From 6dc672fcc756895cd03b1b62883a70bc1a70ca0a Mon Sep 17 00:00:00 2001
From: Hadi <112569860+anotherhadi@users.noreply.github.com>
Date: Wed, 24 Sep 2025 19:54:32 +0200
Subject: [PATCH] Change suggestion strategy

Signed-off-by: Hadi <112569860+anotherhadi@users.noreply.github.com>
---
 leak-utils/go.mod                 |   1 +
 leak-utils/go.sum                 |   2 +
 leak-utils/parquet/suggestions.go | 106 ++++++++++++++++++------------
 nix/leak-utils.nix                |   2 +-
 4 files changed, 68 insertions(+), 43 deletions(-)

diff --git a/leak-utils/go.mod b/leak-utils/go.mod
index ac3a5f2..e3cc2a9 100644
--- a/leak-utils/go.mod
+++ b/leak-utils/go.mod
@@ -7,6 +7,7 @@ require (
 	github.com/charmbracelet/log v0.4.2
 	github.com/marcboeker/go-duckdb/v2 v2.4.0
 	github.com/spf13/pflag v1.0.10
+	golang.org/x/text v0.28.0
 )
 
 require (
diff --git a/leak-utils/go.sum b/leak-utils/go.sum
index 695ae59..85298e4 100644
--- a/leak-utils/go.sum
+++ b/leak-utils/go.sum
@@ -100,6 +100,8 @@ golang.org/x/sync v0.16.0/go.mod h1:1dzgHSNfp02xaA81J2MS99Qcpr2w7fw1gpm99rleRqA=
 golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.35.0 h1:vz1N37gP5bs89s7He8XuIYXpyY0+QlsKmzipCbUtyxI=
 golang.org/x/sys v0.35.0/go.mod h1:BJP2sWEmIv4KK5OTEluFJCKSidICx8ciO85XgH3Ak8k=
+golang.org/x/text v0.28.0 h1:rhazDwis8INMIwQ4tpjLDzUhx6RlXqZNPEM0huQojng=
+golang.org/x/text v0.28.0/go.mod h1:U8nCwOR8jO/marOQ0QbDiOngZVEBB7MAiitBuMjXiNU=
 golang.org/x/tools v0.36.0 h1:kWS0uv/zsvHEle1LbV5LE8QujrxB3wfQyxHfhOk0Qkg=
 golang.org/x/tools v0.36.0/go.mod h1:WBDiHKJK8YgLHlcQPYQzNCkUxUypCaa5ZegCVutKm+s=
 golang.org/x/xerrors v0.0.0-20240903120638-7835f813f4da h1:noIWHXmPHxILtqtCOPIhSt0ABwskkZKjD3bXGnZGpNY=
diff --git a/leak-utils/parquet/suggestions.go b/leak-utils/parquet/suggestions.go
index 2038c67..49e97ec 100644
--- a/leak-utils/parquet/suggestions.go
+++ b/leak-utils/parquet/suggestions.go
@@ -1,15 +1,20 @@
 package parquet
 
 import (
+	"regexp"
 	"slices"
+	"strings"
+	"unicode"
+
+	"golang.org/x/text/unicode/norm"
 )
 
-func getSuggestion(col string) string {
-	col = formatColumnName(col)
-	knownNames := []string{
+var (
+	knownColumnNames = []string{
 		"date",
 		"phone",
 		"username",
+		"iban",
 		"address",
 		"email",
 		"postal_code",
@@ -28,47 +33,49 @@ func getSuggestion(col string) string {
 		"url",
 		"ip",
 	}
-	if slices.Contains(knownNames, col) {
-		return col
+
+	suggestions = map[string]string{
+		"user":           "username",
+		"login":          "username",
+		"sex":            "gender",
+		"genre":          "gender",
+		"ipaddress":      "ip",
+		"firstname":      "first_name",
+		"prenom":         "first_name",
+		"lastname":       "last_name",
+		"nom":            "last_name",
+		"fullname":       "full_name",
+		"nomcomplet":     "full_name",
+		"adresse":        "address",
+		"streetaddress":  "address",
+		"ville":          "city",
+		"pays":           "country",
+		"mail":           "email",
+		"zip":            "postal_code",
+		"postalcode":     "postal_code",
+		"zipcode":        "postal_code",
+		"postal":         "postal_code",
+		"codepostal":     "postal_code",
+		"hash":           "password_hash",
+		"hashedpassword": "password_hash",
+		"hashpassword":   "password_hash",
+		"passwordhashed": "password_hash",
+		"birthdate":      "birth_date",
+		"dob":            "birth_date",
+		"dateofbirth":    "birth_date",
 	}
-	if col == "user" {
-		return "username"
+)
+
+func getSuggestion(col string) string {
+	colFormated := formatColumnName(col)
+	if slices.Contains(knownColumnNames, colFormated) {
+		return colFormated
 	}
-	if col == "login" {
-		return "username"
-	}
-	if col == "sex" {
-		return "gender"
-	}
-	if col == "ip_address" {
-		return "ip"
-	}
-	if col == "password_hashed" {
-		return "password_hash"
-	}
-	if col == "firstname" {
-		return "first_name"
-	}
-	if col == "lastname" {
-		return "last_name"
-	}
-	if col == "fullname" {
-		return "full_name"
-	}
-	if col == "mail" {
-		return "email"
-	}
-	if col == "zip" || col == "postalcode" || col == "zipcode" || col == "postal" || col == "zip_code" {
-		return "postal_code"
-	}
-	if col == "street_address" {
-		return "address"
-	}
-	if col == "hash" || col == "hashed_password" || col == "hash_password" {
-		return "password_hash"
-	}
-	if col == "birthdate" || col == "dob" || col == "date_of_birth" {
-		return "birth_date"
+
+	col = cleanString(col)
+
+	if val, ok := suggestions[col]; ok {
+		return val
 	}
 
 	return ""
@@ -79,3 +86,18 @@ func getSuggestion(col string) string {
 // url: _url, link
 // address: _address
 //
+
+func cleanString(input string) string {
+	t := norm.NFD.String(input)
+	var sb strings.Builder
+	for _, r := range t {
+		if unicode.Is(unicode.Mn, r) {
+			continue
+		}
+		sb.WriteRune(r)
+	}
+	s := strings.ToLower(sb.String())
+	reg, _ := regexp.Compile("[^a-z]+")
+	s = reg.ReplaceAllString(s, "")
+	return s
+}
diff --git a/nix/leak-utils.nix b/nix/leak-utils.nix
index 6e2653c..e467dfa 100644
--- a/nix/leak-utils.nix
+++ b/nix/leak-utils.nix
@@ -9,7 +9,7 @@
     pname = name;
     version = "0.1.0";
     src = ../leak-utils;
-    vendorHash = "sha256-rTfbXCiwv/+tVXZmgztt088Zhz0OQaVTfvxXVzw4o4Q=";
+    vendorHash = "sha256-qgDqmEgL7B8FvoKNwLG0buLmg9Yt54cyWwmXBifgr/g=";
 
     buildInputs = [
       pkgs.duckdb