parser": "^1.4.0",
47
- "@uniweb/runtime": "^0.20.3"
46
+ "@uniweb/runtime": "^0.20.3",
47
+ "@uniweb/semantic-parser": "^1.4.0"
48
48
  },
49
49
  "peerDependencies": {
50
+ "@uniweb/build": "^0.47.0",
50
51
  "@uniweb/content-reader": "^1.2.4",
51
- "@uniweb/semantic-parser": "^1.4.0",
52
- "@uniweb/build": "^0.47.0"
52
+ "@uniweb/semantic-parser": "^1.4.0"
53
53
  },
54
54
  "peerDependenciesMeta": {
55
55
  "@uniweb/build": {
@@ -9,12 +9,27 @@
9
9
  * the answer.
10
10
  *
11
11
  * ⛔ THERE IS NO PRE-FLIGHT, AND ADDING ONE BACK IS A REGRESSION. A
12
- * `GET …/can-go-live` probe used to run before go-live; it called a route no
13
- * backend serves, and it folded every failure 404 included into
14
- * "proceed". A check that answers "fine" when it cannot reach the server is not
15
- * a check, and it made the CLI assume a posture it has no business assuming. A
16
- * pre-flight also cannot be authoritative: the backend re-evaluates at publish
17
- * time regardless, so a second asker is a second producer of one decision.
12
+ * `can-go-live` probe used to run before go-live, and it folded every failure
13
+ * 404 included into "proceed". A check that answers "fine" when it cannot
14
+ * reach the server is not a check, and it made the CLI assume a posture it has no
15
+ * business assuming. A pre-flight also cannot be authoritative: the backend
16
+ * re-evaluates at publish time regardless, so a second asker is a second producer
17
+ * of one decision which is the reason that would stand even if the probe had
18
+ * worked.
19
+ *
20
+ * ⚠️ **THIS COMMENT USED TO SAY THE PROBE "called a route no backend serves".
21
+ * That was FALSE, corrected 2026-09-12.** What it called was
22
+ * `/dev/site/{uuid}/can-go-live` — and the 404 came from the PATH being wrong,
23
+ * not from the capability being absent. So the probe never worked at any version
24
+ * that shipped it, and every failure was swallowed by the fold above, which is
25
+ * exactly why nobody noticed.
26
+ *
27
+ * ⭐ Kept as a warning about the SHAPE of the old justification, not just the
28
+ * fact: a correct rule was resting on a claim about another system that this
29
+ * repo cannot see, and the claim was wrong. The rule survives because its real
30
+ * reasons are local — the fold, and the second-producer argument. **If you find
31
+ * yourself defending a rule here with an assertion about what a server does,
32
+ * check it or drop it.**
18
33
  *
19
34
  * WHAT THE CLI KNOWS ABOUT PAYMENT: nothing. It opens whatever settlement URL
20
35
  * the backend hands it, VERBATIM — provider-agnostic, and route-agnostic. The
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "schemaVersion": 1,
3
- "generatedAt": "2026-09-12T17:05:58.363Z",
3
+ "generatedAt": "2026-09-12T17:35:34.207Z",
4
4
  "packages": {
5
5
  "@uniweb/api": {
6
6
  "version": "0.3.3",
@@ -110,7 +110,7 @@
110
110
  "deps": []
111
111
  },
112
112
  "@uniweb/templates": {
113
- "version": "0.13.0",
113
+ "version": "0.13.1",
114
114
  "path": "framework/templates",
115
115
  "deps": []
116
116
  },
@@ -120,7 +120,7 @@
120
120
  "deps": []
121
121
  },
122
122
  "@uniweb/unipress": {
123
- "version": "0.9.14",
123
+ "version": "0.9.15",
124
124
  "path": "framework/unipress",
125
125
  "deps": [
126
126
  "@uniweb/build",