power-bi 2.6.0 → 2.6.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: a22ca0a37fc03e4fd19214f0ecb71c8fedf820fabbbc9e7a7e5add198e0ab25a
-  data.tar.gz: 61fc121ec95aa7cc815b7ee4e84a9fe28facbb2c5177dbe91e519e17ebcfae19
+  metadata.gz: ad399c89996fb5bc9d8503233c2827c7f57366cd849e16c3e780048a8f1973ae
+  data.tar.gz: a2ffbbac846e08cc9e9fefd52b86e9a996802ec51e5a4f748f0ab8c8799f4b87
 SHA512:
-  metadata.gz: 1c69b92f3b68c19b67a417ba7deb402b4bcc8130ebf074260e97b36c62f773ac27e3720d4fa3946060ad1d8d062bbcf95768c436f7be207e773f849a818a633f
-  data.tar.gz: 9454fcb89c9de930616e8ea944fb9aba73f5cfa558e2bef1e755c4f16ec66164dff386154532242ebd55242bc6ccd6ecbcc59fce3f64bf5e8fca554ed309ee59
+  metadata.gz: 88c0f92c90ae4392b20ea6a85f7be737302857735135b73cba780c2afd31349b672772ac2d95e38fa40a4cd01d8ae226edf49acab7caf4219c118c99021559c8
+  data.tar.gz: 78f8483bfc046ef783b852ec66b6ce7c5645c123e793dd475416de440ac83fb2ae65f3968553e02e8474e983ba06f9b27f32bf0e6ff57f3c406586b8e693427c

data/lib/power-bi/admin.rb

@@ -30,28 +30,11 @@ module PowerBI
     end
 
     def get_workspaces(filter: nil, expand: nil)
-      params = {}
-      params[:$filter] = filter if filter
-      params[:$expand] = expand if expand
+      base_params = {}
+      base_params[:$filter] = filter if filter
+      base_params[:$expand] = expand if expand
 
-      url = '/admin/groups'
-
-      nr_records = 5000
-      count = 0
-
-      data = []
-
-      loop do
-        params[:$top] = nr_records
-        params[:$skip] = count
-        resp = @tenant.get(url, params)
-        data += resp[:value]
-        batch_count = resp[:value].size
-        count += batch_count
-        break if batch_count < nr_records
-      end
-
-      data
+      @tenant.get_paginated('/admin/groups', base_params: base_params)
     end
 
     def force_delete_workspace_by_workspace_name(user_email, workspace_name)

data/lib/power-bi/dataset.rb

@@ -87,7 +87,7 @@ module PowerBI
     end
 
     def get_data
-      @tenant.get("/groups/#{@workspace.id}/datasets")[:value]
+      @tenant.get_paginated("/groups/#{@workspace.id}/datasets")
     end
   end
 end

data/lib/power-bi/profile.rb

@@ -38,7 +38,7 @@ module PowerBI
     end
 
     def get_data
-      @tenant.get("/profiles", use_profile: false)[:value]
+      @tenant.get_paginated("/profiles", use_profile: false)
     end
   end
 end

data/lib/power-bi/report.rb

@@ -106,7 +106,7 @@ module PowerBI
     end
 
     def get_data
-      @tenant.get("/groups/#{@workspace.id}/reports")[:value]
+      @tenant.get_paginated("/groups/#{@workspace.id}/reports")
     end
   end
 end

data/lib/power-bi/tenant.rb

@@ -198,6 +198,69 @@ module PowerBI
       JSON.parse(response.body, symbolize_names: true)
     end
 
+    # Fetches paginated data from the Power BI API using OData-style pagination.
+    #
+    # Power BI API has a documented limit of 5000 records per request.
+    # This method handles pagination automatically by:
+    # 1. Requesting 5000 records at a time using $top and $skip
+    # 2. If exactly 5000 records are returned, fetching the next page
+    # 3. Accumulating all results across pages
+    # 4. Deduplicating records by ID (to handle insertions between requests)
+    #
+    # Note: $skip-based pagination is not fully protected against deletions
+    # between requests (a deleted record may cause a subsequent record to go
+    # unseen). This risk is acceptable given the short pagination window.
+    MAX_PAGE_SIZE = 5000
+    MAX_ITERATIONS = 100
+    def get_paginated(url, page_size: MAX_PAGE_SIZE, base_params: {}, use_profile: true, max_iterations: MAX_ITERATIONS)
+      page_size = [page_size, MAX_PAGE_SIZE].min
+
+      skip = 0
+      all_data = []
+      iteration = 0
+
+      loop do
+        iteration += 1
+        if iteration > max_iterations
+          log "WARNING: Reached maximum iteration limit (#{max_iterations}). " \
+              "Fetched #{all_data.size} records so far. This may indicate an API issue or " \
+              "an extremely large dataset. Consider using API filters to reduce the result set.",
+              level: :warn
+          break
+        end
+
+        params = base_params.merge({
+          '$top' => page_size,
+          '$skip' => skip
+        })
+
+        log "Fetching paginated data from #{url} (skip: #{skip}, top: #{page_size}, iteration: #{iteration})"
+
+        resp = get(url, params, use_profile: use_profile)
+        batch = resp[:value] || []
+        all_data += batch
+        batch_count = batch.size
+
+        log "Received #{batch_count} records (total so far: #{all_data.size})"
+
+        # If we got fewer records than requested, we've reached the last page
+        break if batch_count < page_size
+
+        skip += batch_count
+      end
+
+      # Deduplicate by ID to handle any records that were inserted between requests.
+      # Insertions before the current $skip position shift items right, which can
+      # cause duplicates across pages.
+      deduplicated_data = all_data.uniq{|r| r[:id]}
+
+      if deduplicated_data.size < all_data.size
+        log "Removed #{all_data.size - deduplicated_data.size} duplicate records during deduplication"
+      end
+
+      deduplicated_data
+    end
+
     private
 
     def add_spp_header(req)

data/lib/power-bi/workspace.rb

@@ -113,7 +113,7 @@ module PowerBI
     end
 
     def get_data
-      @tenant.get("/groups")[:value]
+      @tenant.get_paginated("/groups")
     end
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: power-bi
 version: !ruby/object:Gem::Version
-  version: 2.6.0
+  version: 2.6.1
 platform: ruby
 authors:
 - Lode Cools
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2025-07-10 00:00:00.000000000 Z
+date: 2026-02-11 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: faraday