Right-to-Left text and the spelling checker

Right-to-Left (RTL) text editing is by no means solved by supporting Unicode! Most of the prolems stem from the fact that RTL text is never alone. There are numbers which are always LTR, not to mention LTR foreign language (e.g. English) interspersed with RTL text such as Hebrew. Special symbols such as ‘+’, ‘_’ and others are also problematic.

To properly support an RTL language, one needs to distinguish between a short RTL text embedded within a larger LTR paragraph or the other way around: a short LTR string embedded within a mainly RTL paragraph. But I am not proposing to turn EmEditor to a full fledged word processor, by no means! All I need is better support for embedding small RTL text in mainly LTR documents. A good example is the MS Notepad, which does it fairly well.

I know how difficult is it for you to work with a totally foreign language that even the fonts are not familiar. I have the same problem even with Arabic, not to mention Japanese or Chinese.

Support of RTL text in EmEditor leaves a lot to be desired. But to show you the problems, I need you to look at my desktop while I am doing some work and verbally explaining the issues. Can we workout something to do that?

However one small thing that I discovered might be easily shown.

Take please the following line and put it in an EmEditor text file:

Name: עמדה ניהול קרנות נאמנות בע”מ (mutual fund)

The right order of words on the screen should be as follows: The first word on the left is ‘Name: ‘ then the Hebrew (which is RTL) string starts on the right with the first word being ‘עמדה’, then to the left of it ‘ניהול’, then to the left of that is ‘קרנות’, etc. Please also note the double quote in the last word. Then the text switches back to LTR English ‘(mutual fund)’.

And now switch-on please the spelling checker. Well, at least on my machine (with Hebrew support) the order of words is reversed! Mind you, the problem is not with the order of the letters within words. These appear in the right order and every word can be read properly. The problem is the order of WORDS. What was before the rightmost word within the Hebrew string, is now the leftmost. Please also note the word with the embedded double quote: The part that was to the right of it is now to the left and vise versa.

On my machine it is totally consistent. I sure hope you can reproduce it on yours… Please let me know if you can.

Regards
Meir