Improving the Nested Spin Syntax Parser for Power Users
Hello Sven,
First of all, thank you for creating such a powerful and versatile tool. I've been using GSA SER for a long time and it's an essential part of my workflow.
I'm writing to make a feature request regarding the spin syntax parser, specifically its handling of complex, deeply nested articles.
The Problem:
When working with very large and complex source texts (e.g., 50kb+ with multiple levels of nesting), the current parser often fails silently. It doesn't throw an error message, but in the preview (and in the final submission), it stops processing the spin syntax midway through the article. The first part of the text is spun correctly, but the rest is output as raw text, including all the {, |, and } characters.
Example of problematic structure:
The parser seems to struggle with structures that are either deeply nested (more than 2-3 levels) or have many nested blocks at the same level, like this:
{{Option A {1|2}}|{Option B {3|4}}} which leads to {many|a lot of} {possibilities|outcomes}.
The Impact:
This limitation forces advanced users to aggressively pre-process their source texts with external scripts to "flatten" the syntax. This process often reduces the overall uniqueness and variety of the source article, which counteracts the goal of using high-quality spun content. It also adds a complicated extra step to the workflow.
The Request:
Would it be possible to consider improving the spin syntax parser in a future update? A more robust parser that could handle at least 4-5 levels of nesting, or a higher complexity of blocks on the same level, would be a massive quality-of-life improvement for many power users. This would allow us to use much more complex and unique source articles directly within GSA SER.
Thank you for your time and consideration. GSA SER is a fantastic product, and this improvement would make it even better.