PeriodArray#

class pandas::PeriodArray#

Extension array type for specialized data storage.

Example#

#include <pandas/pandas.h>
using namespace pandas;

// Use PeriodArray
PeriodArray obj;
// ... operations ...

Constructors#

Signature

Location

Example

PeriodArray(const numpy::NDArray<numpy::int64>& data, const numpy::NDArray<numpy::bool_>& mask, const std::string& freq, bool copy = false)

pd_period_array.h:106

View

PeriodArray(const numpy::NDArray<numpy::int64>& data, const std::string& freq)

pd_period_array.h:122

View

PeriodArray(const std::vector<std::optional<numpy::int64>>& values, const std::string& freq)

pd_period_array.h:164

View

PeriodArray(const std::vector<std::string>& period_strings, const std::string& freq)

pd_period_array.h:184

View

Construction#

Signature

Return Type

Location

Example

static PeriodArray from_sequence(const std::vector<std::optional<numpy::int64>>& scalars, const std::string& freq)

static PeriodArray

pd_period_array.h:1059

Indexing / Selection#

Signature

Return Type

Location

Example

numpy::int64 at(size_t index) const

numpy::int64

pd_period_array.h:303

View

const numpy::NDArray<numpy::bool_>& mask() const

const numpy::NDArray<numpy::bool_>&

pd_period_array.h:283

View

Data Manipulation#

Signature

Return Type

Location

Example

PeriodArray dropna() const

PeriodArray

pd_period_array.h:396

View

Missing Data#

Signature

Return Type

Location

Example

PeriodArray fillna(numpy::int64 value) const

PeriodArray

pd_period_array.h:382

View

numpy::NDArray<numpy::bool_> isna() const

numpy::NDArray<numpy::bool_>

pd_period_array.h:332

View

numpy::NDArray<numpy::bool_> notna() const

numpy::NDArray<numpy::bool_>

pd_period_array.h:339

View

Statistics#

Signature

Return Type

Location

Example

size_t count() const

size_t

pd_period_array.h:357

View

std::optional<numpy::int64> max() const

std::optional<numpy::int64>

pd_period_array.h:984

View

std::optional<numpy::int64> min() const

std::optional<numpy::int64>

pd_period_array.h:975

View

IntegerArray<numpy::int32> minute() const

IntegerArray<numpy::int32>

pd_period_array.h:563

View

Comparison#

Signature

Return Type

Location

Example

size_t len() const

size_t

pd_period_array.h:251

View

Sorting#

Signature

Return Type

Location

Example

numpy::NDArray<size_t> argsort(bool ascending = true, const std::string& na_position = "last") const

numpy::NDArray<size_t>

pd_period_array.h:905

View

Combining#

Signature

Return Type

Location

Example

static PeriodArray concat(const std::vector<PeriodArray>& arrays)

static PeriodArray

pd_period_array.h:1067

View

Time Series#

Signature

Return Type

Location

Example

PeriodArray asfreq(const std::string& new_freq, const std::string& how = "end") const

PeriodArray

pd_period_array.h:675

View

DatetimeArray to_timestamp(const std::string& how = "start") const

DatetimeArray

pd_period_array.h:642

View

I/O#

Signature

Return Type

Location

Example

std::string to_string() const

std::string

pd_period_array.h:1104

View

Conversion#

Signature

Return Type

Location

Example

PeriodArray copy() const

PeriodArray

pd_period_array.h:350

View

Iteration#

Signature

Return Type

Location

Example

DatetimeArray end_time() const

DatetimeArray

pd_period_array.h:666

View

Set Operations#

Signature

Return Type

Location

Example

PeriodArray unique() const

PeriodArray

pd_period_array.h:997

View

Type Checking#

Signature

Return Type

Location

Example

BooleanArray is_leap_year() const

BooleanArray

pd_period_array.h:619

View

bool is_na(size_t index) const

bool

pd_period_array.h:314

View

Other Methods#

Signature

Return Type

Location

Example

std::optional<size_t> argmax() const

std::optional<size_t>

pd_period_array.h:956

View

std::optional<size_t> argmin() const

std::optional<size_t>

pd_period_array.h:937

View

const numpy::NDArray<numpy::int64>& data() const

const numpy::NDArray<numpy::int64>&

pd_period_array.h:276

View

IntegerArray<numpy::int32> day() const

IntegerArray<numpy::int32>

pd_period_array.h:488

View

IntegerArray<numpy::int32> dayofweek() const

IntegerArray<numpy::int32>

pd_period_array.h:505

View

IntegerArray<numpy::int32> dayofyear() const

IntegerArray<numpy::int32>

pd_period_array.h:529

View

IntegerArray<numpy::int32> days_in_month() const

IntegerArray<numpy::int32>

pd_period_array.h:597

View

days_since_epoch_to_date(days, year, month, day)

pd_period_array.h:697

dtype_type dtype() const

dtype_type

pd_period_array.h:209

View

bool empty() const

bool

pd_period_array.h:244

View

std::pair<numpy::NDArray<numpy::int64>, PeriodArray> factorize() const

std::pair<numpy::NDArray<numpy::int64>, PeriodArray>

pd_period_array.h:1023

View

PeriodFrequency freq() const

PeriodFrequency

pd_period_array.h:258

View

std::string freqstr() const

std::string

pd_period_array.h:265

View

bool has_na() const

bool

pd_period_array.h:370

View

IntegerArray<numpy::int32> hour() const

IntegerArray<numpy::int32>

pd_period_array.h:546

View

IntegerArray<numpy::int32> month() const

IntegerArray<numpy::int32>

pd_period_array.h:447

View

size_t nbytes() const

size_t

pd_period_array.h:223

View

constexpr int ndim() const

constexpr int

pd_period_array.h:230

View

numpy::int64 ordinal(size_t index) const

numpy::int64

pd_period_array.h:321

View

ordinal_to_day(data_.getElementAt({i}), freq_)))

pd_period_array.h:496

ordinal_to_dayofweek(data_.getElementAt({i}), freq_)))

pd_period_array.h:513

ordinal_to_dayofyear(data_.getElementAt({i}), freq_)))

pd_period_array.h:537

ordinal_to_hour(data_.getElementAt({i}), freq_)))

pd_period_array.h:554

ordinal_to_minute(data_.getElementAt({i}), freq_)))

pd_period_array.h:571

ordinal_to_month(data_.getElementAt({i}), freq_)))

pd_period_array.h:455

ordinal_to_quarter(data_.getElementAt({i}), freq_)))

pd_period_array.h:438

ordinal_to_second(data_.getElementAt({i}), freq_)))

pd_period_array.h:588

ordinal_to_week(data_.getElementAt({i}), freq_)))

pd_period_array.h:472

ordinal_to_year(data_.getElementAt({i}), freq_)))

pd_period_array.h:421

IntegerArray<numpy::int32> quarter() const

IntegerArray<numpy::int32>

pd_period_array.h:430

View

std::string repr() const

std::string

pd_period_array.h:1126

View

IntegerArray<numpy::int32> second() const

IntegerArray<numpy::int32>

pd_period_array.h:580

View

std::vector<size_t> shape() const

std::vector<size_t>

pd_period_array.h:237

View

size_t size() const

size_t

pd_period_array.h:216

View

DatetimeArray start_time() const

DatetimeArray

pd_period_array.h:659

View

void validate_arrays()

void

pd_period_array.h:66

IntegerArray<numpy::int32> week() const

IntegerArray<numpy::int32>

pd_period_array.h:464

View

IntegerArray<numpy::int32> weekday() const

IntegerArray<numpy::int32>

pd_period_array.h:522

View

IntegerArray<numpy::int32> weekofyear() const

IntegerArray<numpy::int32>

pd_period_array.h:481

IntegerArray<numpy::int32> year() const

IntegerArray<numpy::int32>

pd_period_array.h:413

View

Code Examples#

The following examples are extracted from the test suite.

PeriodArray (pd_test_1_all.cpp:2927)
2917        }
2918
2919        // Subtract integer
2920        auto unshifted = shifted - 3;
2921        auto un0 = unshifted[0];
2922        if (!un0.has_value() || un0.value() != orig0.value()) {
2923            std::cout << "  [FAIL] : unshifted should equal original" << std::endl;
2924            throw std::runtime_error("pd_test_period_array_arithmetic failed: unshift");
2925        }
2926
2927        // Subtract PeriodArray (get differences)
2928        pandas::PeriodArray arr2(std::vector<std::string>{
2929            "2024-04",
2930            "2024-03",
2931            "2024-01"
2932        }, "M");
2933
2934        auto diff = arr2 - arr;
2935        auto d0 = diff[0];
2936        if (!d0.has_value() || d0.value() != 3) {
2937            std::cout << "  [FAIL] : diff[0] should be 3 (Apr - Jan)" << std::endl;
PeriodArray (pd_test_1_all.cpp:2927)
2917        }
2918
2919        // Subtract integer
2920        auto unshifted = shifted - 3;
2921        auto un0 = unshifted[0];
2922        if (!un0.has_value() || un0.value() != orig0.value()) {
2923            std::cout << "  [FAIL] : unshifted should equal original" << std::endl;
2924            throw std::runtime_error("pd_test_period_array_arithmetic failed: unshift");
2925        }
2926
2927        // Subtract PeriodArray (get differences)
2928        pandas::PeriodArray arr2(std::vector<std::string>{
2929            "2024-04",
2930            "2024-03",
2931            "2024-01"
2932        }, "M");
2933
2934        auto diff = arr2 - arr;
2935        auto d0 = diff[0];
2936        if (!d0.has_value() || d0.value() != 3) {
2937            std::cout << "  [FAIL] : diff[0] should be 3 (Apr - Jan)" << std::endl;
PeriodArray (pd_test_1_all.cpp:2927)
2917        }
2918
2919        // Subtract integer
2920        auto unshifted = shifted - 3;
2921        auto un0 = unshifted[0];
2922        if (!un0.has_value() || un0.value() != orig0.value()) {
2923            std::cout << "  [FAIL] : unshifted should equal original" << std::endl;
2924            throw std::runtime_error("pd_test_period_array_arithmetic failed: unshift");
2925        }
2926
2927        // Subtract PeriodArray (get differences)
2928        pandas::PeriodArray arr2(std::vector<std::string>{
2929            "2024-04",
2930            "2024-03",
2931            "2024-01"
2932        }, "M");
2933
2934        auto diff = arr2 - arr;
2935        auto d0 = diff[0];
2936        if (!d0.has_value() || d0.value() != 3) {
2937            std::cout << "  [FAIL] : diff[0] should be 3 (Apr - Jan)" << std::endl;
PeriodArray (pd_test_1_all.cpp:2927)
2917        }
2918
2919        // Subtract integer
2920        auto unshifted = shifted - 3;
2921        auto un0 = unshifted[0];
2922        if (!un0.has_value() || un0.value() != orig0.value()) {
2923            std::cout << "  [FAIL] : unshifted should equal original" << std::endl;
2924            throw std::runtime_error("pd_test_period_array_arithmetic failed: unshift");
2925        }
2926
2927        // Subtract PeriodArray (get differences)
2928        pandas::PeriodArray arr2(std::vector<std::string>{
2929            "2024-04",
2930            "2024-03",
2931            "2024-01"
2932        }, "M");
2933
2934        auto diff = arr2 - arr;
2935        auto d0 = diff[0];
2936        if (!d0.has_value() || d0.value() != 3) {
2937            std::cout << "  [FAIL] : diff[0] should be 3 (Apr - Jan)" << std::endl;
at (pd_test_1_all.cpp:6581)
6571            // Test isna/notna with float data
6572            {
6573                std::map<std::string, std::vector<numpy::float64>> float_data;
6574                float_data["X"] = {1.0, std::nan(""), 3.0};
6575                float_data["Y"] = {4.0, 5.0, std::nan("")};
6576                pandas::DataFrame df_na(float_data);
6577
6578                auto na_mask = df_na.isna();
6579                // Row 1, col 0 (X) should be NA
6580                if (!na_mask.getElementAt({1, 0})) {
6581                    std::cout << "  [FAIL] : in pd_test_dataframe_manipulation() : isna at (1,0) should be true" << std::endl;
6582                    throw std::runtime_error("pd_test_dataframe_manipulation failed: isna at (1,0)");
6583                }
6584                // Row 2, col 1 (Y) should be NA
6585                if (!na_mask.getElementAt({2, 1})) {
6586                    std::cout << "  [FAIL] : in pd_test_dataframe_manipulation() : isna at (2,1) should be true" << std::endl;
6587                    throw std::runtime_error("pd_test_dataframe_manipulation failed: isna at (2,1)");
6588                }
6589                // Row 0, col 0 should NOT be NA
6590                if (na_mask.getElementAt({0, 0})) {
6591                    std::cout << "  [FAIL] : in pd_test_dataframe_manipulation() : isna at (0,0) should be false" << std::endl;
mask (pd_test_1_all.cpp:9119)
9109void pd_test_datetime_mixin_array_constructor() {
9110    std::cout << "========= DatetimeTDMixin array constructor =========================";
9111
9112    // Create DatetimeArray with some values
9113    numpy::NDArray<numpy::datetime64> data(std::vector<size_t>{3});
9114    data.setElementAt({0}, numpy::datetime64(1000000000000000000LL, numpy::DateTimeUnit::Nanosecond));  // ~2001
9115    data.setElementAt({1}, numpy::datetime64(1500000000000000000LL, numpy::DateTimeUnit::Nanosecond));  // ~2017
9116    data.setElementAt({2}, numpy::datetime64(1600000000000000000LL, numpy::DateTimeUnit::Nanosecond));  // ~2020
9117
9118    numpy::NDArray<numpy::bool_> mask(std::vector<size_t>{3});
9119    mask.setElementAt({0}, numpy::bool_(false));
9120    mask.setElementAt({1}, numpy::bool_(false));
9121    mask.setElementAt({2}, numpy::bool_(false));
9122
9123    pandas::DatetimeArray arr(data, mask);
9124    pandas::DatetimeTDMixin idx(arr, "timestamps");
9125
9126    bool passed = (idx.size() == 3 && !idx.empty() &&
9127                   idx.name().has_value() && *idx.name() == "timestamps" &&
9128                   idx.inferred_type() == "datetime");
dropna (pd_test_1_all.cpp:531)
521        }
522
523        // Test isna array
524        numpy::NDArray<numpy::bool_> na_mask = arr.isna();
525        if (na_mask.getSize() != 4) {
526            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : isna size != 4" << std::endl;
527            throw std::runtime_error("pd_test_categorical_array_na_handling failed: isna size != 4");
528        }
529
530        // Test dropna
531        pandas::CategoricalArray dropped = arr.dropna();
532        if (dropped.size() != 2) {
533            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : dropna size != 2" << std::endl;
534            throw std::runtime_error("pd_test_categorical_array_na_handling failed: dropna size != 2");
535        }
536
537        // Test fillna (fill with existing category)
538        pandas::CategoricalArray filled = arr.fillna("a");  // 'a' is in categories
539        if (filled.has_na()) {
540            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : fillna should have no NA" << std::endl;
541            throw std::runtime_error("pd_test_categorical_array_na_handling failed: fillna should have no NA");
fillna (pd_test_1_all.cpp:537)
527            throw std::runtime_error("pd_test_categorical_array_na_handling failed: isna size != 4");
528        }
529
530        // Test dropna
531        pandas::CategoricalArray dropped = arr.dropna();
532        if (dropped.size() != 2) {
533            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : dropna size != 2" << std::endl;
534            throw std::runtime_error("pd_test_categorical_array_na_handling failed: dropna size != 2");
535        }
536
537        // Test fillna (fill with existing category)
538        pandas::CategoricalArray filled = arr.fillna("a");  // 'a' is in categories
539        if (filled.has_na()) {
540            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : fillna should have no NA" << std::endl;
541            throw std::runtime_error("pd_test_categorical_array_na_handling failed: fillna should have no NA");
542        }
543
544        std::cout << " -> tests passed" << std::endl;
545    }
546
547    void pd_test_categorical_array_add_categories() {
isna (pd_test_1_all.cpp:524)
514            throw std::runtime_error("pd_test_categorical_array_na_handling failed: has_na() should be true");
515        }
516
517        // Test count (non-NA)
518        if (arr.count() != 2) {
519            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : count() != 2" << std::endl;
520            throw std::runtime_error("pd_test_categorical_array_na_handling failed: count() != 2");
521        }
522
523        // Test isna array
524        numpy::NDArray<numpy::bool_> na_mask = arr.isna();
525        if (na_mask.getSize() != 4) {
526            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : isna size != 4" << std::endl;
527            throw std::runtime_error("pd_test_categorical_array_na_handling failed: isna size != 4");
528        }
529
530        // Test dropna
531        pandas::CategoricalArray dropped = arr.dropna();
532        if (dropped.size() != 2) {
533            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : dropna size != 2" << std::endl;
534            throw std::runtime_error("pd_test_categorical_array_na_handling failed: dropna size != 2");
notna (pd_test_1_all.cpp:6595)
6585                if (!na_mask.getElementAt({2, 1})) {
6586                    std::cout << "  [FAIL] : in pd_test_dataframe_manipulation() : isna at (2,1) should be true" << std::endl;
6587                    throw std::runtime_error("pd_test_dataframe_manipulation failed: isna at (2,1)");
6588                }
6589                // Row 0, col 0 should NOT be NA
6590                if (na_mask.getElementAt({0, 0})) {
6591                    std::cout << "  [FAIL] : in pd_test_dataframe_manipulation() : isna at (0,0) should be false" << std::endl;
6592                    throw std::runtime_error("pd_test_dataframe_manipulation failed: isna at (0,0)");
6593                }
6594
6595                auto notna_mask = df_na.notna();
6596                if (notna_mask.getElementAt({1, 0})) {
6597                    std::cout << "  [FAIL] : in pd_test_dataframe_manipulation() : notna at (1,0) should be false" << std::endl;
6598                    throw std::runtime_error("pd_test_dataframe_manipulation failed: notna at (1,0)");
6599                }
6600            }
6601
6602            // Test fillna
6603            {
6604                std::map<std::string, std::vector<numpy::float64>> float_data;
6605                float_data["X"] = {1.0, std::nan(""), 3.0};
count (pd_test_1_all.cpp:66)
56        if (arr.is_na(0)) {
57            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : is_na(0) should be false" << std::endl;
58            throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(0) should be false");
59        }
60
61        if (!arr.has_na()) {
62            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : has_na() should be true" << std::endl;
63            throw std::runtime_error("pd_test_boolean_array_na_handling failed: has_na() should be true");
64        }
65
66        if (arr.count() != 2) {
67            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : count() should be 2" << std::endl;
68            throw std::runtime_error("pd_test_boolean_array_na_handling failed: count() should be 2");
69        }
70
71        std::cout << " -> tests passed" << std::endl;
72    }
73
74    void pd_test_boolean_array_kleene_and() {
75        std::cout << "========= BooleanArray: Kleene AND ======================= ";
max (pd_test_1_all.cpp:771)
761        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats, true);  // ordered
762
763        // Test min
764        std::optional<std::string> min_val = arr.min();
765        if (!min_val.has_value() || *min_val != "low") {
766            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : min != 'low'" << std::endl;
767            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: min != 'low'");
768        }
769
770        // Test max
771        std::optional<std::string> max_val = arr.max();
772        if (!max_val.has_value() || *max_val != "high") {
773            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : max != 'high'" << std::endl;
774            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: max != 'high'");
775        }
776
777        // Test unordered throws for min/max
778        pandas::CategoricalArray unordered = arr.as_unordered();
779        bool threw = false;
780        try {
781            unordered.min();
min (pd_test_1_all.cpp:764)
754    }
755
756    void pd_test_categorical_array_ordered_operations() {
757        std::cout << "========= CategoricalArray: ordered operations (min/max) ======================= ";
758
759        std::vector<std::string> cats = {"low", "medium", "high"};
760        std::vector<numpy::int32> codes = {0, 2, 1, 0, -1};  // low, high, medium, low, NA
761        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats, true);  // ordered
762
763        // Test min
764        std::optional<std::string> min_val = arr.min();
765        if (!min_val.has_value() || *min_val != "low") {
766            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : min != 'low'" << std::endl;
767            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: min != 'low'");
768        }
769
770        // Test max
771        std::optional<std::string> max_val = arr.max();
772        if (!max_val.has_value() || *max_val != "high") {
773            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : max != 'high'" << std::endl;
774            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: max != 'high'");
minute (pd_test_1_all.cpp:7505)
7495    std::cout << "========= minute property =============================";
7496
7497    std::vector<std::optional<numpy::datetime64>> values = {
7498        make_dt(0),                    // Minute 0
7499        make_dt(30 * NS_PER_MIN),      // Minute 30
7500        make_dt(59 * NS_PER_MIN)       // Minute 59
7501    };
7502    pandas::DatetimeArray arr(values);
7503    pandas::DatetimeIndex idx(arr);
7504
7505    auto minutes = idx.minute();
7506
7507    bool passed = (minutes.size() == 3);
7508    auto m0 = minutes[0];
7509    auto m1 = minutes[1];
7510    auto m2 = minutes[2];
7511    passed = passed && m0.has_value() && *m0 == 0;
7512    passed = passed && m1.has_value() && *m1 == 30;
7513    passed = passed && m2.has_value() && *m2 == 59;
7514
7515    if (!passed) {
len (pd_test_3_all.cpp:20867)
20857    auto title_result = s.str().title();
20858    if (title_result[0] != "Hello World" || title_result[1] != "Hello World" || title_result[2] != "Hello World") {
20859        std::cout << "  [FAIL] : title() failed" << std::endl;
20860        throw std::runtime_error("pd_test_str_capitalize_title: title() failed");
20861    }
20862
20863    std::cout << " -> tests passed" << std::endl;
20864}
20865
20866// ============================================================================
20867// Test str().len()
20868// ============================================================================
20869
20870void pd_test_str_len() {
20871    std::cout << "========= Series.str().len() ============================";
20872
20873    pandas::Series<std::string> s({"a", "bb", "ccc", ""});
20874
20875    auto lens = s.str().len();
20876    if (lens[0] != 1 || lens[1] != 2 || lens[2] != 3 || lens[3] != 0) {
20877        std::cout << "  [FAIL] : len() failed" << std::endl;
argsort (pd_test_1_all.cpp:1304)
1294        std::cout << "========= DatetimeArray: sorting ======================= ";
1295
1296        pandas::DatetimeArray arr(std::vector<std::string>{
1297            "2023-06-15",
1298            "NaT",
1299            "2023-01-01",
1300            "2023-12-31"
1301        });
1302
1303        // argsort ascending
1304        auto indices = arr.argsort(true, "last");
1305        // Expected order: 2023-01-01(2), 2023-06-15(0), 2023-12-31(3), NaT(1)
1306        if (indices.getElementAt({0}) != 2) {
1307            std::cout << "  [FAIL] : argsort: first should be index 2 (2023-01-01)" << std::endl;
1308            throw std::runtime_error("pd_test_datetime_array_sorting failed: argsort first");
1309        }
1310        if (indices.getElementAt({3}) != 1) {
1311            std::cout << "  [FAIL] : argsort: last should be index 1 (NaT)" << std::endl;
1312            throw std::runtime_error("pd_test_datetime_array_sorting failed: NaT position");
1313        }
concat (pd_test_1_all.cpp:17717)
17707}
17708
17709void pd_test_period_index_concat() {
17710    std::cout << "========= concat factory ==============================";
17711
17712    std::vector<int64_t> ordinals1 = {0, 1};
17713    std::vector<int64_t> ordinals2 = {2, 3};
17714    pandas::PeriodIndex idx1(ordinals1, "D");
17715    pandas::PeriodIndex idx2(ordinals2, "D");
17716
17717    pandas::PeriodIndex concatenated = pandas::PeriodIndex::concat({idx1, idx2});
17718
17719    bool passed = (concatenated.size() == 4);
17720    if (!passed) {
17721        std::cout << "  [FAIL] : in pd_test_period_index_concat()" << std::endl;
17722        throw std::runtime_error("pd_test_period_index_concat failed");
17723    }
17724
17725    std::cout << " -> tests passed" << std::endl;
17726}
asfreq (pd_test_1_all.cpp:2869)
2859        std::cout << "========= PeriodArray: asfreq ======================= ";
2860
2861        // Monthly to quarterly
2862        pandas::PeriodArray arr_m(std::vector<std::string>{
2863            "2024-01",
2864            "2024-04",
2865            "2024-07",
2866            "NaT"
2867        }, "M");
2868
2869        auto arr_q = arr_m.asfreq("Q");
2870        if (arr_q.size() != 4) {
2871            std::cout << "  [FAIL] : asfreq size should be 4" << std::endl;
2872            throw std::runtime_error("pd_test_period_array_asfreq failed: size");
2873        }
2874        if (arr_q.freqstr() != "Q") {
2875            std::cout << "  [FAIL] : asfreq freqstr should be 'Q'" << std::endl;
2876            throw std::runtime_error("pd_test_period_array_asfreq failed: freqstr");
2877        }
2878
2879        // Check NaT is preserved
to_timestamp (pd_test_1_all.cpp:2830)
2820    void pd_test_period_array_to_timestamp() {
2821        std::cout << "========= PeriodArray: to_timestamp ======================= ";
2822
2823        pandas::PeriodArray arr(std::vector<std::string>{
2824            "2024-01",
2825            "2024-06",
2826            "NaT"
2827        }, "M");
2828
2829        // to_timestamp with start
2830        auto ts_start = arr.to_timestamp("start");
2831        if (ts_start.size() != 3) {
2832            std::cout << "  [FAIL] : to_timestamp size should be 3" << std::endl;
2833            throw std::runtime_error("pd_test_period_array_to_timestamp failed: size");
2834        }
2835
2836        auto ts0 = ts_start[0];
2837        if (!ts0.has_value()) {
2838            std::cout << "  [FAIL] : ts_start[0] should have value" << std::endl;
2839            throw std::runtime_error("pd_test_period_array_to_timestamp failed: ts_start[0]");
2840        }
to_string (pd_test_1_all.cpp:2693)
2683        pandas::PeriodArray arr_m(std::vector<std::string>{
2684            "2020-01",
2685            "NaT",
2686            "2025-06"
2687        }, "M");
2688
2689        // Year
2690        auto years = arr_m.year();
2691        auto y0 = years[0];
2692        if (!y0.has_value() || y0.value() != 2020) {
2693            std::cout << "  [FAIL] : year[0] should be 2020, got " << (y0.has_value() ? std::to_string(y0.value()) : "NA") << std::endl;
2694            throw std::runtime_error("pd_test_period_array_year_month_quarter failed: year[0]");
2695        }
2696
2697        auto y1 = years[1];
2698        if (y1.has_value()) {
2699            std::cout << "  [FAIL] : year[1] should be NA (NaT)" << std::endl;
2700            throw std::runtime_error("pd_test_period_array_year_month_quarter failed: year[1] should be NA");
2701        }
2702
2703        auto y2 = years[2];
copy (pd_test_1_all.cpp:5798)
5788// ============================================================================
5789// Copy/Rename Tests
5790// ============================================================================
5791
5792void pd_test_categorical_index_copy() {
5793    std::cout << "========= copy ========================================";
5794
5795    pandas::CategoricalArray arr({"a", "b", "c"});
5796    pandas::CategoricalIndex idx(arr, "original");
5797
5798    pandas::CategoricalIndex copied = idx.copy();
5799
5800    bool passed = (copied.size() == idx.size() && copied.name() == idx.name() &&
5801                   copied.categories() == idx.categories() && copied.ordered() == idx.ordered());
5802    if (!passed) {
5803        std::cout << "  [FAIL] : in pd_test_categorical_index_copy()" << std::endl;
5804        throw std::runtime_error("pd_test_categorical_index_copy failed");
5805    }
5806
5807    std::cout << " -> tests passed" << std::endl;
5808}
end_time (pd_test_1_all.cpp:17146)
17136    std::cout << " -> tests passed" << std::endl;
17137}
17138
17139void pd_test_period_index_end_time() {
17140    std::cout << "========= end_time property ===========================";
17141
17142    std::vector<int64_t> ordinals = {0};  // 1970-01-01 daily period
17143    pandas::PeriodIndex idx = pandas::PeriodIndex::from_ordinals(ordinals, "D");
17144
17145    pandas::DatetimeArray end_times = idx.end_time();
17146
17147    bool passed = (end_times.size() == 1 && !end_times.is_na(0));
17148    if (!passed) {
17149        std::cout << "  [FAIL] : in pd_test_period_index_end_time()" << std::endl;
17150        throw std::runtime_error("pd_test_period_index_end_time failed");
17151    }
17152
17153    std::cout << " -> tests passed" << std::endl;
17154}
unique (pd_test_1_all.cpp:1345)
1335        pandas::DatetimeArray arr(std::vector<std::string>{
1336            "2023-01-01",
1337            "2023-06-15",
1338            "2023-01-01",
1339            "NaT",
1340            "2023-06-15",
1341            "NaT"
1342        });
1343
1344        // unique
1345        auto uniq = arr.unique();
1346        // Should have: NaT, 2023-01-01, 2023-06-15 (3 unique values)
1347        if (uniq.size() != 3) {
1348            std::cout << "  [FAIL] : unique size should be 3, got " << uniq.size() << std::endl;
1349            throw std::runtime_error("pd_test_datetime_array_unique failed: size");
1350        }
1351
1352        // factorize
1353        auto [codes, uniques] = arr.factorize();
1354        // Codes for NaT should be -1
1355        if (codes.getElementAt({3}) != -1) {
is_leap_year (pd_test_1_all.cpp:1280)
1270        }
1271
1272        // is_month_end
1273        auto me = arr.is_month_end();
1274        if (!me[1].has_value() || !me[1].value()) {
1275            std::cout << "  [FAIL] : 2023-03-31 should be month end" << std::endl;
1276            throw std::runtime_error("pd_test_datetime_array_boolean_props failed: month end");
1277        }
1278
1279        // is_leap_year
1280        auto ly = arr.is_leap_year();
1281        if (!ly[2].has_value() || !ly[2].value()) {
1282            std::cout << "  [FAIL] : 2024 should be leap year" << std::endl;
1283            throw std::runtime_error("pd_test_datetime_array_boolean_props failed: leap year");
1284        }
1285        if (!ly[0].has_value() || ly[0].value()) {
1286            std::cout << "  [FAIL] : 2023 should not be leap year" << std::endl;
1287            throw std::runtime_error("pd_test_datetime_array_boolean_props failed: not leap year");
1288        }
1289
1290        std::cout << " -> tests passed" << std::endl;
is_na (pd_test_1_all.cpp:51)
41    void pd_test_boolean_array_na_handling() {
42        std::cout << "========= BooleanArray: NA handling ======================= ";
43
44        pandas::BooleanArray arr({
45            std::optional<bool>(true),
46            std::nullopt,  // NA at index 1
47            std::optional<bool>(false)
48        });
49
50        if (!arr.is_na(1)) {
51            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : is_na(1) should be true" << std::endl;
52            throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(1) should be true");
53        }
54
55        if (arr.is_na(0)) {
56            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : is_na(0) should be false" << std::endl;
57            throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(0) should be false");
58        }
59
60        if (!arr.has_na()) {
argmax (pd_test_1_all.cpp:1323)
1313        }
1314
1315        // argmin
1316        auto min_idx = arr.argmin();
1317        if (!min_idx.has_value() || min_idx.value() != 2) {
1318            std::cout << "  [FAIL] : argmin should be 2 (2023-01-01)" << std::endl;
1319            throw std::runtime_error("pd_test_datetime_array_sorting failed: argmin");
1320        }
1321
1322        // argmax
1323        auto max_idx = arr.argmax();
1324        if (!max_idx.has_value() || max_idx.value() != 3) {
1325            std::cout << "  [FAIL] : argmax should be 3 (2023-12-31)" << std::endl;
1326            throw std::runtime_error("pd_test_datetime_array_sorting failed: argmax");
1327        }
1328
1329        std::cout << " -> tests passed" << std::endl;
1330    }
1331
1332    void pd_test_datetime_array_unique() {
1333        std::cout << "========= DatetimeArray: unique/factorize ======================= ";
argmin (pd_test_1_all.cpp:1316)
1306        if (indices.getElementAt({0}) != 2) {
1307            std::cout << "  [FAIL] : argsort: first should be index 2 (2023-01-01)" << std::endl;
1308            throw std::runtime_error("pd_test_datetime_array_sorting failed: argsort first");
1309        }
1310        if (indices.getElementAt({3}) != 1) {
1311            std::cout << "  [FAIL] : argsort: last should be index 1 (NaT)" << std::endl;
1312            throw std::runtime_error("pd_test_datetime_array_sorting failed: NaT position");
1313        }
1314
1315        // argmin
1316        auto min_idx = arr.argmin();
1317        if (!min_idx.has_value() || min_idx.value() != 2) {
1318            std::cout << "  [FAIL] : argmin should be 2 (2023-01-01)" << std::endl;
1319            throw std::runtime_error("pd_test_datetime_array_sorting failed: argmin");
1320        }
1321
1322        // argmax
1323        auto max_idx = arr.argmax();
1324        if (!max_idx.has_value() || max_idx.value() != 3) {
1325            std::cout << "  [FAIL] : argmax should be 3 (2023-12-31)" << std::endl;
1326            throw std::runtime_error("pd_test_datetime_array_sorting failed: argmax");
data (pd_test_1_all.cpp:9114)
9104        throw std::runtime_error("pd_test_datetime_mixin_default_constructor failed");
9105    }
9106
9107    std::cout << " -> tests passed" << std::endl;
9108}
9109
9110void pd_test_datetime_mixin_array_constructor() {
9111    std::cout << "========= DatetimeTDMixin array constructor =========================";
9112
9113    // Create DatetimeArray with some values
9114    numpy::NDArray<numpy::datetime64> data(std::vector<size_t>{3});
9115    data.setElementAt({0}, numpy::datetime64(1000000000000000000LL, numpy::DateTimeUnit::Nanosecond));  // ~2001
9116    data.setElementAt({1}, numpy::datetime64(1500000000000000000LL, numpy::DateTimeUnit::Nanosecond));  // ~2017
9117    data.setElementAt({2}, numpy::datetime64(1600000000000000000LL, numpy::DateTimeUnit::Nanosecond));  // ~2020
9118
9119    numpy::NDArray<numpy::bool_> mask(std::vector<size_t>{3});
9120    mask.setElementAt({0}, numpy::bool_(false));
9121    mask.setElementAt({1}, numpy::bool_(false));
9122    mask.setElementAt({2}, numpy::bool_(false));
9123
9124    pandas::DatetimeArray arr(data, mask);
day (pd_test_1_all.cpp:1193)
1183            std::cout << "  [FAIL] : month[0] should be 3" << std::endl;
1184            throw std::runtime_error("pd_test_datetime_array_component_month_day failed: month[0]");
1185        }
1186        auto m1 = months[1];
1187        if (!m1.has_value() || m1.value() != 12) {
1188            std::cout << "  [FAIL] : month[1] should be 12" << std::endl;
1189            throw std::runtime_error("pd_test_datetime_array_component_month_day failed: month[1]");
1190        }
1191
1192        // Day
1193        auto days = arr.day();
1194        auto d0 = days[0];
1195        if (!d0.has_value() || d0.value() != 15) {
1196            std::cout << "  [FAIL] : day[0] should be 15" << std::endl;
1197            throw std::runtime_error("pd_test_datetime_array_component_month_day failed: day[0]");
1198        }
1199        auto d1 = days[1];
1200        if (!d1.has_value() || d1.value() != 25) {
1201            std::cout << "  [FAIL] : day[1] should be 25" << std::endl;
1202            throw std::runtime_error("pd_test_datetime_array_component_month_day failed: day[1]");
1203        }
dayofweek (pd_test_1_all.cpp:7565)
7555    // 1970-01-01 was a Thursday (day 3)
7556    std::vector<std::optional<numpy::datetime64>> values = {
7557        make_dt(0),                // Thursday (3)
7558        make_dt(NS_PER_DAY),       // Friday (4)
7559        make_dt(2 * NS_PER_DAY),   // Saturday (5)
7560        make_dt(3 * NS_PER_DAY)    // Sunday (6)
7561    };
7562    pandas::DatetimeArray arr(values);
7563    pandas::DatetimeIndex idx(arr);
7564
7565    auto dow = idx.dayofweek();
7566
7567    bool passed = (dow.size() == 4);
7568    if (!passed) {
7569        std::cout << "  [FAIL] : in pd_test_datetime_index_dayofweek()" << std::endl;
7570        throw std::runtime_error("pd_test_datetime_index_dayofweek failed");
7571    }
7572
7573    std::cout << " -> tests passed" << std::endl;
7574}
dayofyear (pd_test_3_all.cpp:18582)
18572    auto seconds = s.dt().second();
18573    if (seconds[0] != 45 || seconds[1] != 30 || seconds[2] != 59) {
18574        std::cout << "  [FAIL] : second() failed" << std::endl;
18575        throw std::runtime_error("pd_test_dt_time_components: second() failed");
18576    }
18577
18578    std::cout << " -> tests passed" << std::endl;
18579}
18580
18581// ============================================================================
18582// Test dt().dayofweek(), dt().dayofyear(), dt().quarter()
18583// ============================================================================
18584
18585void pd_test_dt_derived_properties() {
18586    std::cout << "========= Series.dt().dayofweek/dayofyear/quarter() ======";
18587
18588    // 2020-01-01 is a Wednesday (dayofweek=2), dayofyear=1, Q1
18589    // 2020-07-04 is a Saturday (dayofweek=5), dayofyear=186, Q3
18590    pandas::Series<std::string> s({"2020-01-01", "2020-07-04"});
18591
18592    auto dow = s.dt().dayofweek();
days_in_month (pd_test_1_all.cpp:2766)
2756            std::cout << "  [FAIL] : day[0] should be 15" << std::endl;
2757            throw std::runtime_error("pd_test_period_array_day_components failed: day[0]");
2758        }
2759        auto d1 = days[1];
2760        if (!d1.has_value() || d1.value() != 25) {
2761            std::cout << "  [FAIL] : day[1] should be 25" << std::endl;
2762            throw std::runtime_error("pd_test_period_array_day_components failed: day[1]");
2763        }
2764
2765        // Days in month
2766        auto dim = arr.days_in_month();
2767        auto dim0 = dim[0];
2768        if (!dim0.has_value() || dim0.value() != 31) {
2769            std::cout << "  [FAIL] : days_in_month[0] should be 31 (March)" << std::endl;
2770            throw std::runtime_error("pd_test_period_array_day_components failed: days_in_month[0]");
2771        }
2772        auto dim1 = dim[1];
2773        if (!dim1.has_value() || dim1.value() != 31) {
2774            std::cout << "  [FAIL] : days_in_month[1] should be 31 (December)" << std::endl;
2775            throw std::runtime_error("pd_test_period_array_day_components failed: days_in_month[1]");
2776        }
dtype (pd_test_1_all.cpp:295)
285            throw std::runtime_error("pd_test_boolean_array_reductions failed: mean");
286        }
287
288        std::cout << " -> tests passed" << std::endl;
289    }
290
291    void pd_test_boolean_array_dtype() {
292        std::cout << "========= BooleanArray: dtype ======================= ";
293
294        pandas::BooleanArray arr;
295        if (arr.dtype().name() != "boolean") {
296            std::cout << "  [FAIL] : in pd_test_boolean_array_dtype() : dtype name should be 'boolean'" << std::endl;
297            throw std::runtime_error("pd_test_boolean_array_dtype failed: dtype name");
298        }
299
300        if (arr.dtype().kind() != "b") {
301            std::cout << "  [FAIL] : in pd_test_boolean_array_dtype() : dtype kind should be 'b'" << std::endl;
302            throw std::runtime_error("pd_test_boolean_array_dtype failed: dtype kind");
303        }
304
305        std::cout << " -> tests passed" << std::endl;
empty (pd_test_1_all.cpp:941)
931#include "../pandas/pd_config.h"
932
933namespace dataframe_tests {
934
935namespace dataframe_tests_config {
936
937    void pd_test_config_version() {
938        std::cout << "========= df_config: version info ======================= ";
939        const char* version = pandas::DataFrameInfo::version();
940        if (version == nullptr || std::string(version).empty()) {
941            std::cout << "[FAIL] : in pd_test_config_version() : version is null or empty" << std::endl;
942            throw std::runtime_error("pd_test_config_version failed: version is null or empty");
943        }
944        std::cout << "-> tests passed" << std::endl;
945    }
946
947    void pd_test_config_na_repr() {
948        std::cout << "========= df_config: NA representation ======================= ";
949        const char* na_repr = pandas::DataFrameConfig::get_na_repr();
950        if (na_repr == nullptr) {
factorize (pd_test_1_all.cpp:1353)
1343        // unique
1344        auto uniq = arr.unique();
1345        // Should have: NaT, 2023-01-01, 2023-06-15 (3 unique values)
1346        if (uniq.size() != 3) {
1347            std::cout << "  [FAIL] : unique size should be 3, got " << uniq.size() << std::endl;
1348            throw std::runtime_error("pd_test_datetime_array_unique failed: size");
1349        }
1350
1351        // factorize
1352        auto [codes, uniques] = arr.factorize();
1353        // Codes for NaT should be -1
1354        if (codes.getElementAt({3}) != -1) {
1355            std::cout << "  [FAIL] : factorize: NaT code should be -1" << std::endl;
1356            throw std::runtime_error("pd_test_datetime_array_unique failed: NaT code");
1357        }
1358        // Same values should have same codes
1359        if (codes.getElementAt({0}) != codes.getElementAt({2})) {
1360            std::cout << "  [FAIL] : factorize: 2023-01-01 values should have same code" << std::endl;
1361            throw std::runtime_error("pd_test_datetime_array_unique failed: same code");
1362        }
freq (pd_test_1_all.cpp:8233)
8223    std::cout << "========= freq property ===============================";
8224
8225    std::vector<std::optional<numpy::datetime64>> values = {
8226        numpy::datetime64(0LL, numpy::DateTimeUnit::Nanosecond),
8227        numpy::datetime64(86400000000000LL, numpy::DateTimeUnit::Nanosecond)  // 1 day
8228    };
8229    pandas::DatetimeArray arr(values);
8230    pandas::DatetimeMixinIndex idx(arr);
8231
8232    // Default freq is nullopt or inferred
8233    auto f = idx.freq();
8234    std::string fs = idx.freqstr();
8235
8236    bool passed = true;  // freq may or may not be set
8237    if (!passed) {
8238        std::cout << "  [FAIL] : in pd_test_datetime_mixin_freq()" << std::endl;
8239        throw std::runtime_error("pd_test_datetime_mixin_freq failed");
8240    }
8241
8242    std::cout << " -> tests passed" << std::endl;
8243}
freqstr (pd_test_1_all.cpp:2671)
2661        }
2662
2663        pandas::PeriodDtype dtype_y("Y");
2664        if (dtype_y.name() != "period[Y]") {
2665            std::cout << "  [FAIL] : dtype_y.name() should be 'period[Y]'" << std::endl;
2666            throw std::runtime_error("pd_test_period_array_freq_validation failed: dtype name Y");
2667        }
2668
2669        // Test frequency string
2670        pandas::PeriodArray arr(std::vector<std::string>{"2024-01-15"}, "D");
2671        if (arr.freqstr() != "D") {
2672            std::cout << "  [FAIL] : arr.freqstr() should be 'D'" << std::endl;
2673            throw std::runtime_error("pd_test_period_array_freq_validation failed: freqstr");
2674        }
2675
2676        std::cout << " -> tests passed" << std::endl;
2677    }
2678
2679    void pd_test_period_array_year_month_quarter() {
2680        std::cout << "========= PeriodArray: year/month/quarter components ======================= ";
has_na (pd_test_1_all.cpp:61)
51        if (!arr.is_na(1)) {
52            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : is_na(1) should be true" << std::endl;
53            throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(1) should be true");
54        }
55
56        if (arr.is_na(0)) {
57            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : is_na(0) should be false" << std::endl;
58            throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(0) should be false");
59        }
60
61        if (!arr.has_na()) {
62            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : has_na() should be true" << std::endl;
63            throw std::runtime_error("pd_test_boolean_array_na_handling failed: has_na() should be true");
64        }
65
66        if (arr.count() != 2) {
67            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : count() should be 2" << std::endl;
68            throw std::runtime_error("pd_test_boolean_array_na_handling failed: count() should be 2");
69        }
70
71        std::cout << " -> tests passed" << std::endl;
hour (pd_test_1_all.cpp:7476)
7466    std::cout << "========= hour property ===============================";
7467
7468    std::vector<std::optional<numpy::datetime64>> values = {
7469        make_dt(0),                    // Hour 0
7470        make_dt(6 * NS_PER_HOUR),      // Hour 6
7471        make_dt(23 * NS_PER_HOUR)      // Hour 23
7472    };
7473    pandas::DatetimeArray arr(values);
7474    pandas::DatetimeIndex idx(arr);
7475
7476    auto hours = idx.hour();
7477
7478    bool passed = (hours.size() == 3);
7479    auto h0 = hours[0];
7480    auto h1 = hours[1];
7481    auto h2 = hours[2];
7482    passed = passed && h0.has_value() && *h0 == 0;
7483    passed = passed && h1.has_value() && *h1 == 6;
7484    passed = passed && h2.has_value() && *h2 == 23;
7485
7486    if (!passed) {
month (pd_test_1_all.cpp:1180)
1170    void pd_test_datetime_array_component_month_day() {
1171        std::cout << "========= DatetimeArray: month/day components ======================= ";
1172
1173        pandas::DatetimeArray arr(std::vector<std::string>{
1174            "2023-03-15",
1175            "2023-12-25",
1176            "NaT"
1177        });
1178
1179        // Month
1180        auto months = arr.month();
1181        auto m0 = months[0];
1182        if (!m0.has_value() || m0.value() != 3) {
1183            std::cout << "  [FAIL] : month[0] should be 3" << std::endl;
1184            throw std::runtime_error("pd_test_datetime_array_component_month_day failed: month[0]");
1185        }
1186        auto m1 = months[1];
1187        if (!m1.has_value() || m1.value() != 12) {
1188            std::cout << "  [FAIL] : month[1] should be 12" << std::endl;
1189            throw std::runtime_error("pd_test_datetime_array_component_month_day failed: month[1]");
1190        }
nbytes (pd_test_1_all.cpp:6214)
6204            }
6205
6206            // Test empty DataFrame
6207            pandas::DataFrame empty_df;
6208            if (!empty_df.empty()) {
6209                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : should be empty" << std::endl;
6210                throw std::runtime_error("pd_test_dataframe_properties failed: should be empty");
6211            }
6212
6213            // Test nbytes > 0 for non-empty
6214            if (df.nbytes() == 0) {
6215                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : nbytes should be > 0" << std::endl;
6216                throw std::runtime_error("pd_test_dataframe_properties failed: nbytes should be > 0");
6217            }
6218
6219            // Test columns index
6220            if (df.columns().size() != 3) {
6221                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : columns size != 3" << std::endl;
6222                throw std::runtime_error("pd_test_dataframe_properties failed: columns size != 3");
6223            }
ndim (pd_test_1_all.cpp:6195)
6185            pandas::DataFrame df(data);
6186
6187            // Test shape
6188            auto shape = df.shape();
6189            if (shape.size() != 2 || shape[0] != 4 || shape[1] != 3) {
6190                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : shape mismatch" << std::endl;
6191                throw std::runtime_error("pd_test_dataframe_properties failed: shape mismatch");
6192            }
6193
6194            // Test ndim
6195            if (df.ndim() != 2) {
6196                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : ndim != 2" << std::endl;
6197                throw std::runtime_error("pd_test_dataframe_properties failed: ndim != 2");
6198            }
6199
6200            // Test empty
6201            if (df.empty()) {
6202                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : should not be empty" << std::endl;
6203                throw std::runtime_error("pd_test_dataframe_properties failed: should not be empty");
6204            }
ordinal (pd_test_5_all.cpp:41796)
41786    std::cout << "----- case_11_period_nat_sentinel_api_self_check -----\n";
41787    pandas::Period nat1 = pandas::Period::NaT("D");
41788    pandas::Period nat2;  // default ctor = NaT
41789    pandas::Period real(123, std::string("D"));
41790
41791    pandas_tests::check(nat1.isNaT(), "case11.NaT_factory_isNaT", local_fail);
41792    pandas_tests::check(nat2.isNaT(),
41793                        "case11.default_ctor_isNaT", local_fail);
41794    pandas_tests::check(!real.isNaT(),
41795                        "case11.real_period_is_NOT_NaT", local_fail);
41796    pandas_tests::check(nat1.ordinal() == pandas::PERIOD_NAT,
41797                        "case11.NaT_ordinal_equals_PERIOD_NAT_sentinel", local_fail);
41798    pandas_tests::check(real.ordinal() == 123,
41799                        "case11.real_period_ordinal_value", local_fail);
41800
41801    std::cout << "  NaT.ordinal=" << nat1.ordinal()
41802              << " PERIOD_NAT=" << pandas::PERIOD_NAT
41803              << " real.ordinal=" << real.ordinal() << "\n";
41804}
41805
41806void case_12_period_oracle_parity_subset(int& local_fail) {
quarter (pd_test_1_all.cpp:1218)
1208    void pd_test_datetime_array_quarter() {
1209        std::cout << "========= DatetimeArray: quarter ======================= ";
1210
1211        pandas::DatetimeArray arr(std::vector<std::string>{
1212            "2023-01-15",  // Q1
1213            "2023-05-20",  // Q2
1214            "2023-09-10",  // Q3
1215            "2023-11-25"   // Q4
1216        });
1217
1218        auto quarters = arr.quarter();
1219
1220        auto q0 = quarters[0];
1221        if (!q0.has_value() || q0.value() != 1) {
1222            std::cout << "  [FAIL] : quarter[0] should be 1" << std::endl;
1223            throw std::runtime_error("pd_test_datetime_array_quarter failed: quarter[0]");
1224        }
1225        auto q1 = quarters[1];
1226        if (!q1.has_value() || q1.value() != 2) {
1227            std::cout << "  [FAIL] : quarter[1] should be 2" << std::endl;
1228            throw std::runtime_error("pd_test_datetime_array_quarter failed: quarter[1]");
repr (pd_test_1_all.cpp:10906)
10896    std::cout << " -> tests passed" << std::endl;
10897}
10898
10899void pd_test_extension_index_repr() {
10900    std::cout << "========= repr =========================";
10901
10902    pandas::CategoricalArray arr({"a", "b", "c"});
10903    // Use ExtensionIndex<CategoricalArray> directly to test base class repr
10904    pandas::ExtensionIndex<pandas::CategoricalArray> idx(arr, "test");
10905
10906    std::string repr_str = idx.repr();
10907
10908    bool passed = (!repr_str.empty() && repr_str.find("ExtensionIndex") != std::string::npos);
10909    if (!passed) {
10910        std::cout << "  [FAIL] : in pd_test_extension_index_repr() : repr check failed" << std::endl;
10911        throw std::runtime_error("pd_test_extension_index_repr failed");
10912    }
10913
10914    std::cout << " -> tests passed" << std::endl;
10915}
second (pd_test_1_all.cpp:7534)
7524    std::cout << "========= second property =============================";
7525
7526    std::vector<std::optional<numpy::datetime64>> values = {
7527        make_dt(0),                    // Second 0
7528        make_dt(30 * NS_PER_SEC),      // Second 30
7529        make_dt(59 * NS_PER_SEC)       // Second 59
7530    };
7531    pandas::DatetimeArray arr(values);
7532    pandas::DatetimeIndex idx(arr);
7533
7534    auto seconds = idx.second();
7535
7536    bool passed = (seconds.size() == 3);
7537    auto s0 = seconds[0];
7538    auto s1 = seconds[1];
7539    auto s2 = seconds[2];
7540    passed = passed && s0.has_value() && *s0 == 0;
7541    passed = passed && s1.has_value() && *s1 == 30;
7542    passed = passed && s2.has_value() && *s2 == 59;
7543
7544    if (!passed) {
shape (pd_test_1_all.cpp:6188)
6178            std::cout << "========= properties =======================";
6179
6180            std::map<std::string, std::vector<numpy::float64>> data;
6181            data["A"] = {1.0, 2.0, 3.0, 4.0};
6182            data["B"] = {5.0, 6.0, 7.0, 8.0};
6183            data["C"] = {9.0, 10.0, 11.0, 12.0};
6184
6185            pandas::DataFrame df(data);
6186
6187            // Test shape
6188            auto shape = df.shape();
6189            if (shape.size() != 2 || shape[0] != 4 || shape[1] != 3) {
6190                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : shape mismatch" << std::endl;
6191                throw std::runtime_error("pd_test_dataframe_properties failed: shape mismatch");
6192            }
6193
6194            // Test ndim
6195            if (df.ndim() != 2) {
6196                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : ndim != 2" << std::endl;
6197                throw std::runtime_error("pd_test_dataframe_properties failed: ndim != 2");
6198            }
size (pd_test_1_all.cpp:22)
12#include "../pandas/pd_boolean_array.h"
13
14namespace dataframe_tests {
15
16namespace dataframe_tests_boolean_array {
17    void pd_test_boolean_array_constructors() {
18        std::cout << "========= BooleanArray: constructors ======================= ";
19
20        // Default constructor
21        pandas::BooleanArray arr1;
22        if (arr1.size() != 0) {
23            std::cout << "  [FAIL] : in pd_test_boolean_array_constructors() : default constructor size != 0" << std::endl;
24            throw std::runtime_error("pd_test_boolean_array_constructors failed: default constructor size != 0");
25        }
26
27        // Initializer list constructor
28        pandas::BooleanArray arr2({
29            std::optional<bool>(true),
30            std::optional<bool>(false),
31            std::nullopt,
32            std::optional<bool>(true)
start_time (pd_test_1_all.cpp:2848)
2838            std::cout << "  [FAIL] : ts_start[0] should have value" << std::endl;
2839            throw std::runtime_error("pd_test_period_array_to_timestamp failed: ts_start[0]");
2840        }
2841
2842        auto ts2 = ts_start[2];
2843        if (ts2.has_value()) {
2844            std::cout << "  [FAIL] : ts_start[2] should be NaT" << std::endl;
2845            throw std::runtime_error("pd_test_period_array_to_timestamp failed: ts_start[2]");
2846        }
2847
2848        // start_time() alias
2849        auto start_times = arr.start_time();
2850        if (start_times.size() != 3) {
2851            std::cout << "  [FAIL] : start_time size should be 3" << std::endl;
2852            throw std::runtime_error("pd_test_period_array_to_timestamp failed: start_time size");
2853        }
2854
2855        std::cout << " -> tests passed" << std::endl;
2856    }
2857
2858    void pd_test_period_array_asfreq() {
week (pd_test_timestamp_scalar.cpp:406)
396      // 2024-06-15 is a Saturday
397      pandas::Timestamp ts(2024, 6, 15);
398
399      if (ts.dayofweek() != 5) { pass = false; fail_msg = "2024-06-15 should be Saturday (5)"; }
400      if (ts.day_of_week() != 5) { pass = false; fail_msg = "day_of_week alias"; }
401      if (ts.dayofyear() != 167) { pass = false; fail_msg = "2024-06-15 should be day 167"; }
402      if (ts.day_of_year() != 167) { pass = false; fail_msg = "day_of_year alias"; }
403      if (ts.quarter() != 2) { pass = false; fail_msg = "June is Q2"; }
404      if (ts.days_in_month() != 30) { pass = false; fail_msg = "June has 30 days"; }
405
406      int week = ts.week();
407      if (week < 1 || week > 53) { pass = false; fail_msg = "week should be 1-53"; }
408
409      if (!pass) {
410        std::cout << "  [FAIL] : in np_test_timestamp_derived() : " << fail_msg;
411        throw std::runtime_error("np_test_timestamp_derived failed: " + fail_msg);
412      }
413
414      std::cout << " -> tests passed" << std::endl;
415    }
weekday (pd_test_3_all.cpp:1471)
1461            std::cout << "  [FAIL] date_range 10-arg form: expected size 10, got "
1462                      << idx.size() << std::endl;
1463            throw std::runtime_error("date_range 10-arg form regressed");
1464        }
1465    }
1466
1467    std::cout << " -> tests passed" << std::endl;
1468}
1469
1470void pd_test_3_all_period_weekday() {
1471    std::cout << "========= PeriodArray.weekday() ======================";
1472
1473    // Create a PeriodArray with some dates
1474    std::vector<std::optional<numpy::int64>> ordinals = {0, 1, 2, 3, 4};  // Days from epoch
1475    pandas::PeriodArray arr(ordinals, "D");
1476
1477    pandas::IntegerArray<numpy::int32> weekdays = arr.weekday();
1478
1479    if (weekdays.size() != 5) {
1480        std::cout << "  [FAIL] : in pd_test_3_all_period_weekday() : size should be 5" << std::endl;
1481        throw std::runtime_error("pd_test_3_all_period_weekday failed: size");
year (pd_test_1_all.cpp:1147)
1137    void pd_test_datetime_array_component_year() {
1138        std::cout << "========= DatetimeArray: year component ======================= ";
1139
1140        pandas::DatetimeArray arr(std::vector<std::string>{
1141            "2020-01-15",
1142            "NaT",
1143            "2025-06-20"
1144        });
1145
1146        auto years = arr.year();
1147
1148        auto y0 = years[0];
1149        if (!y0.has_value() || y0.value() != 2020) {
1150            std::cout << "  [FAIL] : year[0] should be 2020" << std::endl;
1151            throw std::runtime_error("pd_test_datetime_array_component_year failed: year[0]");
1152        }
1153
1154        auto y1 = years[1];
1155        if (y1.has_value()) {
1156            std::cout << "  [FAIL] : year[1] should be NA (NaT)" << std::endl;