A live hole silently stops re-emitting when its markup has a conditional on a property (`{m.usage && …}`)
Los mantenedores suelen responder en 1 día
Nadie ha tomado este issue todavía.
Evaluación
- Dificultad
- 5/5
- Tiempo estimado
- Más de una semana
- Aptitud para principiantes
- 15/100
- Tipo de issue
- Error
- Claridad
- Bastante claro
- Estado de actividad
- Activo
- Stack tecnológico
- javascript, typescript
Línea de trabajo
Start with the repro in the issue: a live GET server component whose markup holds a createMemo over an async source, with a {m.usage && <b>done</b>} conditional inside a .map() row. Compare the streamed hole events with and without the conditional, as the issue's table does. The likely area is how the compiler memoises conditionals whose condition is a property access, and how a memo created inside a re-running hole is tracked. The existing chat example's comment about owner creation is a lead. Done means the hole re-emits, or a dev diagnostic names the expression.
Escrito por el modelo de indexación a partir del texto del issue.
Descripción
Describe the bug
In a live server component, a markup hole stops re-emitting, silently, when the expression it renders contains a conditional whose condition is a property access. The first render is correct. After that the hole is never sent again, however often its source yields. There is no error, no warning and no diagnostic, on the server or in the browser. Slot arguments in the same component keep updating, which makes it look like a client problem.
What it takes
A live(GET(…)) server function that answers with a component, with a memo over an async iterable read in its markup:
const getThread = liveQuery(
live(
GET(async (id: string) => {
"use server";
return (props: { composer: ComposerSlot }) => {
const view = createMemo(() => db.watchThread(id, signal)); // yields the thread on every write
return (
<article>
<ol>
{view().messages.map(m => (
<li data-id={m.id}>
{m.text.length} {m.usage && <b>done</b>}
</li>
))}
</ol>
<props.composer seen={view().messages.map(m => m.id)} />
</article>
);
};
})
),
"thread"
);
m.usage is undefined until a reply finishes; m.text grows about every 50 ms while one is being written.
What happens
I connected to the live address with curl -N for the length of one reply (about 40 writes to the thread) and counted the events on the stream, changing only the expression inside the <li>:
Inside the <li> |
hole events for the list |
slot-argument events |
|---|---|---|
{m.text.length} |
39 | 43 |
{m.text.length} {m.usage && <b>done</b>} |
0 | 43 |
{m.text.length} {m.usage ? <b>done</b> : null} |
0 | 43 |
const usage = m.usage; then {m.text.length} {usage && <b>done</b>} |
39 | 43 |
In the two failing rows the stream carries the initial html event, with the list registered as a hole and given a digest, and then nothing more for it. In the browser the list stays as first rendered while the composer's seen prop keeps changing.
The same happens when the conditional is inside a plain helper function called from the hole, and with <For> in place of .map(). A condition that is a local identifier (generating && …, where const generating = m.status === "generating") is fine.
What I think is going on
Not established; a reading of the symptoms. The compiler memoises a conditional whose condition it cannot see through (a.b && …, a.b ? … : …). The existing chat example's comments say a live hole "can't mint components as it grows" (an owner-creation latch). A memo created inside a hole on each re-run looks like the same thing, with the difference that nothing reports it and the hole goes quiet instead.
Expected
Either the hole re-emits, or, if creating a memo inside a re-running hole is not supported, a dev diagnostic says so and names the expression. As it stands the markup is silently stale, and the workaround (hoist the property into a local) is not something a reader would guess.
Notes
- Found while rebuilding the chat example as a live server component. It first appeared as
{message.reasoning && <details>…}and{message.usage && <span>…}in a message's markup, and took a bisect of the raw stream to find. - Related, and loud rather than silent: a memo over an async source created per row inside the list (
createMemo(() => db.watchMessage(id))in a row component) fails the document render with<Loading> boundary discovery did not converge after 10001 passes. That one at least says something.
Environment
solidjs/solidnextatdf4e36516, dev build ofsolid-jsand@solidjs/webfrom the workspace@solidjs/router2.0.0-next.38,@solidjs/vite-plugin3.0.0-next.49, Vite 8, dev server- macOS, Node 26; the stream was read with curl, the page in Chromium 152
- Lenguaje dominante
- TypeScript
- Estrellas
- 36.1k
- Forks
- 1.1k
- Merge medio
- 11 h 32 min
- PR fusionados (30 d)
- 356
Preparar el entorno
- Sin Dockerfile ni archivo de Docker Compose
- Tiene una plantilla de pull request
- Leer la guía de contribución
Primeros pasos
- Lee el issue completo y luego la guía de contribución del proyecto.
- Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
- Haz un fork del repositorio y trabaja en una rama.
- Abre un pull request que haga referencia al número del issue.
Más de solidjs/solid
-
[2.0 / next] Error formatting and async iterator protocol failures escape memo error handlingAbierto
Dificultad 4/5 3-5 días Aptitud para principiantes 22/100
Los mantenedores suelen responder en 1 día
-
Dificultad 5/5 Más de una semana Aptitud para principiantes 14/100
Los mantenedores suelen responder en 1 día
-
Dificultad 4/5 3-5 días Aptitud para principiantes 35/100
Los mantenedores suelen responder en 1 día
-
Dificultad 4/5 3-5 días Aptitud para principiantes 18/100
Los mantenedores suelen responder en 1 día
-
[2.0 / next] Outside latest reads of a derivation expose errors and values at different timesAbierto
Dificultad 4/5 3-5 días Aptitud para principiantes 22/100
Los mantenedores suelen responder en 1 día
Todos los issues de solidjs/solid
Issues similares
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 74/100
Los mantenedores suelen responder en 1 día
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 74/100
Los mantenedores suelen responder en 1 día
-
bug category:ai category:devices effort:s
Dificultad 2/5 1-3 horas Aptitud para principiantes 72/100
LanternOps/breeze#8943 ·
Los mantenedores suelen responder en 1 día
-
bug
Dificultad 2/5 1-3 horas Aptitud para principiantes 74/100
-
area:reliability difficulty:easy help wanted points:100 Stellar Wave
Dificultad 2/5 1-3 horas Aptitud para principiantes 80/100
Divinity-Labs/Lens#341 · 1 comentario ·
Los mantenedores suelen responder en 1 día