PgTidy/pkg/lint/rules_naming.go

package lint

import (
	"fmt"
	"regexp"
	"strings"

	pg_query "github.com/pganalyze/pg_query_go/v6"

	"git.warky.dev/wdevs/pgtidy/pkg/diagnostics"
	"git.warky.dev/wdevs/pgtidy/pkg/pgast"
)

// reSnakeCase matches valid snake_case identifiers: lowercase letters, digits,
// and underscores, starting with a letter or underscore.
var reSnakeCase = regexp.MustCompile(`^[a-z_][a-z0-9_]*$`)

func isSnakeCase(s string) bool {
	return reSnakeCase.MatchString(s)
}

// NAM001 — table names must be snake_case.
type ruleNAM001 struct{}

func (ruleNAM001) ID() string                     { return "NAM001" }
func (ruleNAM001) Severity() diagnostics.Severity { return diagnostics.SeverityHint }

func (ruleNAM001) Check(stmts []*pg_query.RawStmt, src string) []diagnostics.Diagnostic {
	var out []diagnostics.Diagnostic
	for _, raw := range stmts {
		cs, ok := raw.Stmt.GetNode().(*pg_query.Node_CreateStmt)
		if !ok {
			continue
		}
		name := cs.CreateStmt.Relation.Relname
		if name == "" || isSnakeCase(name) {
			continue
		}
		line, col := pgast.LocationToLineCol(src, pgast.FirstTokenOffset(src, int(raw.StmtLocation)))
		out = append(out, diagnostics.Diagnostic{
			RuleID:   "NAM001",
			Severity: diagnostics.SeverityHint,
			Message:  fmt.Sprintf("table name %q is not snake_case (expected %q)", name, toSnakeCase(name)),
			Line:     line,
			Col:      col,
		})
	}
	return out
}

// NAM002 — column names in CREATE TABLE must be snake_case.
type ruleNAM002 struct{}

func (ruleNAM002) ID() string                     { return "NAM002" }
func (ruleNAM002) Severity() diagnostics.Severity { return diagnostics.SeverityHint }

func (ruleNAM002) Check(stmts []*pg_query.RawStmt, src string) []diagnostics.Diagnostic {
	var out []diagnostics.Diagnostic
	for _, raw := range stmts {
		cs, ok := raw.Stmt.GetNode().(*pg_query.Node_CreateStmt)
		if !ok {
			continue
		}
		for _, elt := range cs.CreateStmt.TableElts {
			cd, ok := elt.GetNode().(*pg_query.Node_ColumnDef)
			if !ok {
				continue
			}
			name := cd.ColumnDef.Colname
			if name == "" || isSnakeCase(name) {
				continue
			}
			line, col := pgast.LocationToLineCol(src, int(cd.ColumnDef.Location))
			out = append(out, diagnostics.Diagnostic{
				RuleID:   "NAM002",
				Severity: diagnostics.SeverityHint,
				Message:  fmt.Sprintf("column name %q is not snake_case (expected %q)", name, toSnakeCase(name)),
				Line:     line,
				Col:      col,
			})
		}
	}
	return out
}

// NAM003 — function/procedure names must be snake_case.
type ruleNAM003 struct{}

func (ruleNAM003) ID() string                     { return "NAM003" }
func (ruleNAM003) Severity() diagnostics.Severity { return diagnostics.SeverityHint }

func (ruleNAM003) Check(stmts []*pg_query.RawStmt, src string) []diagnostics.Diagnostic {
	var out []diagnostics.Diagnostic
	for _, raw := range stmts {
		cf, ok := raw.Stmt.GetNode().(*pg_query.Node_CreateFunctionStmt)
		if !ok {
			continue
		}
		// Funcname is a list of String nodes: [schema, funcname] or [funcname].
		name := lastStringNode(cf.CreateFunctionStmt.Funcname)
		if name == "" || isSnakeCase(name) {
			continue
		}
		line, col := pgast.LocationToLineCol(src, pgast.FirstTokenOffset(src, int(raw.StmtLocation)))
		out = append(out, diagnostics.Diagnostic{
			RuleID:   "NAM003",
			Severity: diagnostics.SeverityHint,
			Message:  fmt.Sprintf("function name %q is not snake_case (expected %q)", name, toSnakeCase(name)),
			Line:     line,
			Col:      col,
		})
	}
	return out
}

// lastStringNode returns the string value of the last node in a name-list
// (schema-qualified names like [schema, name] → name).
func lastStringNode(nodes []*pg_query.Node) string {
	for i := len(nodes) - 1; i >= 0; i-- {
		sv, ok := nodes[i].GetNode().(*pg_query.Node_String_)
		if ok {
			return sv.String_.Sval
		}
	}
	return ""
}

// toSnakeCase converts a CamelCase or mixed-case identifier to snake_case as a
// suggested fix shown in the diagnostic message.
var reWordBoundary = regexp.MustCompile(`([a-z0-9])([A-Z])`)

func toSnakeCase(s string) string {
	s = reWordBoundary.ReplaceAllString(s, "${1}_${2}")
	return strings.ToLower(s)
}