Paola Petrocelli 

Most likely your rules are not correct.  Looking at what you have used I suspect you tried to do something with what was the default suggestion and you probably have some overlap somewhere.  I'd be inclined to use the Multilingual XML filetype for the XLIFF if you need to handle embedded content, and if the Excel is also multilingual I'd use the Multilingual Excel filetype too for at least the same reason.

 Multilingual XML FileType 

 Multilingual Excel Filetype 

If you can't then maybe provide a couple of small sample files that represent the layout of your file and we can have a go at helping you with improved rules.

Paola Petrocelli 

In fact forget the Multilingual filetypes... I missed you only have 2019. So lets just see some small sample files.

Paola Petrocelli 

Your rules are far to general. It is much more advisable to have a separate rule for the p, b, br, a href and other tags. This way you can much better fine tune the output and do not risk overlapping.

Thanks for your reply, Paul!

Here are some sample images of the codes: 

Hi Jerzy! Thanks for your reply. 

I had no idea about rules at all: I thought the ones I found would work, but now I see they should be more specific.

I've attached an image of most of the tags used in my previous reply.

I guess I'll need to copy and paste more specific rules.

Any idea where I could find them?  

 Paola Petrocelli 

The rules are created using regular expressions.  It's worth you learning a little about this syntax as it's useful in many applications.  Most CAT tools use them, decent text editor use them and they can be very powerful.  Often the syntax is nothing more than plain text and for your rules that would most likely be the case. 

The rules you used were catch all rules designed to try and pick up everything, probably as a placeholder.  You need opening and closing tags, for example (and in plain text):

<h3>Some text in here...</h3>

The opening tags rule would be:

<h3>

The closing tag rule would be:

</h3>

Similarly for some of the ones you showed in your screenshots:

<p>
</p>

<h4>
</h4>

If there are many non-translatable attributes in the element like this:

<td style="text-align:right;">Some translatable text here</td>

Then in this case you would need to use proper syntax if you don't want to write out every possible attribute type that might be in the file.  For example:

<td[^>]+?>
</td>

So for the opening tag the syntax would mean:

<td
That's obviously just plain text

[^>]
This means find any character apart from a >

+?
This means find as many of the characters you were looking for as you can but only keep finding until you find the first > that you were looking for

>
Plain text again

So not too tricky and once you start using them you'll feel more comfortable to create your own more complex things.  Regular expressions are not a Trados thing so you can search and find information all over the internet for this.  Important thing to remember is there are many flavours of syntax.  We use C# (.NET) flavour.