Hi Stephen & David,
So, I've done a couple of queries on the netdev patchset that I think
might be useful for you:
First, a breakdown by current state:
state | count
-------------------+-------
Not Applicable | 7588
Changes Requested | 6364
Awaiting Upstream | 705
Under Review | 3
New | 3
Accepted | 21653
Deferred | 386
Superseded | 2661
Rejected | 1679
RFC | 4106
(10 rows)
Second, a breakdown of current state vs. patch size. The columns are:
* the patch size, given as the log (base-2) of the length of the
patch's content.
* the state id and name
* the number of patches with that fit into the (state, size) category
* the total number of patches with the same log(2, size).
For example, the first row tells us that there are 5742 patches that are
have a size in the range [2^8, 2^9) bytes, and 2 of those patches are in
the 'New' state.
log_size | state | name | n | total_n
----------+-------+-------------------+------+---------
8 | 1 | New | 2 | 5742
11 | 1 | New | 1 | 7824
8 | 2 | Under Review | 2 | 5742
9 | 2 | Under Review | 1 | 9474
7 | 3 | Accepted | 14 | 29
8 | 3 | Accepted | 3210 | 5742
9 | 3 | Accepted | 5211 | 9474
10 | 3 | Accepted | 4469 | 8725
11 | 3 | Accepted | 3542 | 7824
12 | 3 | Accepted | 2546 | 6053
13 | 3 | Accepted | 1294 | 3502
14 | 3 | Accepted | 662 | 1928
15 | 3 | Accepted | 290 | 961
16 | 3 | Accepted | 108 | 398
17 | 3 | Accepted | 51 | 170
18 | 3 | Accepted | 4 | 23
7 | 4 | Rejected | 3 | 29
8 | 4 | Rejected | 252 | 5742
9 | 4 | Rejected | 409 | 9474
10 | 4 | Rejected | 351 | 8725
11 | 4 | Rejected | 281 | 7824
12 | 4 | Rejected | 180 | 6053
13 | 4 | Rejected | 97 | 3502
14 | 4 | Rejected | 48 | 1928
15 | 4 | Rejected | 31 | 961
16 | 4 | Rejected | 14 | 398
17 | 4 | Rejected | 8 | 170
18 | 4 | Rejected | 3 | 23
8 | 5 | RFC | 399 | 5742
9 | 5 | RFC | 656 | 9474
10 | 5 | RFC | 730 | 8725
11 | 5 | RFC | 775 | 7824
12 | 5 | RFC | 680 | 6053
13 | 5 | RFC | 451 | 3502
14 | 5 | RFC | 267 | 1928
15 | 5 | RFC | 97 | 961
16 | 5 | RFC | 39 | 398
17 | 5 | RFC | 12 | 170
6 | 6 | Not Applicable | 1 | 1
7 | 6 | Not Applicable | 10 | 29
8 | 6 | Not Applicable | 932 | 5742
9 | 6 | Not Applicable | 1383 | 9474
10 | 6 | Not Applicable | 1392 | 8725
11 | 6 | Not Applicable | 1321 | 7824
12 | 6 | Not Applicable | 1107 | 6053
13 | 6 | Not Applicable | 691 | 3502
14 | 6 | Not Applicable | 412 | 1928
15 | 6 | Not Applicable | 199 | 961
16 | 6 | Not Applicable | 75 | 398
17 | 6 | Not Applicable | 32 | 170
18 | 6 | Not Applicable | 3 | 23
7 | 7 | Changes Requested | 2 | 29
8 | 7 | Changes Requested | 555 | 5742
9 | 7 | Changes Requested | 1067 | 9474
10 | 7 | Changes Requested | 1078 | 8725
11 | 7 | Changes Requested | 1193 | 7824
12 | 7 | Changes Requested | 985 | 6053
13 | 7 | Changes Requested | 611 | 3502
14 | 7 | Changes Requested | 379 | 1928
15 | 7 | Changes Requested | 265 | 961
16 | 7 | Changes Requested | 132 | 398
17 | 7 | Changes Requested | 58 | 170
18 | 7 | Changes Requested | 12 | 23
8 | 8 | Awaiting Upstream | 72 | 5742
9 | 8 | Awaiting Upstream | 130 | 9474
10 | 8 | Awaiting Upstream | 142 | 8725
11 | 8 | Awaiting Upstream | 125 | 7824
12 | 8 | Awaiting Upstream | 111 | 6053
13 | 8 | Awaiting Upstream | 68 | 3502
14 | 8 | Awaiting Upstream | 35 | 1928
15 | 8 | Awaiting Upstream | 16 | 961
16 | 8 | Awaiting Upstream | 3 | 398
17 | 8 | Awaiting Upstream | 1 | 170
8 | 9 | Superseded | 301 | 5742
9 | 9 | Superseded | 541 | 9474
10 | 9 | Superseded | 484 | 8725
11 | 9 | Superseded | 498 | 7824
12 | 9 | Superseded | 378 | 6053
13 | 9 | Superseded | 254 | 3502
14 | 9 | Superseded | 114 | 1928
15 | 9 | Superseded | 59 | 961
16 | 9 | Superseded | 20 | 398
17 | 9 | Superseded | 6 | 170
18 | 9 | Superseded | 1 | 23
8 | 10 | Deferred | 17 | 5742
9 | 10 | Deferred | 76 | 9474
10 | 10 | Deferred | 79 | 8725
11 | 10 | Deferred | 88 | 7824
12 | 10 | Deferred | 66 | 6053
13 | 10 | Deferred | 36 | 3502
14 | 10 | Deferred | 11 | 1928
15 | 10 | Deferred | 4 | 961
16 | 10 | Deferred | 7 | 398
17 | 10 | Deferred | 2 | 170
(94 rows)
I've also attached the SQL used to generate these data, which you may
want to review before using it for anything serious. Also, let me know
if you'd like these queries re-done with different parameters (is the
log-2 suitable?), the output in a different format, or some other
analysis of the patchset.
Like I said, there's no historical log for the updates to a patch, but
it's sounding like that would be a good feature for me to add.
Cheers,
Jeremy
SELECT patchwork_state.name AS state, count(*)
FROM patchwork_patch
INNER JOIN patchwork_state
ON patchwork_state.id = patchwork_patch.state_id
WHERE patchwork_patch.project_id =
(SELECT id FROM patchwork_project where linkname = 'netdev')
GROUP BY state;
/* query plan:
* - create a set of (count, log(2, size)) tuples from the patches for netdev
* - cross with all of the possible states
* - count all of the patches with the same (state, log_size)
*/
SELECT q.log_size, q.state, patchwork_state.name, count(*) AS n, q.total_n
FROM patchwork_patch, patchwork_state, (
SELECT t.*, s.id AS state
FROM (SELECT count(*) AS total_n,
patchwork_patch.project_id AS project,
floor(log(2, length(patchwork_patch.content))) AS log_size
FROM patchwork_patch
WHERE patchwork_patch.project_id =
(SELECT id FROM patchwork_project
WHERE linkname = 'netdev')
AND patchwork_patch.content IS NOT NULL
GROUP BY log_size, project) t,
patchwork_state s) q
WHERE floor(log(2, length(patchwork_patch.content))) = q.log_size
AND patchwork_patch.state_id = q.state
AND patchwork_patch.project_id = q.project
AND patchwork_state.id = q.state
GROUP BY q.state, q.total_n, q.log_size, patchwork_state.name
ORDER BY q.state, q.log_size;
_______________________________________________
Patchwork mailing list
[email protected]
https://lists.ozlabs.org/listinfo/patchwork
