Generate text in case of utf8 descriptions (#5014)

The idea is to mimic the format in which graph convert html into text when generating ics which gets embedded into eml file.

---

#### Does this PR need a docs update or release note?

- [ ] ✅ Yes, it's included
- [x] 🕐 Yes, but in a later PR
- [ ] ⛔ No

#### Type of change

<!--- Please check the type of change your PR introduces: --->
- [x] 🌻 Feature
- [ ] 🐛 Bugfix
- [ ] 🗺️ Documentation
- [ ] 🤖 Supportability/Tests
- [ ] 💻 CI/Deployment
- [ ] 🧹 Tech Debt/Cleanup

#### Issue(s)

<!-- Can reference multiple issues. Use one of the following "magic words" - "closes, fixes" to auto-close the Github issue. -->
* #<issue>

#### Test Plan

<!-- How will this be tested prior to merging.-->
- [ ] 💪 Manual
- [x] ⚡ Unit test
- [ ] 💚 E2E

src/go.mod

@@ -90,6 +90,7 @@ require (
 	github.com/valyala/fasthttp v1.51.0 // indirect
 	go.opentelemetry.io/otel/metric v1.21.0 // indirect
 	google.golang.org/genproto/googleapis/rpc v0.0.0-20231127180814-3a041ad873d4 // indirect
+	jaytaylor.com/html2text v0.0.0-20230321000545-74c2419ad056 // indirect
 )
 
 require (

src/go.sum

@@ -435,3 +435,5 @@ gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
 gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
 gotest.tools/v3 v3.5.1 h1:EENdUnS3pdur5nybKYIh2Vfgc8IUNBjxDPSjtiJcOzU=
 gotest.tools/v3 v3.5.1/go.mod h1:isy3WKz7GK6uNw/sbHzfKBLvlvXwUyV06n6brMxxopU=
+jaytaylor.com/html2text v0.0.0-20230321000545-74c2419ad056 h1:6YFJoB+0fUH6X3xU/G2tQqCYg+PkGtnZ5nMR5rpw72g=
+jaytaylor.com/html2text v0.0.0-20230321000545-74c2419ad056/go.mod h1:OxvTsCwKosqQ1q7B+8FwXqg4rKZ/UG9dUW+g/VL2xH4=

src/internal/converters/ics/ics.go

@@ -7,10 +7,12 @@ import (
 	"fmt"
 	"strings"
 	"time"
+	"unicode"
 
 	"github.com/alcionai/clues"
 	ics "github.com/arran4/golang-ical"
 	"github.com/microsoftgraph/msgraph-sdk-go/models"
+	"jaytaylor.com/html2text"
 
 	"github.com/alcionai/corso/src/internal/common/ptr"
 	"github.com/alcionai/corso/src/internal/common/str"
@@ -251,6 +253,16 @@ func FromJSON(ctx context.Context, body []byte) (string, error) {
 	return cal.Serialize(), nil
 }
 
+func isASCII(s string) bool {
+	for _, c := range s {
+		if c > unicode.MaxASCII {
+			return false
+		}
+	}
+
+	return true
+}
+
 func updateEventProperties(ctx context.Context, event models.Eventable, iCalEvent *ics.VEvent) error {
 	// CREATED - https://www.rfc-editor.org/rfc/rfc5545#section-3.8.7.1
 	created := event.GetCreatedDateTime()
@@ -322,10 +334,6 @@ func updateEventProperties(ctx context.Context, event models.Eventable, iCalEven
 	}
 
 	// DESCRIPTION - https://www.rfc-editor.org/rfc/rfc5545#section-3.8.1.5
-	// TODO: Emojies currently don't seem to be read properly by Outlook
-	// When outlook exports them(in .eml), it exports them in text as it strips down html
-	bodyPreview := ptr.Val(event.GetBodyPreview())
-
 	if event.GetBody() != nil {
 		description := ptr.Val(event.GetBody().GetContent())
 		contentType := event.GetBody().GetContentType().String()
@@ -333,13 +341,33 @@ func updateEventProperties(ctx context.Context, event models.Eventable, iCalEven
 		if len(description) > 0 && contentType == "text" {
 			iCalEvent.SetDescription(description)
 		} else if len(description) > 0 {
-			// https://stackoverflow.com/a/859475
-			iCalEvent.SetDescription(bodyPreview)
-
 			if contentType == "html" {
-				desc := strings.ReplaceAll(description, "\r\n", "")
-				desc = strings.ReplaceAll(desc, "\n", "")
-				iCalEvent.AddProperty("X-ALT-DESC", desc, ics.WithFmtType("text/html"))
+				// If we have html, we have two routes. If we don't have
+				// UTF-8, then we can do an exact reproduction of the
+				// original data in outlook by using X-ALT-DESC field and
+				// using the html there. But if we have UTF-8, then we
+				// have to use DESCRIPTION field and use the content
+				// stripped of html there. This because even though the
+				// field technically supports UTF-8, Outlook does not
+				// seem to work with it.  Exchange does similar things
+				// when it attaches the event to an email.
+
+				// nolint:lll
+				// https://learn.microsoft.com/en-us/openspecs/exchange_server_protocols/ms-oxcical/d7f285da-9c7a-4597-803b-b74193c898a8
+				// X-ALT-DESC field uses "Text" as in https://www.rfc-editor.org/rfc/rfc2445#section-4.3.11
+				if isASCII(description) {
+					// https://stackoverflow.com/a/859475
+					replacer := strings.NewReplacer("\r\n", "\\n", "\n", "\\n")
+					desc := replacer.Replace(description)
+					iCalEvent.AddProperty("X-ALT-DESC", desc, ics.WithFmtType("text/html"))
+				} else {
+					stripped, err := html2text.FromString(description, html2text.Options{PrettyTables: true})
+					if err != nil {
+						return clues.Wrap(err, "converting html to text")
+					}
+
+					iCalEvent.SetDescription(stripped)
+				}
 			}
 		}
 	}

src/internal/converters/ics/ics_test.go

@@ -607,10 +607,30 @@ func (suite *ICSUnitSuite) TestEventConversion() {
 				return e
 			},
 			check: func(out string) {
-				assert.Contains(t, out, "DESCRIPTION:body preview", "body preview")
 				assert.Contains(t, out, "X-ALT-DESC;FMTTYPE=text/html:<html><body>body</body></html>", "body")
 			},
 		},
+		{
+			name: "html body with utf8",
+			event: func() *models.Event {
+				e := baseEvent()
+
+				body := models.NewItemBody()
+				btype, err := models.ParseBodyType("html")
+				require.NoError(t, err, "parse body type")
+
+				body.SetContentType(btype.(*models.BodyType))
+				body.SetContent(ptr.To("<html><body>മലയാളം</body></html>"))
+
+				e.SetBodyPreview(ptr.To("body preview"))
+				e.SetBody(body)
+
+				return e
+			},
+			check: func(out string) {
+				assert.Contains(t, out, "DESCRIPTION:മലയാളം", "body")
+			},
+		},
 		{
 			name: "showas free",
 			event: func() *models.Event {