Trickly transact SQL needed please: going round in circles

I have data like the following

Ftitle               Fvalue                         Freference
------               ----------                     ---------
filename             file1.java                     123
version              2                              123
cvstree              branch1                        123

filename             file2.java                     345
version              4                              345
cvstree              branch2                   开发者_StackOverflow中文版     345

filename             file1.java                     4556
version              3                              4556
cvstree              branch1                        4556


filename             file3.java                     4312
version              77                             4312
cvstree              branch2                        4312

filename             file1.java                     5616
version              1                              5616
cvstree              branch3                        5616

I have given a blank line above between some rows for easy readability. For various reasons, the table structure cannot be changed and the table has tons and tons of data.

Now, what i have is just the Fvalue for example file1.java . i would like to know if i can use a TSQL statement so that, in one single query, i get, for example all the distinct Freference values where the branch matches what i specify.

So, for example, if i give a query where i want matches for file1.java for branch1, then i just want the SQL to return me "Freference" 4556 and 123 .

Is this doable without looping through all "Freference" values for file.java and then further filtering it where Fvalue is branch1? This kind of loop becomes very slow.

INTERSECT is an option
But a direct join has better support across DBMS.

select distinct a.freference
from tbl a
inner join tbl b on a.freference = b.freference and b.fvalue = 'branch1'
where a.fvalue = 'file1.java'

SELECT --DISTINCT intersect removes duplicates
   Freference
FROM
   MyTable M1
WHERE
   M1.Fvalue = 'file1.java'
INTERSECT
SELECT --DISTINCT intersect removes duplicates
   Freference
FROM
   MyTable M2
WHERE
   M2.Fvalue = 'branch1'

OR

SELECT DISTINCT
   M.Freference
FROM
   MyTable M
WHERE
   M.Fvalue = 'file1.java'
   AND
   EXISTS (SELECT * FROM 
            MyTable M2
         WHERE
           M2.Fvalue = 'branch1'
           AND
           M.Freference = M2.Freference)

You just want to join the table on itself, it's pretty easy.

Edit: How about this?

select a.Freference 

from table a inner join table b
        on a.Freference = b.Freference
        and a.Ftitle = 'filename'
        and b.Ftitle = 'cvstree'
where 
    a.Fvalue ='file1.java'
    and b.Fvalue = 'branch1'

select * 
from theTable 
where Ftitle = 'file1.java' 
and Fvalue = 'branch1'

is Freference unique to each triplet? in other words, could any other set of rows have 4312 as Freference?

Yes, Freference is unique to each triplet. Sorry if i didn't mention that before. I am trying out the various ideas here in the meantime