Merge branch 'privacy-markdown-2024' of https://github.com/HTTPArchiv…

…e/almanac.httparchive.org into privacy-markdown-2024

.github/workflows/code-static-analysis.yml

@@ -35,7 +35,7 @@ jobs:
         uses: actions/checkout@v4
       - name: Set up Python 3.12
         if: ${{ matrix.language == 'python' }}
-        uses: actions/setup-python@v5.2.0
+        uses: actions/setup-python@v5.3.0
         with:
           python-version: '3.12'
       - name: Install dependencies

.github/workflows/lintsql.yml

@@ -19,7 +19,7 @@ jobs:
           # Full git history is needed to get a proper list of changed files within `super-linter`
           fetch-depth: 0
       - name: Set up Python 3.12
-        uses: actions/setup-python@v5.2.0
+        uses: actions/setup-python@v5.3.0
         with:
           python-version: '3.12'
       - name: Lint SQL code

.github/workflows/predeploy.yml

@@ -37,7 +37,7 @@ jobs:
       with:
         node-version: '20'
     - name: Set up Python 3.12
-      uses: actions/setup-python@v5.2.0
+      uses: actions/setup-python@v5.3.0
       with:
         python-version: '3.12'
     - name: Install Asian Fonts

.github/workflows/test_website.yml

@@ -30,7 +30,7 @@ jobs:
       with:
         node-version: '20'
     - name: Set up Python 3.12
-      uses: actions/setup-python@v5.2.0
+      uses: actions/setup-python@v5.3.0
       with:
         python-version: '3.12'
     - name: Run the website

README.md

@@ -40,7 +40,7 @@ We do almost all of our project planning here on GitHub. Browse the open [issues
 
 We also have a [`#web-almanac`](https://join.slack.com/t/httparchive/shared_invite/zt-45sgwmnb-eDEatOhqssqNAKxxOSLAaA) channel on the HTTP Archive Slack where we chat about project updates.
 
-For news and announcements, follow [@HTTPArchive](https://twitter.com/HTTPArchive) on Twitter.
+For news and announcements, follow [@HTTPArchive](https://x.com/HTTPArchive) on Twitter.
 
 ## [License](https://github.com/HTTPArchive/almanac.httparchive.org/blob/main/LICENSE)
 

sql/2024/accessibility/a11y_frontend_technology.sql

@@ -0,0 +1,50 @@
+WITH score_data AS (
+  SELECT
+    client,
+    page,
+    CAST(JSON_EXTRACT_SCALAR(lighthouse, '$.categories.performance.score') AS FLOAT64) AS performance_score,
+    CAST(JSON_EXTRACT_SCALAR(lighthouse, '$.categories.accessibility.score') AS FLOAT64) AS accessibility_score,
+    CAST(JSON_EXTRACT_SCALAR(lighthouse, '$.categories.best-practices.score') AS FLOAT64) AS best_practices_score,
+    CAST(JSON_EXTRACT_SCALAR(lighthouse, '$.categories.seo.score') AS FLOAT64) AS seo_score,
+    t.technology AS framework
+  FROM
+    `httparchive.all.pages`,
+    UNNEST(technologies) AS t
+  WHERE
+    date = '2024-06-01' AND
+    lighthouse IS NOT NULL AND
+    lighthouse != '{}' AND
+    is_root_page = TRUE AND
+    ('Web frameworks' IN UNNEST(t.categories) OR 'JavaScript libraries' IN UNNEST(t.categories) OR 'Frontend frameworks' IN UNNEST(t.categories) OR 'JavaScript frameworks' IN UNNEST(t.categories)) AND
+    t.technology IS NOT NULL
+)
+
+SELECT
+  client,
+  framework,
+  AVG(performance_score) AS avg_performance_score,
+  AVG(accessibility_score) AS avg_accessibility_score,
+  AVG(best_practices_score) AS avg_best_practices_score,
+  AVG(seo_score) AS avg_seo_score,
+  COUNT(DISTINCT page) AS total_pages
+FROM (
+  SELECT
+    client,
+    page,
+    framework,
+    AVG(performance_score) AS performance_score, # All scores are the same for one page (we have multiple rows due to unnest), we could also take the first instead of the average
+    AVG(accessibility_score) AS accessibility_score,
+    AVG(best_practices_score) AS best_practices_score,
+    AVG(seo_score) AS seo_score
+  FROM
+    score_data
+  GROUP BY
+    client,
+    page,
+    framework
+  )
+GROUP BY
+  client,
+  framework
+ORDER BY
+  total_pages DESC;

sql/2024/accessibility/a11y_overall_tech_usage_by_domain_rank.sql

@@ -0,0 +1,56 @@
+#standardSQL
+# Overall Accessibility (A11y) technology, ie. Overlays, usage by domain rank
+
+# Main SELECT statement to aggregate results by client and rank grouping.
+SELECT
+  client,
+  is_root_page,
+  rank_grouping, # Grouping of domains by their rank (e.g., top 1000, top 10000, etc.)
+  total_in_rank, # Total number of sites within the rank grouping
+  COUNT(DISTINCT page) AS sites_with_a11y_tech, # Number of unique sites that use accessibility technology
+  COUNT(DISTINCT page) / total_in_rank AS pct_sites_with_a11y_tech # Percentage of sites using accessibility technology within the rank grouping
+FROM
+  (
+    # Subquery to filter and extract relevant pages with A11Y technology
+    SELECT DISTINCT
+      client,
+      is_root_page,
+      page,
+      rank_grouping,
+      category
+    FROM
+      `httparchive.all.pages`,
+      UNNEST([1000, 10000, 100000, 1000000, 10000000, 100000000]) AS rank_grouping, # Expand rank_grouping to cover different rank categories
+      UNNEST(technologies) AS tech,
+      UNNEST(categories) AS category
+    WHERE
+      date = '2024-06-01' AND
+      category = 'Accessibility' AND
+      rank <= rank_grouping # Include only sites within the specified rank grouping
+  )
+JOIN
+  (
+    # Subquery to count total sites in each rank grouping for each client
+    SELECT
+      client,
+      rank_grouping,
+      COUNT(0) AS total_in_rank
+    FROM
+      `httparchive.all.pages`,
+      UNNEST([1000, 10000, 100000, 1000000, 10000000, 100000000]) AS rank_grouping
+    WHERE
+      date = '2024-06-01' AND
+      rank <= rank_grouping
+    GROUP BY
+      client,
+      rank_grouping
+  ) USING (client, rank_grouping)
+GROUP BY
+  client,
+  is_root_page,
+  rank_grouping,
+  total_in_rank
+ORDER BY
+  client,
+  is_root_page,
+  rank_grouping

sql/2024/accessibility/a11y_technology_usage.sql

@@ -0,0 +1,21 @@
+#standardSQL
+# Accessibility (A11y) technology, ie. Overlays, usage by client
+
+SELECT
+  client,  # Client domain
+  is_root_page,
+  COUNT(DISTINCT page) AS total_sites,  # Total number of unique sites for the client
+  COUNT(DISTINCT IF(category = 'Accessibility', page, NULL)) AS sites_with_a11y_tech,  # Number of unique sites that use accessibility technology
+  COUNT(DISTINCT IF(category = 'Accessibility', page, NULL)) / COUNT(DISTINCT page) AS pct_sites_with_a11y_tech  # Percentage of sites using accessibility technology
+FROM
+  `httparchive.all.pages`,
+  UNNEST(technologies) AS tech,
+  UNNEST(categories) AS category
+WHERE
+  date = '2024-06-01' # Specific date for data extraction
+GROUP BY
+  client,
+  is_root_page
+ORDER BY
+  client,
+  is_root_page;

sql/2024/accessibility/a11y_technology_usage_by_domain_rank.sql

@@ -0,0 +1,71 @@
+#standardSQL
+# A11Y technology usage by domain rank
+WITH ranked_sites AS (
+  -- Get the total number of sites within each rank grouping
+  SELECT
+    client,
+    is_root_page,
+    page,
+    rank,
+    technologies,  -- Include technologies field here
+    CASE
+      WHEN rank <= 1000 THEN 1000
+      WHEN rank <= 10000 THEN 10000
+      WHEN rank <= 100000 THEN 100000
+      WHEN rank <= 1000000 THEN 1000000
+      WHEN rank <= 10000000 THEN 10000000
+      WHEN rank <= 100000000 THEN 100000000
+    END AS rank_grouping
+  FROM
+    `httparchive.all.pages`
+  WHERE
+    date = '2024-06-01'  -- Use the relevant date for analysis
+),
+
+rank_totals AS (
+  -- Calculate total sites in each rank grouping
+  SELECT
+    client,
+    is_root_page,
+    rank_grouping,
+    COUNT(DISTINCT page) AS total_in_rank
+  FROM
+    ranked_sites
+  GROUP BY
+    client,
+    is_root_page,
+    rank_grouping
+)
+
+SELECT
+  r.client,
+  r.is_root_page,
+  r.rank_grouping,
+  rt.total_in_rank,  -- Total number of unique sites within the rank grouping
+  tech.technology AS app,  -- Accessibility technology used
+  COUNT(DISTINCT r.page) AS sites_with_app,  -- Number of sites using the specific accessibility technology
+  SAFE_DIVIDE(COUNT(DISTINCT r.page), rt.total_in_rank) AS pct_sites_with_app  -- Percentage of sites using the accessibility technology
+FROM
+  ranked_sites r
+JOIN
+  UNNEST(r.technologies) AS tech  -- Expand technologies array to individual rows
+JOIN
+  rank_totals rt  -- Join to get the total number of sites per rank grouping
+ON r.client = rt.client AND
+  r.is_root_page = rt.is_root_page AND
+  r.rank_grouping = rt.rank_grouping
+JOIN
+  UNNEST(tech.categories) AS category  -- Unnest the categories array to filter for accessibility
+WHERE
+  category = 'Accessibility'  -- Filter to include only accessibility-related technologies
+GROUP BY
+  r.client,
+  r.is_root_page,
+  r.rank_grouping,
+  rt.total_in_rank,
+  tech.technology
+ORDER BY
+  tech.technology,  -- Order results by technology (app)
+  r.rank_grouping,  -- Order results by rank grouping
+  r.client,
+  r.is_root_page;

sql/2024/accessibility/alt_ending_in_image_extension.sql

@@ -0,0 +1,76 @@
+#standardSQL
+# Alt text ending in an image extension
+CREATE TEMPORARY FUNCTION getUsedExtensions(payload STRING)
+RETURNS ARRAY<STRUCT<extension STRING, total INT64>> LANGUAGE js AS '''
+try {
+  const a11y = JSON.parse(payload);
+
+  return Object.entries(a11y.file_extension_alts.file_extensions).map(([extension, total]) => {
+    return {extension, total};
+  });
+} catch (e) {
+  return [];
+}
+''';
+SELECT
+  client,
+  is_root_page,
+  sites_with_non_empty_alt,
+  sites_with_file_extension_alt,
+  total_alts_with_file_extensions,
+
+  # Of sites with a non-empty alt, what % have an alt with a file extension
+  sites_with_file_extension_alt / sites_with_non_empty_alt AS pct_sites_with_file_extension_alt,
+  # Given a random alt, how often will it end in a file extension
+  total_alts_with_file_extensions / total_non_empty_alts AS pct_alts_with_file_extension,
+
+  extension_stat.extension AS extension,
+  COUNT(0) AS total_sites_using,
+  # Of sites with a non-empty alt, what % have an alt with this file extension
+  COUNT(0) / sites_with_non_empty_alt AS pct_applicable_sites_using,
+
+  # Of sites with a non-empty alt, what % have an alt with this file extension
+  SUM(extension_stat.total) AS total_occurances,
+  # Given a random alt ending in a file extension, how often will it end in this file extension
+  SUM(extension_stat.total) / total_alts_with_file_extensions AS pct_total_occurances
+FROM
+  `httparchive.all.pages`,
+  UNNEST(getUsedExtensions(JSON_EXTRACT(custom_metrics, '$.a11y'))) AS extension_stat
+LEFT JOIN (
+  SELECT
+    client,
+    is_root_page,
+    COUNTIF(total_non_empty_alt > 0) AS sites_with_non_empty_alt,
+    COUNTIF(total_with_file_extension > 0) AS sites_with_file_extension_alt,
+
+    SUM(total_non_empty_alt) AS total_non_empty_alts,
+    SUM(total_with_file_extension) AS total_alts_with_file_extensions
+  FROM (
+    SELECT
+      client,
+      is_root_page,
+      CAST(JSON_EXTRACT_SCALAR(custom_metrics, '$.markup.images.img.alt.present') AS INT64) AS total_non_empty_alt,
+      CAST(JSON_EXTRACT_SCALAR(custom_metrics, '$.a11y.file_extension_alts.total_with_file_extension') AS INT64) AS total_with_file_extension
+    FROM
+      `httparchive.all.pages`
+    WHERE
+      date = '2024-06-01'
+  )
+  GROUP BY
+    client,
+    is_root_page
+) USING (client, is_root_page)
+WHERE
+  date = '2024-06-01'
+GROUP BY
+  client,
+  is_root_page,
+  sites_with_non_empty_alt,
+  sites_with_file_extension_alt,
+  total_non_empty_alts,
+  total_alts_with_file_extensions,
+  extension
+ORDER BY
+  client,
+  is_root_page,
+  total_occurances DESC

sql/2024/accessibility/anchors_with_role_button.sql

@@ -0,0 +1,25 @@
+#standardSQL
+# Anchors with role='button'
+SELECT
+  client,
+  is_root_page,
+  COUNTIF(total_anchors > 0) AS sites_with_anchors,
+  COUNTIF(total_anchors_with_role_button > 0) AS sites_with_anchor_role_button,
+
+  # Of sites that have anchors... how many have an anchor with a role='button'
+  COUNTIF(total_anchors_with_role_button > 0) / COUNTIF(total_anchors > 0) AS pct_sites_with_anchor_role_button
+FROM (
+  SELECT
+    client,
+    is_root_page,
+    date,
+    CAST(JSON_EXTRACT_SCALAR(JSON_EXTRACT_SCALAR(payload, '$._a11y'), '$.total_anchors_with_role_button') AS INT64) AS total_anchors_with_role_button,
+    IFNULL(CAST(JSON_EXTRACT_SCALAR(JSON_EXTRACT_SCALAR(payload, '$._element_count'), '$.a') AS INT64), 0) AS total_anchors
+  FROM
+    `httparchive.all.pages`
+)
+WHERE
+  date = '2024-06-01'
+GROUP BY
+  client,
+  is_root_page;

sql/2024/accessibility/audio_track_usage.sql

@@ -0,0 +1,27 @@
+#standardSQL
+# Audio elements track usage
+SELECT
+  client,
+  is_root_page,
+  COUNT(0) AS total_sites,
+  COUNTIF(total_audios > 0) AS total_with_audio,
+  COUNTIF(total_with_track > 0) AS total_with_tracks,
+
+  SUM(total_with_track) / SUM(total_audios) AS pct_audios_with_tracks,
+  COUNTIF(total_audios > 0) / COUNT(0) AS pct_sites_with_audios,
+  COUNTIF(total_with_track > 0) / COUNTIF(total_audios > 0) AS pct_audio_sites_with_tracks
+FROM (
+  SELECT
+    client,
+    is_root_page,
+    date,
+    CAST(JSON_EXTRACT_SCALAR(JSON_EXTRACT_SCALAR(payload, '$._almanac'), '$.audios.total') AS INT64) AS total_audios,
+    CAST(JSON_EXTRACT_SCALAR(JSON_EXTRACT_SCALAR(payload, '$._almanac'), '$.audios.total_with_track') AS INT64) AS total_with_track
+  FROM
+    `httparchive.all.pages`
+  WHERE
+    date = '2024-06-01'
+)
+GROUP BY
+  client,
+  is_root_page;