How to create an ID to identify groups within table - PostgreSQL

Question

I have a dataset that looks like this:

key_field	report_year	quarter	report_month	old_status	new_status	old_date	new_date	old_hist_id	new_hist_id	date_range
Unq_010203	2021	2	4	REJECTED	REJECTED	4/1/21 20:17	4/5/21 22:06	1	2	4
Unq_010203	2021	2	4	REJECTED	APPROVED	4/5/21 22:06	4/6/21 16:42	2	3	0
Unq_010203	2021	2	4	APPROVED		4/6/21 16:42		3
Unq_010203	2021	2	5	REJECTED	REJECTED	5/30/21 18:55	5/31/21 9:36	4	5	0
Unq_010203	2021	2	5	REJECTED	APPROVED	5/31/21 9:36	4/14/22 12:00	5	6	317
Unq_010203	2022	2	4	APPROVED		4/14/22 12:00		6

This dataset snippet relates to a single key_field, which is at the level I need the grouping happening.

I need to create a campaign identifier within that dataset using the following criteria. If a key_field gets approved, a new campaign will start. If the date_range of a key_field has gone over 90 days, a new campaign will start. Resulting in a table that looks like this:

key_field	report_year	quarter	report_month	old_status	new_status	old_date	new_date	old_hist_id	new_hist_id	date_range	CAMPAIGN_ID
Unq_010203	2021	2	4	REJECTED	REJECTED	4/1/21 20:17	4/5/21 22:06	1	2	4	1
Unq_010203	2021	2	4	REJECTED	APPROVED	4/5/21 22:06	4/6/21 16:42	2	3	0	1
Unq_010203	2021	2	4	APPROVED		4/6/21 16:42		3			1
Unq_010203	2021	2	5	REJECTED	REJECTED	5/30/21 18:55	5/31/21 9:36	4	5	0	2
Unq_010203	2021	2	5	REJECTED	APPROVED	5/31/21 9:36	4/14/22 12:00	5	6	317	2
Unq_010203	2022	2	4	APPROVED		4/14/22 12:00		6			3

I have tried using RANK() OVER() but I can't seem to get the partitions right.

Table CREATE statement:

CREATE TABLE omg_fml
    (key_field    VARCHAR(10),
     report_year  INT(4),
     quarter      INT(1),
     report_month INT(1),
     old_status   VARCHAR(8),
     new_status   VARCHAR(8),
     old_date     DATE,
     new_date     DATE,
     old_hist_id  INT(2),
     new_hist_id  INT(2),
     date_range   INT(3));

Table INSERT statement:

INSERT
INTO omg_fml
VALUES ( 'Unq_010203', 2021, 2, 4, 'REJECTED', 'REJECTED', '2021-04-04', '2021-04-05', 1, 2, 4 )
     , ( 'Unq_010203', 2021, 2, 4, 'REJECTED', 'APPROVED', '2021-04-05', '2021-04-06', 2, 3, 0 )
     , ( 'Unq_010203', 2021, 2, 4, 'APPROVED', NULL, '2021-04-06', NULL, 3, NULL, NULL )
     , ( 'Unq_010203', 2021, 2, 5, 'REJECTED', 'REJECTED', '2021-05-30', '2021-05-31', 4, 5, 0 )
     , ( 'Unq_010203', 2021, 2, 5, 'REJECTED', 'APPROVED', '2021-05-31', '2022-04-14', 5, 6, 317 )
     , ( 'Unq_010203', 2022, 2, 4, 'APPROVED', NULL, '2022-04-14', NULL, 6, NULL, NULL );

PostgreSQL version 8.0 (done in Redshift)