task_id stringlengths 15 15 | repo stringlengths 11 23 | file_path stringlengths 16 49 | function_name stringlengths 4 33 | qualified_name stringlengths 4 37 | function_type stringclasses 2
values | class_name stringclasses 8
values | prompt stringlengths 198 10.1k | signature stringlengths 11 792 | docstring stringlengths 0 549 | canonical_solution stringlengths 106 2.37k | full_function stringlengths 129 2.67k | tests stringlengths 563 4.68M | setup stringlengths 201 225 | metadata stringlengths 74 78 | validation stringlengths 36 72 | original_task_id stringlengths 15 15 | full_context stringlengths 422 16.4k |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
repo_patch/0001 | Comfy-Org/ComfyUI | comfy_execution/jobs.py | normalize_output_item | normalize_output_item | function | null | """
Job utilities for the /api/jobs endpoint.
Provides normalization and helper functions for job status tracking.
"""
from typing import Optional
from comfy_api.internal import prune_dict
class JobStatus:
"""Job status constants."""
PENDING = 'pending'
IN_PROGRESS = 'in_progress'
COMPLETED = 'compl... | def normalize_output_item(item):
"""Normalize a single output list item for the jobs API.
Returns the normalized item, or None to exclude it.
String items with 3D extensions become {filename, type, subfolder} dicts.
""" | Normalize a single output list item for the jobs API.
Returns the normalized item, or None to exclude it.
String items with 3D extensions become {filename, type, subfolder} dicts. | if item is None:
return None
if isinstance(item, str):
if has_3d_extension(item):
return {'filename': item, 'type': 'output', 'subfolder': '', 'mediaType': '3d'}
return None
if isinstance(item, dict):
return item
return None | def normalize_output_item(item):
"""Normalize a single output list item for the jobs API.
Returns the normalized item, or None to exclude it.
String items with 3D extensions become {filename, type, subfolder} dicts.
"""
if item is None:
return None
if isinstance(item, str):
if h... | [{"test_file": "tests/execution/test_jobs.py", "test_function": "TestNormalizeOutputItem.test_none_returns_none", "test_content": "\"\"\"Unit tests for comfy_execution/jobs.py\"\"\"\n\nfrom comfy_execution.jobs import (\n JobStatus,\n is_previewable,\n normalize_queue_item,\n normalize_history_item,\n no... | {"repo_url": "https://github.com/Comfy-Org/ComfyUI", "install_cmd": "pip install -e .", "commit_sha": "dff0a4a15887383c90a031e3fd48ebc41f6928e7", "frozen_requirements": "frozen_requirements/Comfy-Org_ComfyUI.txt"} | {"body_lines": 9, "file_lines": 390, "has_docstring": true, "num_tests": 6} | {"status": "passed", "tests_run": 6} | repo_patch/0001 | """
Job utilities for the /api/jobs endpoint.
Provides normalization and helper functions for job status tracking.
"""
from typing import Optional
from comfy_api.internal import prune_dict
class JobStatus:
"""Job status constants."""
PENDING = 'pending'
IN_PROGRESS = 'in_progress'
COMPLETED = 'compl... |
repo_patch/0002 | Comfy-Org/ComfyUI | comfy_execution/jobs.py | normalize_queue_item | normalize_queue_item | function | null | "\"\"\"\nJob utilities for the /api/jobs endpoint.\nProvides normalization and helper functions for (...TRUNCATED) | "def normalize_queue_item(item: tuple, status: str) -> dict:\n \"\"\"Convert queue item tuple to (...TRUNCATED) | "Convert queue item tuple to unified job dict.\n\nExpects item with sensitive data already removed ((...TRUNCATED) | " priority, prompt_id, _, extra_data, _ = item\n create_time, workflow_id = _extract_job_metad(...TRUNCATED) | "def normalize_queue_item(item: tuple, status: str) -> dict:\n \"\"\"Convert queue item tuple to (...TRUNCATED) | "[{\"test_file\": \"tests/execution/test_jobs.py\", \"test_function\": \"TestNormalizeQueueItem.test(...TRUNCATED) | "{\"repo_url\": \"https://github.com/Comfy-Org/ComfyUI\", \"install_cmd\": \"pip install -e .\", \"c(...TRUNCATED) | {"body_lines": 10, "file_lines": 390, "has_docstring": true, "num_tests": 1} | {"status": "passed", "tests_run": 1} | repo_patch/0002 | "\"\"\"\nJob utilities for the /api/jobs endpoint.\nProvides normalization and helper functions for (...TRUNCATED) |
repo_patch/0003 | Comfy-Org/ComfyUI | comfy_api/feature_flags.py | get_connection_feature | get_connection_feature | function | null | "\"\"\"\nFeature flags module for ComfyUI WebSocket protocol negotiation.\n\nThis module handles cap(...TRUNCATED) | "def get_connection_feature(\n sockets_metadata: dict[str, dict[str, Any]],\n sid: str,\n f(...TRUNCATED) | "Get a feature flag value for a specific connection.\n\nArgs:\n sockets_metadata: Dictionary of s(...TRUNCATED) | " if sid not in sockets_metadata:\n return default\n\n return sockets_metadata[sid].get(...TRUNCATED) | "def get_connection_feature(\n sockets_metadata: dict[str, dict[str, Any]],\n sid: str,\n f(...TRUNCATED) | "[{\"test_file\": \"tests-unit/feature_flags_test.py\", \"test_function\": \"TestFeatureFlags.test_g(...TRUNCATED) | "{\"repo_url\": \"https://github.com/Comfy-Org/ComfyUI\", \"install_cmd\": \"pip install -e .\", \"c(...TRUNCATED) | {"body_lines": 3, "file_lines": 72, "has_docstring": true, "num_tests": 5} | {"status": "passed", "tests_run": 5} | repo_patch/0003 | "\"\"\"\nFeature flags module for ComfyUI WebSocket protocol negotiation.\n\nThis module handles cap(...TRUNCATED) |
repo_patch/0004 | Comfy-Org/ComfyUI | comfy_execution/jobs.py | apply_sorting | apply_sorting | function | null | "\"\"\"\nJob utilities for the /api/jobs endpoint.\nProvides normalization and helper functions for (...TRUNCATED) | "def apply_sorting(jobs: list[dict], sort_by: str, sort_order: str) -> list[dict]:\n \"\"\"Sort j(...TRUNCATED) | Sort jobs list by specified field and order. | " reverse = (sort_order == 'desc')\n\n if sort_by == 'execution_duration':\n def get_so(...TRUNCATED) | "def apply_sorting(jobs: list[dict], sort_by: str, sort_order: str) -> list[dict]:\n \"\"\"Sort j(...TRUNCATED) | "[{\"test_file\": \"tests/execution/test_jobs.py\", \"test_function\": \"TestApplySorting.test_sort_(...TRUNCATED) | "{\"repo_url\": \"https://github.com/Comfy-Org/ComfyUI\", \"install_cmd\": \"pip install -e .\", \"c(...TRUNCATED) | {"body_lines": 10, "file_lines": 390, "has_docstring": true, "num_tests": 4} | {"status": "passed", "tests_run": 4} | repo_patch/0004 | "\"\"\"\nJob utilities for the /api/jobs endpoint.\nProvides normalization and helper functions for (...TRUNCATED) |
repo_patch/0005 | Comfy-Org/ComfyUI | comfy_execution/jobs.py | get_outputs_summary | get_outputs_summary | function | null | "\"\"\"\nJob utilities for the /api/jobs endpoint.\nProvides normalization and helper functions for (...TRUNCATED) | "def get_outputs_summary(outputs: dict) -> tuple[int, Optional[dict]]:\n \"\"\"\n Count output(...TRUNCATED) | "Count outputs and find preview in a single pass.\nReturns (outputs_count, preview_output).\n\nPrevi(...TRUNCATED) | " count = 0\n preview_output = None\n fallback_preview = None\n\n for node_id, node_outp(...TRUNCATED) | "def get_outputs_summary(outputs: dict) -> tuple[int, Optional[dict]]:\n \"\"\"\n Count output(...TRUNCATED) | "[{\"test_file\": \"tests/execution/test_jobs.py\", \"test_function\": \"TestGetOutputsSummary.test_(...TRUNCATED) | "{\"repo_url\": \"https://github.com/Comfy-Org/ComfyUI\", \"install_cmd\": \"pip install -e .\", \"c(...TRUNCATED) | {"body_lines": 49, "file_lines": 390, "has_docstring": true, "num_tests": 13} | {"status": "passed", "tests_run": 13} | repo_patch/0005 | "\"\"\"\nJob utilities for the /api/jobs endpoint.\nProvides normalization and helper functions for (...TRUNCATED) |
repo_patch/0006 | Comfy-Org/ComfyUI | comfy_execution/jobs.py | normalize_outputs | normalize_outputs | function | null | "\"\"\"\nJob utilities for the /api/jobs endpoint.\nProvides normalization and helper functions for (...TRUNCATED) | "def normalize_outputs(outputs: dict) -> dict:\n \"\"\"Normalize raw node outputs for the jobs AP(...TRUNCATED) | "Normalize raw node outputs for the jobs API.\n\nTransforms string 3D filenames into file output dic(...TRUNCATED) | " normalized = {}\n for node_id, node_outputs in outputs.items():\n if not isinstance(n(...TRUNCATED) | "def normalize_outputs(outputs: dict) -> dict:\n \"\"\"Normalize raw node outputs for the jobs AP(...TRUNCATED) | "[{\"test_file\": \"tests/execution/test_jobs.py\", \"test_function\": \"TestNormalizeOutputs.test_e(...TRUNCATED) | "{\"repo_url\": \"https://github.com/Comfy-Org/ComfyUI\", \"install_cmd\": \"pip install -e .\", \"c(...TRUNCATED) | {"body_lines": 19, "file_lines": 390, "has_docstring": true, "num_tests": 6} | {"status": "passed", "tests_run": 6} | repo_patch/0006 | "\"\"\"\nJob utilities for the /api/jobs endpoint.\nProvides normalization and helper functions for (...TRUNCATED) |
repo_patch/0007 | Comfy-Org/ComfyUI | comfy_execution/jobs.py | is_previewable | is_previewable | function | null | "\"\"\"\nJob utilities for the /api/jobs endpoint.\nProvides normalization and helper functions for (...TRUNCATED) | "def is_previewable(media_type: str, item: dict) -> bool:\n \"\"\"\n Check if an output item i(...TRUNCATED) | "Check if an output item is previewable.\nMatches frontend logic in ComfyUI_frontend/src/stores/queu(...TRUNCATED) | " if media_type in PREVIEWABLE_MEDIA_TYPES:\n return True\n\n # Check format field (MIM(...TRUNCATED) | "def is_previewable(media_type: str, item: dict) -> bool:\n \"\"\"\n Check if an output item i(...TRUNCATED) | "[{\"test_file\": \"tests/execution/test_jobs.py\", \"test_function\": \"TestIsPreviewable.test_prev(...TRUNCATED) | "{\"repo_url\": \"https://github.com/Comfy-Org/ComfyUI\", \"install_cmd\": \"pip install -e .\", \"c(...TRUNCATED) | {"body_lines": 12, "file_lines": 390, "has_docstring": true, "num_tests": 7} | {"status": "passed", "tests_run": 7} | repo_patch/0007 | "\"\"\"\nJob utilities for the /api/jobs endpoint.\nProvides normalization and helper functions for (...TRUNCATED) |
repo_patch/0008 | Comfy-Org/ComfyUI | middleware/cache_middleware.py | cache_control | cache_control | function | null | "\"\"\"Cache control middleware for ComfyUI server\"\"\"\n\nfrom aiohttp import web\nfrom typing imp(...TRUNCATED) | "async def cache_control(\n request: web.Request, handler: Callable[[web.Request], Awaitable[web.(...TRUNCATED) | "Cache control middleware that sets appropriate cache headers based on file type and response status(...TRUNCATED) | " response: web.Response = await handler(request)\n\n path_filename = request.path.rsplit(\"/\(...TRUNCATED) | "async def cache_control(\n request: web.Request, handler: Callable[[web.Request], Awaitable[web.(...TRUNCATED) | "[{\"test_file\": \"tests-unit/server_test/test_cache_control.py\", \"test_function\": \"TestCacheCo(...TRUNCATED) | "{\"repo_url\": \"https://github.com/Comfy-Org/ComfyUI\", \"install_cmd\": \"pip install -e .\", \"c(...TRUNCATED) | {"body_lines": 22, "file_lines": 54, "has_docstring": true, "num_tests": 9} | {"status": "passed", "tests_run": 9} | repo_patch/0008 | "\"\"\"Cache control middleware for ComfyUI server\"\"\"\n\nfrom aiohttp import web\nfrom typing imp(...TRUNCATED) |
repo_patch/0009 | docling-project/docling | docling/datamodel/asr_model_specs.py | _get_whisper_base_model | _get_whisper_base_model | function | null | "import logging\nfrom enum import Enum\n\nfrom pydantic import (\n AnyUrl,\n)\n\nfrom docling.dat(...TRUNCATED) | "def _get_whisper_base_model():\n \"\"\"\n Get the best Whisper Base model for the current har(...TRUNCATED) | "Get the best Whisper Base model for the current hardware.\n\nAutomatically selects MLX Whisper Base(...TRUNCATED) | " try:\n import torch\n\n has_mps = torch.backends.mps.is_built() and torch.backend(...TRUNCATED) | "def _get_whisper_base_model():\n \"\"\"\n Get the best Whisper Base model for the current har(...TRUNCATED) | "[{\"test_file\": \"tests/test_asr_mlx_whisper.py\", \"test_function\": \"TestMlxWhisperIntegration.(...TRUNCATED) | "{\"repo_url\": \"https://github.com/docling-project/docling\", \"install_cmd\": \"pip install -e .\(...TRUNCATED) | {"body_lines": 34, "file_lines": 495, "has_docstring": true, "num_tests": 2} | {"status": "passed", "tests_run": 2} | repo_patch/0009 | "import logging\nfrom enum import Enum\n\nfrom pydantic import (\n AnyUrl,\n)\n\nfrom docling.dat(...TRUNCATED) |
repo_patch/0010 | docling-project/docling | docling/datamodel/asr_model_specs.py | _get_whisper_tiny_model | _get_whisper_tiny_model | function | null | "import logging\nfrom enum import Enum\n\nfrom pydantic import (\n AnyUrl,\n)\n\nfrom docling.dat(...TRUNCATED) | "def _get_whisper_tiny_model():\n \"\"\"\n Get the best Whisper Tiny model for the current har(...TRUNCATED) | "Get the best Whisper Tiny model for the current hardware.\n\nAutomatically selects MLX Whisper Tiny(...TRUNCATED) | " try:\n import torch\n\n has_mps = torch.backends.mps.is_built() and torch.backend(...TRUNCATED) | "def _get_whisper_tiny_model():\n \"\"\"\n Get the best Whisper Tiny model for the current har(...TRUNCATED) | "[{\"test_file\": \"tests/test_asr_mlx_whisper.py\", \"test_function\": \"TestMlxWhisperIntegration.(...TRUNCATED) | "{\"repo_url\": \"https://github.com/docling-project/docling\", \"install_cmd\": \"pip install -e .\(...TRUNCATED) | {"body_lines": 34, "file_lines": 495, "has_docstring": true, "num_tests": 1} | {"status": "passed", "tests_run": 1} | repo_patch/0010 | "import logging\nfrom enum import Enum\n\nfrom pydantic import (\n AnyUrl,\n)\n\nfrom docling.dat(...TRUNCATED) |
End of preview. Expand in Data Studio
CodeDP Repo-Patch Benchmark (CPT-friendly)
Repository-level code completion benchmark for evaluating continual pre-training (CPT) models. 55 tasks from 12 real-world repositories, each requiring the model to generate a function body given file-level context.
Prompt Format
The prompt field contains the file context up to the target function's signature and docstring (truncated at the original # TODO: Implement this function marker). This format is directly usable by base/completion models — the model simply continues generating the function body.
The full_context field preserves the original full-file prompt (including code after the target function) for reference or fill-in-the-middle approaches.
Base/Completion Models
from datasets import load_dataset
ds = load_dataset("melihcatal/codedp-bench-repo-patch-cpt", split="train")
# prompt is ready for completion — ends with signature + docstring
prompt = ds[0]["prompt"]
# Model generates the function body from here
Instruction Models
For chat/instruction models, wrap the prompt in a chat template:
msg = f"Complete the implementation of `{ds[0]['function_name']}`. Return ONLY the function body.\n\n```python\n{ds[0]['prompt']}\n```"
Fields
| Field | Description |
|---|---|
prompt |
File context up to function signature + docstring (CPT-ready) |
full_context |
Full file with # TODO marker and downstream code |
canonical_solution |
Reference function body |
signature |
Function signature |
docstring |
Function docstring (empty for 22/55 tasks) |
function_name |
Target function name |
class_name |
Enclosing class (if method, else null) |
tests |
JSON list of pytest test cases |
setup |
JSON with repo URL, install command, commit SHA |
full_function |
Complete function (signature + docstring + body) |
metadata |
JSON with body_lines, file_lines, has_docstring, num_tests |
validation |
Test validation status |
Statistics
- 55 tasks from 12 repositories
- 13 class methods, 42 standalone functions
- 33 with docstrings, 22 without
- Prompt lengths (after truncation): median ~2,800 chars (vs ~6,300 before)
- Reference body lengths: median 437 chars
Metrics
Reference-based metrics (no repo setup needed):
- BLEU-4: Token-level BLEU score
- CodeBLEU: Syntax-aware code similarity
- Edit Similarity: 1 - normalized Levenshtein distance
- Exact Match: Normalized whitespace comparison
Evaluation
python -m evaluation.utility.run_repo_patch \
--model_path ./output/model/checkpoint-final \
--benchmark_path melihcatal/codedp-bench-repo-patch-cpt \
--output_dir results/repo_patch/model/variant \
--devices auto --batch_size 4
# For instruction models, add --chat_template
- Downloads last month
- 95