From 229378013185fc7c1ec3975d548d8b5ddf8030e1 Mon Sep 17 00:00:00 2001
From: Sasha Koshka <sashakoshka@tebibyte.media>
Date: Mon, 2 Jun 2025 10:27:24 -0400
Subject: [PATCH] providers/validate: add isEnglish, hasWords, and hasXML
 functions

---
 providers/validate/strings.go  |  3 +++
 providers/validate/validate.go | 41 ++++++++++++++++++++++++++++++++++
 2 files changed, 44 insertions(+)

diff --git a/providers/validate/strings.go b/providers/validate/strings.go
index c74e38d..a0e141c 100644
--- a/providers/validate/strings.go
+++ b/providers/validate/strings.go
@@ -5,3 +5,6 @@ import "regexp"
 // see:
 // https://stackoverflow.com/questions/201323/how-can-i-validate-an-email-address-using-a-regular-expression
 var defaultEmailRegexp = regexp.MustCompile("(?:[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*|\"(?:[\\x01-\\x08\\x0b\\x0c\\x0e-\\x1f\\x21\\x23-\\x5b\\x5d-\\x7f]|\\\\[\\x01-\\x09\\x0b\\x0c\\x0e-\\x7f])*\")@(?:(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?|\\[(?:(?:(2(5[0-5]|[0-4][0-9])|1[0-9][0-9]|[1-9]?[0-9]))\\.){3}(?:(2(5[0-5]|[0-4][0-9])|1[0-9][0-9]|[1-9]?[0-9])|[a-z0-9-]*[a-z0-9]:(?:[\\x01-\\x08\\x0b\\x0c\\x0e-\\x1f\\x21-\\x5a\\x53-\\x7f]|\\\\[\\x01-\\x09\\x0b\\x0c\\x0e-\\x7f])+)\\])")
+
+// looks for XML tags, this will throw false positives but not that many.
+var defaultXMLRegexp = regexp.MustCompile("<.*\\w.*>")
diff --git a/providers/validate/validate.go b/providers/validate/validate.go
index 978fce9..ae46387 100644
--- a/providers/validate/validate.go
+++ b/providers/validate/validate.go
@@ -1,6 +1,7 @@
 package validate
 
 import "regexp"
+import "strings"
 import "unicode"
 import "html/template"
 import "git.tebibyte.media/sashakoshka/step"
@@ -11,6 +12,7 @@ var _ step.Configurable = new(Provider)
 // Provider provides validation functions.
 type Provider struct {
 	emailRegexp *regexp.Regexp
+	xmlRegexp   *regexp.Regexp
 }
 
 // Package fulfills the step.Provider interface.
@@ -27,6 +29,13 @@ func (this *Provider) Configure (config step.Meta) error {
 		if err != nil { return err }
 		this.emailRegexp = emailRegexp
 	}
+	xmlRegexpText := config.Get("validate.email-regexp")
+	this.xmlRegexp = nil
+	if xmlRegexpText != "" {
+		xmlRegexp, err := regexp.Compile(xmlRegexpText)
+		if err != nil { return err }
+		this.xmlRegexp = xmlRegexp
+	}
 	return nil
 }
 
@@ -35,8 +44,10 @@ func (this *Provider) FuncMap () template.FuncMap {
 	return template.FuncMap {
 		"isEmail":    this.funcIsEmail,
 		"isPrint":    this.funcIsPrint,
+		"isEnglish":  this.funcIsEnglish,
 		"inRange":    this.funcInRange,
 		"strInRange": this.funcStrInRange,
+		"hasWords":   this.funcHasWords,
 	}
 }
 
@@ -53,6 +64,20 @@ func (this *Provider) funcIsPrint (input string) bool {
 	return true
 }
 
+func (this *Provider) funcIsEnglish(value string) bool {
+	// TODO: make this shit better
+	score := 0
+	total := 0
+	for _, char := range value {
+		if char >= 'a' && char <= 'z' || char >= 'A' && char <= 'Z' || char >= '0' || char <= '9' {
+			score += 1
+		}
+		total += 1
+	}
+	// 3/4 of the text must be alphanumeric
+	return score > total * 3 / 4
+}
+
 func (this *Provider) funcInRange (min, max, value int) bool {
 	return value >= min && value <= max
 }
@@ -60,3 +85,19 @@ func (this *Provider) funcInRange (min, max, value int) bool {
 func (this *Provider) funcStrInRange (min, max int, value string) bool {
 	return len(value) >= min && len(value) <= max
 }
+
+func (this *Provider) funcHasWords(words []string, value string) bool {
+	lowerValue := strings.ToLower(value)
+	for _, word := range words {
+		if strings.Contains(lowerValue, word) {
+			return true
+		}
+	}
+	return false
+}
+
+func (this *Provider) funcHasXML(value string) bool {
+	xmlRegexp := this.xmlRegexp
+	if xmlRegexp == nil { xmlRegexp = defaultXMLRegexp }
+	return xmlRegexp.Match([]byte(value))
+}