adityamparikh opened a new pull request, #203: URL: https://github.com/apache/solr-mcp/pull/203
### Summary `index-markdown-documents` treated the whole input as one document (#144), so a client with sixty documents needed sixty tool calls, each a full model round trip. That is what made the Markdown path slow: indexing 61 documents took over two minutes end to end while Solr's share was under a second. A new document now starts at every YAML front-matter block, so a whole dataset is one call. Text before the first block is its own document; a lone `---` followed by prose is a thematic break and does not split; input with a single block, or none, parses exactly as before, including the content-derived id when no `id` is given. ### What changed - `MarkdownDocumentCreator.splitDocuments` finds front-matter blocks (a `---` line, one or more `key: value` or list-item lines, a closing `---`) and parses each chunk with the existing single-document logic. - Tool and parameter descriptions say one or more documents; README row. - `MarkdownDocumentCreatorTest` covers three blocks in one string, a horizontal rule inside a body, per-document ids (equal to the id the same document gets on its own), and a preface before the first block. `MarkdownIndexingTest` is unchanged and still passes. Part of a set with `feat/json-documents-array` and `fix/xml-record-fields`; independent, any merge order. ### Verification - `./gradlew build` on Java 25: green. - `./gradlew nativeTest -Pnative` on GraalVM CE 25.0.2: 260 successful, 0 failed. 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_01CiUHyyXLTo9ATdgg8eRFZJ -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
