New Charting Tool Validates AI-Generated Code to Prevent Errors
Gigasoft has released ProEssentials v10, which includes a tool designed to solve errors in AI-generated charting code. The tool, `pe_query.py`, validates code suggestions against the charting library's compiled DLL binary. This process aims to eliminate hallucinated property names, which the company identifies as the most common source of errors when using AI assistants to write data visualization code.
- Gigasoft, a company founded in 1993, has a long history of focusing on high-performance, mission-critical charting components, releasing the first version of ProEssentials in 1995. - The `pe_query.py` tool validates against a comprehensive API surface, giving AI assistants access to 1,104 properties, 80 methods, 40 events, and 167 enumerations extracted directly from the compiled library. - This validation is deterministic, checking every property path against the compiled binary, which differs from probabilistic approaches that rely on searching documentation to reduce hallucinations. - The problem of "hallucination" is widespread in AI code generation; one university study found that commercial AI models invent non-existent package names 5.2% of the time, while open-source models do so in 21.7% of cases. - The tool is designed to work with a range of popular AI assistants, including ChatGPT, GitHub Copilot, Gemini, and Claude. - A 2026 comparison of charting libraries noted that competitors offer different AI solutions, such as SciChart's documentation chatbot, while others like LightningChart and DevExpress provide no chart-specific AI tools. - The tool was launched as part of ProEssentials v10.0.0.20 on February 25, 2026, a release that also included expanded GPU compute shader rendering to improve performance. - ProEssentials supports a variety of development platforms from a single native engine, including WPF, WinForms, C++ MFC, Delphi VCL, and ActiveX.
