Skip to content

Fix build_chart_query producing empty SELECT when pagination is enabled #1442

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
siddhant3030 wants to merge 6 commits into
base: main
Choose a base branch
from
Open

Fix build_chart_query producing empty SELECT when pagination is enabled #1442

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
85f074a
Fix build_chart_query producing empty SELECT when pagination is active
claude Jul 18, 2026
66f0e2e
Add regression tests for paginated chart query generation
Aug 10, 2026
bdb5574
Fold pagination regression tests into query-generation test file
Aug 10, 2026
54d0514
Merge branch 'main' into fix/chart-pagination-empty-select
siddhant3030 Aug 11, 2026
270c17d
Merge branch 'main' into fix/chart-pagination-empty-select
siddhant3030 Aug 11, 2026
1dd9194
Merge branch 'main' into fix/chart-pagination-empty-select
siddhant3030 Aug 18, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
264 changes: 132 additions & 132 deletions ddpui/core/charts/charts_service.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -533,155 +533,155 @@ def build_chart_query(
query_builder = AggQueryBuilder()
query_builder.fetch_from(payload.table_name, payload.schema_name)

# Pivot table charts have their own query builder with ROLLUP support
if payload.chart_type == "pivot_table":
query_builder = build_pivot_table_query(payload, query_builder)

# Apply filters
if payload.dashboard_filters:
query_builder = apply_dashboard_filters(query_builder, payload.dashboard_filters)
if payload.extra_config and payload.extra_config.get("filters"):
query_builder = apply_chart_filters(query_builder, payload.extra_config["filters"])

return query_builder

# Now build the rest of the query logic on top of the (possibly paginated) data source
# Table charts can work with just dimensions (no metrics) - non-aggregated query
# Other charts require metrics for aggregation
if payload.chart_type != "table":
if not payload.metrics or len(payload.metrics) == 0:
raise ValueError("At least one metric is required for aggregated charts")
elif payload.chart_type == "table":
# Table charts: if no metrics, just select dimensions (non-aggregated)
dimensions = normalize_dimensions(payload)
if not dimensions:
raise ValueError("At least one dimension is required for table charts")
# Pivot table charts have their own query builder with ROLLUP support
if payload.chart_type == "pivot_table":
query_builder = build_pivot_table_query(payload, query_builder)

if not payload.metrics or len(payload.metrics) == 0:
# Non-aggregated query: just select dimension columns
for dim_col in dimensions:
if not dim_col or not dim_col.strip():
continue
dim_expr = column(dim_col)
# Always label to ensure consistent key access
dim_expr = dim_expr.label(dim_col)
query_builder.add_column(dim_expr)
# No GROUP BY needed for non-aggregated queries
else:
# Aggregated query: use multi-metric query builder
query_builder = build_multi_metric_query(payload, query_builder, org_warehouse)

# Apply filters and sorting before returning
if payload.dashboard_filters:
query_builder = apply_dashboard_filters(query_builder, payload.dashboard_filters)
if payload.extra_config and payload.extra_config.get("filters"):
query_builder = apply_chart_filters(query_builder, payload.extra_config["filters"])
if payload.extra_config and payload.extra_config.get("sort"):
query_builder = apply_chart_sorting(
query_builder, payload.extra_config["sort"], payload
)
# Apply filters
if payload.dashboard_filters:
query_builder = apply_dashboard_filters(query_builder, payload.dashboard_filters)
if payload.extra_config and payload.extra_config.get("filters"):
query_builder = apply_chart_filters(query_builder, payload.extra_config["filters"])

return query_builder
return query_builder

# For number charts, we don't need dimension columns
if payload.chart_type == "number":
# Use first metric for number charts
metric = payload.metrics[0]
# Now build the rest of the query logic on top of the (possibly paginated) data source
# Table charts can work with just dimensions (no metrics) - non-aggregated query
# Other charts require metrics for aggregation
if payload.chart_type != "table":
if not payload.metrics or len(payload.metrics) == 0:
raise ValueError("At least one metric is required for aggregated charts")
elif payload.chart_type == "table":
# Table charts: if no metrics, just select dimensions (non-aggregated)
dimensions = normalize_dimensions(payload)
if not dimensions:
raise ValueError("At least one dimension is required for table charts")

# Expression metric: inline raw SQL (e.g. "SUM(a)/SUM(b)") — no aggregation/column.
if metric.column_expression:
alias = metric.alias or "expression_metric"
query_builder.add_column(literal_column(metric.column_expression).label(alias))
else:
# Handle count with None column case
if (
metric.aggregation
and metric.aggregation.lower() == "count"
and metric.column is None
):
alias = f"count_all_{metric.alias}" if metric.alias else "count_all"
else:
if not metric.column:
raise ValueError(f"Column is required for {metric.aggregation} aggregation")
alias = metric.alias or f"{metric.aggregation}_{metric.column}"
if not payload.metrics or len(payload.metrics) == 0:
# Non-aggregated query: just select dimension columns
for dim_col in dimensions:
if not dim_col or not dim_col.strip():
continue
dim_expr = column(dim_col)
# Always label to ensure consistent key access
dim_expr = dim_expr.label(dim_col)
query_builder.add_column(dim_expr)
# No GROUP BY needed for non-aggregated queries
else:
# Aggregated query: use multi-metric query builder
query_builder = build_multi_metric_query(payload, query_builder, org_warehouse)

# Just add the aggregate column without any grouping
query_builder.add_aggregate_column(
metric.column,
metric.aggregation,
alias,
)
elif payload.chart_type == "pie":
# Pie charts need dimension and one metric
if not payload.dimension_col:
raise ValueError("dimension_col is required for pie charts")
# Apply filters and sorting before returning
if payload.dashboard_filters:
query_builder = apply_dashboard_filters(query_builder, payload.dashboard_filters)
if payload.extra_config and payload.extra_config.get("filters"):
query_builder = apply_chart_filters(query_builder, payload.extra_config["filters"])
if payload.extra_config and payload.extra_config.get("sort"):
query_builder = apply_chart_sorting(
query_builder, payload.extra_config["sort"], payload
)

# Add dimension column with time grain if specified
dimension_column = column(payload.dimension_col)
return query_builder

# Apply time grain if specified and warehouse type is available
time_grain = payload.extra_config.get("time_grain") if payload.extra_config else None
if time_grain and org_warehouse:
warehouse_type = org_warehouse.wtype.lower()
dimension_column = apply_time_grain(dimension_column, time_grain, warehouse_type)
# Add label to preserve original column name for data access
dimension_column = dimension_column.label(payload.dimension_col)
# For number charts, we don't need dimension columns
if payload.chart_type == "number":
# Use first metric for number charts
metric = payload.metrics[0]

query_builder.add_column(dimension_column)
# Expression metric: inline raw SQL (e.g. "SUM(a)/SUM(b)") — no aggregation/column.
if metric.column_expression:
alias = metric.alias or "expression_metric"
query_builder.add_column(literal_column(metric.column_expression).label(alias))
else:
# Handle count with None column case
if (
metric.aggregation
and metric.aggregation.lower() == "count"
and metric.column is None
):
alias = f"count_all_{metric.alias}" if metric.alias else "count_all"
else:
if not metric.column:
raise ValueError(f"Column is required for {metric.aggregation} aggregation")
alias = metric.alias or f"{metric.aggregation}_{metric.column}"

# Add extra dimension if specified (for combination slices)
if payload.extra_dimension:
query_builder.add_column(column(payload.extra_dimension))
# Just add the aggregate column without any grouping
query_builder.add_aggregate_column(
metric.column,
metric.aggregation,
alias,
)
elif payload.chart_type == "pie":
# Pie charts need dimension and one metric
if not payload.dimension_col:
raise ValueError("dimension_col is required for pie charts")

# Use first metric for pie charts
metric = payload.metrics[0]
# Add dimension column with time grain if specified
dimension_column = column(payload.dimension_col)

# Expression metric: inline raw SQL (e.g. "SUM(a)/SUM(b)") — no aggregation/column.
if metric.column_expression:
alias = metric.alias or "expression_metric"
query_builder.add_column(literal_column(metric.column_expression).label(alias))
else:
# Handle count with None column case
if (
metric.aggregation
and metric.aggregation.lower() == "count"
and metric.column is None
):
alias = f"count_all_{metric.alias}" if metric.alias else "count_all"
else:
if not metric.column:
raise ValueError(f"Column is required for {metric.aggregation} aggregation")
alias = metric.alias or f"{metric.aggregation}_{metric.column}"
# Apply time grain if specified and warehouse type is available
time_grain = payload.extra_config.get("time_grain") if payload.extra_config else None
if time_grain and org_warehouse:
warehouse_type = org_warehouse.wtype.lower()
dimension_column = apply_time_grain(dimension_column, time_grain, warehouse_type)
# Add label to preserve original column name for data access
dimension_column = dimension_column.label(payload.dimension_col)

# Add aggregate column
query_builder.add_aggregate_column(
metric.column,
metric.aggregation,
alias,
)
query_builder.add_column(dimension_column)

# Group by dimension column and extra dimension if provided
if time_grain and org_warehouse:
# When time grain is applied, group by the time grain expression (without label)
warehouse_type = org_warehouse.wtype.lower()
time_grain_expr = apply_time_grain(
column(payload.dimension_col), time_grain, warehouse_type
)
query_builder.group_cols_by(time_grain_expr)
# Add extra dimension if specified (for combination slices)
if payload.extra_dimension:
query_builder.add_column(column(payload.extra_dimension))

# Use first metric for pie charts
metric = payload.metrics[0]

# Expression metric: inline raw SQL (e.g. "SUM(a)/SUM(b)") — no aggregation/column.
if metric.column_expression:
alias = metric.alias or "expression_metric"
query_builder.add_column(literal_column(metric.column_expression).label(alias))
else:
# Handle count with None column case
if (
metric.aggregation
and metric.aggregation.lower() == "count"
and metric.column is None
):
alias = f"count_all_{metric.alias}" if metric.alias else "count_all"
else:
# Normal grouping by column name
query_builder.group_cols_by(payload.dimension_col)
if not metric.column:
raise ValueError(f"Column is required for {metric.aggregation} aggregation")
alias = metric.alias or f"{metric.aggregation}_{metric.column}"

if payload.extra_dimension:
query_builder.group_cols_by(payload.extra_dimension)
# Add aggregate column
query_builder.add_aggregate_column(
metric.column,
metric.aggregation,
alias,
)

# Add default ordering by time grain column when time grain is applied
if time_grain and org_warehouse:
# Order by the dimension column (which will have time grain applied) in ascending order (chronological)
query_builder.order_cols_by([(payload.dimension_col, "asc")])
# Group by dimension column and extra dimension if provided
if time_grain and org_warehouse:
# When time grain is applied, group by the time grain expression (without label)
warehouse_type = org_warehouse.wtype.lower()
time_grain_expr = apply_time_grain(
column(payload.dimension_col), time_grain, warehouse_type
)
query_builder.group_cols_by(time_grain_expr)
else:
# Bar, line, and other charts - use multi-metric query
query_builder = build_multi_metric_query(payload, query_builder, org_warehouse)
# Normal grouping by column name
query_builder.group_cols_by(payload.dimension_col)

if payload.extra_dimension:
query_builder.group_cols_by(payload.extra_dimension)

# Add default ordering by time grain column when time grain is applied
if time_grain and org_warehouse:
# Order by the dimension column (which will have time grain applied) in ascending order (chronological)
query_builder.order_cols_by([(payload.dimension_col, "asc")])
else:
# Bar, line, and other charts - use multi-metric query
query_builder = build_multi_metric_query(payload, query_builder, org_warehouse)
Comment thread
coderabbitai[bot] marked this conversation as resolved.

# Apply dashboard filters if provided
if payload.dashboard_filters:
Expand Down
Loading
Loading