docs: four sentences say history snapshots stay complete inside unobserved:, and gather refuses a buffer index on a list matrix
Les mainteneurs répondent en général sous 1 jour
Personne n'a encore pris cette issue.
Évaluation
- Difficulté
- 4/5
- Temps estimé
- 3-5 jours
- Accessibilité débutants
- 55/100
- Type d'issue
- Documentation
- Clarté
- Clairement spécifiée
- Activité
- Active
- Stack technique
- c
- Domaine
- documentation, testing
Piste de recherche
Read RULING R31-2 and the four cited passages in README.md, docs/COMPARISON.md, docs/llms.txt, and docs/PREDICATES.md; compare them with the example in the issue. For gather, decide whether a buffer index vector with a nested list matrix is accepted, then update docs/BUILTINS.md or add a row under tests/phase2/ accordingly. Run tests/test_doc_examples.py and tools/docs_claims_check.sh; done means the documentation matches the ruling and the gather behavior is recorded.
Rédigé par le modèle d'indexation à partir du texte de l'issue.
Description
Found by the blind review of #2468 and by the runs of its fix round, outside the list of #2466. Both were seen on phase2-numbers at 05e6d6fd.
1. Four documents say history snapshots stay complete inside unobserved:; the runtime and its ruling say they are not written
README.md(near line 438): "Numeric relative/raw steps, exact flags, running-best state and history snapshots still update inside the block."docs/COMPARISON.md(near line 791): "Numeric samples, interrogation/history snapshots and D2 state remain complete"docs/llms.txt(near line 315): "Numeric samples, exact flags, running-best state and interrogation snapshots remain complete"docs/PREDICATES.md(near line 475): "Numeric value records remain complete, including exact flags, D2 state, the interrogation triple, and history snapshots."
The record: RULING R31-2 on #1758, as the fragment changes/fixed/1758-r312-runaway-contract.md states it: "Elided numeric samples inside unobserved: retain live value and running-best updates without writing line-qualified observer-history snapshots".
Shows it:
x is 3
unobserved:
x is 4
print of [when is x, where is x, why is x, how is x]
print of [where is x at 3, why is x at 3, how is x at 3]
prints [2, 4, 1, 0.00398406374501992] and then [null, null, null]. The same program with if true: in place of unobserved: prints [4, 1, 0.00398406374501992] on its second line. The live interrogatives and trajectory of x are the same in both.
So the live answers after the block are as the documents say, and the line-qualified ones at an elided line are not there. The four sentences are stale against the ruling.
2. gather refuses a buffer index vector when the matrix is a nested list
nl is [[1.0, 2.0], [3.0, 4.0]]
sb is reshape of [buf_from_list of ([1.0, 2.0, 3.0, 4.0]), 2, 2]
ib is buf_from_list of ([0.0, 1.0])
print of (type of (gather of [nl, [0, 1]]))
print of (type of (gather of [sb, [0, 1]]))
print of (type of (gather of [sb, ib]))
print of (type of (gather of [nl, ib]))
prints list, buffer, buffer, and then raises type_mismatch: gather: expected a tensor and an index or index vector. The row in docs/BUILTINS.md says "matrix may be a shaped buffer and indices a list or a buffer"; the _rows/_cols builtins take a buffer index vector with either container. Either the fourth form is accepted (it then returns a list, as every mixed call does), or the row says that a buffer index vector needs a buffer matrix.
Done when
- The four sentences of item 1 say what the ruling says (live value, running best and the live interrogatives are kept; line-qualified snapshots at an elided line are not written), each next to a run of the program above.
- Item 2 is decided on this issue (accept, or document), and the decision is in the tree: a row of
tests/phase2/for the fourth form if it is accepted, or the row's sentence if it is not. -
tests/test_doc_examples.pyandtools/docs_claims_check.shpass on the changed documents.
The bool case of the rows that name a lenient answer is #2409, not this issue.
— Claude DevBox
- Langage dominant
- C
- Étoiles
- 3
- Forks
- 7
- Merge moyen
- 3 h 28 min
- PR mergées (30 j)
- 270
Préparer son environnement
Lance le conteneur de développement du projet dans votre navigateur, avec votre propre compte GitHub.
- Fournit un Dockerfile ou un fichier Docker Compose
- Propose un modèle de pull request
- Lire le guide de contribution
Par où commencer
- Lisez l'issue en entier, puis le guide de contribution du projet.
- Signalez en commentaire que vous la prenez — cela évite que deux personnes fassent le même travail.
- Forkez le dépôt et travaillez sur une branche.
- Ouvrez une pull request qui référence le numéro de l'issue.
Autres issues de InauguralSystems/EigenScript
-
area:gates kind:gate-defect
Difficulté 2/5 1-3 heures Accessibilité débutants 83/100
InauguralSystems/EigenScript#2476 ·
Les mainteneurs répondent en général sous 1 jour
-
area:gates needs-triage
Difficulté 2/5 1-3 heures Accessibilité débutants 78/100
InauguralSystems/EigenScript#2462 · 1 commentaire ·
Les mainteneurs répondent en général sous 1 jour
-
The hosted-RNG check refuses a clean tree that has a tracked file with a double quote in its nameOuvertearea:gates kind:gate-defect needs-triage
Difficulté 2/5 1-3 heures Accessibilité débutants 78/100
InauguralSystems/EigenScript#2461 · 1 commentaire ·
Les mainteneurs répondent en général sous 1 jour
-
area:gates kind:gate-defect needs-triage
Difficulté 2/5 1-3 heures Accessibilité débutants 86/100
InauguralSystems/EigenScript#2458 · 1 commentaire ·
Les mainteneurs répondent en général sous 1 jour
-
`http_post`: `headers` text with something after its JSON value now means no headers, without a wordOuvertearea:http kind:silent-wrong
Difficulté 2/5 1-3 heures Accessibilité débutants 72/100
InauguralSystems/EigenScript#2394 · 1 commentaire ·
Les mainteneurs répondent en général sous 1 jour
Toutes les issues de InauguralSystems/EigenScript
Issues similaires
-
Difficulté 2/5 1-3 heures Accessibilité débutants 62/100
Les mainteneurs répondent en général sous 1 jour
-
Difficulté 1/5 Moins d'une heure Accessibilité débutants 92/100
flipperdevices/flipperzero-firmware#4520 ·
Les mainteneurs répondent en général sous 1 jour
-
area/build-framework kind/enhancement priority/low
Difficulté 2/5 1-3 heures Accessibilité débutants 77/100
yugabyte/yugabyte-db#38826 ·
Les mainteneurs répondent en général sous 1 jour
-
Difficulté 1/5 Moins d'une heure Accessibilité débutants 88/100
meshcore-dev/MeshCore#3891 · 1 commentaire ·
Les mainteneurs répondent en général sous 1 jour
-
Bug
Difficulté 2/5 1-3 heures Accessibilité débutants 66/100
vlang/v#32675 · 1 commentaire ·
Les mainteneurs répondent en général sous 1 jour