@nextsparkjs/ai-workflow 0.1.0-beta.100

package/claude/skills/entity-system/scripts/generate-child-migration.py

@@ -0,0 +1,298 @@
+#!/usr/bin/env python3
+"""
+Generate Child Entity Migration Script
+
+Generates SQL migration for child entity table with parent reference.
+
+Usage:
+    python generate-child-migration.py --parent PARENT --child CHILD [--theme THEME]
+
+Options:
+    --parent PARENT     Parent entity name (kebab-case, e.g., 'orders')
+    --child CHILD       Child entity name (kebab-case, e.g., 'items')
+    --theme THEME       Theme name (default: from NEXT_PUBLIC_ACTIVE_THEME or 'default')
+    --output FILE       Output file path
+    --fields FIELDS     Comma-separated field:type pairs (e.g., 'name:text,quantity:number')
+    --with-rls          Include RLS policies (default: true)
+    --no-rls            Exclude RLS policies
+    --dry-run           Print SQL without writing file
+"""
+
+import os
+import sys
+import argparse
+import re
+from pathlib import Path
+from datetime import datetime
+from typing import List, Dict, Tuple
+
+
+def get_active_theme() -> str:
+    """Get active theme from environment or default."""
+    return os.environ.get('NEXT_PUBLIC_ACTIVE_THEME', 'default')
+
+
+def to_snake_case(name: str) -> str:
+    """Convert kebab-case to snake_case."""
+    return name.replace('-', '_')
+
+
+def to_camel_case(name: str) -> str:
+    """Convert kebab-case to camelCase."""
+    components = name.split('-')
+    return components[0] + ''.join(x.title() for x in components[1:])
+
+
+# Field type to SQL type mapping
+FIELD_TYPE_MAP = {
+    'text': 'VARCHAR(255)',
+    'textarea': 'TEXT',
+    'number': 'DECIMAL(10, 2)',
+    'integer': 'INTEGER',
+    'boolean': 'BOOLEAN DEFAULT FALSE',
+    'date': 'DATE',
+    'datetime': 'TIMESTAMPTZ',
+    'email': 'VARCHAR(255)',
+    'url': 'TEXT',
+    'json': 'JSONB DEFAULT \'{}\'::jsonb',
+    'select': 'VARCHAR(100)',
+    'uuid': 'TEXT',
+}
+
+
+def parse_fields(fields_str: str) -> List[Dict]:
+    """Parse comma-separated field:type pairs."""
+    if not fields_str:
+        return []
+
+    fields = []
+    for field_def in fields_str.split(','):
+        parts = field_def.strip().split(':')
+        if len(parts) >= 2:
+            name = parts[0].strip()
+            field_type = parts[1].strip()
+            required = len(parts) > 2 and parts[2].strip().lower() == 'required'
+            fields.append({
+                'name': name,
+                'type': field_type,
+                'required': required
+            })
+    return fields
+
+
+def to_pascal_case(name: str) -> str:
+    """Convert kebab-case to PascalCase."""
+    return ''.join(x.title() for x in name.split('-'))
+
+
+def generate_child_migration(
+    parent_slug: str,
+    child_slug: str,
+    fields: List[Dict],
+    with_rls: bool = True
+) -> str:
+    """Generate child entity table migration SQL matching project conventions."""
+
+    parent_table = to_snake_case(parent_slug)
+    child_table = f"{parent_table}_{to_snake_case(child_slug)}"
+    parent_pascal = to_pascal_case(parent_slug)
+    child_pascal = to_pascal_case(child_slug)
+    date_str = datetime.now().strftime('%Y-%m-%d')
+
+    # Build column definitions with proper alignment
+    columns = [
+        '  -- Primary Key',
+        '  id             TEXT PRIMARY KEY DEFAULT gen_random_uuid()::text,',
+        '',
+        '  -- Parent Reference',
+        f'  "parentId"     TEXT NOT NULL REFERENCES public."{parent_table}"(id) ON DELETE CASCADE,',
+    ]
+
+    # Add custom fields if any
+    if fields:
+        columns.append('')
+        columns.append('  -- Entity-specific fields')
+        for field in fields:
+            name = field['name']
+            field_type = field['type']
+            required = field.get('required', False)
+
+            sql_type = FIELD_TYPE_MAP.get(field_type, 'TEXT')
+            not_null = ' NOT NULL' if required else ''
+            # Pad the name for alignment
+            padded_name = f'"{name}"'.ljust(14) if name[0].isupper() or '-' in name else name.ljust(14)
+            columns.append(f'  {padded_name} {sql_type}{not_null},')
+
+    # System fields
+    columns.append('')
+    columns.append('  -- System fields')
+    columns.append('  "createdAt"    TIMESTAMPTZ NOT NULL DEFAULT now(),')
+    columns.append('  "updatedAt"    TIMESTAMPTZ NOT NULL DEFAULT now()')
+
+    columns_sql = '\n'.join(columns)
+
+    sql = f'''-- Migration: 003_{child_table}.sql
+-- Description: {parent_pascal} {child_pascal} child entity (table, indexes, RLS)
+-- Date: {date_str}
+
+-- ============================================
+-- TABLE
+-- ============================================
+CREATE TABLE IF NOT EXISTS public."{child_table}" (
+{columns_sql}
+);
+
+COMMENT ON TABLE  public."{child_table}"              IS 'Child entity: {child_slug} for {parent_slug}';
+COMMENT ON COLUMN public."{child_table}"."parentId"   IS 'Reference to parent {parent_slug}';
+
+-- ============================================
+-- TRIGGER updatedAt
+-- ============================================
+DROP TRIGGER IF EXISTS {child_table}_set_updated_at ON public."{child_table}";
+CREATE TRIGGER {child_table}_set_updated_at
+BEFORE UPDATE ON public."{child_table}"
+FOR EACH ROW EXECUTE FUNCTION public.set_updated_at();
+
+-- ============================================
+-- INDEXES
+-- ============================================
+CREATE INDEX IF NOT EXISTS idx_{child_table}_parent_id  ON public."{child_table}"("parentId");
+CREATE INDEX IF NOT EXISTS idx_{child_table}_created_at ON public."{child_table}"("createdAt" DESC);
+'''
+
+    if with_rls:
+        sql += f'''
+-- ============================================
+-- RLS
+-- ============================================
+ALTER TABLE public."{child_table}" ENABLE ROW LEVEL SECURITY;
+
+-- Cleanup existing policies
+DROP POLICY IF EXISTS "{parent_pascal} {child_pascal} team can do all" ON public."{child_table}";
+
+-- ============================
+-- RLS: TEAM ISOLATION VIA PARENT
+-- ============================
+-- Hereda el aislamiento del parent via teamId
+CREATE POLICY "{parent_pascal} {child_pascal} team can do all"
+ON public."{child_table}"
+FOR ALL TO authenticated
+USING (
+  -- Superadmin bypass
+  public.is_superadmin()
+  OR
+  -- Team isolation via parent
+  EXISTS (
+    SELECT 1 FROM public."{parent_table}" t
+    WHERE t.id = "parentId"
+      AND t."teamId" = ANY(public.get_user_team_ids())
+  )
+)
+WITH CHECK (
+  public.is_superadmin()
+  OR
+  EXISTS (
+    SELECT 1 FROM public."{parent_table}" t
+    WHERE t.id = "parentId"
+      AND t."teamId" = ANY(public.get_user_team_ids())
+  )
+);
+'''
+
+    return sql
+
+
+def find_next_migration_number(migrations_dir: Path) -> int:
+    """Find the next available migration number."""
+    if not migrations_dir.exists():
+        return 1
+
+    existing = list(migrations_dir.glob('*.sql'))
+    if not existing:
+        return 1
+
+    numbers = []
+    for f in existing:
+        try:
+            num = int(f.name.split('_')[0])
+            numbers.append(num)
+        except (ValueError, IndexError):
+            continue
+
+    return max(numbers, default=0) + 1
+
+
+def main():
+    parser = argparse.ArgumentParser(description='Generate child entity migration')
+    parser.add_argument('--parent', required=True, help='Parent entity name (kebab-case)')
+    parser.add_argument('--child', required=True, help='Child entity name (kebab-case)')
+    parser.add_argument('--theme', default=None, help='Theme name')
+    parser.add_argument('--output', help='Output file path')
+    parser.add_argument('--fields', help='Comma-separated field:type pairs')
+    parser.add_argument('--with-rls', action='store_true', default=True)
+    parser.add_argument('--no-rls', action='store_true', help='Exclude RLS policies')
+    parser.add_argument('--dry-run', action='store_true', help='Print SQL without writing')
+
+    args = parser.parse_args()
+
+    theme = args.theme or get_active_theme()
+    parent_slug = args.parent.lower()
+    child_slug = args.child.lower()
+    with_rls = not args.no_rls
+
+    # Parse fields
+    fields = parse_fields(args.fields) if args.fields else []
+
+    print(f"\n{'='*60}")
+    print(f"Generating child entity migration")
+    print(f"Parent: {parent_slug}")
+    print(f"Child: {child_slug}")
+    print(f"Theme: {theme}")
+    print(f"With RLS: {with_rls}")
+    print(f"Fields: {len(fields)}")
+    print(f"{'='*60}")
+
+    # Generate SQL
+    sql = generate_child_migration(parent_slug, child_slug, fields, with_rls)
+
+    if args.dry_run:
+        print("\n" + sql)
+        print(f"\n{'='*60}")
+        print("DRY RUN - No file written")
+        return 0
+
+    # Determine output path
+    if args.output:
+        output_path = Path(args.output)
+    else:
+        # Auto-generate path in parent entity's migrations folder
+        entity_dir = Path(f'contents/themes/{theme}/entities/{parent_slug}')
+        migrations_dir = entity_dir / 'migrations'
+
+        if not entity_dir.exists():
+            print(f"\nError: Parent entity directory not found: {entity_dir}")
+            print("Please create the parent entity first using scaffold-entity.py")
+            return 1
+
+        migrations_dir.mkdir(exist_ok=True)
+
+        next_num = find_next_migration_number(migrations_dir)
+        parent_table = to_snake_case(parent_slug)
+        child_table = to_snake_case(child_slug)
+        output_path = migrations_dir / f'{next_num:03d}_{parent_table}_{child_table}.sql'
+
+    # Write file
+    with open(output_path, 'w', encoding='utf-8') as f:
+        f.write(sql)
+
+    print(f"\nMigration written to: {output_path}")
+    print(f"\nNext steps:")
+    print(f"  1. Add fields to the migration if not specified via --fields")
+    print(f"  2. Add childEntities config to {parent_slug}.config.ts")
+    print(f"  3. Run: pnpm db:migrate")
+
+    return 0
+
+
+if __name__ == '__main__':
+    sys.exit(main())

package/claude/skills/entity-system/scripts/generate-metas-migration.py

@@ -0,0 +1,233 @@
+#!/usr/bin/env python3
+"""
+Generate Metas Migration Script
+
+Generates SQL migration for entity metadata table.
+
+Usage:
+    python generate-metas-migration.py --entity ENTITY_NAME [--theme THEME] [--output FILE]
+
+Options:
+    --entity ENTITY_NAME  Name of the entity (kebab-case, e.g., 'products', 'blog-posts')
+    --theme THEME         Theme name (default: from NEXT_PUBLIC_ACTIVE_THEME or 'default')
+    --output FILE         Output file (default: auto-generate path)
+    --with-rls            Include RLS policies (default: true)
+    --dry-run             Print SQL without writing file
+"""
+
+import os
+import sys
+import argparse
+from pathlib import Path
+from datetime import datetime
+
+
+def get_active_theme() -> str:
+    """Get active theme from environment or default."""
+    return os.environ.get('NEXT_PUBLIC_ACTIVE_THEME', 'default')
+
+
+def to_camel_case(name: str) -> str:
+    """Convert kebab-case to camelCase."""
+    components = name.split('-')
+    return components[0] + ''.join(x.title() for x in components[1:])
+
+
+def to_singular(name: str) -> str:
+    """Simple pluralization - remove trailing 's' for common cases."""
+    if name.endswith('ies'):
+        return name[:-3] + 'y'
+    elif name.endswith('es') and not name.endswith('ses'):
+        return name[:-2]
+    elif name.endswith('s') and not name.endswith('ss'):
+        return name[:-1]
+    return name
+
+
+def to_pascal_case(name: str) -> str:
+    """Convert kebab-case to PascalCase."""
+    return ''.join(x.title() for x in name.split('-'))
+
+
+def generate_metas_migration(entity_slug: str, with_rls: bool = True) -> str:
+    """Generate metadata table migration SQL matching project conventions."""
+
+    # Entity naming
+    table_name = entity_slug.replace('-', '_')
+    singular = to_singular(entity_slug).replace('-', '_')
+    pascal = to_pascal_case(entity_slug)
+    date_str = datetime.now().strftime('%Y-%m-%d')
+
+    sql = f'''-- Migration: 002_{table_name}_metas.sql
+-- Description: {pascal} metas (table, indexes, RLS)
+-- Date: {date_str}
+
+-- ============================================
+-- TABLE
+-- ============================================
+CREATE TABLE IF NOT EXISTS public."{table_name}_metas" (
+  id             TEXT PRIMARY KEY DEFAULT gen_random_uuid()::text,
+  "entityId"     TEXT NOT NULL REFERENCES public."{table_name}"(id) ON DELETE CASCADE,
+  "metaKey"      TEXT NOT NULL,
+  "metaValue"    JSONB NOT NULL DEFAULT '{{}}'::jsonb,
+  "dataType"     TEXT DEFAULT 'json',
+  "isPublic"     BOOLEAN NOT NULL DEFAULT false,
+  "isSearchable" BOOLEAN NOT NULL DEFAULT false,
+  "createdAt"    TIMESTAMPTZ NOT NULL DEFAULT now(),
+  "updatedAt"    TIMESTAMPTZ NOT NULL DEFAULT now(),
+  CONSTRAINT {table_name}_metas_unique_key UNIQUE ("entityId", "metaKey")
+);
+
+COMMENT ON TABLE  public."{table_name}_metas"               IS 'Key-value metadata for {table_name}';
+COMMENT ON COLUMN public."{table_name}_metas"."entityId"    IS 'Reference to parent {singular}';
+COMMENT ON COLUMN public."{table_name}_metas"."metaKey"     IS 'Metadata key identifier';
+COMMENT ON COLUMN public."{table_name}_metas"."metaValue"   IS 'Metadata value in JSONB format';
+COMMENT ON COLUMN public."{table_name}_metas"."dataType"    IS 'Type hint: json, string, number, boolean';
+COMMENT ON COLUMN public."{table_name}_metas"."isPublic"    IS 'Whether this metadata is publicly readable';
+COMMENT ON COLUMN public."{table_name}_metas"."isSearchable" IS 'Whether this metadata is searchable';
+
+-- ============================================
+-- TRIGGER updatedAt
+-- ============================================
+DROP TRIGGER IF EXISTS {table_name}_metas_set_updated_at ON public."{table_name}_metas";
+CREATE TRIGGER {table_name}_metas_set_updated_at
+BEFORE UPDATE ON public."{table_name}_metas"
+FOR EACH ROW EXECUTE FUNCTION public.set_updated_at();
+
+-- ============================================
+-- INDEXES
+-- ============================================
+CREATE INDEX IF NOT EXISTS idx_{table_name}_metas_entity_id   ON public."{table_name}_metas"("entityId");
+CREATE INDEX IF NOT EXISTS idx_{table_name}_metas_key         ON public."{table_name}_metas"("metaKey");
+CREATE INDEX IF NOT EXISTS idx_{table_name}_metas_is_public   ON public."{table_name}_metas"("isPublic") WHERE "isPublic" = true;
+CREATE INDEX IF NOT EXISTS idx_{table_name}_metas_searchable  ON public."{table_name}_metas"("isSearchable") WHERE "isSearchable" = true;
+CREATE INDEX IF NOT EXISTS idx_{table_name}_metas_value_gin   ON public."{table_name}_metas" USING GIN ("metaValue");
+CREATE INDEX IF NOT EXISTS idx_{table_name}_metas_value_ops   ON public."{table_name}_metas" USING GIN ("metaValue" jsonb_path_ops);
+'''
+
+    if with_rls:
+        sql += f'''
+-- ============================================
+-- RLS
+-- ============================================
+ALTER TABLE public."{table_name}_metas" ENABLE ROW LEVEL SECURITY;
+
+-- Cleanup existing policies
+DROP POLICY IF EXISTS "{pascal} metas team can do all" ON public."{table_name}_metas";
+
+-- ============================
+-- RLS: TEAM ISOLATION VIA PARENT
+-- ============================
+-- Hereda el aislamiento del parent via teamId
+CREATE POLICY "{pascal} metas team can do all"
+ON public."{table_name}_metas"
+FOR ALL TO authenticated
+USING (
+  -- Superadmin bypass
+  public.is_superadmin()
+  OR
+  -- Team isolation via parent
+  EXISTS (
+    SELECT 1 FROM public."{table_name}" t
+    WHERE t.id = "entityId"
+      AND t."teamId" = ANY(public.get_user_team_ids())
+  )
+)
+WITH CHECK (
+  public.is_superadmin()
+  OR
+  EXISTS (
+    SELECT 1 FROM public."{table_name}" t
+    WHERE t.id = "entityId"
+      AND t."teamId" = ANY(public.get_user_team_ids())
+  )
+);
+'''
+
+    return sql
+
+
+def find_next_migration_number(migrations_dir: Path) -> int:
+    """Find the next available migration number."""
+    if not migrations_dir.exists():
+        return 1
+
+    existing = list(migrations_dir.glob('*.sql'))
+    if not existing:
+        return 1
+
+    numbers = []
+    for f in existing:
+        try:
+            num = int(f.name.split('_')[0])
+            numbers.append(num)
+        except (ValueError, IndexError):
+            continue
+
+    return max(numbers, default=0) + 1
+
+
+def main():
+    parser = argparse.ArgumentParser(description='Generate entity metas migration')
+    parser.add_argument('--entity', required=True, help='Entity name (kebab-case)')
+    parser.add_argument('--theme', default=None, help='Theme name')
+    parser.add_argument('--output', help='Output file path')
+    parser.add_argument('--with-rls', action='store_true', default=True, help='Include RLS policies')
+    parser.add_argument('--no-rls', action='store_true', help='Exclude RLS policies')
+    parser.add_argument('--dry-run', action='store_true', help='Print SQL without writing')
+
+    args = parser.parse_args()
+
+    theme = args.theme or get_active_theme()
+    entity_slug = args.entity.lower()
+    with_rls = not args.no_rls
+
+    print(f"\n{'='*60}")
+    print(f"Generating metas migration for: {entity_slug}")
+    print(f"Theme: {theme}")
+    print(f"With RLS: {with_rls}")
+    print(f"{'='*60}")
+
+    # Generate SQL
+    sql = generate_metas_migration(entity_slug, with_rls)
+
+    if args.dry_run:
+        print("\n" + sql)
+        print(f"\n{'='*60}")
+        print("DRY RUN - No file written")
+        return 0
+
+    # Determine output path
+    if args.output:
+        output_path = Path(args.output)
+    else:
+        # Auto-generate path in entity's migrations folder
+        entity_dir = Path(f'contents/themes/{theme}/entities/{entity_slug}')
+        migrations_dir = entity_dir / 'migrations'
+
+        if not entity_dir.exists():
+            print(f"\nError: Entity directory not found: {entity_dir}")
+            print("Please create the entity first using scaffold-entity.py")
+            return 1
+
+        migrations_dir.mkdir(exist_ok=True)
+
+        next_num = find_next_migration_number(migrations_dir)
+        table_name = entity_slug.replace('-', '_')
+        output_path = migrations_dir / f'{next_num:03d}_{table_name}_metas.sql'
+
+    # Write file
+    with open(output_path, 'w', encoding='utf-8') as f:
+        f.write(sql)
+
+    print(f"\nMigration written to: {output_path}")
+    print(f"\nNext steps:")
+    print(f"  1. Review the generated migration")
+    print(f"  2. Run: pnpm db:migrate")
+    print(f"  3. Update entity config: access.metadata = true")
+
+    return 0
+
+
+if __name__ == '__main__':
+    sys.exit(main())