I guess there would be no point in a program that searched for the first _, replaced it with e.g., searched for the second, replaced it with e.g.? That was my first thought coming back to this but the problem is that it would miss out any unbalanced pairs. Anyone remember the ^S sequences for underline on/off in early WordStar? You'd always got them wrong at least some of the time, and only discover after printing!At 8/8/2008 07:32 PM, you wrote:CI /[S]_/≪MDIT≫/ CI /_[S]/≪MDNM≫/ should be CI /[S]_/[S]≪MDIT≫/ CI /_[S]/≪MDNM≫[S]/and the first could snare ",_ [space]" and convert it to ",≪MDIT≫[space]" -- so you might want to use CV instead of CI, unless you're not in the habit of putting punctuation inside italics or you're not concerned about the occasional misconversion. If there are a lot of emails with a lot of italics to convert, a program that focused only on "[S]_[S]" strings first and asked for input, then autoconverted the rest of the underscores [and perhaps then checked for any remaining underscores], might be the best way to go.Paul Lagasse Harry Binswanger wrote:Offhand, why not do a CI (change invisible)? You could use: CI /[S]_/≪MDIT≫/ CI /_[S]/≪MDNM≫/where the [S] is the "separator" character usually mapped to ctrl-alt-S, and the double angle brackets indicate the guillemets.You might want to do CV (change verify) instead of CI.Is there a facility to change email style _italics_ to mdit/mdnm pairs or something else such as Quark /pairs? Some simple pgm?
Harry Binswanger hb@xxxxxxxx