Pfthb

I'm trying to match (if possible, only) the values of coordinates contained in lines like:

function f is described by the (x,y) couples: 0.000000E+00 0.000000E+00 5.00000 0.500000E-01 1.0000 0.290000 2.0000 1.56000 3.0000 5.47000 4.0000 17.3000 4.50000 31.2000 5.0000 52.6000

The first couple is matched as wished, that is to say in two different groups, by

(?<=bcouples:s)(S+)s+(S+)s+

Then,

 (?<=bcouples:s)((S+)s+(S+)s+)+

matches the whole line, but only splits the last two coordinates in separate groups.

Precision: the number of couples of coordinates is not known, so just adding several times

(S+)s+(S+)s+

in the end of the regex is not an option.

Thanks for your input!

Use findall():

re.findall(r"(?:s+([d.Ee+-]+)s+([d.Ee+-]+))+?",s)

([d.Ee+-]+)s+([d.Ee+-]+) --> two float numbers,
 () each of grouped;
 (?:s+ ... )+? --> +? there can be more couples, ? means non-greedy matching,
 (?: the outer group is not interesting;

Edit:
You can select the appropriate line:

 if "couples:" in s:
 coords= re.findall(...)

If your text contains more "couples", you can split it. In the following example, we can apply the regex for the 2nd or 3rd, or both part of the splitted string:

s="function f is described by the (x,y) couples: 0.000000E+00 0.000000E+00 5.00000 0.500000E-01 function g is described by the (x,y) couples: 0.1E+00 0.2E+00 9.00000 0.900000E-01"

ls=s.split("couples")
print(ls)
['function f is described by the (x,y) ',
 ': 0.000000E+00 0.000000E+00 5.00000 0.500000E-01 function g is described by the (x,y) ',
 ': 0.1E+00 0.2E+00 9.00000 0.900000E-01']

 re.findall(r"(?:s+([d.Ee+-]+)s+([d.Ee+-]+))+?",ls[1])
 [('0.000000E+00', '0.000000E+00'), ('5.00000', '0.500000E-01')]