wgtmac commented on code in PR #873:
URL: https://github.com/apache/iceberg-cpp/pull/873#discussion_r4026226616


##########
src/iceberg/manifest/manifest_group.cc:
##########
@@ -124,13 +123,320 @@ ManifestGroup::ManifestGroup(
       data_filter_(True::Instance()),
       file_filter_(True::Instance()),
       partition_filter_(True::Instance()),
-      manifest_entry_predicate_([](const ManifestEntry&) { return true; }) {}
+      manifest_entry_predicate_([](const ManifestEntry&) { return true; }),
+      columns_{std::string(Schema::kAllColumns)} {}
 
 ManifestGroup::~ManifestGroup() = default;
 
 ManifestGroup::ManifestGroup(ManifestGroup&&) noexcept = default;
 ManifestGroup& ManifestGroup::operator=(ManifestGroup&&) noexcept = default;
 
+class ManifestGroup::FilePlanningStream final : public FileScanTaskStream {
+ public:
+  static Result<FileScanTaskStreamPtr> Make(std::unique_ptr<ManifestGroup> 
group) {
+    ICEBERG_RETURN_UNEXPECTED(group->CheckErrors());
+
+    group->delete_index_builder_.WithScanMetrics(group->scan_metrics_);
+    ICEBERG_ASSIGN_OR_RAISE(auto delete_index, 
group->delete_index_builder_.Build());
+
+    auto stats_projection =
+        group->PrepareStatsProjection(delete_index->has_equality_deletes());
+
+    std::unique_ptr<Evaluator> data_file_evaluator;
+    if (group->file_filter_ &&
+        group->file_filter_->op() != Expression::Operation::kTrue) {
+      ICEBERG_ASSIGN_OR_RAISE(
+          data_file_evaluator,
+          Evaluator::Make(*DataFileFilterSchema(), group->file_filter_,
+                          group->case_sensitive_));
+    }
+    const bool drop_stats = stats_projection.drop_stats;
+
+    return FileScanTaskStreamPtr(new FilePlanningStream(
+        std::move(group), std::move(delete_index), 
std::move(data_file_evaluator),
+        std::move(stats_projection.columns), drop_stats));
+  }
+
+  Result<std::optional<std::shared_ptr<FileScanTask>>> NextImpl() override {
+    while (true) {
+      ICEBERG_ASSIGN_OR_RAISE(auto entry, NextEntry());
+      if (!entry.has_value()) {
+        return std::nullopt;
+      }
+
+      auto [spec_id, value] = std::move(entry).value();
+      if (group_->ignore_existing_ && value.status == 
ManifestStatus::kExisting) {
+        IncrementSkippedDataFiles();
+        continue;
+      }
+
+      ICEBERG_DCHECK(value.data_file != nullptr, "Data file cannot be null");
+      if (data_file_evaluator_) {
+        DataFileStructLike data_file(*value.data_file);
+        ICEBERG_ASSIGN_OR_RAISE(bool should_match,
+                                data_file_evaluator_->Evaluate(data_file));
+        if (!should_match) {
+          IncrementSkippedDataFiles();
+          continue;
+        }
+      }
+
+      if (!group_->manifest_entry_predicate_(value)) {
+        IncrementSkippedDataFiles();
+        continue;
+      }
+
+      ICEBERG_ASSIGN_OR_RAISE(auto delete_files, 
delete_index_->ForEntry(value));
+
+      // Equality-delete matching uses data-file statistics. Drop unrequested 
stats only
+      // after the delete index has finished matching this entry.
+      if (drop_stats_) {
+        ContentFileUtil::DropAllStats(*value.data_file);
+      } else if (!group_->columns_to_keep_stats_.empty()) {
+        ContentFileUtil::DropUnselectedStats(*value.data_file,
+                                             group_->columns_to_keep_stats_);
+      }
+
+      UpdateResultMetrics(*value.data_file, delete_files);
+
+      ICEBERG_ASSIGN_OR_RAISE(auto residuals, GetResidualEvaluator(spec_id));
+      ICEBERG_ASSIGN_OR_RAISE(auto residual,
+                              
residuals->ResidualFor(value.data_file->partition));
+
+      return 
std::optional<std::shared_ptr<FileScanTask>>{std::make_shared<FileScanTask>(
+          std::move(value.data_file), std::move(delete_files), 
std::move(residual))};
+    }
+  }
+
+ private:
+  FilePlanningStream(std::unique_ptr<ManifestGroup> group,
+                     std::unique_ptr<DeleteFileIndex> delete_index,
+                     std::unique_ptr<Evaluator> data_file_evaluator,
+                     std::vector<std::string> columns, bool drop_stats)
+      : group_(std::move(group)),
+        delete_index_(std::move(delete_index)),
+        data_file_evaluator_(std::move(data_file_evaluator)),
+        columns_(std::move(columns)),
+        drop_stats_(drop_stats) {}
+
+  using TaggedEntry = std::pair<int32_t, ManifestEntry>;
+  using TaggedStream = std::pair<int32_t, ManifestEntryStreamPtr>;
+
+  // TODO: Perhaps refactor this concurrent/sequential stream state machine 
into a generic

Review Comment:
   ```suggestion
     // FIXME: Perhaps refactor this concurrent/sequential stream state machine 
into a generic
   ```



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to