[Due for payment 2025-09-23][$250] workspace - Multiline markdown is not applied in employee workspace chat invite message

[lanitochka17]

If you haven’t already, check out our contributing guidelines for onboarding and email contributors@expensify.com to request to join our Slack channel!

---

Version Number: 9 1.65-0
Reproducible in staging?: Y
Reproducible in production?: Y
If this was caught during regression testing, add the test name, ID and link from TestRail: N/A
Issue reported by: Applause - Internal Team

Action Performed:

Pre-condition: Keep logged in
Mweb - admin
Android - employee

1. Go to https://staging.new.expensify.com/home
2. In mweb - create a workspace - Invite a user
3. In invite message, send multi line markdown message
   Happy
   happy

Happy
happy

4. In android, open employee workspace chat and note invited message
5. From mweb, send invite message to same user as Happy - single line message
6. In android, open employee workspace chat and note invited message

Expected Result:

In employee workspace chat invite message, single/ multiline line markdown text must show consistent behaviour

Actual Result:

If we send markdown text in multiline in certain order, markdown is not applied in employee workspace chat invite message
If we single line markdown text, markdown is applied in employee workspace chat invite message

Workaround:

Unknown

Platforms:

Select the officially supported platforms where the issue was reproduced:

• Android: App
• Android: mWeb Chrome
• iOS: App
• iOS: mWeb Safari
• iOS: mWeb Chrome
• Windows: Chrome
• MacOS: Chrome / Safari
• MacOS: Desktop

Platforms Tested:

On which of our officially supported platforms was this issue tested:

• Android: App
• Android: mWeb Chrome
• iOS: App
• iOS: mWeb Safari
• iOS: mWeb Chrome
• Windows: Chrome
• MacOS: Chrome / Safari
• MacOS: Desktop

Screenshots/Videos

Add any screenshot/video evidence

Bug6858177_1749743309396.Screenrecorder-2025-06-12-21-11-07-513.mp4

View all open jobs on GitHub

Upwork Automation - Do Not Edit

• Upwork Job URL: https://www.upwork.com/jobs/~021933530880478374810
• Upwork Job ID: 1933530880478374810
• Last Price Increase: 2025-08-01
• Automatic offers:
• daledah | Contributor | 108316319

Issue Owner

Current Issue Owner: @VictoriaExpensify

[melvin-bot]

Triggered auto assignment to @maddylewis (Bug), see https://stackoverflow.com/c/expensify/questions/14418 for more details. Please add this bug to a GH project, as outlined in the SO.

[bernhardoj]

Proposal

Please re-state the problem that we are trying to solve in this issue.

Markdown in ws invite message is not applied.

What is the root cause of that problem?

When we send the invite message, we convert it to markdown with Parser.replace with shouldEscapeText as false.

App/src/libs/actions/Policy/Member.ts

Lines 982 to 984 in cfa3ffc

	welcomeNote: Parser.replace(welcomeNote, {
	shouldEscapeText: false,
	}),

But some markdown rules expect the input to be escaped, for example, inline code block regex accepts the escaped (), that is ``.

https://github.com/bernhardoj/expensify-common/blob/17b212f2a4969ff6cfabf912ec8c56c21b51f054/lib/ExpensiMark.ts#L230

Also, the unescaped blockquote makes the parsing fails.

The shouldEscapeText was added in #52684 to solve an issue where the invitation SMS received doesn't unescape the text. The WS chat invitation message shows the unescaped message. So, it's the SMS-specific issue.

What changes do you think we should make in order to solve the problem?

The SMS issue is a BE issue where the text isn't unescaped before sending it to the user, so we should remove the shouldEscapeText: false and let the text be escaped.

What specific scenarios should we cover in automated tests to prevent reintroducing this issue in the future?

This is just enabling/disabling a param from a separate library, so I think a test is not needed here. If we still want to add test, we can assert that the API is called with the parsed welcomeNote, but we will compare it like:

expect(actualWelcomeNote).toBe(Parser.replace(welcomeNote))

which is kinda pointless IMO because the parsing is handled by expensify-common.

[daledah]

Proposal

Please re-state the problem that we are trying to solve in this issue.

If we send markdown text in multiline in certain order, markdown is not applied in employee workspace chat invite message
If we single line markdown text, markdown is applied in employee workspace chat invite message

What is the root cause of that problem?

When adding member, we parse invite comment with shouldEscapeText flag set to false

App/src/libs/actions/Policy/Member.ts

Lines 982 to 984 in cfa3ffc

	welcomeNote: Parser.replace(welcomeNote, {
	shouldEscapeText: false,
	}),

This will prevent replacing some HTML characters into escaped characters, including > and `

https://github.com/Expensify/expensify-common/blob/51d90c04a228101d105a10cafb8c26e9e482c495/lib/utils.ts#L13-L20

And when parsing, we only check for escaped characters for code block and quote

https://github.com/Expensify/expensify-common/blob/51d90c04a228101d105a10cafb8c26e9e482c495/lib/ExpensiMark.ts#L428

https://github.com/Expensify/expensify-common/blob/51d90c04a228101d105a10cafb8c26e9e482c495/lib/ExpensiMark.ts#L230

So the parser is not processed as expected

What changes do you think we should make in order to solve the problem?

shouldEscapeText flag set to false was added in #52684 to solve text sms issues, so we should keep that behaviour.

Instead we should add normal character to the rules regex as well

Add ` for code blocks:

https://github.com/Expensify/expensify-common/blob/51d90c04a228101d105a10cafb8c26e9e482c495/lib/ExpensiMark.ts#L230

regex: /(\B|_|)(?:&#x60;|`)(.*?)(?:&#x60;|`)(\B|_|)(?!(?!<pre>)[^<]*(?:<(?!pre>)[^<]*)*<\/pre>|[^<]*<\/video>)/gm

Add > for quotes

https://github.com/Expensify/expensify-common/blob/51d90c04a228101d105a10cafb8c26e9e482c495/lib/ExpensiMark.ts#L428

const regex = /^(?:&gt;|>)+ +(?! )(?![^<]*(?:<\/pre>|<\/code>|<\/video>))([^\v\n\r]*)/gm;

And here

https://github.com/Expensify/expensify-common/blob/51d90c04a228101d105a10cafb8c26e9e482c495/lib/ExpensiMark.ts#L1290

const textToReplace = text.replace(/^(?:>|>)( )?/gm, handleMatch);

What specific scenarios should we cover in automated tests to prevent reintroducing this issue in the future?

We should add additional test for ExpensiMark tests for the cases where shouldEscapeText is false.

What alternative solutions did you explore? (Optional)

Reminder: Please use plain English, be brief and avoid jargon. Feel free to use images, charts or pseudo-code if necessary. Do not post large multi-line diffs or write walls of text. Do not create PRs unless you have been hired for this job.

[melvin-bot]

Job added to Upwork: https://www.upwork.com/jobs/~021933530880478374810

[melvin-bot]

Triggered auto assignment to Contributor-plus team member for initial proposal review - @ZhenjaHorbach (External)

[melvin-bot]

@ZhenjaHorbach Whoops! This issue is 2 days overdue. Let's get this updated quick!

[ZhenjaHorbach]

Not overdue
I will check today