Closes #484

We want Actor developers to be able to easily test their full permission
Actors with limited permissions without changing the Actor configuration
for everybody or redeploying the Actor under a different name. For that
reason, we have introduced a new forcePermissionLevel run option. This
PR adds the option to the client.

Full context here: apify/apify-core#22681

This has been already done in Apify JS client:
apify/apify-client-js#743

Co-authored-by: Vlada Dusek <[email protected]>

### Description

- Allow the possibility to define a custom public URL; if not used, the
default API public URL will be used.
- Python version of the following fix JS client fix
apify/apify-client-js#745

### Issues

- Closes: #496

### Testing

- Added new tests for relevant kvs and datasets clients' methods.

---------

Co-authored-by: Jindřich Bär <[email protected]>

### Description
- Add `KeyValueStoreClient.get_record_public_url`.
- Add `KeyValueStoreClientAsync.get_record_public_url`.
- Add tests.
### Issues

Closes: #497

Sending `1` and `0` is working, but the validator is complaining about
that.
Lets align with the JS client. [JS client uses `false`,
`true`](https://github.com/apify/apify-client-js/blob/v2.22.2/src/utils.ts)
[OpenAPI spec does not
define](https://spec.openapis.org/oas/v3.2.0.html#boolean-query-parameter-examples)
serialization of Booleans, but tools frequently choose `false/true`

### Issues:

Related to: apify/apify-docs#2286

This PR contains the following updates:

| Update | Change |
|---|---|
| lockFileMaintenance | All locks refreshed |

🔧 This Pull Request updates lock files to use the latest dependency
versions.

---

### Configuration

📅 **Schedule**: Branch creation - "before 4am on monday" (UTC),
Automerge - At any time (no schedule defined).

🚦 **Automerge**: Enabled.

♻ **Rebasing**: Whenever PR is behind base branch, or you tick the
rebase/retry checkbox.

👻 **Immortal**: This PR will be recreated if closed unmerged. Get
[config
help](https://redirect.github.com/renovatebot/renovate/discussions) if
that's undesired.

---

- [ ] <!-- rebase-check -->If you want to rebase/retry this PR, check
this box

---

This PR was generated by [Mend Renovate](https://mend.io/renovate/).
View the [repository job
log](https://developer.mend.io/github/apify/apify-client-python).

<!--renovate-debug:eyJjcmVhdGVkSW5WZXIiOiI0My42Ni40IiwidXBkYXRlZEluVmVyIjoiNDMuNjYuNCIsInRhcmdldEJyYW5jaCI6Im1hc3RlciIsImxhYmVscyI6W119-->

Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>

- Update Renovate and project settings to use only package versions
older than 24 hours
- Motivation is to add minimal defence to supply chain attacks.

…688)

)

## Summary

- Replace `logger.warning()` with `warnings.warn(DeprecationWarning)`
for deprecated params in `batch_add_requests()` (both sync and async) to
match the pattern used in `DatasetClient.download_items()`
- `DeprecationWarning` is the Python standard — visible by default,
caught by `pytest -W error::DeprecationWarning`, and filterable via
`warnings.filterwarnings()`
- Remove unused `logging` import

---

Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>

## Summary

- **Fix error handling examples** — `except Exception as ApifyApiError`
was catching all exceptions and shadowing the class name; now properly
uses `from apify_client.errors import ApifyApiError` with `except
ApifyApiError as err`
- **Fix retries page** — default retry count corrected from 8 to 4
(matches `DEFAULT_MAX_RETRIES`); updated outdated
`min_delay_between_retries_millis` to `min_delay_between_retries` (takes
`timedelta`)
- **Fix convenience methods page** — `start()` was described as waiting
for finish (it doesn't); added missing `wait_for_finish()` method
- **Fix nested client examples** — replaced `ty: ignore` comments with
proper `ActorJobStatus.SUCCEEDED` enum usage
- **Fix README** — updated Quick Start to typed attribute access
(`actor_call.default_dataset_id`), corrected retry count and parameter
name
- **Fix comment grammar** — "waits" → "wait" in code examples
- **Add `description` frontmatter** to all 20 doc pages for SEO
- **Standardize guide titles** to imperative form ("Pass input to an
Actor", "Integrate with data libraries", "Use HTTPX as the HTTP client")

---

Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>

…696)

### Description

Adds an `ownership` query parameter (`ownedByMe | sharedWithMe`) to the
`list()` method on `DatasetCollectionClient`,
`KeyValueStoreCollectionClient`, and `RequestQueueCollectionClient`
(both sync and async variants). This allows filtering listed storages by
whether they are owned by the user or shared with the user.

### Testing

Includes unit tests verifying the parameter is correctly passed through
to the API request for all three storage types.

…697)

## Summary

- Standardize opening paragraphs to follow "what it is + why it matters"
pattern
- Unify example intro phrasing to "The following example shows how
to..."
- Migrate all API reference links from raw markdown to `<ApiLink>`
component
- Trim verbose "benefits" paragraphs in pagination and streaming pages
- Add `:::tip` admonitions to convenience methods and pagination pages

---------

Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>

## Summary

- **Pagination page**: Added "Generator-based iteration" section
documenting `iterate_items()` as the recommended approach, with
sync/async code examples and a mention of `iterate_keys()` for key-value
stores.
- **Async support & Error handling pages**: Added closing notes with
`<ApiLink>` references to `ApifyClientAsync` and `ApifyApiError` so
pages no longer end abruptly after the code block.
- **Timeouts page** (new): Added `docs/02_concepts/11_timeouts.mdx`
covering the tiered timeout system (short/medium/long/no_timeout),
global configuration via constructor, per-call overrides with
`timedelta` or tier literals, and interaction with the retry system.

---------

Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>

## Summary

- Add `workflow_dispatch` workflow that regenerates Pydantic models from
an `apify-docs` PR's OpenAPI spec
- Builds the spec locally, runs `datamodel-codegen`, and opens a PR if
models changed
- Comments back on the source `apify-docs` PR with a link

## Issues

- Closes: #618

## Context

- apify/apify-docs#2381
- #694

---------

Co-authored-by: Claude Opus 4.6 <[email protected]>

## Summary

Fixes two bugs in the `manual_regenerate_models` workflow that caused
the [manual
run](https://github.com/apify/apify-client-python/actions/runs/24075156216/job/70221653926)
to fail:

- **Pre-commit hook blocked commit**: `git commit` triggered the
type-check hook, which failed because regenerated models removed symbols
(`ScheduleCreateActions`, `TaskOptions`) still imported by client code.
Added `commit: --no-verify` to bypass hooks — the resulting PR will get
proper CI checks anyway.
  - pre commit failures should be addressed manually
- **Push failed for new branch**: `push: --force` ran `git push --force`
without setting upstream tracking. Changed to `push: -u origin $BRANCH
--force` so the new branch is correctly pushed.

---------

Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>

## Summary

- Expanded the v3 upgrade guide to document all 6 breaking changes from
the changelog: Python version support, fully typed clients (Pydantic
models for responses and parameters), pluggable HTTP client
architecture, and tiered timeout system.
- Converted both upgrading guides (`upgrading_to_v2`, `upgrading_to_v3`)
from `.md` to `.mdx` and replaced plain markdown API references with
`ApiLink` components for consistent linking to the API reference docs.
- Fixed the guide link to use the correct `custom-http-client-httpx` id.

---------

Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>

## Summary

- Add `generate-models-from-file` poe task that accepts a local file
path, working around `datamodel-codegen` ignoring `--input` when `url`
is set in `pyproject.toml`
- Update CI workflow to call poe tasks instead of `datamodel-codegen`
directly, ensuring the downloaded artifact from apify-docs PRs is
actually used

---------

Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>

This PR updates the auto-generated Pydantic models based on OpenAPI specification changes in [apify-docs PR #2398](apify/apify-docs#2398).

## Summary

- Regenerated `_models.py` from updated OpenAPI spec
- Adapted resource clients to renamed/restructured models:
  - `TaskOptions` → `OptionalRunOptions`
  - `ScheduleCreateActions` → discriminated union in `ScheduleCreate`
  - `EnvVar` → `EnvVarRequest` for create/update requests

---------

Co-authored-by: Vlada Dusek <[email protected]>

This PR updates the auto-generated Pydantic models from the published OpenAPI specification.

## Changes

- Regenerated `src/apify_client/_models.py` using `datamodel-codegen`

## Summary

- Add reviewer, assignee, and label to auto-created PRs
- Add `@docs_group('Models')` post-processing step for API docs
- Replace force push with incremental commits to preserve reviewer changes
- Simplify PR description body

### Issues

- Closes: #642

---------

Co-authored-by: Claude Opus 4.6 (1M context) <[email protected]>