Update unified search providers for unified contacts model (company queries contacts with type=company)

This commit is contained in:
Agent Zero
2026-07-19 21:32:24 +02:00
parent 3177daf47f
commit c670846cbd
5 changed files with 50 additions and 40 deletions
@@ -181,7 +181,7 @@ async def index_entity(
table_map = { table_map = {
"contact": "contacts", "contact": "contacts",
"company": "companies", "company": "contacts",
"mail": "mails", "mail": "mails",
"file": "files", "file": "files",
"event": "calendar_entries", "event": "calendar_entries",
+3 -3
View File
@@ -131,7 +131,7 @@ async def index_company(ctx: dict[str, Any], company_id: str) -> None:
try: try:
eid = _parse_id(company_id) eid = _parse_id(company_id)
result = await db.execute( result = await db.execute(
text("SELECT tenant_id FROM companies WHERE id = :cid"), text("SELECT tenant_id FROM contacts WHERE id = :cid"),
{"cid": eid}, {"cid": eid},
) )
row = result.mappings().first() row = result.mappings().first()
@@ -170,7 +170,7 @@ async def reindex(ctx: dict[str, Any], entity_type: str) -> None:
table_map = { table_map = {
"contact": "contacts", "contact": "contacts",
"company": "companies", "company": "contacts",
"mail": "mails", "mail": "mails",
"file": "files", "file": "files",
"event": "calendar_entries", "event": "calendar_entries",
@@ -216,7 +216,7 @@ async def embedding_batch(ctx: dict[str, Any]) -> None:
table_map = { table_map = {
"contact": "contacts", "contact": "contacts",
"company": "companies", "company": "contacts",
"mail": "mails", "mail": "mails",
"file": "files", "file": "files",
"event": "calendar_entries", "event": "calendar_entries",
@@ -1,4 +1,4 @@
"""Company search provider.""" """Company search provider — queries contacts with type='company'."""
from __future__ import annotations from __future__ import annotations
@@ -13,7 +13,7 @@ logger = logging.getLogger(__name__)
class CompanySearchProvider: class CompanySearchProvider:
"""Search provider for Company entities.""" """Search provider for Company entities (contacts with type='company')."""
entity_type = "company" entity_type = "company"
@@ -24,13 +24,14 @@ class CompanySearchProvider:
tenant_id: uuid.UUID, tenant_id: uuid.UUID,
limit: int, limit: int,
) -> list[dict[str, Any]]: ) -> list[dict[str, Any]]:
"""Full-text search on companies.search_tsv.""" """Full-text search on contacts.search_tsv where type='company'."""
sql = text( sql = text(
""" """
SELECT c.*, ts_rank(c.search_tsv, to_tsquery('pg_catalog.german', :q)) AS rank SELECT c.*, ts_rank(c.search_tsv, to_tsquery('pg_catalog.german', :q)) AS rank
FROM companies c FROM contacts c
WHERE c.tenant_id = :tid WHERE c.tenant_id = :tid
AND c.deleted_at IS NULL AND c.deleted_at IS NULL
AND c.type = 'company'
AND c.search_tsv @@ to_tsquery('pg_catalog.german', :q) AND c.search_tsv @@ to_tsquery('pg_catalog.german', :q)
ORDER BY rank DESC ORDER BY rank DESC
LIMIT :lim LIMIT :lim
@@ -50,13 +51,14 @@ class CompanySearchProvider:
tenant_id: uuid.UUID, tenant_id: uuid.UUID,
limit: int, limit: int,
) -> list[dict[str, Any]]: ) -> list[dict[str, Any]]:
"""Semantic search on companies.embedding.""" """Semantic search on contacts.embedding where type='company'."""
sql = text( sql = text(
""" """
SELECT c.*, 1 - (c.embedding <=> cast(:emb AS vector)) AS score SELECT c.*, 1 - (c.embedding <=> cast(:emb AS vector)) AS score
FROM companies c FROM contacts c
WHERE c.tenant_id = :tid WHERE c.tenant_id = :tid
AND c.deleted_at IS NULL AND c.deleted_at IS NULL
AND c.type = 'company'
AND c.embedding IS NOT NULL AND c.embedding IS NOT NULL
ORDER BY c.embedding <=> cast(:emb AS vector) ORDER BY c.embedding <=> cast(:emb AS vector)
LIMIT :lim LIMIT :lim
@@ -78,9 +80,10 @@ class CompanySearchProvider:
"""Get text for embedding generation.""" """Get text for embedding generation."""
sql = text( sql = text(
""" """
SELECT name, description, industry SELECT displayname, name, email_1, email_2, phone_1, website,
FROM companies mailing_city, tags, vat_code
WHERE id = :eid AND tenant_id = :tid FROM contacts
WHERE id = :eid AND tenant_id = :tid AND type = 'company'
""" """
) )
result = await db.execute(sql, {"eid": entity_id, "tid": tenant_id}) result = await db.execute(sql, {"eid": entity_id, "tid": tenant_id})
@@ -88,27 +91,31 @@ class CompanySearchProvider:
if not row: if not row:
return "" return ""
parts = [ parts = [
row.get("displayname", ""),
row.get("name", ""), row.get("name", ""),
row.get("description", ""), row.get("email_1", ""),
row.get("industry", ""), row.get("phone_1", ""),
row.get("website", ""),
row.get("mailing_city", ""),
row.get("tags", ""),
] ]
return " ".join(str(p) for p in parts if p) return " ".join(str(p) for p in parts if p)
def to_search_result(self, entity: object) -> dict[str, Any]: def to_search_result(self, entity: object) -> dict[str, Any]:
"""Convert company to search result dict.""" """Convert company to search result dict."""
if isinstance(entity, dict): if isinstance(entity, dict):
name = entity.get("name", "") name = entity.get("name", "") or entity.get("displayname", "")
description = entity.get("description", "") or "" email = entity.get("email_1", "") or ""
entity_id = str(entity.get("id", "")) entity_id = str(entity.get("id", ""))
else: else:
name = getattr(entity, "name", "") name = getattr(entity, "name", "") or getattr(entity, "displayname", "")
description = getattr(entity, "description", "") or "" email = getattr(entity, "email_1", "") or ""
entity_id = str(getattr(entity, "id", "")) entity_id = str(getattr(entity, "id", ""))
return { return {
"entity_type": self.entity_type, "entity_type": self.entity_type,
"entity_id": entity_id, "entity_id": entity_id,
"title": name, "title": name,
"snippet": description[:200], "snippet": email[:200],
"score": 0.0, "score": 0.0,
"data": {}, "data": {},
} }
@@ -1,4 +1,4 @@
"""Contact search provider.""" """Contact search provider — works with unified contacts table."""
from __future__ import annotations from __future__ import annotations
@@ -9,13 +9,11 @@ from typing import Any
from sqlalchemy import text from sqlalchemy import text
from sqlalchemy.ext.asyncio import AsyncSession from sqlalchemy.ext.asyncio import AsyncSession
from app.models.contact import Contact
logger = logging.getLogger(__name__) logger = logging.getLogger(__name__)
class ContactSearchProvider: class ContactSearchProvider:
"""Search provider for Contact entities.""" """Search provider for Contact entities (all types)."""
entity_type = "contact" entity_type = "contact"
@@ -80,7 +78,8 @@ class ContactSearchProvider:
"""Get text for embedding generation.""" """Get text for embedding generation."""
sql = text( sql = text(
""" """
SELECT first_name, last_name, email, phone, mobile, notes SELECT displayname, name, firstname, surname, email_1, email_2,
phone_1, phone_2, mailing_city, tags, contact_warning
FROM contacts FROM contacts
WHERE id = :eid AND tenant_id = :tid WHERE id = :eid AND tenant_id = :tid
""" """
@@ -90,32 +89,36 @@ class ContactSearchProvider:
if not row: if not row:
return "" return ""
parts = [ parts = [
row.get("first_name", ""), row.get("displayname", ""),
row.get("last_name", ""), row.get("name", ""),
row.get("email", ""), row.get("firstname", ""),
row.get("phone", ""), row.get("surname", ""),
row.get("mobile", ""), row.get("email_1", ""),
row.get("notes", ""), row.get("email_2", ""),
row.get("phone_1", ""),
row.get("phone_2", ""),
row.get("mailing_city", ""),
row.get("tags", ""),
] ]
return " ".join(str(p) for p in parts if p) return " ".join(str(p) for p in parts if p)
def to_search_result(self, entity: object) -> dict[str, Any]: def to_search_result(self, entity: object) -> dict[str, Any]:
"""Convert contact to search result dict.""" """Convert contact to search result dict."""
if isinstance(entity, dict): if isinstance(entity, dict):
first = entity.get("first_name", "") displayname = entity.get("displayname", "")
last = entity.get("last_name", "") email = entity.get("email_1", "")
email = entity.get("email", "")
entity_id = str(entity.get("id", "")) entity_id = str(entity.get("id", ""))
contact_type = entity.get("type", "")
else: else:
first = getattr(entity, "first_name", "") displayname = getattr(entity, "displayname", "")
last = getattr(entity, "last_name", "") email = getattr(entity, "email_1", "")
email = getattr(entity, "email", "")
entity_id = str(getattr(entity, "id", "")) entity_id = str(getattr(entity, "id", ""))
contact_type = getattr(entity, "type", "")
return { return {
"entity_type": self.entity_type, "entity_type": self.entity_type,
"entity_id": entity_id, "entity_id": entity_id,
"title": f"{first} {last}".strip(), "title": displayname,
"snippet": email or "", "snippet": email or "",
"score": 0.0, "score": 0.0,
"data": {}, "data": {"type": contact_type},
} }
@@ -265,7 +265,7 @@ async def search_stats(
stats: dict = {} stats: dict = {}
tables = { tables = {
"contacts": "embedding", "contacts": "embedding",
"companies": "embedding", "contacts": "embedding",
"mails": "embedding", "mails": "embedding",
"files": "embedding", "files": "embedding",
"calendar_entries": "embedding", "calendar_entries": "embedding",