opensolr-haystack 0.2.3__tar.gz → 0.2.5__tar.gz

{opensolr_haystack-0.2.3 → opensolr_haystack-0.2.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: opensolr-haystack
-Version: 0.2.3
+Version: 0.2.5
 Summary: Haystack integration for Opensolr — managed Apache Solr DocumentStore with server-side embeddings and hybrid BM25+kNN retrieval
 Author-email: Opensolr <support@opensolr.com>
 License: MIT
@@ -98,8 +98,10 @@ Writes go through Opensolr's [Data Ingestion API](https://opensolr.com/learn/api
 — the same pipeline the Drupal and WordPress connectors use. It is
 **asynchronous**: documents are queued, then embeddings, sentiment, language
 and all crawler-identical derived fields are computed **server-side**, and
-documents become searchable within about a minute. Progress is visible in the
-Opensolr Control Panel and via the `ingest_status` API. Each document's
+documents become searchable within about a minute. Progress is visible in
+**Control Panel → Data Ingestion** — a per-job status board (queued /
+processing / completed / failed, with processed / success / failed document
+counts per job) — and via the `ingest_status` API. Each document's
 identity is its `uri` (the Solr id is `md5(uri)`): pass a real URL in
 metadata (`{"uri": "https://..."}`), or a deterministic one is synthesized
 from your id. Re-submitting the same `uri` updates the document. Pass
@@ -136,6 +138,23 @@ answer = store.ai_answer(
 )
 ```
 
+
+### Search tuning
+
+Retrieval (search and RAG grounding) runs through the platform's tuned
+pipeline: global defaults → your index's saved **Search Tuning** (Control
+Panel → Index Settings → Search Tuning: semantic↔lexical balance, field
+weights, minimum match, search mode, vector candidate pool, content quality
+boost) → optional per-call overrides via `tuning`:
+
+```
+tuning={"search_mode": "keywords_required", "fw_title": 0.2,
+        "mm": "strict", "vector_topk": 500, "quality_boost": 0.3}
+```
+
+Defaults match the platform's PHP configuration exactly — customize in the
+Control Panel once, or per call from code.
+
 ## How it's tested
 
 Every release is validated against **live Opensolr infrastructure** — no mocks:

{opensolr_haystack-0.2.3 → opensolr_haystack-0.2.5}/README.md

@@ -80,8 +80,10 @@ Writes go through Opensolr's [Data Ingestion API](https://opensolr.com/learn/api
 — the same pipeline the Drupal and WordPress connectors use. It is
 **asynchronous**: documents are queued, then embeddings, sentiment, language
 and all crawler-identical derived fields are computed **server-side**, and
-documents become searchable within about a minute. Progress is visible in the
-Opensolr Control Panel and via the `ingest_status` API. Each document's
+documents become searchable within about a minute. Progress is visible in
+**Control Panel → Data Ingestion** — a per-job status board (queued /
+processing / completed / failed, with processed / success / failed document
+counts per job) — and via the `ingest_status` API. Each document's
 identity is its `uri` (the Solr id is `md5(uri)`): pass a real URL in
 metadata (`{"uri": "https://..."}`), or a deterministic one is synthesized
 from your id. Re-submitting the same `uri` updates the document. Pass
@@ -118,6 +120,23 @@ answer = store.ai_answer(
 )
 ```
 
+
+### Search tuning
+
+Retrieval (search and RAG grounding) runs through the platform's tuned
+pipeline: global defaults → your index's saved **Search Tuning** (Control
+Panel → Index Settings → Search Tuning: semantic↔lexical balance, field
+weights, minimum match, search mode, vector candidate pool, content quality
+boost) → optional per-call overrides via `tuning`:
+
+```
+tuning={"search_mode": "keywords_required", "fw_title": 0.2,
+        "mm": "strict", "vector_topk": 500, "quality_boost": 0.3}
+```
+
+Defaults match the platform's PHP configuration exactly — customize in the
+Control Panel once, or per call from code.
+
 ## How it's tested
 
 Every release is validated against **live Opensolr infrastructure** — no mocks:

{opensolr_haystack-0.2.3 → opensolr_haystack-0.2.5}/haystack_integrations/document_stores/opensolr/client.py

@@ -233,7 +233,20 @@ class OpensolrClient:
             raise OpensolrError(f"ingest_status: non-JSON response: {resp.text[:200]}") from exc
 
     def embed_and_search(self, index: str, query: str, rows: int = 10, **params: Any) -> Dict[str, Any]:
-        """Server-side one-shot: embed the query, run hybrid search, return docs."""
+        """Server-side one-shot: embed the query, run the platform's tuned
+        hybrid search, return ranked docs.
+
+        Retrieval uses the same pipeline as the hosted search UI: global
+        defaults, overridden by the index's saved Search Tuning (Control
+        Panel → Index Settings → Search Tuning), overridden by any of these
+        per-call knobs passed as extra params: ``fw_title``,
+        ``fw_description``, ``fw_uri``, ``fw_text``, ``fw_text_t``,
+        ``lexical_weight``, ``vector_weight``, ``vector_topk``,
+        ``search_mode`` (union / keywords_required / meaning_required /
+        intersection), ``quality_boost``, ``min_score``,
+        ``freshness_boost``, ``lexical_norm_k``, ``mm`` (flexible /
+        balanced / strict or raw Solr mm syntax).
+        """
         body = self.ai(
             "embed_and_search",
             index_name=index,
@@ -308,6 +321,7 @@ class OpensolrClient:
         fq: Optional[str] = None,
         docs: Optional[int] = None,
         words: Optional[int] = None,
+        tuning: Optional[Dict[str, Any]] = None,
     ) -> str:
         """Build the LLM context from the top hybrid search hits.
 
@@ -329,7 +343,7 @@ class OpensolrClient:
         hits: List[Dict[str, Any]] = []
         if not fq:
             try:
-                body = self.embed_and_search(index, query, rows=docs)
+                body = self.embed_and_search(index, query, rows=docs, **(tuning or {}))
                 if isinstance(body, dict):
                     hits = body.get("results", {}).get("docs", []) or []
             except (OpensolrError, httpx.HTTPError):
@@ -357,6 +371,7 @@ class OpensolrClient:
         rag_docs: Optional[int] = None,
         rag_words: Optional[int] = None,
         instruction: Optional[str] = None,
+        tuning: Optional[Dict[str, Any]] = None,
         **params: Any,
     ) -> str:
         """Grounded RAG answer: hybrid retrieval over the index feeds the LLM.
@@ -380,7 +395,8 @@ class OpensolrClient:
         if "context" not in data:
             try:
                 context = self._rag_context(
-                    index, query, fq=filter_query, docs=rag_docs, words=rag_words
+                    index, query, fq=filter_query, docs=rag_docs, words=rag_words,
+                    tuning=tuning,
                 )
             except (OpensolrError, httpx.HTTPError):
                 context = ""

{opensolr_haystack-0.2.3 → opensolr_haystack-0.2.5}/haystack_integrations/document_stores/opensolr/store.py

@@ -255,6 +255,7 @@ class OpensolrDocumentStore:
         rag_docs: int = 3,
         rag_words: int = 1500,
         instruction: Optional[str] = None,
+        tuning: Optional[Dict[str, Any]] = None,
         **kwargs: Any,
     ) -> str:
         """Grounded RAG answer generated only from this index's content.
@@ -264,13 +265,17 @@ class OpensolrDocumentStore:
         title/description/text become the LLM context — the same pipeline as
         Opensolr's hosted search UI. Pass ``instruction`` to fully control
         the prompt (e.g. "Answer in German, cite the sources you used").
-        Returns plain text.
+        Retrieval uses the platform's tuned pipeline: your index's saved
+        Search Tuning (Control Panel) applies automatically; ``tuning``
+        overrides any knob per call (fw_title, lexical_weight, search_mode,
+        mm, vector_topk, quality_boost, ...). Returns plain text.
         """
         fqs = _filters_to_fq(filters)
         fq = " AND ".join(f"({f})" for f in fqs) if fqs else None
         return self.client.ai_summary(
             self.index, query, filter_query=fq,
             rag_docs=rag_docs, rag_words=rag_words, instruction=instruction,
+            tuning=tuning,
             **kwargs,
         )
 

{opensolr_haystack-0.2.3 → opensolr_haystack-0.2.5}/opensolr_haystack.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: opensolr-haystack
-Version: 0.2.3
+Version: 0.2.5
 Summary: Haystack integration for Opensolr — managed Apache Solr DocumentStore with server-side embeddings and hybrid BM25+kNN retrieval
 Author-email: Opensolr <support@opensolr.com>
 License: MIT
@@ -98,8 +98,10 @@ Writes go through Opensolr's [Data Ingestion API](https://opensolr.com/learn/api
 — the same pipeline the Drupal and WordPress connectors use. It is
 **asynchronous**: documents are queued, then embeddings, sentiment, language
 and all crawler-identical derived fields are computed **server-side**, and
-documents become searchable within about a minute. Progress is visible in the
-Opensolr Control Panel and via the `ingest_status` API. Each document's
+documents become searchable within about a minute. Progress is visible in
+**Control Panel → Data Ingestion** — a per-job status board (queued /
+processing / completed / failed, with processed / success / failed document
+counts per job) — and via the `ingest_status` API. Each document's
 identity is its `uri` (the Solr id is `md5(uri)`): pass a real URL in
 metadata (`{"uri": "https://..."}`), or a deterministic one is synthesized
 from your id. Re-submitting the same `uri` updates the document. Pass
@@ -136,6 +138,23 @@ answer = store.ai_answer(
 )
 ```
 
+
+### Search tuning
+
+Retrieval (search and RAG grounding) runs through the platform's tuned
+pipeline: global defaults → your index's saved **Search Tuning** (Control
+Panel → Index Settings → Search Tuning: semantic↔lexical balance, field
+weights, minimum match, search mode, vector candidate pool, content quality
+boost) → optional per-call overrides via `tuning`:
+
+```
+tuning={"search_mode": "keywords_required", "fw_title": 0.2,
+        "mm": "strict", "vector_topk": 500, "quality_boost": 0.3}
+```
+
+Defaults match the platform's PHP configuration exactly — customize in the
+Control Panel once, or per call from code.
+
 ## How it's tested
 
 Every release is validated against **live Opensolr infrastructure** — no mocks:

{opensolr_haystack-0.2.3 → opensolr_haystack-0.2.5}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "opensolr-haystack"
-version = "0.2.3"
+version = "0.2.5"
 description = "Haystack integration for Opensolr — managed Apache Solr DocumentStore with server-side embeddings and hybrid BM25+kNN retrieval"
 readme = "README.md"
 license = { text = "MIT" }