AIOSAIandClaude Opus 4.6
ba51e747cf
Complete implementation of a plugin-based code standards checker. Core Framework: - CheckResult/CheckItem/Severity data models - Three-source plugin discovery (builtin, entry_points, local) - Config system (.seedgo/config.json) with profiles and overrides - Severity-weighted scoring (errors block, warnings degrade, info reports) CLI: - seedgo init — create project config - seedgo check — run all enabled plugins - seedgo list — show available plugins - Three output formats: human, json, github (CI annotations) 5 Starter Plugins: - no-bare-except, type-hints-required, docstring-coverage - function-length (no linter equivalent) - file-structure (no linter equivalent) 354 tests, all passing. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
77 lines
2.6 KiB
Python
77 lines
2.6 KiB
Python
"""
|
|
Seed Go Core Data Models
|
|
|
|
Defines the plugin contract types: Severity, CheckItem, and CheckResult.
|
|
All plugin check() functions must return a CheckResult instance.
|
|
|
|
These are pure dataclasses with no external dependencies — safe to import
|
|
anywhere without pulling in AIPass or other infrastructure.
|
|
"""
|
|
|
|
from dataclasses import dataclass, field
|
|
from enum import Enum
|
|
from typing import Optional
|
|
|
|
|
|
class Severity(Enum):
|
|
"""Severity levels for individual check items.
|
|
|
|
Determines how a failed check impacts the overall score and pass/fail:
|
|
- ERROR: Full score deduction. Any unresolved error blocks pass regardless of score.
|
|
- WARNING: Half score deduction (configurable). Degrades score but does not block pass.
|
|
- INFO: No score deduction. Informational only — shown in output, zero score impact.
|
|
"""
|
|
|
|
ERROR = "error"
|
|
WARNING = "warning"
|
|
INFO = "info"
|
|
|
|
|
|
@dataclass
|
|
class CheckItem:
|
|
"""One individual check within a plugin result.
|
|
|
|
Represents a single rule evaluation. A plugin may return many CheckItems,
|
|
one per rule or per violation found in the file.
|
|
|
|
Attributes:
|
|
name: Short identifier for this check (e.g., "bare-except", "missing-type-hint").
|
|
passed: Whether this specific check passed.
|
|
message: Human-readable explanation of the result or violation.
|
|
severity: How serious the violation is. Defaults to ERROR.
|
|
line: Source line number where the violation was found, if applicable.
|
|
fix_hint: Actionable suggestion for how to resolve the violation.
|
|
"""
|
|
|
|
name: str
|
|
passed: bool
|
|
message: str
|
|
severity: Severity = Severity.ERROR
|
|
line: Optional[int] = None
|
|
fix_hint: Optional[str] = None
|
|
|
|
|
|
@dataclass
|
|
class CheckResult:
|
|
"""Return type for all plugin check() functions.
|
|
|
|
Every plugin must return exactly one CheckResult per file checked.
|
|
The result aggregates all individual CheckItem results and provides
|
|
an overall pass/fail verdict and score.
|
|
|
|
Attributes:
|
|
plugin: The plugin's PLUGIN_NAME string (e.g., "no-bare-except").
|
|
passed: Overall pass/fail for this plugin against this file.
|
|
checks: List of individual check results. Empty list means no checks ran.
|
|
score: Weighted score from 0-100. Calculated by the scoring engine.
|
|
file_path: Absolute path to the file that was checked.
|
|
metadata: Plugin-defined extra data (e.g., AST stats, timing). Serializable to JSON.
|
|
"""
|
|
|
|
plugin: str
|
|
passed: bool
|
|
checks: list[CheckItem] = field(default_factory=list)
|
|
score: int = 0
|
|
file_path: str = ""
|
|
metadata: dict = field(default_factory=dict)
|