Hi, Right now, with tmpl_process3.pl, there are still some strings that are extracted out of context because an <INPUT> tag is in the middle of the sentence, making it produce strings that are not translatable. I think I have "almost" got it right so that tmpl_process3 can recognize that some INPUT tags should not break the strings. However, before I finish and commit the code, I need to think about a question. I wonder what kind of placeholders I should use for these INPUT tags (radio and text): - Right now, TMPL_VAR is represented by %s. This makes sense. - Right now, A elements are represented by <A1>, <A2>, etc. Some time after I committed it, I felt that this is perhaps a mistake because the original strings may have <A1> etc. (though they should not), and I now feel that such things are best represented by some sort of c-format code, perhaps one that is abused or made up. Do you think it makes more sense that, for the INPUT tags, we 1. Reuse some c format to represent INPUT (e.g., %i for INPUT)? This makes the strings shorter, but doesn't make sense. 2. Invent some c format to represent the different kinds of INPUT (eg., %R for radio, %T for test) This also makes the strings shorter, make more sense I think, but perhaps a bit weird. 3. Invent some new tags to represent the INPUT elements (e.g., <INPUT1>, or <RADIO1> or <TEXT1> etc) If we invent some c format to represent the INPUT tags, then I think the meanings of the 1$, 2$ etc modifiers would change. What do you think should be used to represent <INPUT type=radio> and <INPUT type=text> ? If you think we should use pseudo-c-format codes, do you think the <A> tags should also be changed to use such codes instead of <A1>, <A2>, etc? or are <A1>, <A2>, etc. easier to read? Note: These placeholders are only used by the tmpl_process3.pl translation script. Translators will see them in the strings to translate if they decide to use this translation script, but they will not appear in the translated files. I hope I have asked my question coherently. Thanks, -- Ambrose LI Cheuk-Wing <a.c.li@ieee.org> http://ada.dhs.org/~acli/