SQLAlchemy Migration Guides

How to replace string loader options in SQLAlchemy 2.0

Replace joinedload("addresses") and similar string loader options with mapped attributes when the relationship root is obvious.

Start here

The safe first move

For straightforward paths, replace the string with the mapped attribute, like joinedload(User.addresses).

Stop before automation when: Dotted paths like joinedload("orders.items"). contains_eager and duplicate-join interactions.

Target shape

query = session.query(User).options(joinedload(User.addresses))
query = query.options(selectinload(User.orders))

Repo fit check

Want to know if your repo has this pattern?

Run the matching scanner locally first. Read the free findings and manual alternatives; consider paid cleanup only if supported patterns repeat enough to justify the price.

  • Local scan; no repository upload.
  • Supported findings are separated from manual-review findings.
  • Buy only when the repeated pattern is worth automating.
Run the SQLAlchemy scan first View example report See when to buy
Example scan summary
supported_findings: 38
manual_review_findings: 6
files_uploaded: 0
confidence: reviewable

Why this page exists

How to replace string loader options in SQLAlchemy 2.0

Simple loader strings are cleanup work. Dotted paths and query-shape-dependent options are not.

Direct answer

What changes

For straightforward paths, replace the string with the mapped attribute, like joinedload(User.addresses).

Example

Before and after

Before

query = session.query(User).options(joinedload("addresses"))
query = query.options(selectinload("orders"))

After

query = session.query(User).options(joinedload(User.addresses))
query = query.options(selectinload(User.orders))

Decision path

If this repeats across the repo, open the evaluation pages next.

Start with the free scan, inspect the report, and only move to the cleanup pack if the repo has repeated supported patterns that are still expensive to fix by hand.

  • The free scan shows repeated Query.get, select([..]), string join, string loader, declarative import, or DML constructor findings.
  • The remaining supported cleanup is still expensive enough to justify a controlled local migration workflow.
  • If the scan output is ambiguous, review its findings and upstream guidance. Do not buy expecting unsupported cases to be resolved.

Typical symptoms

  • Loader options still depend on legacy string paths.
  • The easy cases are repetitive but not conceptually hard.
  • Teams need a clean stop signal before dotted or aliased paths.

What the product covers

  • Simple joinedload/lazyload/selectinload/subqueryload string cleanup.
  • Mapped-attribute replacements when the root entity is obvious.
  • Cross-links to the dotted-path and joinedload_all manual buckets.

Manual-review boundary

  • Dotted paths like joinedload("orders.items").
  • contains_eager and duplicate-join interactions.
  • Cases where the owning entity is not obvious from the local code.

FAQ

Fast answers before you decide

What is the safest fix for joinedload string path sqlalchemy 2.0?

For straightforward paths, replace the string with the mapped attribute, like joinedload(User.addresses).

Can joinedload string path sqlalchemy 2.0 be automated?

Automate only the supported subset: Simple joinedload/lazyload/selectinload/subqueryload string cleanup. Mapped-attribute replacements when the root entity is obvious. Cross-links to the dotted-path and joinedload_all manual buckets. Keep these cases in manual review: Dotted paths like joinedload("orders.items"). contains_eager and duplicate-join interactions. Cases where the owning entity is not obvious from the local code.

When should I buy SQLAlchemy 1.4 to 2.0 Migration Cleanup Pack?

Use SQLAlchemy 1.4 to 2.0 Migration Cleanup Pack only when this pattern repeats across enough files that manual cleanup is still costly. Use the free report and upstream guide to evaluate fit. Buy only if repeated supported edits justify the price; no paid assessment is required.

Purchase fit

Choose the lowest-risk next step.

One matching page is not enough by itself. Run the local scan, then buy only when the same supported pattern appears often enough to matter.

Buy if

  • The free scan shows repeated Query.get, select([..]), string join, string loader, declarative import, or DML constructor findings.
  • The remaining supported cleanup is still expensive enough to justify a controlled local migration workflow.
  • The team can run the migration on a branch and validate locally before merge.

What the workflow includes

  • Installable local CLI workflow, not a hosted black box.
  • Preview/apply modes, diff output, and JSON migration report.
  • Coverage notes, rollback checklist, manager summary, and buyer terms.

Before checkout

  • Stripe Checkout handles secure payment and receipts.
  • Download your purchased ZIP after payment.
  • Runs locally; no hosted API, repo upload, or production credentials needed.
  • The free report includes the evidence needed to evaluate fit. Paid summary and template add-ons are optional.
  • 14-day refund review for published-scope or delivery mismatches.