PrestoSoft Community

And they are marked with the ignored line color.

They are marked with the ignored line colour whether the setting is on or off.
Whereas with the other regex, there is a difference in marking added blocks, depending on the setting.
Thus my claim of inconsistent behaviour.

Also, please note that line 33 in the left file was matched with different lines on the right, based on the value of the regex.
It was matched with an empty line using the 2nd regex, effectively treating it an added line, but it was marked green regardless of the setting.
And it was matched with line 48 using the 1st regex, treating it as a changed line.

I don't understand this difference in behaviour, since the only lines that should be matched by the 1st regex but not the second are:
Lines 26, 31 on the left,
Lines 7, 10, 50, 55 on the right.

Incidentally, the line matching with the 2nd regex looks more "natural".
Playing with the settings, I was able to achieve the "better" matching by turning off "Optimize diff block alignment" (which seems counter-intuitive), but I was not able to achieve this result without an ignore regex. All the combinations of the various diff algorithms (other than line-by-line) and the diff block optimization setting resulted in exactly the same matching.

A better-looking matching on XML files with multiple similar blocks will be much appreciated (if at all possible).

I really don't want this conversation to continue in this circular manner. Your assumptions about how EDP ignores lines are clearly incorrect.

When EDP is told to ignore lines, it considers them non-existent for the purpose of comparison. No file alignment is based on them. They are simply shown in the result but they are never compared. When you change your Ignore Regex, you get different lines to be ignored, and the results are obviously different.

I hope this will conclude this topic.