Hacktoberfest 2026 : les issues que les mainteneurs ont marquées pour octobre, ouvertes et accessibles aux débutants. Parcourir les issues Hacktoberfest

json-type tutorial teaches client-side filtering as a property of JSON

Ouverte
  • Quelque chose ne va pas ?
#292 0 commentaires 0 réactions 0 personnes assignées Voir sur GitHub

Les mainteneurs répondent en général sous 4 jours

Personne n'a encore pris cette issue.

Évaluation

Difficulté
3/5
Temps estimé
1-2 jours
Accessibilité débutants
45/100
Type d'issue
Documentation
Clarté
Clairement spécifiée
Activité
Active
Stack technique
jupyter-notebook, python
Domaine
documentation

Piste de recherche

This work is in src/tutorials/advanced/json-type.ipynb and is blocked on datajoint/datajoint-python#1564. After that issue merges, read the filtering reference added in #291, then update the “Filtering on JSON Content” section and Design Guidelines table. Re-execute the notebook so its committed outputs match, and cross-link the reference.

Rédigé par le modèle d'indexation à partir du texte de l'issue.

Description

documentation

Blocked on datajoint/datajoint-python#1564. Deliberately deferred so the tutorial is not rewritten twice.

The problem

src/tutorials/advanced/json-type.ipynb is the page a user opens to learn how to work with JSON. Its filtering section reads:

Filtering on JSON Content

Fetch then filter in Python:

followed by pulling the whole table into a list comprehension:

calibrated = [
    e for e in Equipment.to_dicts()
    if e['specs'] and e['specs'].get('calibrated')
]

and its Design Guidelines table lists this as an inherent property of the type:

JSON Normalized Tables
Filter in Python Filter in SQL

Why it is wrong

Server-side filtering works, on both backends. Equipment & {"specs.vendor": "Acme"} is translated to json_value() on MySQL and jsonb_extract_path_text() on PostgreSQL; ordering comparisons work through a typed projection, proj(ch="specs.channels:unsigned") & "ch > 32". The syntax is documented in #291.

So the tutorial teaches a full table scan into Python for something the database does, and presents it as a limitation of JSON rather than a choice.

Why it probably says that

The tutorial's own example is specs.calibrated, a boolean — and & {"specs.calibrated": True} returns no rows on MySQL and raises on PostgreSQL (datajoint-python#1564). Whoever wrote the section likely tried exactly that, got nothing back, and concluded JSON cannot be filtered server-side.

That is why this waits: writing the section against the string workaround ({"specs.calibrated": "true"}) would teach an idiom that becomes wrong as soon as #1564 lands.

When #1564 merges

  1. Rewrite Filtering on JSON Content to filter server-side, keeping one short client-side example for genuinely non-SQL-expressible predicates.
  2. Correct the Design Guidelines table — "Filter in Python" is not a property of JSON. The honest contrast is about indexing and type enforcement, not about where filtering happens.
  3. Cross-link the reference section added in #291.
  4. Re-execute the notebook: these are code cells with committed outputs.
Langage dominant
Jupyter Notebook
Étoiles
7
Forks
12
Merge moyen
3 j 16 h
PR mergées (30 j)
15

Préparer son environnement

  • Fournit un Dockerfile ou un fichier Docker Compose
  • Aucun modèle de pull request
  • Aucun guide de contribution

Par où commencer

  1. Lisez l'issue en entier, puis le guide de contribution du projet.
  2. Signalez en commentaire que vous la prenez — cela évite que deux personnes fassent le même travail.
  3. Forkez le dépôt et travaillez sur une branche.
  4. Ouvrez une pull request qui référence le numéro de l'issue.

Autres issues de datajoint/datajoint-docs

Toutes les issues de datajoint/datajoint-docs

Issues similaires

Plus d'issues Documentation

Recevez les nouvelles issues par e-mail

Un résumé court des issues GitHub adaptées aux débutants.