chore: more work done and planning with AI.

docs/plan.md

@@ -71,6 +71,89 @@ to the intended style below.
   `::`, `->`, `->>`, array `[...]`, or before a call's `(`.
 - Dollar-quote tags preserved verbatim (`$$`, `$S$`, `$Z$`, …).
 
+## DataGrip settings mapping
+
+`PostgresCodeStyleSettings` (DataGrip / JetBrains) is the reference for all configurable
+style options. The table below maps every relevant DataGrip key to its `.pgtidy.yaml`
+counterpart so a user can reproduce their DataGrip style exactly in PgTidy.
+
+DataGrip enum conventions used below:
+- **Case**: 0=preserve, 1=upper, 2=lower
+- **Comma**: 1=leading (`,col`), 2=trailing (`col,`)
+- **Placement**: 1=same_line, 2=new_line
+- **Wrap**: 0=never, 1=when_long, 2=always
+
+### Casing
+
+| DataGrip key | PgTidy key | Default | Notes |
+|---|---|---|---|
+| `KEYWORD_CASE` | `keyword_case` | `upper` | SELECT, FROM, WHERE, … |
+| `IDENTIFIER_CASE` | `ident_case` | `lower` | unquoted column/variable names |
+| `TYPE_CASE` | `type_case` | `lower` | built-in type names (text, integer, …) |
+| `CUSTOM_TYPE_CASE` | `custom_type_case` | `lower` | user-defined / domain types |
+| `ALIAS_CASE` | `alias_case` | `lower` | column and table aliases |
+| `BUILT_IN_CASE` | `builtin_case` | `lower` | built-in functions (COALESCE, MAX, …) |
+
+### Query layout
+
+| DataGrip key | PgTidy key | Default | Notes |
+|---|---|---|---|
+| `QUERY_EL_COMMA` | `commas` | `leading` | applies to all clause element lists |
+| `QUERY_ALIGN_ELEMENTS` | `align_columns` | `false` | align SELECT list items to same column |
+| `QUERY_ALIGN_LINE_COMMENTS` | `align_line_comments` | `false` | align `--` inline comments in a block |
+| `SELECT_ALIGN_AS` | `select_align_as` | `false` | align `AS` keyword across SELECT list |
+| `FROM_INDENT_JOIN` | `indent_join` | `false` | indent JOIN relative to FROM |
+| `FROM_ONLY_JOIN_INDENT` | `join_indent_size` | `1` | extra indent levels for JOINs |
+| `SET_ALIGN_EQUAL_SIGN` | `set_align_equal` | `false` | align `=` in UPDATE SET list |
+| `WHERE_EL_WRAP` + `WHERE_EL_LINE` | `where_wrap` | `always` | always \| when_long \| never — each AND/OR condition on its own line |
+| _(no DataGrip equivalent)_ | `where_and_or_indent` | `true` | when true, AND/OR are indented one level under WHERE, not at WHERE's column |
+
+### Subqueries
+
+| DataGrip key | PgTidy key | Default | Notes |
+|---|---|---|---|
+| `SUBQUERY_OPENING` | `subquery_opening` | `same_line` | opening `(` placement |
+| `SUBQUERY_CONTENT` | `subquery_content` | `new_line` | content indentation inside paren |
+| `SUBQUERY_CLOSING` | `subquery_closing` | `new_line` | closing `)` placement |
+| `SUBQUERY_PAR_SPACE_BEFORE` | `subquery_space_before_paren` | `false` | space before `(` |
+
+### INSERT
+
+| DataGrip key | PgTidy key | Default | Notes |
+|---|---|---|---|
+| `INSERT_COLLAPSE_MULTI_ROW_VALUES` | `insert_collapse_values` | `true` | fold VALUES rows into fewer lines |
+
+### Routine (function / procedure)
+
+| DataGrip key | PgTidy key | Default | Notes |
+|---|---|---|---|
+| `ROUTINE_ARG_COMMA` | uses `commas` | `leading` | same setting as query lists |
+| `ROUTINE_ARG_ALIGN_TYPES` | `align_param_types` | `true` | align type column in param list |
+| `ROUTINE_AS_WRAP` | `routine_as_wrap` | `true` | newline before `AS $$` |
+
+### PL/pgSQL body
+
+| DataGrip key | PgTidy key | Default | Notes |
+|---|---|---|---|
+| `IMP_COMMON_KEEP_BLANK_LINES_IN_CODE` | `plpgsql_max_blank_lines` | `1` | max consecutive blank lines in body |
+| `IMP_DECLARE_ALIGN_TYPE` | `plpgsql_declare_align_type` | `false` | align type column in DECLARE block |
+| `IMP_DECLARE_ALIGN_EQ` | `plpgsql_declare_align_eq` | `false` | align `:=` / `=` in DECLARE block |
+| `IMP_IF_THEN_WRAP_THEN` | `plpgsql_if_then_newline` | `true` | THEN on its own line |
+| `IMP_LOOP_COLLAPSE` | `plpgsql_loop_collapse` | `true` | collapse empty loop bodies |
+
+### Expressions
+
+| DataGrip key | PgTidy key | Default | Notes |
+|---|---|---|---|
+| `EXPR_BINARY_OP_ALIGN` | `binary_op_align` | `false` | align `=`, `<>`, `||`, … vertically in WHERE/expression lists; default false — must not be hardcoded |
+| `EXPR_CALL_SPACE_AFTER_COMMA` | `space_after_comma_in_calls` | `false` | space after `,` in function calls |
+| `EXPR_CASE_WHEN_WRAP` | `case_when_wrap` | `false` | each WHEN on its own line |
+| `EXPR_CASE_END` | `case_end` | `new_line` | same_line \| new_line |
+| `EXPR_CASE_COLLAPSE` | `case_collapse` | `false` | collapse short CASE to one line |
+| `CORTEGE_SPACE_BEFORE_L_PAREN` | `record_space_before_paren` | `false` | space before `(` in ROW/record constructors |
+
+---
+
 ## Milestones
 
 ### V1 — Formatter + CLI (priority)
@@ -88,6 +171,8 @@ to the intended style below.
 5. **CLI** (`cmd/pgtidy fmt`): `--check`, `--write`/`-w`, stdin→stdout, `--diff`; config
    discovery walking up to `.pgtidy.yaml`; CI-friendly exit codes.
 6. **Config** (`pkg/config`): load/merge style config; defaults = house style above.
+   Full field set defined in the DataGrip settings mapping section above — covers casing
+   (6 keys), query layout, subqueries, INSERT, routines, PL/pgSQL body, and expressions.
 
 **Safety guarantees (tested):** semantic equivalence (re-lex output, compare non-trivia token
 stream to input), and idempotence (`fmt(fmt(x)) == fmt(x)`). The corpus is the