Here is what I am trying to do (in MySQL 4.0.25)

Table 1: t_city (Has the population of each city_id)
Fields:  city_id, population

Table 2: t_city_zip (links each city_id to all the zipcodes that are within it)
Fields:  city_id, zipcode

Answering this question involves writing one or more queries that will
do the following:

for every row in t_city where population=0, find if there is
another row in t_city that that has the following criteria:

- It has the exact same number and content (number and zip code values)
of "daughter" rows in t_city_zipcode regardless of their order

- It has a population > 0

When the query finds this "matching" row in t_city then update the "0"
population to that "non-zero' population of the matching row.

Below is an example of what I am looking to do...

t_city: (city_id, population)
Row 1: (1, 25000)
Row 2: (2, 0)
Row 3: (3, 0)

t_city_zipcode: (city_id, zipcode)
Row 1: (1, 21146)
Row 2: (1, 21401)
Row 3: (2, 21401)
Row 4: (2, 21146)
Row 5: (3, 21401)
Row 6: (3, 21402)

Successfully running this query would have the following impact on t_city:

Row 1: Not evaluated because population > 0

Row 2: Set "population" for row 2 of t_city to 25,000 because:

1) population for row 2 is 0
2) it found a matching row in t_city (Row 1) which has the exact same makeup of
zipcode values (despite being in a different order) and that has a
population <> 0

Leaves population for t_city Row 3 unchanged because it has no
matching t_city rows that meet the criteria.

Why Row 3: (3, 0) in t_city table should not be updated even there is 
Row 5: (3, 21401) in t_city_zipcode table which is having the same zip
code number of Row 2: (1, 21401) ?

I need all the "subrow" area codes to be the same.

Row 1 (1,25000) in t_city links to the following areacode subrows (21146, 21401)
Row 3 (3,0) in t_city links to the following areacode subrows (21401, 21402)

Since [set of (21401,21146)] <>  [set of (21401,21402)] then I should
not update row 3.

Note that row 2 updates because the content of the "subrow" sets is
the same despite being in different order (before sorting)

You want to perform relational division. This is going to be a huge
pain because you cannot do subqueries in your version of MySQL. Here's
the idea with subqueries:

select distinct A_KEY
from TABLE_A A
where not exists ( 
    select B_KEY
    from TABLE_B B
    where not exists ( 
        select * 
        from TABLE_C C
        where A.A_KEY = C.A_KEY
        and B.B_KEY = C.B_KEY ))

The double negated existential query is needed because there is no
universal quantification (e.g., FORALL) in SQL.

I tried to work out the formulation in left outer joins, but I cannot.
I hope this points you in the right direction. Good luck.