Inconsistent formatting of {{token}} blocks in draft emails will cause token replacement to fail for HTML

[benkloester]

I am not certain I got the root cause of this issue right, but I noticed that the HTML version of the email being sent out inn my test merge had a blank where the name should have been swapped in. Inspecting the html showed that there were tags in the middle of the token - possibly in between some of the parentheses? So naive replacement might have resulted in the replacedment html being interpreted as a malformed tag or something.

Removing formatting from the {{token}} text in my draft fixed it.

It might be worth documenting that this can be an issue.

[knmnyn]

Thanks for the bug report @benkloester . The script uses the global replace on the stringifyied version of the template to do the replacements (Lines 314–316 in the live script; Lines 319–321 in the repo's Code.gs). The replacement ordering is not clearly defined by this method so if your {{token}} replacement requires a certain ordering it may be problematic.

Can you send a screenshot of the issue you face so that I can diagnose it appropriate?

In the meantime, I've updated the README.md to mention the issue.