When the text contained in an mjml tag contains the character '<' an exception is triggered

[qjoseph]

When the text contained in an mjml tag contains the character '<' an exception is triggered.

Sample:

<mjml>
    <mj-body>
        <mj-text>test: &, <, >, ', \"</mj-text>
    </mj-body>
</mjml>

You should add the conversion of the '<' to '& lt;' as for the '&'

I added this condition in XmlFixer, it seems to work:

else if (span.Length >= 2 && span[0] == '<' && span[1] != '/' // not closing tag
    && !((span.Length >= 4 && span[1] == 'm' && span[2] == 'j') // ignore start tag
        || (span.Length >= 3 && span[2] == 'm' && span[3] == 'j'))) // ignore end tag
{
    if (!IsEntity(span))
    {
        sb.Append("&lt;");
    }
    else
    {
        sb.Append(first);
    }
}

@SebastianStehle what do you think?

[SebastianStehle]

You can also use custom html tags, so your fix would not work.

[qjoseph]

Oh, you're right, I have another implementation based on your xml fixer but it's a bit complex and doesn't necessarily handle all cases.

internal static string EscapeLowerThanCharacterInTags(string html)
        {
            var span = html.AsSpan();
            var result = StringBuilders.Get();

            try
            {
                for (var i = 0; i < span.Length; i++)
                {
                    if (span[i] != '<')
                    {
                        result.Append(span.Slice(i, 1));
                        continue;
                    }

                    if (i > 0 && span[i - 1] == span[i]) // same character '<'
                    {
                        result.Append("&lt;");
                        continue;
                    }

                    // find end tag
                    var indexOfEndTag = span[i..].IndexOf('>');

                    if (indexOfEndTag != -1)
                    {
                        var tag = span.Slice(i, indexOfEndTag + 1);

                        if (tag[..2] is "</" || tag[..3] is "<br")
                        {
                            result.Append(tag);
                            i += tag.Length - 1;
                        }
                        else
                        {
                            bool notATag = false;

                            if (!(tag.Trim().IndexOfAny('!', '=') != -1 && tag.Trim().Length > 3)) // not found
                            {
                                foreach (var tagCharacter in tag)
                                    if (char.IsWhiteSpace(tagCharacter) || (char.IsPunctuation(tagCharacter) && tagCharacter is not '-' or '_'))
                                    {
                                        notATag = true;
                                        break;
                                    }
                            }

                            if (notATag)
                                result.Append("&lt;");
                            else
                            {
                                result.Append(tag);
                                i += tag.Length - 1;
                            }
                        }
                    }
                    else
                        result.Append(span[i]);
                }

                return result.ToString();
            }
            finally
            {
                StringBuilders.Return(result);
            }
        }

[SebastianStehle]

Could work.

Can you:

1. Create a PR
2. Integrate it into this class: https://github.com/SebastianStehle/mjml-net/blob/main/Mjml.Net/XmlFixer.cs?
3. Write a few tests?