Shell Programming and Scripting

I have a large numbers of files containing data that look like this:

Code:

(ID31:0.01682,(ID-123:0.00000,(ID_24:0.00000,ID&890:0.00000):0.00000):0.00000,ID12876:0.00000);
(ID_24:-0.00052,(ID31:0.01697,(ID-123:-0.00059,ID&890:0.03528):0.00037):0.00027,ID12876:0.03484);

I need to find ":" anywhere in the expression and erase it along with the number that follows till I find ",", "(" or ")". The ";" indicates the end of the expression and the next expression needs to be processed and placed in the following row. Thus, I will end up with something like this:

Code:

(ID31,(ID-123,ID_24,ID&890)),ID12876);
(ID_24,(ID31,(ID-123,ID&890)),ID12876);

Thanks in advance!

I think your expected output is missing a bracket.

Code:

$ sed "s/:[0-9\.-]*//g" file1
(ID31,(ID-123,(ID_24,ID&890)),ID12876);
(ID_24,(ID31,(ID-123,ID&890)),ID12876);

Sorry about missing one of the brackets.

No need to apologise, I just wondered why my output lines were the same length, and yours weren't

Scottn,
I am processing the input file using your code

Code:

sed "s/:[0-9\.-]*//g" infile

However, in the output file the 'expression' is not linear. So, I tried to linearize it using the following codes one after another

Code:

sed -e :a -e '/^>/!N;s/\n[^>]/ /;ta' infile

Code:

sed 's! !!g' infile

I am using the second one because the output from my first one contain 'gaps/blanks' that are not supposed to be there. Unfortunately, my output file is slightly different from the expected file. I am uploading the 5 files (input, the one Iget using your code, the file I get after linearizing the file. My final output and theexpected output file).
I will be really thankful if you could please take a look at it.

Code:

sed "s/:[0-9\.-]*//g" infile.txt |awk '{printf $0}'

Thanks!