OK you should read the entire therad theres a lot fo hints here.
But basically:
. = any character
* = any number of preceeding characters wich is like
$ = end of line (^ beginning except se the [] below)
see matches command referenc4e more on basic structures
thus .* match everything .*: is m match everything ending with a colon, note matchers are greedy so line
"aargh: c:\temp\aargh.tif" would match the bold part not just the first aargh:
[^ ]* = match anything that isnt space [] = anyoff but ^ inverts the meaning
() is a grouping and $1 is a backwards substitution, match cant use backwards substitutions
grouping can also be used a bit differently namely (hey|ho) wich matches either hey or ho its still a back substitution again match cant use all general regexp rules but substitute can.
the jodaspeak is in plain englsh is
$1= "([^ ]*)"= all chars from beginning that arent space = first word
" "= match space and discard it
$2= "(.*)=everything untill space of last word (bacause its greedy it moves backwatds till it can find the last word)
" "= match space and discard it
$3= (.*)= lastword
" "= match space and discard it
$4= [.!?]= punctuation either .! or ?
ok read
s/What to match/substitute with/
s styands fore substitute
s/a/b/ = find first a and substitute it with b
s/a/b/g =find all a's and substitute them with b (this is also known as transliterate, tough many a trasliterate function works on bigger sets)
s/author/Janne Ojala/g findanrelpace all authors with Janne Ojala wich is sometimes handy but handier is:
s/koira(|. |lle|lta)/kissa$1/g wich would change the finnish word dog and a few of its possible variations into a cat with same wordform.
http://en.wikipedia.org/wiki/Finnish\_language
especialy thereoff to understand why a regexp can be hendy for finns here the same word has been formed in diifferent purpose 2,253 times:
http://www.ling.helsinki.fi/~fkarlsso/genkau2.html
this menas that a normal find and replace in english can do a miracle but in finish its usefullness is limitted