Skip to content

Commit badf415

Browse files
committed
fix linter errors for green_third_party
1 parent c5a6a4e commit badf415

1 file changed

Lines changed: 49 additions & 45 deletions

File tree

Lines changed: 49 additions & 45 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,5 @@
11
#standardSQL
2+
# Median third-parties & green third-party requests per websites by rank
23

34
WITH requests AS (
45
SELECT
@@ -34,109 +35,112 @@ pages AS (
3435

3536
third_party AS (
3637
SELECT
37-
domain,
38-
COUNT(DISTINCT page) AS page_usage
38+
tp.domain,
39+
COUNT(DISTINCT r.page) AS page_usage
3940
FROM
4041
`httparchive.almanac.third_parties` AS tp
4142
INNER JOIN
4243
requests AS r
4344
ON NET.HOST(r.url) = NET.HOST(tp.domain)
4445
WHERE
45-
date = '2025-06-01' AND
46-
category NOT IN ('hosting')
46+
tp.date = '2025-06-01' AND
47+
tp.category NOT IN ('hosting')
4748
GROUP BY
48-
domain
49+
tp.domain
4950
HAVING
5051
page_usage >= 50
5152
),
5253

5354
green_tp AS (
54-
SELECT domain
55+
SELECT tp.domain
5556
FROM
5657
`httparchive.almanac.third_parties` AS tp
5758
INNER JOIN
5859
green AS g
5960
ON NET.HOST(g.host) = NET.HOST(tp.domain)
6061
WHERE
61-
date = '2025-06-01' AND
62-
category NOT IN ('hosting')
62+
tp.date = '2025-06-01' AND
63+
tp.category NOT IN ('hosting')
6364
GROUP BY
64-
domain
65+
tp.domain
6566
),
6667

6768
base AS (
6869
SELECT
69-
client,
70-
page,
71-
rank,
72-
COUNT(domain) AS third_parties_per_page
70+
r.client,
71+
r.page,
72+
p.rank,
73+
COUNT(tp.domain) AS third_parties_per_page
7374
FROM
74-
requests
75+
requests AS r
7576
LEFT JOIN
76-
third_party
77+
third_party AS tp
7778
ON
78-
NET.HOST(requests.url) = NET.HOST(third_party.domain)
79+
NET.HOST(r.url) = NET.HOST(tp.domain)
7980
INNER JOIN
80-
pages
81-
USING (client, page)
81+
pages AS p
82+
ON r.client = p.client AND r.page = p.page
8283
GROUP BY
83-
client,
84-
page,
85-
rank
84+
r.client,
85+
r.page,
86+
p.rank
8687
),
8788

8889
base_green AS (
8990
SELECT
90-
client,
91-
page,
92-
rank,
93-
COUNT(domain) AS green_third_parties_per_page
91+
r.client,
92+
r.page,
93+
p.rank,
94+
COUNT(gtp.domain) AS green_third_parties_per_page
9495
FROM
95-
requests
96+
requests AS r
9697
LEFT JOIN
97-
green_tp
98+
green_tp AS gtp
9899
ON
99-
NET.HOST(requests.url) = NET.HOST(green_tp.domain)
100+
NET.HOST(r.url) = NET.HOST(gtp.domain)
100101
INNER JOIN
101-
pages
102-
USING (client, page)
102+
pages AS p
103+
ON r.client = p.client AND r.page = p.page
103104
GROUP BY
104-
client,
105-
page,
106-
rank
105+
r.client,
106+
r.page,
107+
p.rank
107108
)
108109

109110
SELECT
110-
client,
111+
b.client,
111112
rank_grouping,
112113
CASE
113114
WHEN rank_grouping = 0 THEN ''
114115
WHEN rank_grouping = 100000000 THEN 'all'
115116
ELSE FORMAT("%'d", rank_grouping)
116117
END AS ranking,
117118
APPROX_QUANTILES(
118-
third_parties_per_page, 1000
119+
b.third_parties_per_page, 1000
119120
) [OFFSET(500)] AS p50_third_parties_per_page,
120121
APPROX_QUANTILES(
121-
green_third_parties_per_page, 1000
122+
bg.green_third_parties_per_page, 1000
122123
) [OFFSET(500)] AS p50_green_third_parties_per_page,
123124
APPROX_QUANTILES(
124-
SAFE_DIVIDE(green_third_parties_per_page, third_parties_per_page), 1000
125+
SAFE_DIVIDE(
126+
bg.green_third_parties_per_page,
127+
b.third_parties_per_page
128+
), 1000
125129
) [OFFSET(500)] AS pct_green
126130
FROM
127-
base,
131+
base AS b,
128132
UNNEST([1000, 10000, 100000, 1000000, 10000000, 100000000]) AS rank_grouping
129133
INNER JOIN
130-
base_green
134+
base_green AS bg
131135
ON
132-
base.client = base_green.client AND
133-
base.page = base_green.page AND
134-
base.rank = base_green.rank
136+
b.client = bg.client AND
137+
b.page = bg.page AND
138+
b.rank = bg.rank
135139
WHERE
136-
rank <= rank_grouping
140+
b.rank <= rank_grouping
137141
GROUP BY
138-
client,
142+
b.client,
139143
rank_grouping
140144
ORDER BY
141-
client,
145+
b.client,
142146
rank_grouping

0 commit comments

Comments
 (0)