Skip to content

fix(parser): preserve Jinja templates in SQL extraction #445

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
dataders wants to merge 2 commits into
base: main
Choose a base branch
from
+185 −3
Open

fix(parser): preserve Jinja templates in SQL extraction #445

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2888f85
fix(parser): preserve Jinja templates in SQL extraction
dataders May 8, 2026
efebea9
Fix Jinja source parsing boundaries
dataders May 9, 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
33 changes: 33 additions & 0 deletions src/parser/source_tree.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -264,6 +264,18 @@ mod tests {
assert!(viz.starts_with("visualise"));
}

#[test]
fn test_extract_sql_preserves_jinja_ref() {
let query = "SELECT order_date, region, revenue FROM {{ ref('fct_orders') }}\nVISUALISE order_date AS x, revenue AS y, region AS color\nDRAW point";
let tree = SourceTree::new(query).unwrap();

let sql = tree.extract_sql().unwrap();
assert_eq!(
sql,
"SELECT order_date, region, revenue FROM {{ ref('fct_orders') }}"
);
}

#[test]
fn test_extract_sql_no_visualise() {
let query = "SELECT * FROM data WHERE x > 5";
Expand All @@ -289,6 +301,18 @@ mod tests {
assert!(viz.starts_with("VISUALISE FROM mtcars"));
}

#[test]
fn test_extract_sql_visualise_from_jinja_ref() {
let query = "VISUALISE FROM {{ ref('fct_orders') }} DRAW point MAPPING x AS x, y AS y";
let tree = SourceTree::new(query).unwrap();

let sql = tree.extract_sql().unwrap();
assert_eq!(sql, "SELECT * FROM {{ ref('fct_orders') }}");

let viz = tree.extract_visualise().unwrap();
assert!(viz.starts_with("VISUALISE FROM {{ ref('fct_orders') }}"));
}

#[test]
fn test_extract_sql_visualise_from_with_cte() {
let query =
Expand Down Expand Up @@ -407,6 +431,15 @@ mod tests {
assert!(sql.contains("SELECT * FROM 'mtcars.csv'"));
}

#[test]
fn test_extract_sql_from_first_jinja_ref() {
let query = "FROM {{ ref('fct_orders') }} VISUALISE DRAW point MAPPING x AS x, y AS y";
let tree = SourceTree::new(query).unwrap();

let sql = tree.extract_sql().unwrap();
assert_eq!(sql, "SELECT * FROM {{ ref('fct_orders') }}");
}

#[test]
fn test_extract_sql_from_first_case_insensitive() {
let query = "from sales visualise DRAW point MAPPING x AS x, y AS y";
Expand Down
36 changes: 33 additions & 3 deletions tree-sitter-ggsql/grammar.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,10 @@ function caseInsensitive(keyword) {
module.exports = grammar({
name: 'ggsql',

inline: $ => [
$.source_ref,
],

conflicts: $ => [
[$.sql_portion],
],
Expand Down Expand Up @@ -63,6 +67,7 @@ module.exports = grammar({
$.case_expression,
$.cast_expression,
$.function_call,
$.jinja_template,
$.non_from_sql_keyword,
$.string,
$.number,
Expand All @@ -84,6 +89,7 @@ module.exports = grammar({
$.case_expression, // CASE WHEN ... THEN ... END
$.cast_expression, // CAST(expr AS type), TRY_CAST(expr AS type)
$.function_call, // Regular function calls like COUNT(), SUM()
$.jinja_template,
$.sql_keyword,
$.string,
$.number,
Expand Down Expand Up @@ -128,6 +134,7 @@ module.exports = grammar({
$.identifier,
$.string,
$.number,
$.jinja_template,
$.subquery,
',', '(', ')', '*', '.', '=',
/[^\s;(),'"]+/
Expand All @@ -143,6 +150,7 @@ module.exports = grammar({
$.identifier,
$.string,
$.number,
$.jinja_template,
$.subquery,
',', '(', ')', '*', '.', '=',
/[^\s;(),'"]+/
Expand All @@ -157,6 +165,7 @@ module.exports = grammar({
$.identifier,
$.string,
$.number,
$.jinja_template,
$.subquery,
',', '(', ')', '*', '.', '=',
/[^\s;(),'"]+/
Expand All @@ -171,6 +180,7 @@ module.exports = grammar({
$.identifier,
$.string,
$.number,
$.jinja_template,
$.subquery,
',', '(', ')', '*', '.', '=',
/[^\s;(),'"]+/
Expand All @@ -179,6 +189,7 @@ module.exports = grammar({

other_sql_statement: $ => prec(-1, repeat1(choice(
$.non_from_sql_keyword,
$.jinja_template,
/[^\s;(),'"]+/,
$.string,
$.number,
Expand Down Expand Up @@ -218,6 +229,7 @@ module.exports = grammar({
$.sql_keyword,
$.string,
$.number,
$.jinja_template,
$.identifier,
$.subquery,
',', '*', '.', '=', '<', '>', '!', '::',
Expand All @@ -242,6 +254,7 @@ module.exports = grammar({
$.cast_expression,
$.function_call,
$.subquery, // also handles IN-lists like ('a', 'b')
$.jinja_template,
token('='), token('!='), token('<>'), token('<='), token('>='),
token('<'), token('>'),
token('+'), token('-'), token('*'), token('/'), token('%'), token('||'), token('::'),
Expand Down Expand Up @@ -396,6 +409,7 @@ module.exports = grammar({
$.qualified_name, // Handles both simple identifiers and table.column
$.number,
$.string,
$.jinja_template,
'*',
// CASE expression
$.case_expression,
Expand Down Expand Up @@ -470,9 +484,16 @@ module.exports = grammar({
repeat(seq('.', $.identifier))
)),

source_ref: $ => choice(
$.qualified_name,
$.string,
$.namespaced_identifier,
$.jinja_template
),

table_ref: $ => prec.right(seq(
choice(
field('table', choice($.qualified_name, $.string, $.namespaced_identifier)),
field('table', $.source_ref),
$.subquery,
),
optional(seq(
Expand Down Expand Up @@ -591,14 +612,14 @@ module.exports = grammar({
// Option 1: Just FROM (inherit global mappings)
seq(
caseInsensitive('FROM'),
field('layer_source', choice($.qualified_name, $.string, $.namespaced_identifier))
field('layer_source', $.source_ref)
),
// Option 2: Mapping list (uses shared structure), optionally followed by FROM
seq(
$.mapping_list,
optional(seq(
caseInsensitive('FROM'),
field('layer_source', choice($.qualified_name, $.string, $.namespaced_identifier))
field('layer_source', $.source_ref)
))
)
)
Expand Down Expand Up @@ -928,6 +949,15 @@ module.exports = grammar({
$.quoted_identifier
),

// Jinja templates are opaque SQL-side tokens. dbt/fusion renders these
// before ggsql executes SQL, but the parser must preserve them while
// splitting SQL from VISUALISE.
jinja_template: $ => token(choice(
seq('{{', repeat(choice(/[^}]+/, /}[^}]/)), '}}'),
seq('{%', repeat(choice(/[^%]+/, /%[^%]/)), '%}'),
seq('{#', repeat(choice(/[^#]+/, /#[^#]/)), '#}')
)),

// Identifier for use in filter expressions - uses lower precedence so that
// keywords like PARTITION and ORDER can take priority and end the filter
filter_identifier: $ => token(prec(-1, /[a-zA-Z_][a-zA-Z0-9_]*/)),
Expand Down
119 changes: 119 additions & 0 deletions tree-sitter-ggsql/test/corpus/basic.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3596,3 +3596,122 @@ SELECT grade, ROUND(COUNT(CASE WHEN status = 'Default' THEN 1 END) * 100.0 / COU
(viz_clause
(draw_clause
(geom_type)))))

================================================================================
SQL source with Jinja ref
================================================================================

SELECT order_date, region, revenue FROM {{ ref('fct_orders') }}
VISUALISE order_date AS x, revenue AS y, region AS color
DRAW point

--------------------------------------------------------------------------------

(query
(sql_portion
(sql_statement
(select_statement
(select_body
(identifier
(bare_identifier))
(identifier
(bare_identifier))
(identifier
(bare_identifier))
(from_clause
(table_ref
table: (jinja_template)))))))
(visualise_statement
(visualise_keyword)
(global_mapping
(mapping_list
(mapping_element
(explicit_mapping
value: (mapping_value
(column_reference
(identifier
(bare_identifier))))
name: (aesthetic_name)))
(mapping_element
(explicit_mapping
value: (mapping_value
(column_reference
(identifier
(bare_identifier))))
name: (aesthetic_name)))
(mapping_element
(explicit_mapping
value: (mapping_value
(column_reference
(identifier
(bare_identifier))))
name: (aesthetic_name)))))
(viz_clause
(draw_clause
(geom_type)))))

================================================================================
SQL source with Jinja var containing dict literal
================================================================================

SELECT * FROM {{ var('table', {'fallback': 'orders'}) }}
VISUALISE x AS x, y AS y
DRAW point

--------------------------------------------------------------------------------

(query
(sql_portion
(sql_statement
(select_statement
(select_body
(from_clause
(table_ref
table: (jinja_template)))))))
(visualise_statement
(visualise_keyword)
(global_mapping
(mapping_list
(mapping_element
(explicit_mapping
value: (mapping_value
(column_reference
(identifier
(bare_identifier))))
name: (aesthetic_name)))
(mapping_element
(explicit_mapping
value: (mapping_value
(column_reference
(identifier
(bare_identifier))))
name: (aesthetic_name)))))
(viz_clause
(draw_clause
(geom_type)))))

================================================================================
Layer source with Jinja ref
================================================================================

VISUALISE
DRAW point MAPPING x AS x FROM {{ ref('fct_orders') }}

--------------------------------------------------------------------------------

(query
(visualise_statement
(visualise_keyword)
(viz_clause
(draw_clause
(geom_type)
(mapping_clause
(mapping_list
(mapping_element
(explicit_mapping
value: (mapping_value
(column_reference
(identifier
(bare_identifier))))
name: (aesthetic_name))))
layer_source: (jinja_template))))))