Fix #63

Author: cmutel

docs-src/index.md

@@ -46,7 +46,12 @@ The following parameters must be either specified as environment variables, or g
 * `PyST_typesense_url` : Typesense host URL
 * `PyST_typesense_api_key` : Typesense API key. Must have collection creation rights.
 * `PyST_typesense_embedding_model` : [Typesense embedding model](https://typesense.org/docs/28.0/api/vector-search.html#using-built-in-models) for semantic search. Default is "ts/all-MiniLM-L12-v2"
-* `PyST_languages` : List of language codes used in the search engine and web UI. Should be a JSON _string_, e.g. `'["en", "de"]'`. Default is `'["en", "de", "es", "fr", "pt", "it", "da"]'`
+* `PyST_typesense_prefix` : Optional prefix for Typesense [collection](https://typesense.org/docs/28.0/api/collections.html#create-a-collection) labels.
+* `PyST_languages` : List of language codes used in the search engine and web UI. Should be a JSON _string_, e.g. `'["en", "de"]'`. Default is `'["en", "de", "es", "fr", "pt", "it", "da"]'`.
+
+!!! Note
+
+    If you are deploying more than one PyST instance, you can set `PyST_typesense_prefix` to a different value for each instance. This will keep the search results for each instance separate. The `PyST_typesense_prefix` should only include letters and numbers, and should start with a letter.
 
 3\. Run the server
 

docs/index.html

@@ -513,8 +513,13 @@ <h2 id="quickstart">Quickstart<a class="headerlink" href="#quickstart" title="Pe
 <li><code>PyST_typesense_url</code> : Typesense host URL</li>
 <li><code>PyST_typesense_api_key</code> : Typesense API key. Must have collection creation rights.</li>
 <li><code>PyST_typesense_embedding_model</code> : <a href="https://typesense.org/docs/28.0/api/vector-search.html#using-built-in-models">Typesense embedding model</a> for semantic search. Default is "ts/all-MiniLM-L12-v2"</li>
-<li><code>PyST_languages</code> : List of language codes used in the search engine and web UI. Should be a JSON <em>string</em>, e.g. <code>'["en", "de"]'</code>. Default is <code>'["en", "de", "es", "fr", "pt", "it", "da"]'</code></li>
+<li><code>PyST_typesense_prefix</code> : Optional prefix for Typesense <a href="https://typesense.org/docs/28.0/api/collections.html#create-a-collection">collection</a> labels.</li>
+<li><code>PyST_languages</code> : List of language codes used in the search engine and web UI. Should be a JSON <em>string</em>, e.g. <code>'["en", "de"]'</code>. Default is <code>'["en", "de", "es", "fr", "pt", "it", "da"]'</code>.</li>
 </ul>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>If you are deploying more than one PyST instance, you can set <code>PyST_typesense_prefix</code> to a different value for each instance. This will keep the search results for each instance separate. The <code>PyST_typesense_prefix</code> should only include letters and numbers, and should start with a letter.</p>
+</div>
 <p>3. Run the server</p>
 <p>PyST is a <a href="https://fastapi.tiangolo.com/">FastAPI app</a>; it can be run <a href="https://fastapi.tiangolo.com/deployment/manually/">as any python ASGI app</a>, e.g. with <code>uvicorn</code>:</p>
 <div class="highlight"><pre><span></span><code><span class="kn">import</span><span class="w"> </span><span class="nn">uvicorn</span>

docs/search/search_index.json

@@ -9,10 +9,6 @@
 logger = structlog.get_logger("py-semantic-taxonomy")
 
 
-def c(language: str) -> str:
-    return f"pyst-concepts-{language}"
-
-
 class TypesenseSearchEngine:
     def __init__(self, url: str, api_key: str, embedding_model: str):
         url = urlparse(url)

src/py_semantic_taxonomy/adapters/persistence/search_engine.py

@@ -14,14 +14,19 @@
 logger = structlog.get_logger("py-semantic-taxonomy")
 
 
-def c(language: str) -> str:
-    return f"pyst-concepts-{language}"
+def c(language: str, prefix: str = "") -> str:
+    if prefix:
+        return f"{prefix}-pyst-concepts-{language}"
+    else:
+        return f"pyst-concepts-{language}"
 
 
 class SearchService:
     def __init__(self, engine: SearchEngine | None = None):
         self.settings = get_settings()
-        self.languages = {lang: c(lang) for lang in self.settings.languages}
+        self.languages = {
+            lang: c(lang, self.settings.typesense_prefix) for lang in self.settings.languages
+        }
 
         if self.settings.typesense_url == "missing" or not self.settings.typesense_url:
             self.configured = False

src/py_semantic_taxonomy/application/search_service.py

@@ -17,6 +17,7 @@ class Settings(BaseSettings):
     typesense_api_key: str = "missing"
     typesense_embedding_model: str = "ts/all-MiniLM-L12-v2"
     typesense_exclude_if_missing_for_language: bool = True
+    typesense_prefix: str = ""
 
     languages: list[str] = ["en", "de", "es", "fr", "pt", "it", "da"]
 

src/py_semantic_taxonomy/cfg.py

@@ -280,6 +280,44 @@ async def typesense(typesense_container, entities):
         await ts.create_concept(entities[i])
 
 
+@pytest.fixture
+async def typesense_container_with_prefix(monkeypatch, test_password: str = "123abc"):
+    container = DockerContainer("typesense/typesense:28.0")
+    container.with_exposed_ports(8108)
+    # Giving the API key in `with_command` in a pytest fixture doesn't work...
+    container.with_env("TYPESENSE_API_KEY", test_password)
+    container.with_command("--data-dir /home --enable-cors")
+    container.start()
+    wait_for_logs(container, "Peer refresh succeeded")
+
+    ip = container.get_container_host_ip()
+    port = container.get_exposed_port(8108)
+
+    check_response = httpx.get(f"http://{ip}:{port}/health")
+    if not check_response.status_code == 200:
+        raise OSError("Typesense container can't start")
+
+    monkeypatch.setenv("PyST_typesense_url", f"http://{ip}:{port}")
+    monkeypatch.setenv("PyST_typesense_api_key", test_password)
+    monkeypatch.setenv("PyST_languages", '["en", "de"]')
+    monkeypatch.setenv("PyST_typesense_prefix", "prefixtest")
+
+    yield
+
+    container.stop()
+
+
+@pytest.fixture
+async def typesense_with_prefix(typesense_container_with_prefix, entities):
+    from py_semantic_taxonomy.dependencies import get_search_service
+
+    ts = get_search_service()
+    await ts.initialize()
+
+    for i in [0, 1, 5, 6]:
+        await ts.create_concept(entities[i])
+
+
 @pytest.fixture
 async def client() -> TestClient:
     from httpx import ASGITransport, AsyncClient

tests/conftest.py

@@ -51,3 +51,10 @@ async def test_search_engine(typesense, entities):
 
     results = await engine.search("diesel", "pyst-concepts-en", True, False)
     assert not results
+
+
+@pytest.mark.typesense
+async def test_search_engine_collection_prefix(typesense_with_prefix, entities):
+    engine = get_search_engine()
+    collections = await engine._collection_labels()
+    assert collections == ["prefixtest-pyst-concepts-de", "prefixtest-pyst-concepts-en"]