On Sat, Mar 28, 2009 at 11:08 AM, Aggelidis Nikos <aggelidis.news@gmail.com> wrote:

hi to all,i am interested in implementing some string processing algorithms{eg boyer-moore , knuth morris prat} for educational purposes.

This is the first "real" exposure i will have with boost. I've done a bit of searching and i think Xpressive is the corresponding library of boost, for string processing. From what i read Xpressive is a header-only template library, so i will have to study xpressive.hpp .

Any other directions i should follow?

thanks in advance for your trouble,

If you are wanting to do string searching and matching and all such, yep, boost.xpressive is designed for that. If you are interesting in PEG grammars, can do the same stuff as xpressive, but xpressive is a regex style, Spirit2(x) can do the same stuff, but of a PEG syntax. A PEG syntax has no ambiguities and is capable of recursively calling itself (that is the one main thing that regex style grammars cannot do). Boost.Xpressive will be perfect if you are more comfortable with regex, but Boost.Spirit2(x) is more powerful for compile time language constructs. If you are interested in Boost.Spirit2(x), feel free to join the mailing list, lot of helpful people there too. However, both Xpressive and Spirit2(x) already do string searches as part of their very language, from what you said (the boyer-moore method and so forth), it sounds like just doing a standalone thing might work best. If you were to implement it as a C++ grammar, then Spirit2x might fit better (you can get a link to the ltaest in the Spirit mailing list) as it can do matching styles exactly that of what you stated. Still though, might be better to do it manually since those are more low-level ways...