62 static Result<std::unique_ptr<ManifestGroup>> Make(
63 std::shared_ptr<FileIO> io, std::shared_ptr<Schema> schema,
64 std::unordered_map<int32_t, std::shared_ptr<PartitionSpec>> specs_by_id_,
65 std::vector<ManifestFile> manifests);
74 static Result<std::unique_ptr<ManifestGroup>> Make(
75 std::shared_ptr<FileIO> io, std::shared_ptr<Schema> schema,
76 std::unordered_map<int32_t, std::shared_ptr<PartitionSpec>> specs_by_id,
77 std::vector<ManifestFile> data_manifests,
78 std::vector<ManifestFile> delete_manifests);
88 ManifestGroup& FilterData(std::shared_ptr<Expression> filter);
91 ManifestGroup& FilterFiles(std::shared_ptr<Expression> filter);
94 ManifestGroup& FilterPartitions(std::shared_ptr<Expression> filter);
125 ManifestGroup& ColumnsToKeepStats(std::unordered_set<int32_t> column_ids);
134 Result<std::vector<std::shared_ptr<FileScanTask>>> PlanFiles();
137 Result<std::vector<ManifestEntry>> Entries();
139 using CreateTasksFunction =
140 std::function<Result<std::vector<std::shared_ptr<ScanTask>>>(
141 std::vector<ManifestEntry>&&,
const TaskContext&)>;
147 Result<std::vector<std::shared_ptr<ScanTask>>> Plan(
148 const CreateTasksFunction& create_tasks);
151 ManifestGroup(std::shared_ptr<FileIO> io, std::shared_ptr<Schema> schema,
152 std::unordered_map<int32_t, std::shared_ptr<PartitionSpec>> specs_by_id,
153 std::vector<ManifestFile> data_manifests,
154 DeleteFileIndex::Builder&& delete_index_builder);
156 Result<std::unordered_map<int32_t, std::vector<ManifestEntry>>> ReadEntries();
158 Result<std::unique_ptr<ManifestReader>> MakeReader(
const ManifestFile& manifest);
160 std::shared_ptr<FileIO> io_;
161 std::shared_ptr<Schema> schema_;
162 std::unordered_map<int32_t, std::shared_ptr<PartitionSpec>> specs_by_id_;
163 std::vector<ManifestFile> data_manifests_;
164 DeleteFileIndex::Builder delete_index_builder_;
165 std::shared_ptr<Expression> data_filter_;
166 std::shared_ptr<Expression> file_filter_;
167 std::shared_ptr<Expression> partition_filter_;
168 std::function<bool(
const ManifestEntry&)> manifest_entry_predicate_;
169 std::vector<std::string> columns_;
170 std::unordered_set<int32_t> columns_to_keep_stats_;
171 OptionalExecutor executor_;
172 bool case_sensitive_ =
true;
173 bool ignore_deleted_ =
false;
174 bool ignore_existing_ =
false;
175 bool ignore_residuals_ =
false;
A manifest is an immutable Avro file that lists data files or delete files, along with each file's pa...
Definition manifest_entry.h:311
Context passed to task creation functions.
Definition manifest_group.h:44