gsd-trae 1.0.1 → 1.0.2

package/refs/vbenchmark/tasks/glue-code/data-transform/excel-to-json/PROMPT.md

@@ -1,67 +0,0 @@
-# Task: Excel to JSON Transformer
-
-## Objective
-Create a Python script that transforms messy Excel files into clean, schema-compliant JSON.
-
-## Input Data Issues
-The input Excel file (`data/sales.xlsx`) has these common problems:
-- Merged cells in headers
-- Inconsistent date formats (MM/DD/YYYY, DD-MM-YYYY, YYYY.MM.DD)
-- Currency values with mixed symbols ($, €, £)
-- Empty rows and columns
-- Leading/trailing whitespace
-- Duplicate entries
-
-## Requirements
-
-1. **Read Excel File**
-   - Handle merged cells (unmerge and fill)
-   - Skip empty rows
-   - Detect header row automatically
-
-2. **Clean Data**
-   - Normalize dates to ISO 8601 format (YYYY-MM-DD)
-   - Convert currency to numeric (float, USD)
-   - Trim whitespace from strings
-   - Remove duplicate rows (based on order_id)
-
-3. **Output JSON Schema**
-   ```json
-   {
-     "orders": [
-       {
-         "order_id": "string",
-         "customer_name": "string",
-         "order_date": "string (YYYY-MM-DD)",
-         "amount": "number",
-         "status": "string (pending|completed|cancelled)"
-       }
-     ],
-     "metadata": {
-       "total_orders": "number",
-       "total_amount": "number",
-       "generated_at": "string (ISO 8601)"
-     }
-   }
-   ```
-
-4. **Error Handling**
-   - Log rows that couldn't be parsed
-   - Continue processing despite individual row errors
-   - Include error summary in output
-
-## Files to Create
-- `transform.py` - Main transformation script
-- `schema.json` - JSON Schema for validation
-
-## Usage
-```bash
-python transform.py data/sales.xlsx output.json
-```
-
-## Success Criteria
-- [ ] Correctly parses all valid rows from test Excel file
-- [ ] Output matches JSON schema exactly
-- [ ] Handles all date format variations
-- [ ] Currency conversion is accurate to 2 decimal places
-- [ ] Duplicates are removed

package/refs/vbenchmark/tasks/glue-code/data-transform/excel-to-json/task.yaml

@@ -1,28 +0,0 @@
-name: Excel to JSON Transformer
-category: glue-code
-subcategory: data-transform
-description: Create a Python script that cleans and transforms messy Excel data to strict JSON schema
-difficulty: easy
-stack: python
-timeout: 120
-tokenLimit: 50000
-
-promptFile: PROMPT.md
-
-tests:
-  functional: tests/transform.test.py
-
-weight: 1.0
-tags:
-  - python
-  - excel
-  - json
-  - data-transform
-  - pandas
-
-evaluation:
-  functional: 0.6
-  visual: 0.0
-  quality: 0.25
-  cost: 0.1
-  speed: 0.05

package/refs/vbenchmark/tasks/glue-code/data-transform/excel-to-json/tests/transform.test.py

@@ -1,137 +0,0 @@
-import pytest
-import json
-import subprocess
-import os
-from pathlib import Path
-
-SCRIPT_PATH = Path(__file__).parent.parent / "transform.py"
-TEST_DATA_PATH = Path(__file__).parent / "fixtures" / "sales.xlsx"
-OUTPUT_PATH = Path(__file__).parent / "output.json"
-
-
-@pytest.fixture(autouse=True)
-def cleanup():
-    yield
-    if OUTPUT_PATH.exists():
-        OUTPUT_PATH.unlink()
-
-
-def test_script_runs_without_error():
-    """Script should execute successfully"""
-    result = subprocess.run(
-        ["python", str(SCRIPT_PATH), str(TEST_DATA_PATH), str(OUTPUT_PATH)],
-        capture_output=True,
-        text=True
-    )
-    assert result.returncode == 0, f"Script failed: {result.stderr}"
-
-
-def test_output_is_valid_json():
-    """Output should be valid JSON"""
-    subprocess.run(
-        ["python", str(SCRIPT_PATH), str(TEST_DATA_PATH), str(OUTPUT_PATH)],
-        capture_output=True
-    )
-
-    with open(OUTPUT_PATH) as f:
-        data = json.load(f)
-
-    assert "orders" in data
-    assert "metadata" in data
-
-
-def test_orders_have_required_fields():
-    """Each order should have all required fields"""
-    subprocess.run(
-        ["python", str(SCRIPT_PATH), str(TEST_DATA_PATH), str(OUTPUT_PATH)],
-        capture_output=True
-    )
-
-    with open(OUTPUT_PATH) as f:
-        data = json.load(f)
-
-    required_fields = ["order_id", "customer_name", "order_date", "amount", "status"]
-
-    for order in data["orders"]:
-        for field in required_fields:
-            assert field in order, f"Missing field: {field}"
-
-
-def test_dates_are_iso_format():
-    """All dates should be in YYYY-MM-DD format"""
-    subprocess.run(
-        ["python", str(SCRIPT_PATH), str(TEST_DATA_PATH), str(OUTPUT_PATH)],
-        capture_output=True
-    )
-
-    with open(OUTPUT_PATH) as f:
-        data = json.load(f)
-
-    import re
-    date_pattern = re.compile(r"^\d{4}-\d{2}-\d{2}$")
-
-    for order in data["orders"]:
-        assert date_pattern.match(order["order_date"]), \
-            f"Invalid date format: {order['order_date']}"
-
-
-def test_amounts_are_numeric():
-    """All amounts should be numeric (float)"""
-    subprocess.run(
-        ["python", str(SCRIPT_PATH), str(TEST_DATA_PATH), str(OUTPUT_PATH)],
-        capture_output=True
-    )
-
-    with open(OUTPUT_PATH) as f:
-        data = json.load(f)
-
-    for order in data["orders"]:
-        assert isinstance(order["amount"], (int, float)), \
-            f"Amount is not numeric: {order['amount']}"
-
-
-def test_no_duplicate_order_ids():
-    """There should be no duplicate order IDs"""
-    subprocess.run(
-        ["python", str(SCRIPT_PATH), str(TEST_DATA_PATH), str(OUTPUT_PATH)],
-        capture_output=True
-    )
-
-    with open(OUTPUT_PATH) as f:
-        data = json.load(f)
-
-    order_ids = [o["order_id"] for o in data["orders"]]
-    assert len(order_ids) == len(set(order_ids)), "Duplicate order IDs found"
-
-
-def test_metadata_is_accurate():
-    """Metadata should match actual data"""
-    subprocess.run(
-        ["python", str(SCRIPT_PATH), str(TEST_DATA_PATH), str(OUTPUT_PATH)],
-        capture_output=True
-    )
-
-    with open(OUTPUT_PATH) as f:
-        data = json.load(f)
-
-    assert data["metadata"]["total_orders"] == len(data["orders"])
-
-    total_amount = sum(o["amount"] for o in data["orders"])
-    assert abs(data["metadata"]["total_amount"] - total_amount) < 0.01
-
-
-def test_status_is_valid_enum():
-    """Status should be one of: pending, completed, cancelled"""
-    subprocess.run(
-        ["python", str(SCRIPT_PATH), str(TEST_DATA_PATH), str(OUTPUT_PATH)],
-        capture_output=True
-    )
-
-    with open(OUTPUT_PATH) as f:
-        data = json.load(f)
-
-    valid_statuses = {"pending", "completed", "cancelled"}
-
-    for order in data["orders"]:
-        assert order["status"] in valid_statuses, \
-            f"Invalid status: {order['status']}"

package/refs/vbenchmark/tasks/glue-code/data-transform/json-to-xml/PROMPT.md

@@ -1,45 +0,0 @@
-# Task: Bidirectional JSON/XML Converter
-
-## Objective
-Build a Go tool that converts between JSON and XML while preserving data types and structure.
-
-## Requirements
-
-1. **JSON to XML**
-   - Objects become elements
-   - Arrays become repeated elements
-   - Preserve attribute hints (@attr)
-   - Handle namespaces
-   - Pretty print or minified output
-
-2. **XML to JSON**
-   - Elements become object keys
-   - Attributes become @attr keys
-   - Text content becomes #text
-   - Handle mixed content
-   - Preserve order where possible
-
-3. **Type Preservation**
-   - Numbers stay numbers
-   - Booleans stay booleans
-   - Nulls are explicit
-   - Arrays vs single elements
-
-4. **CLI Interface**
-   - `converter json2xml input.json output.xml`
-   - `converter xml2json input.xml output.json`
-   - `--pretty` for formatted output
-   - `--schema` for validation
-
-## Files to Create
-- `cmd/converter/main.go` - CLI entry
-- `internal/json2xml/convert.go` - JSON to XML
-- `internal/xml2json/convert.go` - XML to JSON
-- `internal/types/node.go` - Common types
-
-## Success Criteria
-- [ ] JSON→XML→JSON roundtrip preserves data
-- [ ] XML→JSON→XML roundtrip preserves data
-- [ ] Attributes handled correctly
-- [ ] Namespaces preserved
-- [ ] Arrays vs single elements distinguished

package/refs/vbenchmark/tasks/glue-code/data-transform/json-to-xml/task.yaml

@@ -1,24 +0,0 @@
-name: Bidirectional JSON/XML Converter
-category: glue-code
-subcategory: data-transform
-description: Convert between JSON and XML with schema preservation in Go
-difficulty: medium
-stack: go
-timeout: 600
-tokenLimit: 150000
-
-tests:
-  functional: tests/converter_test.go
-
-tags:
-  - go
-  - json
-  - xml
-  - conversion
-  - data-transform
-
-evaluation:
-  functional: 0.5
-  quality: 0.25
-  cost: 0.1
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/data-transform/protobuf-converter/PROMPT.md

@@ -1,44 +0,0 @@
-# Task: Protocol Buffer Serialization
-
-## Objective
-Build a high-performance protobuf serialization library in Rust with JSON interop.
-
-## Requirements
-
-1. **Proto Schema**
-   - Define message types for common entities
-   - User, Order, Product, Event messages
-   - Support nested messages
-   - Support repeated fields and maps
-
-2. **Serialization**
-   - Serialize Rust structs to protobuf binary
-   - Deserialize protobuf binary to Rust structs
-   - Zero-copy deserialization where possible
-   - Streaming support for large messages
-
-3. **JSON Interop**
-   - Convert protobuf to JSON
-   - Convert JSON to protobuf
-   - Handle field name mapping (camelCase vs snake_case)
-   - Optional fields handling
-
-4. **CLI Tool**
-   - `proto encode --schema user.proto input.json output.bin`
-   - `proto decode --schema user.proto input.bin output.json`
-   - `proto validate --schema user.proto input.bin`
-
-## Files to Create
-- `src/main.rs` - CLI entry
-- `src/schema/mod.rs` - Proto definitions
-- `src/encoder.rs` - Serialization
-- `src/decoder.rs` - Deserialization
-- `src/json_interop.rs` - JSON conversion
-- `protos/` - .proto files
-
-## Success Criteria
-- [ ] Binary serialization matches protobuf spec
-- [ ] Deserialization is correct
-- [ ] JSON roundtrip works
-- [ ] Performance better than JSON for large data
-- [ ] Streaming works for large messages

package/refs/vbenchmark/tasks/glue-code/data-transform/protobuf-converter/task.yaml

@@ -1,24 +0,0 @@
-name: Protocol Buffer Serialization
-category: glue-code
-subcategory: data-transform
-description: Implement protobuf serialization/deserialization in Rust
-difficulty: hard
-stack: rust
-timeout: 900
-tokenLimit: 200000
-
-tests:
-  functional: tests/protobuf_test.rs
-
-tags:
-  - rust
-  - protobuf
-  - serialization
-  - performance
-  - binary
-
-evaluation:
-  functional: 0.5
-  quality: 0.25
-  cost: 0.1
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/etl/cdc-pipeline/PROMPT.md

@@ -1,52 +0,0 @@
-# Task: Change Data Capture Pipeline
-
-## Objective
-Build a CDC pipeline using Debezium that streams database changes to downstream systems.
-
-## Requirements
-
-1. **Debezium Setup**
-   - Configure PostgreSQL connector
-   - Capture inserts, updates, deletes
-   - Stream to Kafka topics
-   - Handle schema changes
-
-2. **Consumer Service**
-   - Consume CDC events from Kafka
-   - Parse Debezium event format
-   - Apply changes to target system
-   - Maintain ordering guarantees
-
-3. **Target Systems**
-   - Elasticsearch (search index)
-   - Cache invalidation (Redis)
-   - Webhook notifications
-   - Configurable per-table targets
-
-4. **Operations**
-   - Snapshot for initial sync
-   - Resume from offset on restart
-   - Monitor lag and health
-   - Dead letter handling
-
-## Technical Stack
-- TypeScript/Node.js
-- Debezium (Kafka Connect)
-- Kafka (KafkaJS)
-- PostgreSQL
-- Elasticsearch/Redis
-
-## Files to Create
-- `src/consumer/index.ts` - Kafka consumer
-- `src/consumer/parser.ts` - Event parser
-- `src/handlers/elasticsearch.ts` - ES handler
-- `src/handlers/cache.ts` - Cache handler
-- `src/handlers/webhook.ts` - Webhook handler
-- `docker-compose.yaml` - Infrastructure
-
-## Success Criteria
-- [ ] Database changes captured in real-time
-- [ ] Events correctly parsed
-- [ ] Elasticsearch updated on changes
-- [ ] Cache invalidated correctly
-- [ ] Consumer resumes after restart

package/refs/vbenchmark/tasks/glue-code/etl/cdc-pipeline/task.yaml

@@ -1,27 +0,0 @@
-name: Change Data Capture Pipeline
-category: glue-code
-subcategory: etl
-description: Implement CDC with Debezium and TypeScript
-difficulty: hard
-stack: nodejs
-timeout: 900
-tokenLimit: 200000
-
-tests:
-  functional: tests/cdc.test.ts
-
-docker:
-  compose: docker-compose.yaml
-
-tags:
-  - typescript
-  - debezium
-  - cdc
-  - kafka
-  - streaming
-
-evaluation:
-  functional: 0.5
-  quality: 0.25
-  cost: 0.1
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/etl/database-sync/PROMPT.md

@@ -1,51 +0,0 @@
-# Task: PostgreSQL to MongoDB Sync
-
-## Objective
-Build a data sync pipeline that replicates data from PostgreSQL to MongoDB with transformation support.
-
-## Requirements
-
-1. **Initial Sync**
-   - Full table dump from PostgreSQL
-   - Transform relational to document model
-   - Batch inserts to MongoDB
-   - Progress tracking and resumability
-
-2. **Incremental Sync**
-   - Track changes via updated_at timestamp
-   - Detect inserts, updates, deletes
-   - Sync only changed records
-   - Handle deletions (soft delete support)
-
-3. **Transformation**
-   - Map table columns to document fields
-   - Embed related data (denormalize)
-   - Handle foreign keys → embedded docs
-   - Custom field transformations
-
-4. **Configuration**
-   ```yaml
-   tables:
-     - name: users
-       collection: users
-       transform:
-         - embed: orders
-           via: user_id
-     - name: products
-       collection: products
-   ```
-
-## Files to Create
-- `sync/main.py` - Entry point
-- `sync/extractor.py` - PostgreSQL extraction
-- `sync/transformer.py` - Data transformation
-- `sync/loader.py` - MongoDB loading
-- `sync/config.py` - Configuration parser
-- `config.yaml` - Sync configuration
-
-## Success Criteria
-- [ ] Initial sync transfers all data
-- [ ] Incremental sync captures changes
-- [ ] Transformations apply correctly
-- [ ] Embedded documents work
-- [ ] Sync is resumable on failure

package/refs/vbenchmark/tasks/glue-code/etl/database-sync/task.yaml

@@ -1,24 +0,0 @@
-name: PostgreSQL to MongoDB Sync
-category: glue-code
-subcategory: etl
-description: Sync data between PostgreSQL and MongoDB in Python
-difficulty: medium
-stack: python
-timeout: 600
-tokenLimit: 150000
-
-tests:
-  functional: tests/test_db_sync.py
-
-tags:
-  - python
-  - postgresql
-  - mongodb
-  - sync
-  - etl
-
-evaluation:
-  functional: 0.5
-  quality: 0.2
-  cost: 0.15
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/etl/s3-to-warehouse/PROMPT.md

@@ -1,50 +0,0 @@
-# Task: S3 to Data Warehouse Loader
-
-## Objective
-Build an ETL pipeline in Go that loads files from S3 into a data warehouse (PostgreSQL/Redshift).
-
-## Requirements
-
-1. **S3 Source**
-   - List files in S3 bucket/prefix
-   - Support multiple formats: CSV, JSON, Parquet
-   - Handle partitioned data (year=2025/month=01/)
-   - Track processed files to avoid reprocessing
-
-2. **Data Loading**
-   - Schema inference from files
-   - Create tables if not exist
-   - Bulk COPY for performance
-   - Handle schema evolution
-
-3. **Processing**
-   - Parallel file processing
-   - Memory-efficient streaming
-   - Error handling per file
-   - Dead letter queue for failed files
-
-4. **CLI Interface**
-   - `loader run --source s3://bucket/path --target warehouse.table`
-   - `loader status` - Show processing status
-   - `loader retry` - Retry failed files
-
-## Technical Stack
-- Go 1.21+
-- AWS SDK v2
-- PostgreSQL (or Redshift compatible)
-- Parquet-go for Parquet files
-
-## Files to Create
-- `cmd/loader/main.go` - CLI entry
-- `internal/source/s3.go` - S3 reader
-- `internal/parser/csv.go` - CSV parser
-- `internal/parser/parquet.go` - Parquet parser
-- `internal/loader/warehouse.go` - DB loader
-- `internal/state/tracker.go` - State tracking
-
-## Success Criteria
-- [ ] CSV files loaded correctly
-- [ ] Parquet files loaded correctly
-- [ ] Partitioned data handled
-- [ ] Files not reprocessed
-- [ ] Failed files tracked for retry

package/refs/vbenchmark/tasks/glue-code/etl/s3-to-warehouse/task.yaml

@@ -1,24 +0,0 @@
-name: S3 to Data Warehouse Loader
-category: glue-code
-subcategory: etl
-description: Load S3 files into data warehouse with Go
-difficulty: medium
-stack: go
-timeout: 600
-tokenLimit: 150000
-
-tests:
-  functional: tests/loader_test.go
-
-tags:
-  - go
-  - s3
-  - data-warehouse
-  - etl
-  - parquet
-
-evaluation:
-  functional: 0.5
-  quality: 0.2
-  cost: 0.15
-  speed: 0.15

package/refs/vbenchmark/tasks/glue-code/file-processing/image-resizer/PROMPT.md

@@ -1,52 +0,0 @@
-# Task: High-Performance Image Resizer
-
-## Objective
-Build a high-performance image processing service in Rust for resizing, converting, and optimizing images.
-
-## Requirements
-
-1. **Operations**
-   - Resize (fit, fill, cover)
-   - Crop (center, smart)
-   - Format conversion (JPEG, PNG, WebP, AVIF)
-   - Quality adjustment
-   - Strip metadata
-
-2. **Performance**
-   - Process images in memory
-   - SIMD acceleration
-   - Parallel processing
-   - Streaming for large images
-
-3. **API**
-   - `POST /process` - Process image
-   - Query params: width, height, format, quality
-   - Accept file upload or URL
-   - Return processed image
-
-4. **Presets**
-   - Thumbnail: 150x150, crop center
-   - Preview: 800x600, fit
-   - Full: max 2000px, preserve aspect
-   - Configurable custom presets
-
-## Technical Stack
-- Rust 1.70+
-- image crate or libvips bindings
-- Actix-web or Axum
-- Tokio for async
-
-## Files to Create
-- `src/main.rs` - HTTP server
-- `src/processor/mod.rs` - Image processor
-- `src/processor/resize.rs` - Resize operations
-- `src/processor/convert.rs` - Format conversion
-- `src/handlers/process.rs` - HTTP handlers
-- `src/presets.rs` - Preset definitions
-
-## Success Criteria
-- [ ] Resize works correctly
-- [ ] Format conversion works
-- [ ] WebP output is optimized
-- [ ] Processes 100+ images/sec
-- [ ] Memory usage stays bounded

package/refs/vbenchmark/tasks/glue-code/file-processing/image-resizer/task.yaml

@@ -1,24 +0,0 @@
-name: High-Performance Image Resizer
-category: glue-code
-subcategory: file-processing
-description: Build high-performance image processing service in Rust
-difficulty: hard
-stack: rust
-timeout: 900
-tokenLimit: 200000
-
-tests:
-  functional: tests/image_test.rs
-
-tags:
-  - rust
-  - image-processing
-  - performance
-  - thumbnails
-  - webp
-
-evaluation:
-  functional: 0.5
-  quality: 0.25
-  cost: 0.1
-  speed: 0.15