dvt-dask 2026.3.11__tar.gz

dvt_dask-2026.3.11/.codex/config.toml

@@ -0,0 +1,18 @@
+[mcp_servers.project_mcp]
+command = "python"
+args = [
+    "-m",
+    "scripts.services.run_project_mcp"
+]
+
+[mcp_servers.pycharm]
+command = "npx"
+args = [
+	"mcp-remote",
+	"http://localhost:64342/sse",
+	"--header",
+	"IJ_MCP_SERVER_PROJECT_PATH: ${PROJECT_PATH}"
+]
+startup_timeout_sec = 20
+tool_timeout_sec = 45
+enabled = true

dvt_dask-2026.3.11/.codex/skills/experiment-mapping-lab/SKILL.md

@@ -0,0 +1,64 @@
+---
+name: experiment-mapping-lab
+description: Run iterative experiments for cross-source event matching accuracy in arb_monitor. Use when the user asks to plan, implement, execute, and document a new mapping experiment between sources (especially polymarket and bcgame), including updates to tmp/experiments/* README files and INSIGHTS.md.
+---
+
+# Experiment Mapping Lab
+
+Execute the full experiment cycle end-to-end with precision-first defaults.
+
+## Core workflow
+
+1. Read `tmp/experiments/README.md`, `tmp/experiments/INSIGHTS.md`, and the latest 1-3 experiment package `README.md` files.
+2. Recheck at least one previous experiment on new data:
+- read sources and `parsed_at` listed in that experiment `README.md`;
+- query current `MAX(parsed_at)` for the same sources in DB;
+- if current `MAX(parsed_at)` is newer, rerun that previous experiment on the new snapshot.
+3. If rerun shows divergences or precision drop:
+- write root cause notes in `## Дополнительно` of that previous experiment `README.md`;
+- add transferable findings to `tmp/experiments/INSIGHTS.md`.
+4. Determine the next package name: `tmp/experiments/{experiment_number:04}_{experiment_name}`.
+5. Create the new package and write initial `README.md` with:
+- hypothesis
+- difference from previous experiment
+- method
+- run command
+- sources with exact `parsed_at`
+- planned metrics
+6. Implement experiment code inside that package (`experiment.py` by default).
+7. Run the experiment on current data.
+8. Save artifacts in the package (`results.json`, `matched_pairs.csv`, and optional diagnostics CSV).
+9. Update the package `README.md` with actual numbers, findings, limitations, exact sources+`parsed_at`, and next-step proposal.
+10. Add transferable learnings to `tmp/experiments/INSIGHTS.md`.
+
+## Precision-first defaults
+
+- Use latest snapshot per source from `market_items` (`MAX(parsed_at)` grouped by `source`).
+- Restrict candidate pool by canonical sport and start-time window.
+- Prefer one-to-one mapping (`mutual best`) before lowering thresholds.
+- Report both coverage and risk indicators:
+- accepted matches count and share per source
+- threshold/margin values used
+- ambiguity indicators (top1-top2 gap)
+- manual sanity sample (at least 10-20 pairs when no ground truth)
+
+## Repository-specific rules
+
+- Keep each experiment isolated; do not edit old experiment code unless explicitly requested.
+- Keep numbering monotonic (`0001`, `0002`, ...).
+- Keep all experiment files under `tmp/experiments`.
+- If `tmp/experiments/README.md` or `INSIGHTS.md` is missing, initialize both before starting a new experiment.
+
+## Helper script
+
+Use `scripts/create_experiment.py` in this skill to scaffold the next experiment package quickly.
+
+Example:
+```powershell
+python .codex/skills/experiment-mapping-lab/scripts/create_experiment.py --repo-root C:\dev\projects\arb_monitor --experiment-name sport-time-title-v2
+```
+
+The script creates:
+- package directory with next numeric prefix
+- `__init__.py`
+- starter `README.md`

dvt_dask-2026.3.11/.codex/skills/experiment-mapping-lab/agents/openai.yaml

@@ -0,0 +1,4 @@
+interface:
+  display_name: "Experiment Mapping Lab"
+  short_description: "Iterative event-matching experiment runner"
+  default_prompt: "Use $experiment-mapping-lab to run the next event-mapping experiment and document results in tmp/experiments."

dvt_dask-2026.3.11/.codex/skills/experiment-mapping-lab/scripts/create_experiment.py

@@ -0,0 +1,111 @@
+#!/usr/bin/env python3
+from __future__ import annotations
+
+import argparse
+import re
+from pathlib import Path
+
+EXPERIMENT_DIR_PATTERN = re.compile(r"^(\d{4})_(.+)$")
+
+README_TEMPLATE = """# {title}
+
+## Hypothesis
+Describe what should improve and why.
+
+## Difference From Previous Experiment
+- Describe concrete changes vs previous experiment.
+
+## Method
+- Data scope:
+- Candidate generation:
+- Scoring or rules:
+- Acceptance criteria:
+
+## Sources And Parsed At
+- source: `parsed_at`
+
+## Run Command
+```powershell
+python tmp/experiments/{package}/experiment.py
+```
+
+## Artifacts
+- `results.json`
+- `matched_pairs.csv`
+- `top2_candidates.csv` (optional)
+
+## Results
+Fill after execution.
+
+## Additional
+- Previous experiment recheck on newer data:
+- Divergences/precision drop analysis (if any):
+"""
+
+
+def find_next_number(experiments_dir: Path) -> int:
+    max_num = 0
+    for child in experiments_dir.iterdir():
+        if not child.is_dir():
+            continue
+        match = EXPERIMENT_DIR_PATTERN.match(child.name)
+        if not match:
+            continue
+        max_num = max(max_num, int(match.group(1)))
+    return max_num + 1
+
+
+def normalize_name(name: str) -> str:
+    cleaned = re.sub(r"[^a-zA-Z0-9\s_]", "", name).strip().lower()
+    cleaned = re.sub(r"[\s\-]+", "_", cleaned)
+    cleaned = re.sub(r"_+", "_", cleaned)
+    if not cleaned:
+        raise ValueError("Experiment name is empty after normalization")
+    return cleaned
+
+
+def create_package(repo_root: Path, experiment_name: str, dry_run: bool) -> Path:
+    experiments_dir = repo_root / "tmp" / "experiments"
+    experiments_dir.mkdir(parents=True, exist_ok=True)
+
+    number = find_next_number(experiments_dir)
+    slug = normalize_name(experiment_name)
+    package_name = f"{number:04}_{slug}"
+    package_dir = experiments_dir / package_name
+
+    if dry_run:
+        return package_dir
+
+    package_dir.mkdir(parents=False, exist_ok=False)
+
+    init_text = f'"""Experiment package: {package_name}."""\n'
+    (package_dir / "__init__.py").write_text(init_text, encoding="utf-8")
+
+    readme_title = f"{number:04} {slug.replace('_', ' ').title()}"
+    readme_text = README_TEMPLATE.format(title=readme_title, package=package_name)
+    (package_dir / "README.md").write_text(readme_text, encoding="utf-8")
+
+    return package_dir
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="Create the next experiment package in tmp/experiments")
+    parser.add_argument("--repo-root", required=True, help="Repository root path")
+    parser.add_argument("--experiment-name", required=True, help="Experiment name slug or phrase")
+    parser.add_argument("--dry-run", action="store_true", help="Only print target directory")
+    return parser.parse_args()
+
+
+def main() -> None:
+    args = parse_args()
+    repo_root = Path(args.repo_root).resolve()
+    package_dir = create_package(
+        repo_root=repo_root,
+        experiment_name=args.experiment_name,
+        dry_run=args.dry_run,
+    )
+    print(str(package_dir))
+
+
+if __name__ == "__main__":
+    main()

dvt_dask-2026.3.11/.git-blame-ignore-revs

@@ -0,0 +1,24 @@
+# .git-blame-ignore-revs
+# absolufy-imports - No relative - PEP8 (#8796)
+cccb9d8d8e33a891396b1275c2448c352ef40c27
+
+# Update `pre-commit` version (#8691)
+510bbc380531cbf56a409f1ae68e6fd84a9599e6
+
+# Run pyupgrade in CI (#8246)
+80a82008d5b02a08f6ff59d802defcc43247eb1a
+
+# Bump pre-commit hook versions (#7676)
+d6bbbb08c92652eae2820e93edc2f3fe502391d3
+
+# Start adding isort (#7370)
+a31c0fc72e1cc59b8b0254965824abb0718c5f56
+
+# Rerun with latest black release (#6568)
+64e2a9b3b9992503221a074a547827501927d1fa
+
+# LINT: Fixup black string normalization (#5227)
+d92f4015a1da3da10c04c682ed2acae8469e9576
+
+# Apply Black formatting (#4983)
+7e4beffb339c69278091d4e305c2ae18ddf8c74f

dvt_dask-2026.3.11/.github/CONTRIBUTING.md

@@ -0,0 +1,2 @@
+See [developer documentation](https://docs.dask.org/en/latest/develop.html)
+for tips on how to get started.

dvt_dask-2026.3.11/.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,3 @@
+- [ ] Closes #xxxx
+- [ ] Tests added / passed
+- [ ] Passes `pre-commit run --all-files`

dvt_dask-2026.3.11/.github/dependabot.yml

@@ -0,0 +1,9 @@
+# Set update schedule for GitHub Actions
+
+version: 2
+updates:
+  - package-ecosystem: "github-actions"
+    directory: "/"
+    schedule:
+      # Check for updates to GitHub Actions every weekday
+      interval: "weekly"

dvt_dask-2026.3.11/.github/release-drafter.yml

@@ -0,0 +1,13 @@
+# These will be overridden by the publish workflow and set to the new tag
+name-template: 'Next Release'
+tag-template: 'next'
+
+change-template: '- $TITLE @$AUTHOR (#$NUMBER)'
+change-title-escapes: '\<*_&' # You can add # and @ to disable mentions, and add ` to disable code blocks.
+
+template: |
+  ## Changes
+
+  $CHANGES  
+
+  See the [Changelog](https://docs.dask.org/en/stable/changelog.html) for more information.

dvt_dask-2026.3.11/.github/release.yml

@@ -0,0 +1,26 @@
+# .github/release.yml
+
+changelog:
+  categories:
+    - title: New Features
+      labels:
+        - feature
+    - title: Enhancements
+      labels:
+        - enhancement
+    - title: Bug Fixes
+      labels:
+        - bug
+    - title: Deprecations
+      labels:
+        - deprecation
+    - title: Documentation
+      labels:
+        - documentation
+    - title: Maintenance
+      labels:
+        - tests
+        - hygiene
+    - title: Misc
+      labels:
+        - "*"

dvt_dask-2026.3.11/.github/workflows/additional.yml

@@ -0,0 +1,32 @@
+name: Additional
+
+on: [push, pull_request]
+
+# Required shell entrypoint to have properly activated conda environments
+defaults:
+  run:
+    shell: bash -l {0}
+
+jobs:
+  doctest:
+    runs-on: "ubuntu-latest"
+    timeout-minutes: 90
+    steps:
+      - name: Checkout source
+        uses: actions/checkout@v6
+
+      - name: Setup Conda Environment
+        uses: conda-incubator/setup-miniconda@v3.3.0
+        with:
+          miniforge-version: latest
+          use-mamba: true
+          channel-priority: strict
+          environment-file: continuous_integration/environment-3.12.yaml
+          activate-environment: test-environment
+          auto-activate-base: false
+
+      - name: Install
+        run: source continuous_integration/scripts/install.sh
+
+      - name: Run tests
+        run: pytest -v --doctest-modules --ignore-glob='*/test_*.py' --ignore='dask/array/_array_expr/' dask

dvt_dask-2026.3.11/.github/workflows/conda.yml

@@ -0,0 +1,64 @@
+name: Conda build
+on:
+  push:
+    branches:
+      - main
+    tags:
+      - "*"
+  pull_request:
+    paths:
+      - setup.py
+      - continuous_integration/recipe/**
+      - .github/workflows/conda.yml
+      - pyproject.toml
+
+# When this workflow is queued, automatically cancel any previous running
+# or pending jobs from the same branch
+concurrency:
+  group: conda-${{ github.ref }}
+  cancel-in-progress: true
+
+# Required shell entrypoint to have properly activated conda environments
+defaults:
+  run:
+    shell: bash -l {0}
+
+jobs:
+  conda:
+    name: Build (and upload)
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+        with:
+          fetch-depth: 0
+      - name: Set up Python
+        uses: conda-incubator/setup-miniconda@v3.3.0
+        with:
+          miniforge-version: latest
+          use-mamba: true
+          python-version: 3.9
+          channel-priority: strict
+      - name: Install dependencies
+        run: |
+          mamba install -c conda-forge boa conda-verify
+
+          which python
+          pip list
+          mamba list
+      - name: Build conda package
+        run: |
+          # suffix for nightly package versions
+          export VERSION_SUFFIX=a`date +%y%m%d`
+
+          conda mambabuild continuous_integration/recipe \
+                           --no-anaconda-upload \
+                           --output-folder .
+      - name: Upload conda package
+        if: github.event_name == 'push' && github.repository == 'dask/dask'
+        env:
+          ANACONDA_API_TOKEN: ${{ secrets.DASK_CONDA_TOKEN }}
+        run: |
+          # install anaconda for upload
+          mamba install -c conda-forge anaconda-client
+
+          anaconda upload --label dev noarch/*.tar.bz2

dvt_dask-2026.3.11/.github/workflows/label-all.yml

@@ -0,0 +1,14 @@
+name: "Issue and PR Labeler"
+on:
+  pull_request:
+    types: [opened]
+  issues:
+    types: [opened, reopened]
+jobs:
+  label-all-on-open:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: andymckay/labeler@1.0.4
+        with:
+          add-labels: "needs triage"
+          ignore-if-labeled: false

dvt_dask-2026.3.11/.github/workflows/pre-commit.yml

@@ -0,0 +1,18 @@
+name: Linting
+
+on:
+  push:
+    branches: main
+  pull_request:
+    branches: main
+
+jobs:
+  checks:
+    name: pre-commit hooks
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+      - uses: actions/setup-python@v6
+        with:
+          python-version: '3.11'
+      - uses: pre-commit/action@v3.0.1

dvt_dask-2026.3.11/.github/workflows/publish-test-results.yaml

@@ -0,0 +1,50 @@
+# Copied from https://github.com/EnricoMi/publish-unit-test-result-action/blob/v1.23/README.md#support-fork-repositories-and-dependabot-branches
+# Warning: changes to this workflow will NOT be picked up until they land in the main branch!
+# See https://docs.github.com/en/actions/using-workflows/events-that-trigger-workflows#workflow_run
+
+name: Publish test results
+
+on:
+  workflow_run:
+    workflows: [Tests]
+    types: [completed]
+
+jobs:
+  publish-test-results:
+    name: Publish test results
+    runs-on: ubuntu-latest
+    if: github.event.workflow_run.conclusion != 'skipped'
+
+    # Needed to post comments on the PR
+    permissions:
+      checks: write
+      pull-requests: write
+
+    steps:
+      - name: Download and extract artifacts
+        env:
+          GITHUB_TOKEN: ${{secrets.GITHUB_TOKEN}}
+        run: |
+          mkdir artifacts && cd artifacts
+
+          artifacts_url=${{ github.event.workflow_run.artifacts_url }}
+
+          gh api --paginate "$artifacts_url" -q '.artifacts[] | [.name, .archive_download_url] | @tsv' | while read artifact
+          do
+            IFS=$'\t' read name url <<< "$artifact"
+            gh api $url > "$name.zip"
+            unzip -d "$name" "$name.zip"
+          done
+
+      - name: Publish Unit Test Results
+        uses: EnricoMi/publish-unit-test-result-action@v2
+        with:
+          check_name: Unit Test Results
+          comment_title: |
+            Unit Test Results
+            _See [test report](https://dask.github.io/dask/test_report.html) for an extended history of previous test failures. This is useful for diagnosing flaky tests._
+          commit: ${{ github.event.workflow_run.head_sha }}
+          event_file: artifacts/Event File/event.json
+          event_name: ${{ github.event.workflow_run.event }}
+          junit_files: artifacts/**/*.xml
+          fail_on: "errors"

dvt_dask-2026.3.11/.github/workflows/release-drafter.yml

@@ -0,0 +1,32 @@
+name: Release Drafter
+
+on:
+  push:
+    branches:
+      - main
+
+permissions:
+  contents: read
+
+jobs:
+  update_release_draft:
+    if: github.repository == 'dask/dask'
+    permissions:
+      # Write permission is required to create a GitHub release
+      contents: write
+      pull-requests: read
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+        with:
+          fetch-depth: 0  # Required to get tag history
+      - name: Check if release commit
+        id: check_release_commit
+        run: git describe --exact-match --tags $(git rev-parse HEAD)
+        continue-on-error: true
+      - uses: release-drafter/release-drafter@v6
+        if: steps.check_release_commit.outcome != 'success'
+        with:
+          disable-autolabeler: true
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

dvt_dask-2026.3.11/.github/workflows/release-publish.yml

@@ -0,0 +1,33 @@
+name: Release Publisher
+
+on:
+  push:
+    tags:
+      - "*.*.*"
+
+permissions:
+  contents: read
+
+jobs:
+  publish_release:
+    if: github.repository == 'dask/dask'
+    permissions:
+      # Write permission is required to publish a GitHub release
+      contents: write
+      pull-requests: read
+    runs-on: ubuntu-latest
+    steps:
+      - name: Set version env
+        # Use a little bit of bash to extract the tag name from the GitHub ref
+        run: echo "RELEASE_VERSION=${GITHUB_REF#refs/*/}" >> $GITHUB_ENV
+      - uses: release-drafter/release-drafter@v6
+        with:
+          disable-autolabeler: true
+          # Override the Release name/tag/version with the actual tag name
+          name: ${{ env.RELEASE_VERSION }}
+          tag: ${{ env.RELEASE_VERSION }}
+          version: ${{ env.RELEASE_VERSION }}
+          # Publish the Release
+          publish: true
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

dvt_dask-2026.3.11/.github/workflows/stale-bot.yaml

@@ -0,0 +1,20 @@
+name: 'Label stale issues and PRs'
+on:
+  schedule:
+    - cron: '30 1 * * 1' # runs once a week
+
+jobs:
+  stale:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/stale@v10
+        with:
+          stale-issue-message: '' # no comment left if string is empty
+          stale-pr-message: '' # no comment left if string is empty
+          days-before-stale: 30
+          days-before-close: -1
+          stale-issue-label: 'needs attention'
+          stale-pr-label: 'needs attention'
+          exempt-issue-labels: 'good intro to dask,good first issue,Good First Issue,good second issue,feature request'
+          exempt-draft-pr: true
+          start-date: '2020-04-18T00:00:00Z' # ignore before this date, ISO 8601 or RFC 2822

dvt_dask-2026.3.11/.github/workflows/test-report.yaml

@@ -0,0 +1,66 @@
+name: Test Report
+
+on:
+  schedule:
+    # Run 2h after the daily tests.yaml
+    - cron: "0 8,20 * * *"
+  workflow_dispatch:
+
+permissions:
+  contents: write
+
+jobs:
+  test-report:
+    name: Test Report
+    # Do not run the report job on forks
+    if: github.repository == 'dask/dask' || github.event_name == 'workflow_dispatch'
+    runs-on: ubuntu-latest
+    env:
+      GITHUB_TOKEN: ${{ github.token }}
+    defaults:
+      run:
+        shell: bash -l {0}
+    steps:
+      - uses: actions/checkout@v6
+        with:
+          repository: dask/distributed
+
+      - name: Setup Conda Environment
+        uses: conda-incubator/setup-miniconda@v3.3.0
+        with:
+          miniforge-version: latest
+          condarc-file: continuous_integration/condarc
+          use-mamba: true
+          # Note: this file is in the dask/distributed repo
+          environment-file: continuous_integration/scripts/test-report-environment.yml
+          activate-environment: dask
+
+      - name: Show conda options
+        run: conda config --show
+
+      - name: mamba list
+        run: mamba list
+
+      - uses: actions/cache@v5
+        id: cache
+        with:
+          # Suffix is depending on the backend / OS. Let's be agnostic here
+          # See https://docs.python.org/3/library/shelve.html#shelve.open
+          path: |
+            test_report*
+            !test_report.html
+          # Note: these files are in the dask/distributed repository
+          key: ${{ hashFiles('continuous_integration/scripts/test_report*') }}
+
+      - name: Generate report
+        run: |
+          python continuous_integration/scripts/test_report.py --repo dask/dask --max-days 90 --max-runs 30 --nfails 1 -o test_report.html
+          python continuous_integration/scripts/test_report.py --repo dask/dask --max-days 7 --max-runs 30 --nfails 2 -o test_short_report.html --title "Test Short Report"
+          mkdir deploy
+          mv test_report.html test_short_report.html deploy/
+
+      - name: Deploy 🚀
+        uses: JamesIves/github-pages-deploy-action@v4.8.0
+        with:
+          branch: gh-pages
+          folder: deploy