Hi @ntdiary - I misunderstood Conor’s comment before, sorry about that! I thought his “workaround” was a proposed solution when it’s actually what the user would have to do to make the whole text line bold.

I did a bit more research and found that it’s possible for URLs to have *'s. For that reason, I think we should leave the current edge-case bug as is.

In the future, after we add live markdown previews, we’ll also add keyboard shortcuts (i.e. Command+b) at some point, which will make this less of an issue.

I can take this and cover in https://github.com/Expensify/App/issues/30823

This seems like dupe of https://github.com/Expensify/App/issues/30823

Proposal

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

URL with hash inside bold markdown does not apply the bold markdown

*text https://expensify.com#abc*

What is the root cause of that problem?

The bold rule is after the auto link rule, The auto link markdown parses url with ending * before bold rule is applied, thus the bold rule does not apply.

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

If Url with ending asterisk is part of string which is likely to be parsed as bold (e.g. *text https://expensify.com#abc*), we should not contain ending * as part of url. To do that, we can add following process in modifyTextForUrlLinks function. The process is to exclude ending asterisk if there is string with asterisk leading to url .

                if (numberOfCharsToRemove) {
                    match[0] = match[0].substring(0, match[0].length - numberOfCharsToRemove);
                    url = url.substring(0, url.length - numberOfCharsToRemove);
                }
            }
--------->
            if (match.index !== 0) {
                const stringBeforeUrl = textToCheck.substring(startIndex, match.index);
                const lastChar = url.charAt(url.length - 1);
                if (lastChar === '*' && stringBeforeUrl.includes(lastChar)) {
                    match[0] = match[0].substring(0, match[0].length - 1);
                    url = url.substring(0, url.length - 1);
                }
            }
<-----------
            replacedText = replacedText.concat(textToCheck.substr(startIndex, (match.index - startIndex)));

we can add relevant test case

This approach does not make other test cases failed.

Will you let me know what the original text is?

@s-alves10 please try with this text *text https://expensify.com#abc*