JSON Converter — Transform Data Formats
Instant JSON Validation and Transformation
Convert JSON to YAML, XML, CSV, and SQL inserts in one click. Reverse every conversion. Preserve nested structures, handle type coercion, and apply custom field mappings — all in your browser, zero data leaving your machine.
Live Demo
See the Converter in Action
Paste a JSON payload into the left pane, pick a target format, and watch the output render in real time. The converter flags type mismatches, truncates overflowing strings, and auto-generates column headers for tabular targets.
The demo above shows a 47-field e-commerce order object being converted from JSON to YAML. Nested arrays become properly indented sequences, boolean values stay lowercase, and ISO-8601 timestamps are preserved verbatim. Click "Try the Converter" to load the sandbox with sample datasets from three industries: SaaS subscription plans, IoT sensor telemetry, and healthcare HL7-FHIR excerpts.
Format Coverage
Supported Formats
JSONForge handles five core data formats with bidirectional conversion. Each format supports depth-limited traversal (default 12 levels), null-coalescing, and configurable quoting behavior.
JSON ↔ YAML
Round-trip conversion between JSON and YAML 1.2. Preserves object keys, scalar types, and array ordering. YAML output uses 2-space indentation by default; toggle block vs. flow style per collection. Handles anchors and aliases on import.
JSON ↔ XML
Transform JSON objects into well-formed XML with configurable root elements and namespace prefixes. Arrays become repeated child nodes. XML attributes can be mapped from JSON properties prefixed with `@`. Supports CDATA wrapping for HTML-embedded values.
JSON ↔ CSV
Flatten nested JSON into tabular CSV rows with dot-notation column headers (e.g., `address.city`). Missing fields produce empty cells. Export includes RFC 4180-compliant quoting, configurable delimiters (comma, semicolon, tab), and optional BOM for Excel compatibility.
JSON ↔ SQL Inserts
Generate `INSERT INTO` statements for MySQL, PostgreSQL, and SQLite from JSON arrays. Auto-detects column types (VARCHAR, INT, BOOLEAN, JSONB). Escapes single quotes and handles NULL coalescing. Batch sizes configurable from 1 to 5,000 rows per statement.
YAML ↔ XML
Direct conversion without an intermediate JSON step. YAML mappings become XML elements; YAML sequences become repeated nodes. Preserves scalar typing so that YAML `true` becomes XML `
CSV ↔ SQL Inserts
Read CSV headers as column names and emit parameterized `INSERT` statements. Detects numeric columns automatically. Supports quoted identifiers for reserved keywords and optional `ON CONFLICT` / `ON DUPLICATE KEY UPDATE` clauses.
Advanced Configuration
Custom Mapping Rules
When your data structures don't align neatly, mapping rules let you define field renames, type coercions, nested extractions, and default values — all declaratively, no code required.
A mapping rule is a JSON object that describes how source fields translate to target fields. For example, converting a legacy API response into a modern schema might require renaming `user_id` → `customerId`, coercing `signup_ts` from a Unix epoch to an ISO-8601 string, and extracting `billing.address[0].zip` into a flat `postalCode` field. JSONForge evaluates these rules at conversion time, so you can chain renames, apply regex replacements on values, and set fallback defaults when a source path is missing.
The rule editor supports three rule types:
Field Rename
Map `sourcePath` to `targetPath` using dot notation or bracket accessors. Example: `order.items[*].sku` → `lineItems[].productCode`. Wildcards iterate over arrays; each match produces a row in the output.
Type Coercion
Force a source value into a target type. Supported coercions include string→int, string→float, string→boolean, int→string, and timestamp epoch→ISO-8601. Invalid conversions emit a warning and fall back to the raw string.
Default & Fallback
When a source path resolves to `null` or is absent, inject a default value. Defaults can be static (e.g., `"active": true`), computed (e.g., current timestamp), or pulled from another field in the same record.
Save mapping rule sets as named profiles (e.g., "Shopify → BigQuery", "FHIR R4 → Flat CSV") and reuse them across projects. Profiles are stored locally in your browser's IndexedDB and can be exported as `.json` files for team sharing. The mapping engine processes up to 10,000 records per batch with sub-second latency on modern hardware.