Need fore speed...

[viktoras didziulis]

one more alternative - works very fast - use sqlite in-memory database:
1) download sqlite3 command-line executable
2) run shell (cmd on windows) and create an in-memory database by executing:
 sqlite3 :memory:
3) assuming your data consists of 4 fields (columns) that are tab 
delimited and stored in file table.csv type or copy paste and execute 
the following commands in your sqlite3 session:

create table myTable (a varchar, b varchar, pepMass float, d integer);
.separator \t
.import table.csv myTable

4) Now all you need to do is to store a list of update statements like 
this in text file (lets name it q.sql):
UPDATE myTable SET  a='whatever', b='whatever',  d='100' WHERE pepMass 
BETWEEN   2267.0000 AND 2267.9999;
UPDATE myTable SET a='whatever', b='whatever',  d='1000' WHERE pepMass 
BETWEEN   2268.0000 AND 2268.9999;

and
5) execute the following sql string in sqlite:
.read q.sql

6) Thats it. Now you can get your result:
.output myTable.csv
 SELECT * FROM myTable;
.output stdout

Here you can use Revolution to create a list of multiple UPDATES :-). 
But I do not know how to use in memory sqlite database directly from 
Revolution :-(.

Viktoras


.