Data normalization upon load to DuckDB + cleanup templates

Clean up templates

Templates

Author: amrutha97

frontend/src/components/widgets/MarkdownRendererWidget.jsx

@@ -46,7 +46,7 @@ const MarkdownRendererWidget = ({ markdown, value, error, className }) => {
 
   return (
     <Card className={cn('overflow-hidden', className)}>
-      <CardContent className="prose max-w-none prose-pre:p-0 prose-pre:bg-transparent prose-pre:m-0 prose-code:bg-muted prose-code:px-1.5 prose-code:py-0.5 prose-code:rounded prose-code:text-foreground prose-code:before:content-none prose-code:after:content-none">
+      <CardContent className="prose max-w-none prose-pre:p-0 prose-pre:bg-transparent prose-pre:m-0 prose-code:bg-muted prose-code:px-1.5 prose-code:py-0.5 prose-code:rounded prose-code:text-foreground prose-code:before:content-none prose-code:after:content-none prose-headings:mt-2 prose-headings:mb-2 prose-p:my-0 prose-p:mb-0.5">
         <ReactMarkdown
           remarkPlugins={[remarkGfm, remarkSlug]}
           components={{

preswald/engine/managers/data.py

@@ -38,7 +38,6 @@ def load_json_source(config: dict[str, Any]) -> pd.DataFrame:
                 f"Invalid record_path '{record_path}' for JSON file '{path}': {e}"
             ) from e
 
-
     # Normalize or convert data if "flatten"
     try:
         if flatten:
@@ -51,7 +50,6 @@ def load_json_source(config: dict[str, Any]) -> pd.DataFrame:
         ) from e
 
 
-
 # Database Configs ############################################################
 @dataclass
 class ClickhouseConfig:
@@ -118,6 +116,7 @@ class S3CSVConfig:
     s3_use_ssl: bool = False
     s3_url_style: str = "path"
 
+
 class DataSource:
     """Base class for all data sources"""
 
@@ -182,7 +181,14 @@ def __init__(
         self._table_name = f"csv_{name}_{uuid.uuid4().hex[:8]}"
         self._duckdb.execute(f"""
             CREATE TABLE {self._table_name} AS
-            SELECT * FROM read_csv_auto('{self.path}')
+            SELECT * FROM read_csv_auto('{self.path}',
+                header=true,
+                auto_detect=true,
+                ignore_errors=true,
+                normalize_names=true,
+                sample_size=-1,
+                all_varchar=true
+            )
         """)
 
     def query(self, sql: str) -> pd.DataFrame: