feat(cli): Add script entrypoint

README.md

@@ -211,6 +211,15 @@ adaptor.get_by_id(document_id="id:YOUR_NAMESPACE:YOUR_SCHEMA_NAME::SOME_DOCUMENT
 
 All of the above search functionality assumes that a valid set of vespa credentials is available in `~/.vespa`, or in a directory supplied to the `VespaSearchAdapter` constructor directly. See [the docs](docs/vespa-auth.md) for more information on how vespa expects credentials.
 
+# CLI
+
+There is a simple CLI provided.
+
+```bash
+poetry poetry install --extras "vespa"
+poetry run cpr
+```
+
 # Test setup
 
 Some tests rely on a local running instance of vespa.
@@ -247,4 +256,4 @@ make vespa_dev_down
 - Merge.
 - Tag a release manually in github with a version that matches the latest on main that you just merged.
   - In CI/CD we will check that the latest release matches the versions defined in code.
-- Check in `pypi`.
+- Check in `pypi`.

pyproject.toml

@@ -19,6 +19,9 @@ classifiers = [
 
 license = "LICENSE"
 
+[tool.poetry.scripts]
+cpr = "src.cpr_sdk.cli.run_search_query:app"
+
 [project.urls]
 Homepage = "https://github.com/climatepolicyradar/cpr-sdk"
 Repository = "https://github.com/climatepolicyradar/cpr-sdk"
@@ -44,7 +47,7 @@ poetry = "^1.8.3"
 flatten-dict = "^0.4.2"
 
 [tool.poetry.extras]
-vespa = ["pyvespa", "pyyaml", "sentence-transformers", "torch"]
+vespa = ["pyvespa", "pyyaml"]
 spacy = ["spacy"]
 
 [tool.poetry.group.dev]
@@ -58,6 +61,12 @@ black = "^24.2.0"
 moto = { extras = ["s3"], version = "^5.0.13" }
 pytest-dotenv = "^0.5.2"
 
+[tool.poetry.group.cli]
+optional = true
+
+[tool.poetry.group.cli.dependencies]
+typer = "^0.12.5"
+rich = "^13.8.1"
 
 [tool.pytest.ini_options]
 addopts = "-p no:cacheprovider"
@@ -109,3 +118,7 @@ line-length = 88
 [tool.ruff.per-file-ignores]
 "__init__.py" = ["F401"]
 "tests/*" = ["E501"]
+
+[tool.pyright]
+stubPath = ""
+reportMissingImports = false

src/cpr_sdk/cli/run_search_query.py

@@ -0,0 +1,83 @@
+import json
+from rich.console import Console
+from rich.table import Table
+from rich import print_json
+from rich import print as rprint
+import typer
+from src.cpr_sdk.search_adaptors import VespaSearchAdapter
+from src.cpr_sdk.models.search import SearchParameters
+from tests.conftest import VESPA_TEST_SEARCH_URL
+
+app = typer.Typer()
+
+
+@app.callback()
+def main(
+    instance_url: str = VESPA_TEST_SEARCH_URL,
+    exact_match: bool = False,
+    limit: int = 10,
+):
+    """Run a search query with different rank profiles."""
+    console = Console()
+    search_adapter = VespaSearchAdapter(instance_url)
+
+    while True:
+        query = input("Enter your search query (or 'q' to quit): ")
+        if query.lower() == "q":
+            break
+
+        search_parameters = SearchParameters(
+            query_string=query, exact_match=exact_match, limit=limit
+        )
+        search_response = search_adapter.search(search_parameters)
+
+        for family in search_response.families:
+            family_data = family.hits[0].model_dump()
+            console.rule(
+                title=f"{family_data['family_name']} ({family_data['family_geography']} ,{family_data['family_import_id']})"
+            )
+            print_json(
+                json.dumps(
+                    {
+                        k: v
+                        for k, v in family_data.items()
+                        if not k.startswith("text_block") and "metadata" not in k
+                    },
+                    default=str,
+                )
+            )
+
+            # There's some typing weirdness going on here:
+            # hasattr(family.hits[0], 'text_blocks') can be False, but
+            # family.hits[0].text_block exists
+            try:
+                rprint("Text blocks:")
+                table = Table(title="Hits Table")
+
+                # Add columns to the table
+                table.add_column(
+                    "Text Block ID", justify="right", style="cyan", no_wrap=True
+                )
+                table.add_column("Text Block", style="magenta")
+
+                # Add rows to the table
+                for hit in family.hits:
+                    try:
+                        table.add_row(f"{hit.text_block_id}", f"{hit.text_block}")  # type: ignore
+                    except Exception:
+                        pass
+
+                # Print the table
+                console.print(table)
+
+                if family == search_response.families[-1]:
+                    print("No more families to show.")
+                    break
+            except AttributeError:
+                print("No text blocks found.")
+
+            input("Press any key to show next family")
+
+
+if __name__ == "__main__":
+    app()

src/cpr_sdk/version.py

@@ -1,6 +1,6 @@
 _MAJOR = "1"
 _MINOR = "6"
-_PATCH = "1"
+_PATCH = "2"
 _SUFFIX = ""
 
 VERSION_SHORT = "{0}.{1}".format(_MAJOR, _MINOR)