Extract pattern from text line

[TehOne]

The text line has the following formats:

what.ever.bla.bla.C01G06.BLA.BLA2
what.ever.bla.bla.C11G33.BLA.BLA2
what.ever.bla.bla.01x03.BLA.BLA2
what.ever.bla.bla.03x05.BLA.BLA2
what.ever.bla.bla.Part01.BLA.BLA2

and other similar ones, I need a way to select the "what.ever.bla.bla" part out of the text.

So basically it has to be based on such regex:

(.*?)(C[0-9]+G[0-9]+|[0-9]+x[0-9]+|Part[0-9]+)

where (.*?) is the part I want to extract... any ideas?

[Cabhan]

So it depends a great deal on what the format actually is.

If the format of each line is 7 columns separated by dots and you always want the 5th, then all you need is:

.*\..*\..*\..*\.(.*)\..*\..*

Alternatively, if it is simply some constant string, followed by the desired part, followed by a constant string, you could do:

CONST_STRING_1(part_you_want)CONST_STRING_2

So basically, more information on the format would be nice.

[ghostdog74]

The text line has the following formats:

what.ever.bla.bla.C01G06.BLA.BLA2
what.ever.bla.bla.C11G33.BLA.BLA2
what.ever.bla.bla.01x03.BLA.BLA2
what.ever.bla.bla.03x05.BLA.BLA2
what.ever.bla.bla.Part01.BLA.BLA2

and other similar ones, I need a way to select the "what.ever.bla.bla" part out of the text.

So basically it has to be based on such regex:

(.*?)(C[0-9]+G[0-9]+|[0-9]+x[0-9]+|Part[0-9]+)

where (.*?) is the part I want to extract... any ideas?

it is easier to split them using field delimiters than using regular expressions

 cut -f1-4 -d"." file

[TehOne]

So it depends a great deal on what the format actually is.

If the format of each line is 7 columns separated by dots and you always want the 5th, then all you need is:

.*\..*\..*\..*\.(.*)\..*\..*

Alternatively, if it is simply some constant string, followed by the desired part, followed by a constant string, you could do:

CONST_STRING_1(part_you_want)CONST_STRING_2

So basically, more information on the format would be nice.

The what.ever.bla.bla was just an example, it does differ.
It can be Example1.C11G33 or Bla123.Bla123.C11G33 and so on,
I need to extract always the name that is befor the "C11G33" no matter what it is .*?

The only thing that stays is the C11G33 or 01x03 or Part01 just with different numbers, just look on my regex example again.

EDIT:

solved

sed 's/\(.*\)\(C[0-9][0-9]*G[0-9][0-9]*\)\(.*\)/\1/'