File size: 16,815 Bytes
9c6594c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 |
// Licensed to the Apache Software Foundation (ASF) under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License. You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing,
// software distributed under the License is distributed on an
// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied. See the License for the
// specific language governing permissions and limitations
// under the License.
// This API is EXPERIMENTAL.
#pragma once
#include <functional>
#include <iosfwd>
#include <memory>
#include <optional>
#include <string>
#include <unordered_map>
#include <utility>
#include <vector>
#include "arrow/compute/expression.h"
#include "arrow/dataset/type_fwd.h"
#include "arrow/dataset/visibility.h"
#include "arrow/util/compare.h"
namespace arrow {
namespace dataset {
constexpr char kFilenamePartitionSep = '_';
struct ARROW_DS_EXPORT PartitionPathFormat {
std::string directory, filename;
};
// ----------------------------------------------------------------------
// Partitioning
/// \defgroup dataset-partitioning Partitioning API
///
/// @{
/// \brief Interface for parsing partition expressions from string partition
/// identifiers.
///
/// For example, the identifier "foo=5" might be parsed to an equality expression
/// between the "foo" field and the value 5.
///
/// Some partitionings may store the field names in a metadata
/// store instead of in file paths, for example
/// dataset_root/2009/11/... could be used when the partition fields
/// are "year" and "month"
///
/// Paths are consumed from left to right. Paths must be relative to
/// the root of a partition; path prefixes must be removed before passing
/// the path to a partitioning for parsing.
class ARROW_DS_EXPORT Partitioning : public util::EqualityComparable<Partitioning> {
public:
virtual ~Partitioning() = default;
/// \brief The name identifying the kind of partitioning
virtual std::string type_name() const = 0;
//// \brief Return whether the partitionings are equal
virtual bool Equals(const Partitioning& other) const {
return schema_->Equals(other.schema_, /*check_metadata=*/false);
}
/// \brief If the input batch shares any fields with this partitioning,
/// produce sub-batches which satisfy mutually exclusive Expressions.
struct PartitionedBatches {
RecordBatchVector batches;
std::vector<compute::Expression> expressions;
};
virtual Result<PartitionedBatches> Partition(
const std::shared_ptr<RecordBatch>& batch) const = 0;
/// \brief Parse a path into a partition expression
virtual Result<compute::Expression> Parse(const std::string& path) const = 0;
virtual Result<PartitionPathFormat> Format(const compute::Expression& expr) const = 0;
/// \brief A default Partitioning which is a DirectoryPartitioning
/// with an empty schema.
static std::shared_ptr<Partitioning> Default();
/// \brief The partition schema.
const std::shared_ptr<Schema>& schema() const { return schema_; }
protected:
explicit Partitioning(std::shared_ptr<Schema> schema) : schema_(std::move(schema)) {}
std::shared_ptr<Schema> schema_;
};
/// \brief The encoding of partition segments.
enum class SegmentEncoding : int8_t {
/// No encoding.
None = 0,
/// Segment values are URL-encoded.
Uri = 1,
};
ARROW_DS_EXPORT
std::ostream& operator<<(std::ostream& os, SegmentEncoding segment_encoding);
/// \brief Options for key-value based partitioning (hive/directory).
struct ARROW_DS_EXPORT KeyValuePartitioningOptions {
/// After splitting a path into components, decode the path components
/// before parsing according to this scheme.
SegmentEncoding segment_encoding = SegmentEncoding::Uri;
};
/// \brief Options for inferring a partitioning.
struct ARROW_DS_EXPORT PartitioningFactoryOptions {
/// When inferring a schema for partition fields, yield dictionary encoded types
/// instead of plain. This can be more efficient when materializing virtual
/// columns, and Expressions parsed by the finished Partitioning will include
/// dictionaries of all unique inspected values for each field.
bool infer_dictionary = false;
/// Optionally, an expected schema can be provided, in which case inference
/// will only check discovered fields against the schema and update internal
/// state (such as dictionaries).
std::shared_ptr<Schema> schema;
/// After splitting a path into components, decode the path components
/// before parsing according to this scheme.
SegmentEncoding segment_encoding = SegmentEncoding::Uri;
KeyValuePartitioningOptions AsPartitioningOptions() const;
};
/// \brief Options for inferring a hive-style partitioning.
struct ARROW_DS_EXPORT HivePartitioningFactoryOptions : PartitioningFactoryOptions {
/// The hive partitioning scheme maps null to a hard coded fallback string.
std::string null_fallback;
HivePartitioningOptions AsHivePartitioningOptions() const;
};
/// \brief PartitioningFactory provides creation of a partitioning when the
/// specific schema must be inferred from available paths (no explicit schema is known).
class ARROW_DS_EXPORT PartitioningFactory {
public:
virtual ~PartitioningFactory() = default;
/// \brief The name identifying the kind of partitioning
virtual std::string type_name() const = 0;
/// Get the schema for the resulting Partitioning.
/// This may reset internal state, for example dictionaries of unique representations.
virtual Result<std::shared_ptr<Schema>> Inspect(
const std::vector<std::string>& paths) = 0;
/// Create a partitioning using the provided schema
/// (fields may be dropped).
virtual Result<std::shared_ptr<Partitioning>> Finish(
const std::shared_ptr<Schema>& schema) const = 0;
};
/// \brief Subclass for the common case of a partitioning which yields an equality
/// expression for each segment
class ARROW_DS_EXPORT KeyValuePartitioning : public Partitioning {
public:
/// An unconverted equality expression consisting of a field name and the representation
/// of a scalar value
struct Key {
std::string name;
std::optional<std::string> value;
};
Result<PartitionedBatches> Partition(
const std::shared_ptr<RecordBatch>& batch) const override;
Result<compute::Expression> Parse(const std::string& path) const override;
Result<PartitionPathFormat> Format(const compute::Expression& expr) const override;
const ArrayVector& dictionaries() const { return dictionaries_; }
SegmentEncoding segment_encoding() const { return options_.segment_encoding; }
bool Equals(const Partitioning& other) const override;
protected:
KeyValuePartitioning(std::shared_ptr<Schema> schema, ArrayVector dictionaries,
KeyValuePartitioningOptions options)
: Partitioning(std::move(schema)),
dictionaries_(std::move(dictionaries)),
options_(options) {
if (dictionaries_.empty()) {
dictionaries_.resize(schema_->num_fields());
}
}
virtual Result<std::vector<Key>> ParseKeys(const std::string& path) const = 0;
virtual Result<PartitionPathFormat> FormatValues(const ScalarVector& values) const = 0;
/// Convert a Key to a full expression.
Result<compute::Expression> ConvertKey(const Key& key) const;
Result<std::vector<std::string>> FormatPartitionSegments(
const ScalarVector& values) const;
Result<std::vector<Key>> ParsePartitionSegments(
const std::vector<std::string>& segments) const;
ArrayVector dictionaries_;
KeyValuePartitioningOptions options_;
};
/// \brief DirectoryPartitioning parses one segment of a path for each field in its
/// schema. All fields are required, so paths passed to DirectoryPartitioning::Parse
/// must contain segments for each field.
///
/// For example given schema<year:int16, month:int8> the path "/2009/11" would be
/// parsed to ("year"_ == 2009 and "month"_ == 11)
class ARROW_DS_EXPORT DirectoryPartitioning : public KeyValuePartitioning {
public:
/// If a field in schema is of dictionary type, the corresponding element of
/// dictionaries must be contain the dictionary of values for that field.
explicit DirectoryPartitioning(std::shared_ptr<Schema> schema,
ArrayVector dictionaries = {},
KeyValuePartitioningOptions options = {});
std::string type_name() const override { return "directory"; }
bool Equals(const Partitioning& other) const override;
/// \brief Create a factory for a directory partitioning.
///
/// \param[in] field_names The names for the partition fields. Types will be
/// inferred.
static std::shared_ptr<PartitioningFactory> MakeFactory(
std::vector<std::string> field_names, PartitioningFactoryOptions = {});
private:
Result<std::vector<Key>> ParseKeys(const std::string& path) const override;
Result<PartitionPathFormat> FormatValues(const ScalarVector& values) const override;
};
/// \brief The default fallback used for null values in a Hive-style partitioning.
static constexpr char kDefaultHiveNullFallback[] = "__HIVE_DEFAULT_PARTITION__";
struct ARROW_DS_EXPORT HivePartitioningOptions : public KeyValuePartitioningOptions {
std::string null_fallback = kDefaultHiveNullFallback;
static HivePartitioningOptions DefaultsWithNullFallback(std::string fallback) {
HivePartitioningOptions options;
options.null_fallback = std::move(fallback);
return options;
}
};
/// \brief Multi-level, directory based partitioning
/// originating from Apache Hive with all data files stored in the
/// leaf directories. Data is partitioned by static values of a
/// particular column in the schema. Partition keys are represented in
/// the form $key=$value in directory names.
/// Field order is ignored, as are missing or unrecognized field names.
///
/// For example given schema<year:int16, month:int8, day:int8> the path
/// "/day=321/ignored=3.4/year=2009" parses to ("year"_ == 2009 and "day"_ == 321)
class ARROW_DS_EXPORT HivePartitioning : public KeyValuePartitioning {
public:
/// If a field in schema is of dictionary type, the corresponding element of
/// dictionaries must be contain the dictionary of values for that field.
explicit HivePartitioning(std::shared_ptr<Schema> schema, ArrayVector dictionaries = {},
std::string null_fallback = kDefaultHiveNullFallback)
: KeyValuePartitioning(std::move(schema), std::move(dictionaries),
KeyValuePartitioningOptions()),
hive_options_(
HivePartitioningOptions::DefaultsWithNullFallback(std::move(null_fallback))) {
}
explicit HivePartitioning(std::shared_ptr<Schema> schema, ArrayVector dictionaries,
HivePartitioningOptions options)
: KeyValuePartitioning(std::move(schema), std::move(dictionaries), options),
hive_options_(options) {}
std::string type_name() const override { return "hive"; }
std::string null_fallback() const { return hive_options_.null_fallback; }
const HivePartitioningOptions& options() const { return hive_options_; }
static Result<std::optional<Key>> ParseKey(const std::string& segment,
const HivePartitioningOptions& options);
bool Equals(const Partitioning& other) const override;
/// \brief Create a factory for a hive partitioning.
static std::shared_ptr<PartitioningFactory> MakeFactory(
HivePartitioningFactoryOptions = {});
private:
const HivePartitioningOptions hive_options_;
Result<std::vector<Key>> ParseKeys(const std::string& path) const override;
Result<PartitionPathFormat> FormatValues(const ScalarVector& values) const override;
};
/// \brief Implementation provided by lambda or other callable
class ARROW_DS_EXPORT FunctionPartitioning : public Partitioning {
public:
using ParseImpl = std::function<Result<compute::Expression>(const std::string&)>;
using FormatImpl =
std::function<Result<PartitionPathFormat>(const compute::Expression&)>;
FunctionPartitioning(std::shared_ptr<Schema> schema, ParseImpl parse_impl,
FormatImpl format_impl = NULLPTR, std::string name = "function")
: Partitioning(std::move(schema)),
parse_impl_(std::move(parse_impl)),
format_impl_(std::move(format_impl)),
name_(std::move(name)) {}
std::string type_name() const override { return name_; }
bool Equals(const Partitioning& other) const override { return false; }
Result<compute::Expression> Parse(const std::string& path) const override {
return parse_impl_(path);
}
Result<PartitionPathFormat> Format(const compute::Expression& expr) const override {
if (format_impl_) {
return format_impl_(expr);
}
return Status::NotImplemented("formatting paths from ", type_name(), " Partitioning");
}
Result<PartitionedBatches> Partition(
const std::shared_ptr<RecordBatch>& batch) const override {
return Status::NotImplemented("partitioning batches from ", type_name(),
" Partitioning");
}
private:
ParseImpl parse_impl_;
FormatImpl format_impl_;
std::string name_;
};
class ARROW_DS_EXPORT FilenamePartitioning : public KeyValuePartitioning {
public:
/// \brief Construct a FilenamePartitioning from its components.
///
/// If a field in schema is of dictionary type, the corresponding element of
/// dictionaries must be contain the dictionary of values for that field.
explicit FilenamePartitioning(std::shared_ptr<Schema> schema,
ArrayVector dictionaries = {},
KeyValuePartitioningOptions options = {});
std::string type_name() const override { return "filename"; }
/// \brief Create a factory for a filename partitioning.
///
/// \param[in] field_names The names for the partition fields. Types will be
/// inferred.
static std::shared_ptr<PartitioningFactory> MakeFactory(
std::vector<std::string> field_names, PartitioningFactoryOptions = {});
bool Equals(const Partitioning& other) const override;
private:
Result<std::vector<Key>> ParseKeys(const std::string& path) const override;
Result<PartitionPathFormat> FormatValues(const ScalarVector& values) const override;
};
ARROW_DS_EXPORT std::string StripPrefix(const std::string& path,
const std::string& prefix);
/// \brief Extracts the directory and filename and removes the prefix of a path
///
/// e.g., `StripPrefixAndFilename("/data/year=2019/c.txt", "/data") ->
/// {"year=2019","c.txt"}`
ARROW_DS_EXPORT std::string StripPrefixAndFilename(const std::string& path,
const std::string& prefix);
/// \brief Vector version of StripPrefixAndFilename.
ARROW_DS_EXPORT std::vector<std::string> StripPrefixAndFilename(
const std::vector<std::string>& paths, const std::string& prefix);
/// \brief Vector version of StripPrefixAndFilename.
ARROW_DS_EXPORT std::vector<std::string> StripPrefixAndFilename(
const std::vector<fs::FileInfo>& files, const std::string& prefix);
/// \brief Either a Partitioning or a PartitioningFactory
class ARROW_DS_EXPORT PartitioningOrFactory {
public:
explicit PartitioningOrFactory(std::shared_ptr<Partitioning> partitioning)
: partitioning_(std::move(partitioning)) {}
explicit PartitioningOrFactory(std::shared_ptr<PartitioningFactory> factory)
: factory_(std::move(factory)) {}
PartitioningOrFactory& operator=(std::shared_ptr<Partitioning> partitioning) {
return *this = PartitioningOrFactory(std::move(partitioning));
}
PartitioningOrFactory& operator=(std::shared_ptr<PartitioningFactory> factory) {
return *this = PartitioningOrFactory(std::move(factory));
}
/// \brief The partitioning (if given).
const std::shared_ptr<Partitioning>& partitioning() const { return partitioning_; }
/// \brief The partition factory (if given).
const std::shared_ptr<PartitioningFactory>& factory() const { return factory_; }
/// \brief Get the partition schema, inferring it with the given factory if needed.
Result<std::shared_ptr<Schema>> GetOrInferSchema(const std::vector<std::string>& paths);
private:
std::shared_ptr<PartitioningFactory> factory_;
std::shared_ptr<Partitioning> partitioning_;
};
/// @}
} // namespace dataset
} // namespace arrow
|