From ab7deacadc1ec90f982dbc86e7f448e2b2b1ff26 Mon Sep 17 00:00:00 2001
From: Stephen Young <stephen@customer.io>
Date: Wed, 22 Jan 2025 12:54:56 -0500
Subject: [PATCH] - F add date scrubber

---
 date_scrubber.go                              | 64 +++++++++++++++++++
 date_scrubber_test.go                         | 19 ++++++
 scrubber.go                                   |  2 +-
 .../scrubber_test.TestScrubGuids.approved.txt |  6 +-
 4 files changed, 87 insertions(+), 4 deletions(-)
 create mode 100644 date_scrubber.go
 create mode 100644 date_scrubber_test.go

diff --git a/date_scrubber.go b/date_scrubber.go
new file mode 100644
index 0000000..4aafb12
--- /dev/null
+++ b/date_scrubber.go
@@ -0,0 +1,64 @@
+package approvals
+
+import (
+	"fmt"
+	"regexp"
+)
+
+type DateScrubber struct {
+	pattern     string
+	replacement func(int) string
+}
+
+func NewDateScrubber(pattern string) scrubber {
+	return CreateRegexScrubberWithLabeler(regexp.MustCompile(pattern), func(n int) string {
+		return fmt.Sprintf("[Date%d]", n)
+	})
+}
+
+type SupportedFormat struct {
+	Regex    string
+	Examples []string
+}
+
+func GetSupportedFormats() []SupportedFormat {
+	return []SupportedFormat{
+		{"[a-zA-Z]{3} [a-zA-Z]{3} \\d{2} \\d{2}:\\d{2}:\\d{2}", []string{"Tue May 13 16:30:00"}},
+		{"[a-zA-Z]{3} [a-zA-Z]{3} \\d{2} \\d{2}:\\d{2}:\\d{2} [a-zA-Z]{3,4} \\d{4}", []string{"Wed Nov 17 22:28:33 EET 2021"}},
+		{"(Mon|Tue|Wed|Thu|Fri|Sat|Sun), \\d{2} (Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec) \\d{4} \\d{2}:\\d{2}:\\d{2} GMT", []string{"Wed, 21 Oct 2015 07:28:00 GMT"}},
+		{"[a-zA-Z]{3} [a-zA-Z]{3} \\d{2} \\d{4} \\d{2}:\\d{2}:\\d{2}.\\d{3}", []string{"Tue May 13 2014 23:30:00.789"}},
+		{"[a-zA-Z]{3} [a-zA-Z]{3} \\d{2} \\d{2}:\\d{2}:\\d{2} -\\d{4} \\d{4}", []string{"Tue May 13 16:30:00 -0800 2014"}},
+		{"\\d{2} [a-zA-Z]{3} \\d{4} \\d{2}:\\d{2}:\\d{2},\\d{3}", []string{"13 May 2014 23:50:49,999"}},
+		{"[A-Za-z]{3} \\d{2} \\d{2}:\\d{2}", []string{"Oct 13 15:29"}},
+		{"[a-zA-Z]{3} \\d{2}, \\d{4} \\d{2}:\\d{2}:\\d{2} [a-zA-Z]{2} [a-zA-Z]{3}", []string{"May 13, 2014 11:30:00 PM PST"}},
+		{"\\d{2}:\\d{2}:\\d{2}", []string{"23:30:00"}},
+		{"\\d{4}/\\d{2}/\\d{2} \\d{2}:\\d{2}:\\d{2}(\\.\\d{3})?", []string{"2014/05/13 16:30:59.786", "2014/05/13 16:30:59"}},
+		{"\\d{4}-\\d{1,2}-\\d{1,2}T\\d{1,2}:\\d{2}Z", []string{"2020-9-10T08:07Z", "2020-09-9T08:07Z", "2020-09-10T8:07Z", "2020-09-10T08:07Z"}},
+		{"\\d{4}-\\d{1,2}-\\d{1,2}T\\d{1,2}:\\d{2}:\\d{2}Z", []string{"2020-09-10T08:07:89Z"}},
+		{"\\d{4}-\\d{1,2}-\\d{1,2}T\\d{1,2}:\\d{2}\\:\\d{2}\\.\\d{3}Z", []string{"2020-09-10T01:23:45.678Z"}},
+		{"\\d{8}T\\d{6}Z", []string{"20210505T091112Z"}},
+		{"\\d{4}-\\d{2}-\\d{2}", []string{"2024-12-17"}},
+		{"\\d{4}-\\d{1,2}-\\d{1,2}T\\d{1,2}:\\d{2}:\\d{2}(\\.\\d{1,9})?Z", []string{"2024-12-18T14:04:46.746130Z", "2024-12-18T14:04:46Z", "2024-12-18T14:04:46.746130834Z"}},
+		{"\\d{2}[-/.]\\d{2}[-/.]\\d{4}\\s\\d{2}:\\d{2}(:\\d{2})?( (?:pm|am|PM|AM))?", []string{"13/05/2014 23:50:49", "13.05.2014 23:50:49", "13-05-2014 23:50:49", "13.05.2014 23:50", "05/13/2014 11:50:49 PM"}},
+	}
+}
+
+// func GetScrubberFor(formattedExample string) (*DateScrubber, error) {
+// 	for _, pattern := range GetSupportedFormats() {
+// 		regex := regexp.MustCompile(pattern.Regex)
+// 		if regex.MatchString(formattedExample) {
+// 			return NewDateScrubber(pattern.Regex), nil
+// 		}
+// 	}
+// 	return nil, fmt.Errorf("no match found for %s. Current supported formats are: %s",
+// 		formattedExample, strings.Join(GetSupportedFormatsRegex(), ", "))
+// }
+
+func GetSupportedFormatsRegex() []string {
+	formats := GetSupportedFormats()
+	regexList := make([]string, len(formats))
+	for i, format := range formats {
+		regexList[i] = format.Regex
+	}
+	return regexList
+}
diff --git a/date_scrubber_test.go b/date_scrubber_test.go
new file mode 100644
index 0000000..814e3e7
--- /dev/null
+++ b/date_scrubber_test.go
@@ -0,0 +1,19 @@
+package approvals_test
+
+import (
+	"testing"
+
+	approvals "github.com/approvals/go-approval-tests"
+)
+
+func TestSupportedFormatWorksForExamples(t *testing.T) {
+	for _, supportedFormat := range approvals.GetSupportedFormats() {
+		dateScrubber := approvals.NewDateScrubber(supportedFormat.Regex)
+		for _, example := range supportedFormat.Examples {
+			result := dateScrubber(example)
+			if result != "[Date1]" {
+				t.Errorf("did not work for\nregex: %s\nexample: %s\ngot: %s", supportedFormat.Regex, example, result)
+			}
+		}
+	}
+}
diff --git a/scrubber.go b/scrubber.go
index a39615e..541696d 100644
--- a/scrubber.go
+++ b/scrubber.go
@@ -30,7 +30,7 @@ func CreateRegexScrubberWithLabeler(regex *regexp.Regexp, replacer func(int) str
 				idx = len(seen)
 				seen[s] = idx
 			}
-			return replacer(idx)
+			return replacer(idx + 1)
 		}
 		return regex.ReplaceAllStringFunc(s, replacefn)
 	}
diff --git a/testdata/scrubber_test.TestScrubGuids.approved.txt b/testdata/scrubber_test.TestScrubGuids.approved.txt
index d83bfef..00cdfc7 100644
--- a/testdata/scrubber_test.TestScrubGuids.approved.txt
+++ b/testdata/scrubber_test.TestScrubGuids.approved.txt
@@ -1,8 +1,8 @@
 guids
 
 
-guid_0
 guid_1
 guid_2
-guid_0
-guid_0 and text
\ No newline at end of file
+guid_3
+guid_1
+guid_1 and text
\ No newline at end of file