+search algo to include synonyms bucket (#1626)

Author: LineIndent

.github/workflows/deploy-dev.yml

@@ -47,6 +47,7 @@ jobs:
         run: |
           uv pip install typesense python-frontmatter markdown beautifulsoup4
           python scripts/typesense_indexer.py --docs-path ./docs --blog-path ./blog --force
+          python scripts/synonym_indexer.py
 
       - name: Deploy to Reflex
         id: deploy

.github/workflows/deploy-prd.yml

@@ -45,6 +45,7 @@ jobs:
         run: |
           uv pip install typesense python-frontmatter markdown beautifulsoup4
           python scripts/typesense_indexer.py --docs-path ./docs --blog-path ./blog --force
+          python scripts/synonym_indexer.py
 
       - name: Deploy to Reflex
         id: deploy

.github/workflows/deploy-stg.yml

@@ -47,6 +47,7 @@ jobs:
         run: |
           uv pip install typesense python-frontmatter markdown beautifulsoup4
           python scripts/typesense_indexer.py --docs-path ./docs --blog-path ./blog --force
+          python scripts/synonym_indexer.py
 
       - name: Deploy to Reflex
         id: deploy

scripts/synonym_indexer.py

@@ -0,0 +1,88 @@
+""" Dynamically generate synonyms for docs to better type match"""
+import os
+import pathlib
+import typesense
+
+from typesense_indexer import TYPESENSE_CONFIG
+
+client = typesense.Client(TYPESENSE_CONFIG)
+
+def get_folder_hierarchy(root: str) -> dict:
+    """Recursively build folder hierarchy from root, ignoring unwanted folders."""
+    hierarchy = {}
+    for entry in os.scandir(root):
+        if entry.is_dir() and entry.name not in ("__pycache__", ".git", ".venv"):
+            hierarchy[entry.name] = get_folder_hierarchy(entry.path)
+    return hierarchy
+
+def generate_synonyms(name: str) -> list[str]:
+    """
+    Generate multiple synonym forms for a folder/component name:
+    - flattened lowercase: reactflow
+    - lowercase spaced: react flow
+    - title case spaced: React Flow
+    - underscore: react_flow
+    - hyphen: react-flow
+    - camelcase: ReactFlow
+    """
+    clean_name = name.replace("_", " ").replace("-", " ").strip()
+    words = clean_name.split()
+
+    synonyms = set()
+
+    # 1) Flattened lowercase
+    synonyms.add("".join(w.lower() for w in words))
+    # 2) Lowercase spaced
+    synonyms.add(" ".join(w.lower() for w in words))
+    # 3) Title case spaced
+    synonyms.add(" ".join(w.capitalize() for w in words))
+    # 4) Original underscore
+    synonyms.add("_".join(w.lower() for w in words))
+    # 5) Original hyphen
+    synonyms.add("-".join(w.lower() for w in words))
+    # 6) CamelCase
+    synonyms.add("".join(w.capitalize() for w in words))
+
+    return list(synonyms)
+
+def flatten_hierarchy(hierarchy: dict) -> dict:
+    """
+    Flatten nested folder hierarchy into a single dict with synonyms
+    keyed by flattened lowercase name.
+    """
+    flat = {}
+
+    def recurse(subtree):
+        for key, value in subtree.items():
+            # Flatten key for canonical form
+            key_flat = "".join(key.lower().split("_")).replace("-", "")
+            flat[key_flat] = {"synonyms": generate_synonyms(key)}
+            if value:
+                recurse(value)
+
+    recurse(hierarchy)
+    return flat
+
+def main() -> bool:
+    try:
+        docs_root = pathlib.Path("docs")
+        hierarchy = get_folder_hierarchy(docs_root)
+        SYNONYMS = flatten_hierarchy(hierarchy)
+
+        print("Upserting new synonyms to Typesense ...")
+        for canonical, info in SYNONYMS.items():
+            client.collections["docs"].synonyms.upsert(
+                canonical,
+                {"synonyms": info["synonyms"]}
+            )
+
+        print("Synonyms synced successfully!")
+        return True
+
+    except Exception as e:
+        print("Error syncing synonyms:", e)
+        return False
+
+if __name__ == "__main__":
+    success = main()
+    exit(0 if success else 1)

scripts/typesense_indexer.py

@@ -1,3 +1,5 @@
+"""Generate indexed docs + collection for typesense search"""
+
 import os
 import pathlib
 import datetime
@@ -10,16 +12,15 @@
 
 import typesense
 import reflex as rx
-from reflex.utils.imports import ImportVar
 
-# Add the project root to the sys.path
+
 project_root = pathlib.Path(__file__).resolve().parent.parent
 if str(project_root) not in sys.path:
     sys.path.insert(0, str(project_root))
 
-from pcweb.pages.docs.source import Source, generate_docs
+from pcweb.pages.docs.source import Source
 from pcweb.pages.docs.apiref import modules
-from pcweb.pages.docs.env_vars import env_vars_page, EnvVarDocs
+from pcweb.pages.docs.env_vars import EnvVarDocs
 
 logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
 logger = logging.getLogger(__name__)
@@ -73,12 +74,13 @@ def _extract_headings_from_component(c: Any) -> List[str]:
     'connection_timeout_seconds': 60
 }
 
+
 COLLECTION_SCHEMA = {
     'name': 'docs',
     'fields': [
-        {'name': 'id', 'type': 'string'},
-        {'name': 'title', 'type': 'string'},
-        {'name': 'content', 'type': 'string'},
+        {'name': 'id', 'type': 'string', "infix": True},
+        {'name': 'title', 'type': 'string', "infix": True},
+        {'name': 'content', 'type': 'string', "infix": True},
         {'name': 'headings', 'type': 'string[]'},
         {'name': 'components', 'type': 'string[]', 'optional': True},
         {'name': 'path', 'type': 'string'},
@@ -91,6 +93,8 @@ def _extract_headings_from_component(c: Any) -> List[str]:
     ],
 }
 
+
+
 class SimpleTypesenseIndexer:
     """Simplified indexer using your existing logic."""