Hola, amigo Yutaka.
EM is a great product, but at your request ("earlier rather than later") I want to report here several issues (all versions). I numbered'em to facilitate your answers
Sorry for my bad English, I'll try to be as clear as possible.

1. The OR operator with 'Search Only Word' got confused if an including word goes first at the expression (two or more):
(siempre|sintagma|si) OK, matches 'siempre'/'sintagma'/'si'
(si|siempre|sintagma) BAD, matches 'si' but not 'siempre'/'sintagma'
(If this is the expected behavior, please say it at "Regular Expression Syntax" section inside Help)

2. "Find and Replace in Files" operations treat the special chars ('ñ', 'ç', accented vowels) as word boundaries. For example: "aca" with 'Search Only Word' matches the first part of "acañar" or the last parts of "austríaca", "ilíaca" and so.........
BUT if "Use Regular Expressions" is enabled too, EM works well (!!!). (F&R at open files never have any problem).

3. Please stop the "Escape sequence" auto triggering in F&R dialog (I already said it in a previous post ). Besides the additional click to deactivate it, it always duplicates / and \ chars at Find Line. A nicer behavior could be remember the last user selection of switches.

4. The dialog "Treat the following characters as alphanumerics" (Customize) is working...? (i.e., it applies to what operations?)
If I add any alien char (for example @) EM still treats it as non-letter (using 'Search Only Word' or the \w operator).

5: (Helpfile) Please include two clarifications (for regexp newbies as I... was ):
a) The ordinals (º,ª) are always treated as alphanumeric, not word boundaries, so \l and \w includes'em.
b) The "(?" subexpressions allow several strings with I (OR operator), but the lookbehind ones must be lenght fixed.


Thank you very much for your time, attention and talent!
Saludos desde Perú,
jugaor

----------------
jugaor
The guy with a terrible English