CategoricalArray

class pandas::CategoricalArray

Extension array type for specialized data storage.

Example

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

// Use CategoricalArray
CategoricalArray obj;
// ... operations ...

Constructors

Signature	Location	Example
CategoricalArray(const numpy::NDArray<numpy::int32>& codes, const std::vector<std::string>& categories, bool ordered = false, bool copy = true)	pd_categorical_array.h:103	View
CategoricalArray(const std::vector<std::optional<std::string>>& values, const std::optional<std::vector<std::string>>& categories = std::nullopt, std::optional<bool> ordered = std::nullopt, const std::optional<CategoricalDtype>& dtype = std::nullopt, bool fastpath = false, bool copy = true)	pd_categorical_array.h:126	View
explicit CategoricalArray(const std::vector<std::optional<std::string>>& values, bool ordered)	pd_categorical_array.h:233	View
CategoricalArray(const std::vector<std::optional<std::string>>& values, const std::vector<std::string>& categories, bool ordered = false)	pd_categorical_array.h:251	View

Construction

Signature	Return Type	Location	Example
static CategoricalArray from_codes( const std::vector<numpy::int32>& codes, const std::vector<std::string>& categories, bool ordered = false, const std::string& dtype = "", bool validate = true)	static CategoricalArray	pd_categorical_array.h:275	View
static CategoricalArray from_sequence( const std::vector<std::optional<std::string>>& values, bool ordered = false)	static CategoricalArray	pd_categorical_array.h:304

Indexing / Selection

Signature	Return Type	Location	Example
numpy::int32 get_code(ssize_t index) const	numpy::int32	pd_categorical_array.h:519
CategoricalArray take(const std::vector<numpy::int64>& indices, std::optional<int> axis = std::nullopt, bool allow_fill = false, const std::string& fill_value = "") const	CategoricalArray	pd_categorical_array.h:1305	View

Data Manipulation

Signature	Return Type	Location	Example
CategoricalArray dropna() const	CategoricalArray	pd_categorical_array.h:595	View
CategoricalArray insert(size_t loc, const std::optional<std::string>& item) const	CategoricalArray	pd_categorical_array.h:1594	View
CategoricalArray rename_categories(const std::vector<std::string>& new_categories) const	CategoricalArray	pd_categorical_array.h:740	View
CategoricalArray rename_categories(const std::unordered_map<std::string, std::string>& mapping) const	CategoricalArray	pd_categorical_array.h:760	View

Missing Data

Signature	Return Type	Location	Example
CategoricalArray fillna(const std::string& value, const std::string& method = "", std::optional<size_t> limit = std::nullopt, bool copy = true) const	CategoricalArray	pd_categorical_array.h:566	View
CategoricalArray interpolate( const std::string& method = "pad", int axis = 0, const std::optional<std::vector<size_t>>& index = std::nullopt, std::optional<size_t> limit = std::nullopt, const std::string& limit_direction = "forward", const std::optional<std::string>& limit_area = std::nullopt, bool copy = true ) const	CategoricalArray	pd_categorical_array.h:1643	View
numpy::NDArray<numpy::bool_> isna() const	numpy::NDArray<numpy::bool_>	pd_categorical_array.h:539	View
numpy::NDArray<numpy::bool_> isnull() const	numpy::NDArray<numpy::bool_>	pd_categorical_array.h:1798	View
numpy::NDArray<numpy::bool_> notna() const	numpy::NDArray<numpy::bool_>	pd_categorical_array.h:550	View
numpy::NDArray<numpy::bool_> notnull() const	numpy::NDArray<numpy::bool_>	pd_categorical_array.h:1884	View

Statistics

Signature	Return Type	Location	Example
size_t count() const	size_t	pd_categorical_array.h:609	View
std::map<std::string, std::string> describe() const	std::map<std::string, std::string>	pd_categorical_array.h:1480	View
std::optional<std::string> max(bool skipna = true) const	std::optional<std::string>	pd_categorical_array.h:918	View
std::optional<std::string> min(bool skipna = true) const	std::optional<std::string>	pd_categorical_array.h:884	View
std::pair<std::vector<std::string>, std::vector<numpy::int64>> value_counts() const	std::pair<std::vector<std::string>, std::vector<numpy::int64>>	pd_categorical_array.h:999	View

Aggregation

Signature	Return Type	Location	Example
CategoricalArray map(const std::unordered_map<std::string, std::string>& mapper, const std::string& na_action = "") const	CategoricalArray	pd_categorical_array.h:1808	View
CategoricalArray map(Func func, const std::string& na_action = "") const	CategoricalArray	pd_categorical_array.h:1852	View

Arithmetic

Signature	Return Type	Location	Example
CategoricalArray add_categories(const std::vector<std::string>& new_categories) const	CategoricalArray	pd_categorical_array.h:638	View

Comparison

Signature	Return Type	Location	Example
bool equals(const CategoricalArray& other) const	bool	pd_categorical_array.h:1575	View
size_t len() const	size_t	pd_categorical_array.h:447	View

Sorting

Signature	Return Type	Location	Example
numpy::NDArray<numpy::int64> argsort(bool ascending = true, const std::string& na_position = "last", const std::string& kind = "quicksort") const	numpy::NDArray<numpy::int64>	pd_categorical_array.h:1180	View
size_t searchsorted(const std::string& value, const std::string& side = "left", std::optional<numpy::NDArray<numpy::int64>> sorter = std::nullopt) const	size_t	pd_categorical_array.h:1970	View
CategoricalArray sort_values(bool ascending = true, const std::string& na_position = "last", bool inplace = false) const	CategoricalArray	pd_categorical_array.h:2067	View

Reshaping

Signature	Return Type	Location	Example
CategoricalArray T() const	CategoricalArray	pd_categorical_array.h:2162	View
CategoricalArray swapaxes(int axis1, int axis2) const	CategoricalArray	pd_categorical_array.h:2094	View
CategoricalArray transpose() const	CategoricalArray	pd_categorical_array.h:2155	View

Combining

Signature	Return Type	Location	Example
static CategoricalArray concat(const std::vector<CategoricalArray>& arrays)	static CategoricalArray	pd_categorical_array.h:314	View
static CategoricalArray concat_merge(const std::vector<CategoricalArray>& arrays)	static CategoricalArray	pd_categorical_array.h:351	View

Time Series

Signature	Return Type	Location	Example
CategoricalArray shift(int64_t periods = 1, const std::optional<std::string>& fill_value = std::nullopt) const	CategoricalArray	pd_categorical_array.h:2031	View

I/O

Signature	Return Type	Location	Example
std::vector<std::optional<std::string>> to_list() const	std::vector<std::optional<std::string>>	pd_categorical_array.h:2105	View
numpy::NDArray<U> to_numpy(bool copy = true, U na_value = U{-1}) const	numpy::NDArray<U>	pd_categorical_array.h:2136	View
std::string to_string() const	std::string	pd_categorical_array.h:2212	View
std::vector<std::optional<std::string>> tolist() const	std::vector<std::optional<std::string>>	pd_categorical_array.h:2124	View

Conversion

Signature	Return Type	Location	Example
std::vector<std::optional<T>> astype(const std::string& dtype, bool copy = true) const	std::vector<std::optional<T>>	pd_categorical_array.h:1365	View
numpy::NDArray<numpy::int32> astype_codes() const	numpy::NDArray<numpy::int32>	pd_categorical_array.h:1418	View
CategoricalArray copy() const	CategoricalArray	pd_categorical_array.h:1298	View
CategoricalArray view() const	CategoricalArray	pd_categorical_array.h:2174	View

Set Operations

Signature	Return Type	Location	Example
numpy::NDArray<numpy::bool_> duplicated(const std::string& keep = "first") const	numpy::NDArray<numpy::bool_>	pd_categorical_array.h:1517	View
numpy::NDArray<numpy::bool_> isin(const std::vector<std::string>& values) const	numpy::NDArray<numpy::bool_>	pd_categorical_array.h:1770	View
CategoricalArray unique() const	CategoricalArray	pd_categorical_array.h:951	View

Type Checking

Signature	Return Type	Location	Example
bool is_na(ssize_t index) const	bool	pd_categorical_array.h:527	View

Other Methods

Signature	Return Type	Location	Example
std::optional<size_t> argmax(std::optional<int> axis = std::nullopt, bool skipna = true) const	std::optional<size_t>	pd_categorical_array.h:1258	View
std::optional<size_t> argmin(std::optional<int> axis = std::nullopt, bool skipna = true) const	std::optional<size_t>	pd_categorical_array.h:1220	View
CategoricalArray as_ordered() const	CategoricalArray	pd_categorical_array.h:865	View
CategoricalArray as_unordered() const	CategoricalArray	pd_categorical_array.h:872	View
std::unordered_map<std::string, numpy::int32> build_category_map() const	std::unordered_map<std::string, numpy::int32>	pd_categorical_array.h:59
const std::vector<std::string>& categories() const	const std::vector<std::string>&	pd_categorical_array.h:473	View
const std::string& categories_dtype_str() const	const std::string&	pd_categorical_array.h:492
void check_for_ordered(const std::string& op) const	void	pd_categorical_array.h:1427	View
check_for_ordered("searchsorted")		pd_categorical_array.h:1974	View
check_for_ordered("sort_values")		pd_categorical_array.h:2071	View
const numpy::NDArray<numpy::int32>& codes() const	const numpy::NDArray<numpy::int32>&	pd_categorical_array.h:454	View
std::vector<numpy::int32> codes_vector() const	std::vector<numpy::int32>	pd_categorical_array.h:461
CategoricalArray delete_(size_t loc, std::optional<int> axis = std::nullopt) const	CategoricalArray	pd_categorical_array.h:1438	View
CategoricalArray delete_(const std::vector<size_t>& locs) const	CategoricalArray	pd_categorical_array.h:1461	View
CategoricalDtype dtype() const	CategoricalDtype	pd_categorical_array.h:401	View
bool empty() const	bool	pd_categorical_array.h:440	View
std::pair<IntegerArray<numpy::int64>, CategoricalArray> factorize() const	std::pair<IntegerArray<numpy::int64>, CategoricalArray>	pd_categorical_array.h:969	View
bool has_na() const	bool	pd_categorical_array.h:622	View
std::vector<std::string> internal_get_values() const	std::vector<std::string>	pd_categorical_array.h:2198
size_t memory_usage(bool deep = false) const	size_t	pd_categorical_array.h:1865	View
const std::optional<std::string>& name() const	const std::optional<std::string>&	pd_categorical_array.h:484	View
size_t nbytes() const	size_t	pd_categorical_array.h:415	View
constexpr int ndim() const	constexpr int	pd_categorical_array.h:426	View
bool ordered() const	bool	pd_categorical_array.h:480	View
CategoricalArray ravel() const	CategoricalArray	pd_categorical_array.h:1892	View
CategoricalArray remove_categories(const std::vector<std::string>& removals) const	CategoricalArray	pd_categorical_array.h:657	View
CategoricalArray remove_unused_categories() const	CategoricalArray	pd_categorical_array.h:823	View
CategoricalArray reorder_categories(const std::vector<std::string>& new_categories, std::optional<bool> ordered = std::nullopt) const	CategoricalArray	pd_categorical_array.h:783	View
CategoricalArray repeat(size_t repeats, std::optional<int> axis = std::nullopt) const	CategoricalArray	pd_categorical_array.h:1901	View
CategoricalArray repeat(const std::vector<size_t>& repeats) const	CategoricalArray	pd_categorical_array.h:1921	View
std::string repr() const	std::string	pd_categorical_array.h:2238	View
CategoricalArray reshape(const std::vector<size_t>& new_shape) const	CategoricalArray	pd_categorical_array.h:1950	View
CategoricalArray set_categories(const std::vector<std::string>& new_categories, std::optional<bool> ordered = std::nullopt, bool rename = false) const	CategoricalArray	pd_categorical_array.h:700	View
void set_categories_dtype(const std::string& dtype)	void	pd_categorical_array.h:496
void set_name(const std::string& name)	void	pd_categorical_array.h:488	View
CategoricalArray set_ordered(bool value) const	CategoricalArray	pd_categorical_array.h:2021	View
std::vector<size_t> shape() const	std::vector<size_t>	pd_categorical_array.h:433	View
size_t size() const	size_t	pd_categorical_array.h:408	View
CategoricalArray slice(size_t start, size_t stop, size_t step = 1) const	CategoricalArray	pd_categorical_array.h:1342	View
void validate_codes() const	void	pd_categorical_array.h:70

Internal Methods

1 internal methods (prefixed with underscore)

Code Examples

The following examples are extracted from the test suite.

CategoricalArray (pd_test_3_all.cpp:28514)

static int cgo_check(bool cond, const char* msg) {
    if (!cond) { std::cout << "    FAIL: " << msg << std::endl; return 1; }
    return 0;
}

static pandas::CategoricalArray make_abc() {
    std::vector<std::optional<std::string>> v{
        std::string("a"), std::string("b"), std::string("c"), std::string("a")
    };
    return pandas::CategoricalArray(v, false);
}

void pd_test_cat_rename_dict() {
    std::cout << "  -- pd_test_cat_rename_dict --" << std::endl;
    int fail = 0;
    auto arr = make_abc();
    std::unordered_map<std::string, std::string> m{{"a", "A"}, {"b", "B"}};
    auto r = arr.rename_categories(m);
    const auto& cats = r.categories();
    fail += cgo_check(cats.size() == 3, "size==3");

CategoricalArray (pd_test_3_all.cpp:28514)

static int cgo_check(bool cond, const char* msg) {
    if (!cond) { std::cout << "    FAIL: " << msg << std::endl; return 1; }
    return 0;
}

static pandas::CategoricalArray make_abc() {
    std::vector<std::optional<std::string>> v{
        std::string("a"), std::string("b"), std::string("c"), std::string("a")
    };
    return pandas::CategoricalArray(v, false);
}

void pd_test_cat_rename_dict() {
    std::cout << "  -- pd_test_cat_rename_dict --" << std::endl;
    int fail = 0;
    auto arr = make_abc();
    std::unordered_map<std::string, std::string> m{{"a", "A"}, {"b", "B"}};
    auto r = arr.rename_categories(m);
    const auto& cats = r.categories();
    fail += cgo_check(cats.size() == 3, "size==3");

CategoricalArray (pd_test_3_all.cpp:28514)

static int cgo_check(bool cond, const char* msg) {
    if (!cond) { std::cout << "    FAIL: " << msg << std::endl; return 1; }
    return 0;
}

static pandas::CategoricalArray make_abc() {
    std::vector<std::optional<std::string>> v{
        std::string("a"), std::string("b"), std::string("c"), std::string("a")
    };
    return pandas::CategoricalArray(v, false);
}

void pd_test_cat_rename_dict() {
    std::cout << "  -- pd_test_cat_rename_dict --" << std::endl;
    int fail = 0;
    auto arr = make_abc();
    std::unordered_map<std::string, std::string> m{{"a", "A"}, {"b", "B"}};
    auto r = arr.rename_categories(m);
    const auto& cats = r.categories();
    fail += cgo_check(cats.size() == 3, "size==3");

CategoricalArray (pd_test_3_all.cpp:28514)

static int cgo_check(bool cond, const char* msg) {
    if (!cond) { std::cout << "    FAIL: " << msg << std::endl; return 1; }
    return 0;
}

static pandas::CategoricalArray make_abc() {
    std::vector<std::optional<std::string>> v{
        std::string("a"), std::string("b"), std::string("c"), std::string("a")
    };
    return pandas::CategoricalArray(v, false);
}

void pd_test_cat_rename_dict() {
    std::cout << "  -- pd_test_cat_rename_dict --" << std::endl;
    int fail = 0;
    auto arr = make_abc();
    std::unordered_map<std::string, std::string> m{{"a", "A"}, {"b", "B"}};
    auto r = arr.rename_categories(m);
    const auto& cats = r.categories();
    fail += cgo_check(cats.size() == 3, "size==3");

from_codes (pd_test_1_all.cpp:403)

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_categorical_array_from_codes() {
        std::cout << "========= CategoricalArray: from_codes ======================= ";

        std::vector<std::string> cats = {"a", "b", "c"};
        std::vector<numpy::int32> codes = {0, 1, 2, 0, 1, -1};  // -1 is NA

        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats, false);

        if (arr.size() != 6) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_from_codes() : size != 6" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_from_codes failed: size != 6");
        }

        // Check that code=-1 creates NA
        if (!arr.is_na(5)) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_from_codes() : code -1 should be NA" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_from_codes failed: code -1 should be NA");

take (pd_test_1_all.cpp:5903)

// Inherited Operations Tests
// ============================================================================

void pd_test_categorical_index_take() {
    std::cout << "========= inherited take ==============================";

    pandas::CategoricalArray arr({"a", "b", "c", "d"});
    pandas::CategoricalIndex idx(arr);

    std::vector<size_t> indices = {0, 2, 3};
    pandas::ExtensionIndex<pandas::CategoricalArray> taken = idx.take(indices);

    bool passed = (taken.size() == 3);
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_categorical_index_take()" << std::endl;
        throw std::runtime_error("pd_test_categorical_index_take failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

dropna (pd_test_1_all.cpp:531)

        }

        // Test isna array
        numpy::NDArray<numpy::bool_> na_mask = arr.isna();
        if (na_mask.getSize() != 4) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : isna size != 4" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_na_handling failed: isna size != 4");
        }

        // Test dropna
        pandas::CategoricalArray dropped = arr.dropna();
        if (dropped.size() != 2) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : dropna size != 2" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_na_handling failed: dropna size != 2");
        }

        // Test fillna (fill with existing category)
        pandas::CategoricalArray filled = arr.fillna("a");  // 'a' is in categories
        if (filled.has_na()) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : fillna should have no NA" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_na_handling failed: fillna should have no NA");

insert (pd_test_1_all.cpp:12028)

            }

            std::cout << " -> tests passed" << std::endl;
        }

        void pd_test_index_insert_delete() {
            std::cout << "========= insert and delete ===========================";

            pandas::Index<numpy::int64> idx{1, 2, 4, 5};

            auto inserted = idx.insert(2, 3);
            bool passed = (inserted.size() == 5);
            passed = passed && (inserted[2] == 3);

            auto deleted = inserted.delete_(2);
            passed = passed && (deleted.size() == 4);
            passed = passed && deleted.equals(idx);

            if (!passed) {
                std::cout << "  [FAIL] : in pd_test_index_insert_delete() : insert/delete failed" << std::endl;
                throw std::runtime_error("pd_test_index_insert_delete failed");

rename_categories (pd_test_1_all.cpp:655)

    void pd_test_categorical_array_rename_categories() {
        std::cout << "========= CategoricalArray: rename_categories ======================= ";

        std::vector<std::string> cats = {"a", "b"};
        std::vector<numpy::int32> codes = {0, 1, 0};  // a, b, a
        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);

        // Rename categories
        std::vector<std::string> new_names = {"alpha", "beta"};
        pandas::CategoricalArray result = arr.rename_categories(new_names);

        // Check categories are renamed
        const std::vector<std::string>& result_cats = result.categories();
        if (result_cats[0] != "alpha" || result_cats[1] != "beta") {
            std::cout << "  [FAIL] : in pd_test_categorical_array_rename_categories() : categories not renamed" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_rename_categories failed: categories not renamed");
        }

        // Values should now be renamed
        std::optional<std::string> val = result[0];

rename_categories (pd_test_1_all.cpp:655)

    void pd_test_categorical_array_rename_categories() {
        std::cout << "========= CategoricalArray: rename_categories ======================= ";

        std::vector<std::string> cats = {"a", "b"};
        std::vector<numpy::int32> codes = {0, 1, 0};  // a, b, a
        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);

        // Rename categories
        std::vector<std::string> new_names = {"alpha", "beta"};
        pandas::CategoricalArray result = arr.rename_categories(new_names);

        // Check categories are renamed
        const std::vector<std::string>& result_cats = result.categories();
        if (result_cats[0] != "alpha" || result_cats[1] != "beta") {
            std::cout << "  [FAIL] : in pd_test_categorical_array_rename_categories() : categories not renamed" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_rename_categories failed: categories not renamed");
        }

        // Values should now be renamed
        std::optional<std::string> val = result[0];

fillna (pd_test_1_all.cpp:537)

            throw std::runtime_error("pd_test_categorical_array_na_handling failed: isna size != 4");
        }

        // Test dropna
        pandas::CategoricalArray dropped = arr.dropna();
        if (dropped.size() != 2) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : dropna size != 2" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_na_handling failed: dropna size != 2");
        }

        // Test fillna (fill with existing category)
        pandas::CategoricalArray filled = arr.fillna("a");  // 'a' is in categories
        if (filled.has_na()) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : fillna should have no NA" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_na_handling failed: fillna should have no NA");
        }

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_categorical_array_add_categories() {

interpolate (pd_test_1_all.cpp:24365)

        std::cout << "====================================== [OK] pd_test_idxmax_idxmin test suite ========================== " << std::endl;
        return 0;
    }

} // namespace dataframe_tests
// ------------------- pd_test_idxmax_idxmin.cpp (end) -----------------------------

// ------------------- pd_test_interpolate.cpp (start) -----------------------------
// dataframe_tests/pd_test_interpolate.cpp
// Test file for DataFrame.interpolate() method

#include <iostream>
#include <stdexcept>
#include <cmath>
#include <limits>
#include <map>
#include "../pandas/pd_dataframe.h"

// CRITICAL: No using namespace directives

isna (pd_test_1_all.cpp:524)

            throw std::runtime_error("pd_test_categorical_array_na_handling failed: has_na() should be true");
        }

        // Test count (non-NA)
        if (arr.count() != 2) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : count() != 2" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_na_handling failed: count() != 2");
        }

        // Test isna array
        numpy::NDArray<numpy::bool_> na_mask = arr.isna();
        if (na_mask.getSize() != 4) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : isna size != 4" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_na_handling failed: isna size != 4");
        }

        // Test dropna
        pandas::CategoricalArray dropped = arr.dropna();
        if (dropped.size() != 2) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_na_handling() : dropna size != 2" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_na_handling failed: dropna size != 2");

isnull (pd_test_3_all.cpp:671)

// Category 5: Index Null Detection
// ============================================================================

void pd_test_3_all_index_null_detection() {
    std::cout << "========= Index.isnull/notnull() =====================";

    // Test with float index (can have NaN)
    std::vector<double> vals = {1.0, std::nan(""), 3.0, std::nan("")};
    pandas::Index<double> idx(vals);

    numpy::NDArray<numpy::bool_> isnull_result = idx.isnull();
    if (isnull_result.getSize() != 4) {
        std::cout << "  [FAIL] : in pd_test_3_all_index_null_detection() : isnull() size mismatch" << std::endl;
        throw std::runtime_error("pd_test_3_all_index_null_detection failed: isnull() size");
    }
    // Index 0: 1.0 -> not null
    if (isnull_result.getElementAt({0})) {
        std::cout << "  [FAIL] : in pd_test_3_all_index_null_detection() : index 0 should not be null" << std::endl;
        throw std::runtime_error("pd_test_3_all_index_null_detection failed: index 0");
    }
    // Index 1: NaN -> null

notna (pd_test_1_all.cpp:6595)

                if (!na_mask.getElementAt({2, 1})) {
                    std::cout << "  [FAIL] : in pd_test_dataframe_manipulation() : isna at (2,1) should be true" << std::endl;
                    throw std::runtime_error("pd_test_dataframe_manipulation failed: isna at (2,1)");
                }
                // Row 0, col 0 should NOT be NA
                if (na_mask.getElementAt({0, 0})) {
                    std::cout << "  [FAIL] : in pd_test_dataframe_manipulation() : isna at (0,0) should be false" << std::endl;
                    throw std::runtime_error("pd_test_dataframe_manipulation failed: isna at (0,0)");
                }

                auto notna_mask = df_na.notna();
                if (notna_mask.getElementAt({1, 0})) {
                    std::cout << "  [FAIL] : in pd_test_dataframe_manipulation() : notna at (1,0) should be false" << std::endl;
                    throw std::runtime_error("pd_test_dataframe_manipulation failed: notna at (1,0)");
                }
            }

            // Test fillna
            {
                std::map<std::string, std::vector<numpy::float64>> float_data;
                float_data["X"] = {1.0, std::nan(""), 3.0};

notnull (pd_test_3_all.cpp:665)

    }

    std::cout << " -> tests passed" << std::endl;
}

// ============================================================================
// Category 5: Index Null Detection
// ============================================================================

void pd_test_3_all_index_null_detection() {
    std::cout << "========= Index.isnull/notnull() =====================";

    // Test with float index (can have NaN)
    std::vector<double> vals = {1.0, std::nan(""), 3.0, std::nan("")};
    pandas::Index<double> idx(vals);

    numpy::NDArray<numpy::bool_> isnull_result = idx.isnull();
    if (isnull_result.getSize() != 4) {
        std::cout << "  [FAIL] : in pd_test_3_all_index_null_detection() : isnull() size mismatch" << std::endl;
        throw std::runtime_error("pd_test_3_all_index_null_detection failed: isnull() size");
    }

count (pd_test_1_all.cpp:66)

        if (arr.is_na(0)) {
            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : is_na(0) should be false" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(0) should be false");
        }

        if (!arr.has_na()) {
            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : has_na() should be true" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_na_handling failed: has_na() should be true");
        }

        if (arr.count() != 2) {
            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : count() should be 2" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_na_handling failed: count() should be 2");
        }

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_boolean_array_kleene_and() {
        std::cout << "========= BooleanArray: Kleene AND ======================= ";

describe (pd_test_2_all.cpp:19793)

        ++g_fail;
    }
}

static bool approx_eq(double a, double b, double tol = 1e-9) {
    if (std::isnan(a) && std::isnan(b)) return true;
    return std::abs(a - b) < tol;
}

// =====================================================================
// Test: describe() default mode — numeric columns only
// =====================================================================

void pd_test_describe_numeric_only() {
    std::cout << "  -- pd_test_describe_numeric_only --" << std::endl;

    pandas::DataFrame df;
    df.add_column("A", std::vector<double>{1.0, 2.0, 3.0, 4.0, 5.0});
    df.add_column("B", std::vector<double>{10.0, 20.0, 30.0, 40.0, 50.0});
    df.add_column("Name", std::vector<std::string>{"a", "b", "c", "d", "e"});

max (pd_test_1_all.cpp:771)

        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats, true);  // ordered

        // Test min
        std::optional<std::string> min_val = arr.min();
        if (!min_val.has_value() || *min_val != "low") {
            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : min != 'low'" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: min != 'low'");
        }

        // Test max
        std::optional<std::string> max_val = arr.max();
        if (!max_val.has_value() || *max_val != "high") {
            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : max != 'high'" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: max != 'high'");
        }

        // Test unordered throws for min/max
        pandas::CategoricalArray unordered = arr.as_unordered();
        bool threw = false;
        try {
            unordered.min();

min (pd_test_1_all.cpp:764)

    }

    void pd_test_categorical_array_ordered_operations() {
        std::cout << "========= CategoricalArray: ordered operations (min/max) ======================= ";

        std::vector<std::string> cats = {"low", "medium", "high"};
        std::vector<numpy::int32> codes = {0, 2, 1, 0, -1};  // low, high, medium, low, NA
        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats, true);  // ordered

        // Test min
        std::optional<std::string> min_val = arr.min();
        if (!min_val.has_value() || *min_val != "low") {
            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : min != 'low'" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: min != 'low'");
        }

        // Test max
        std::optional<std::string> max_val = arr.max();
        if (!max_val.has_value() || *max_val != "high") {
            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : max != 'high'" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: max != 'high'");

value_counts (pd_test_1_all.cpp:865)

        std::vector<std::optional<std::string>> values = {
            std::optional<std::string>("a"),
            std::optional<std::string>("b"),
            std::optional<std::string>("a"),
            std::optional<std::string>("a"),
            std::optional<std::string>("b"),
            std::nullopt  // NA not counted
        };
        pandas::CategoricalArray arr(values);

        auto [cats, counts] = arr.value_counts();

        // Should have 2 categories
        if (cats.size() != 2 || counts.size() != 2) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_value_counts() : wrong size" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_value_counts failed: wrong size");
        }

        // Find 'a' count
        int64_t a_count = 0, b_count = 0;
        for (size_t i = 0; i < cats.size(); ++i) {

map (pd_test_1_all.cpp:5839)

// Map Tests
// ============================================================================

void pd_test_categorical_index_map() {
    std::cout << "========= map =========================================";

    pandas::CategoricalArray arr({"yes", "no", "yes"});
    pandas::CategoricalIndex idx(arr);

    std::unordered_map<std::string, std::string> mapping = {{"yes", "1"}, {"no", "0"}};
    pandas::CategoricalIndex mapped = idx.map(mapping);

    bool passed = (mapped.has_category("1") && mapped.has_category("0") &&
                   !mapped.has_category("yes") && !mapped.has_category("no"));
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_categorical_index_map()" << std::endl;
        throw std::runtime_error("pd_test_categorical_index_map failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

map (pd_test_1_all.cpp:5839)

// Map Tests
// ============================================================================

void pd_test_categorical_index_map() {
    std::cout << "========= map =========================================";

    pandas::CategoricalArray arr({"yes", "no", "yes"});
    pandas::CategoricalIndex idx(arr);

    std::unordered_map<std::string, std::string> mapping = {{"yes", "1"}, {"no", "0"}};
    pandas::CategoricalIndex mapped = idx.map(mapping);

    bool passed = (mapped.has_category("1") && mapped.has_category("0") &&
                   !mapped.has_category("yes") && !mapped.has_category("no"));
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_categorical_index_map()" << std::endl;
        throw std::runtime_error("pd_test_categorical_index_map failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

add_categories (pd_test_1_all.cpp:555)

    }

    void pd_test_categorical_array_add_categories() {
        std::cout << "========= CategoricalArray: add_categories ======================= ";

        std::vector<std::string> cats = {"a", "b"};
        std::vector<numpy::int32> codes = {0, 1, 0};
        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);

        // Add new categories
        pandas::CategoricalArray result = arr.add_categories({"c", "d"});
        if (result.categories().size() != 4) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_add_categories() : new categories size != 4" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_add_categories failed: new categories size != 4");
        }

        // Original values should be preserved
        std::optional<std::string> val = result[0];
        if (!val.has_value() || *val != "a") {
            std::cout << "  [FAIL] : in pd_test_categorical_array_add_categories() : value not preserved" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_add_categories failed: value not preserved");

equals (pd_test_1_all.cpp:5866)

    std::cout << "========= equals ======================================";

    pandas::CategoricalArray arr1({"a", "b", "a"});
    pandas::CategoricalArray arr2({"a", "b", "a"});
    pandas::CategoricalArray arr3({"a", "b", "c"});

    pandas::CategoricalIndex idx1(arr1);
    pandas::CategoricalIndex idx2(arr2);
    pandas::CategoricalIndex idx3(arr3);

    bool passed = (idx1.equals(idx2) && !idx1.equals(idx3));
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_categorical_index_equals()" << std::endl;
        throw std::runtime_error("pd_test_categorical_index_equals failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_categorical_index_identical() {
    std::cout << "========= identical ===================================";

len (pd_test_3_all.cpp:20867)

    auto title_result = s.str().title();
    if (title_result[0] != "Hello World" || title_result[1] != "Hello World" || title_result[2] != "Hello World") {
        std::cout << "  [FAIL] : title() failed" << std::endl;
        throw std::runtime_error("pd_test_str_capitalize_title: title() failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

// ============================================================================
// Test str().len()
// ============================================================================

void pd_test_str_len() {
    std::cout << "========= Series.str().len() ============================";

    pandas::Series<std::string> s({"a", "bb", "ccc", ""});

    auto lens = s.str().len();
    if (lens[0] != 1 || lens[1] != 2 || lens[2] != 3 || lens[3] != 0) {
        std::cout << "  [FAIL] : len() failed" << std::endl;

argsort (pd_test_1_all.cpp:1304)

        std::cout << "========= DatetimeArray: sorting ======================= ";

        pandas::DatetimeArray arr(std::vector<std::string>{
            "2023-06-15",
            "NaT",
            "2023-01-01",
            "2023-12-31"
        });

        // argsort ascending
        auto indices = arr.argsort(true, "last");
        // Expected order: 2023-01-01(2), 2023-06-15(0), 2023-12-31(3), NaT(1)
        if (indices.getElementAt({0}) != 2) {
            std::cout << "  [FAIL] : argsort: first should be index 2 (2023-01-01)" << std::endl;
            throw std::runtime_error("pd_test_datetime_array_sorting failed: argsort first");
        }
        if (indices.getElementAt({3}) != 1) {
            std::cout << "  [FAIL] : argsort: last should be index 1 (NaT)" << std::endl;
            throw std::runtime_error("pd_test_datetime_array_sorting failed: NaT position");
        }

searchsorted (pd_test_1_all.cpp:18958)

    // =========================================================================
    // Search Tests
    // =========================================================================

    void pd_test_range_index_searchsorted() {
        std::cout << "========= searchsorted ================================ ";

        pandas::RangeIndex ri(0, 10, 2);  // [0, 2, 4, 6, 8]

        bool passed = (ri.searchsorted(4, "left") == 2 &&
                      ri.searchsorted(4, "right") == 3 &&
                      ri.searchsorted(3, "left") == 2 &&   // 3 would go between 2 and 4
                      ri.searchsorted(-1, "left") == 0 &&  // Before all
                      ri.searchsorted(10, "left") == 5);   // After all

        if (!passed) {
            std::cout << "  [FAIL] : searchsorted" << std::endl;
            throw std::runtime_error("pd_test_range_index_searchsorted failed");
        }

sort_values (pd_test_1_all.cpp:6408)

        void pd_test_dataframe_sorting() {
            std::cout << "========= sorting ==========================";

            std::map<std::string, std::vector<numpy::float64>> data;
            data["A"] = {3.0, 1.0, 4.0, 1.0, 5.0};
            data["B"] = {9.0, 2.0, 6.0, 5.0, 3.0};

            pandas::DataFrame df(data);

            // Test sort_values ascending
            auto sorted_asc = df.sort_values("A", true);
            // First value should be smallest (1.0)
            std::string first_val = sorted_asc["A"].get_value_str(0);
            if (std::stod(first_val) != 1.0) {
                std::cout << "  [FAIL] : in pd_test_dataframe_sorting() : sort_values asc first != 1" << std::endl;
                throw std::runtime_error("pd_test_dataframe_sorting failed: sort_values asc first != 1");
            }

            // Test sort_values descending
            auto sorted_desc = df.sort_values("A", false);
            first_val = sorted_desc["A"].get_value_str(0);

T (pd_test_1_all.cpp:128)

            throw std::runtime_error("pd_test_boolean_array_kleene_and failed: NA & F");
        }

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_boolean_array_kleene_or() {
        std::cout << "========= BooleanArray: Kleene OR ======================= ";

        // Kleene OR truth table:
        // T | T = T, T | F = T, T | NA = T (True dominates)
        // F | T = T, F | F = F, F | NA = NA
        // NA | T = T, NA | F = NA, NA | NA = NA

        pandas::BooleanArray t({std::optional<bool>(true)});
        pandas::BooleanArray f({std::optional<bool>(false)});
        pandas::BooleanArray na({std::nullopt});

        // T | NA = T (True dominates)
        auto tna = (t | na);
        if (!tna[0].has_value() || !tna[0].value()) {

swapaxes (pd_test_3_all.cpp:2276)

    auto sorted_desc = arr.sort_values(false, "last");
    if (*sorted_desc[0] != "c" || *sorted_desc[1] != "b" ||
        *sorted_desc[2] != "a" || sorted_desc[3].has_value()) {
        throw std::runtime_error("sort_values descending failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_swapaxes() {
    std::cout << "========= CategoricalArray.swapaxes() =================";

    std::vector<std::optional<std::string>> values = {"a", "b", "c"};
    pandas::CategoricalArray arr(values);

    auto result = arr.swapaxes(0, 0);
    if (result.size() != 3) {
        throw std::runtime_error("swapaxes failed");
    }

    bool threw = false;

transpose (pd_test_1_all.cpp:16648)

                std::cout << "  [FAIL] : in pd_test_ndframe_transpose() : T_() size" << std::endl;
                throw std::runtime_error("pd_test_ndframe_transpose failed: T_() size");
            }

            passed = transposed[0] == 1 && transposed[1] == 2 && transposed[2] == 3;
            if (!passed) {
                std::cout << "  [FAIL] : in pd_test_ndframe_transpose() : T_() values" << std::endl;
                throw std::runtime_error("pd_test_ndframe_transpose failed: T_() values");
            }

            // Test transpose() alias
            auto transposed2 = s.transpose();
            passed = transposed2.size() == s.size();
            if (!passed) {
                std::cout << "  [FAIL] : in pd_test_ndframe_transpose() : transpose() size" << std::endl;
                throw std::runtime_error("pd_test_ndframe_transpose failed: transpose() size");
            }

            std::cout << " -> tests passed" << std::endl;
        }

concat (pd_test_1_all.cpp:17717)

}

void pd_test_period_index_concat() {
    std::cout << "========= concat factory ==============================";

    std::vector<int64_t> ordinals1 = {0, 1};
    std::vector<int64_t> ordinals2 = {2, 3};
    pandas::PeriodIndex idx1(ordinals1, "D");
    pandas::PeriodIndex idx2(ordinals2, "D");

    pandas::PeriodIndex concatenated = pandas::PeriodIndex::concat({idx1, idx2});

    bool passed = (concatenated.size() == 4);
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_period_index_concat()" << std::endl;
        throw std::runtime_error("pd_test_period_index_concat failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

concat_merge (pd_test_3_all.cpp:26636)

    const auto& mi = result.multiindex();
    if (mi.get_level_values_str(0)[0] != "x") throw std::runtime_error("Key 'x' not found at level 0");
    if (mi.get_level_values_str(0)[2] != "y") throw std::runtime_error("Key 'y' not found at level 0");
    std::cout << "    -> test passed" << std::endl;
}

void test_categorical_array_concat_merge() {
    std::cout << "  test_categorical_array_concat_merge" << std::endl;
    auto cat1 = CategoricalArray::from_codes({0, 1}, {"a", "b"});
    auto cat2 = CategoricalArray::from_codes({0, 1}, {"b", "c"});
    auto result = CategoricalArray::concat_merge({cat1, cat2});
    if (result.categories().size() != 3) throw std::runtime_error("Expected 3 merged categories");
    if (result.categories()[0] != "a" || result.categories()[1] != "b" || result.categories()[2] != "c")
        throw std::runtime_error("Merged categories wrong");
    if (result.size() != 4) throw std::runtime_error("Expected 4 elements");
    std::cout << "    -> test passed" << std::endl;
}

int pd_test_concat_ext_main() {
    try {
        std::cout << "========= concat extension tests =========" << std::endl;

shift (pd_test_1_all.cpp:5188)

            // First element should be NaN
            val = d["A"].get_value_str(0);
            passed = std::isnan(std::stod(val));
            if (!passed) {
                std::cout << "  [FAIL] : in pd_test_arithmetic_dataframe_diff_shift() : diff NaN failed" << std::endl;
                throw std::runtime_error("pd_test_arithmetic_dataframe_diff_shift failed: diff NaN failed");
            }

            // shift: [NaN, 1, 3, 6]
            auto s = df.shift();
            val = s["A"].get_value_str(1);
            passed = std::abs(std::stod(val) - 1.0) < 0.001;
            if (!passed) {
                std::cout << "  [FAIL] : in pd_test_arithmetic_dataframe_diff_shift() : shift failed" << std::endl;
                throw std::runtime_error("pd_test_arithmetic_dataframe_diff_shift failed: shift failed");
            }

            std::cout << " -> tests passed" << std::endl;
        }

to_list (pd_test_1_all.cpp:10247)

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_extension_index_to_list() {
    std::cout << "========= to_list =========================";

    pandas::CategoricalArray arr({"x", "y", "z"});
    pandas::CategoricalIndex idx(arr);

    auto list = idx.to_list();

    bool passed = (list.size() == 3 &&
                   list[0].has_value() && *list[0] == "x" &&
                   list[1].has_value() && *list[1] == "y" &&
                   list[2].has_value() && *list[2] == "z");
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_extension_index_to_list() : to_list check failed" << std::endl;
        throw std::runtime_error("pd_test_extension_index_to_list failed");
    }

to_numpy (pd_test_1_all.cpp:16764)

        // =====================================================================
        // to_numpy Tests
        // =====================================================================

        void pd_test_ndframe_to_numpy() {
            std::cout << "========= to_numpy =============================================" << std::endl;

            pandas::Series<int> s({10, 20, 30});

            auto arr = s.to_numpy();

            bool passed = arr.getSize() == 3;
            if (!passed) {
                std::cout << "  [FAIL] : in pd_test_ndframe_to_numpy() : size" << std::endl;
                throw std::runtime_error("pd_test_ndframe_to_numpy failed: size");
            }

            passed = arr.getElementAt({0}) == 10 && arr.getElementAt({1}) == 20 && arr.getElementAt({2}) == 30;
            if (!passed) {
                std::cout << "  [FAIL] : in pd_test_ndframe_to_numpy() : values" << std::endl;

to_string (pd_test_1_all.cpp:2693)

        pandas::PeriodArray arr_m(std::vector<std::string>{
            "2020-01",
            "NaT",
            "2025-06"
        }, "M");

        // Year
        auto years = arr_m.year();
        auto y0 = years[0];
        if (!y0.has_value() || y0.value() != 2020) {
            std::cout << "  [FAIL] : year[0] should be 2020, got " << (y0.has_value() ? std::to_string(y0.value()) : "NA") << std::endl;
            throw std::runtime_error("pd_test_period_array_year_month_quarter failed: year[0]");
        }

        auto y1 = years[1];
        if (y1.has_value()) {
            std::cout << "  [FAIL] : year[1] should be NA (NaT)" << std::endl;
            throw std::runtime_error("pd_test_period_array_year_month_quarter failed: year[1] should be NA");
        }

        auto y2 = years[2];

tolist (pd_test_3_all.cpp:2300)

        threw = true;
    }
    if (!threw) {
        throw std::runtime_error("swapaxes should throw for invalid axes");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_to_list() {
    std::cout << "========= CategoricalArray.to_list()/tolist() =========";

    std::vector<std::optional<std::string>> values = {"a", "b", std::nullopt, "c"};
    pandas::CategoricalArray arr(values);

    auto list = arr.to_list();
    if (list.size() != 4 || *list[0] != "a" || *list[1] != "b" ||
        list[2].has_value() || *list[3] != "c") {
        throw std::runtime_error("to_list failed");
    }

astype (pd_test_1_all.cpp:21292)

            std::cout << "========= astype all columns to float64 =============";

            // Create DataFrame with int64 columns
            std::map<std::string, std::vector<numpy::int64>> data;
            data["A"] = {1, 2, 3, 4, 5};
            data["B"] = {10, 20, 30, 40, 50};

            pandas::DataFrame df(data);

            // Convert all columns to float64
            pandas::DataFrame df_float = df.astype("float64");

            // Verify dtype changed
            pandas::Series<std::string> dtypes = df_float.dtypes();

            bool passed = true;
            if (dtypes[static_cast<size_t>(0)] != "float64") {
                std::cout << "  [FAIL] : in pd_test_astype_all_columns_to_float64() : column A dtype is " << dtypes[static_cast<size_t>(0)] << ", expected float64" << std::endl;
                passed = false;
            }
            if (dtypes[static_cast<size_t>(1)] != "float64") {

astype_codes (pd_test_3_all.cpp:1822)

    std::cout << "========= CategoricalArray.astype() ==================";

    std::vector<std::optional<std::string>> values = {"a", "b", "c", "a", std::nullopt};
    pandas::CategoricalArray arr(values);

    auto str_result = arr.astype<std::string>("str");
    if (str_result.size() != 5 || !str_result[0].has_value() || *str_result[0] != "a" || str_result[4].has_value()) {
        throw std::runtime_error("astype failed");
    }

    auto codes = arr.astype_codes();
    if (codes.getSize() != 5) {
        throw std::runtime_error("astype_codes failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_check_ordered() {
    std::cout << "========= CategoricalArray.check_for_ordered() ========";

copy (pd_test_1_all.cpp:5798)

// ============================================================================
// Copy/Rename Tests
// ============================================================================

void pd_test_categorical_index_copy() {
    std::cout << "========= copy ========================================";

    pandas::CategoricalArray arr({"a", "b", "c"});
    pandas::CategoricalIndex idx(arr, "original");

    pandas::CategoricalIndex copied = idx.copy();

    bool passed = (copied.size() == idx.size() && copied.name() == idx.name() &&
                   copied.categories() == idx.categories() && copied.ordered() == idx.ordered());
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_categorical_index_copy()" << std::endl;
        throw std::runtime_error("pd_test_categorical_index_copy failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

view (pd_test_3_all.cpp:2147)

        throw std::runtime_error("memory_usage shallow too small");
    }
    if (deep < shallow) {
        throw std::runtime_error("memory_usage deep should be >= shallow");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_ravel_view() {
    std::cout << "========= CategoricalArray.ravel()/view() =============";

    std::vector<std::optional<std::string>> values = {"a", "b", "c"};
    pandas::CategoricalArray arr(values);

    auto raveled = arr.ravel();
    if (raveled.size() != 3 || !raveled.equals(arr)) {
        throw std::runtime_error("ravel failed");
    }

    auto viewed = arr.view();

duplicated (pd_test_1_all.cpp:10583)

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_extension_index_duplicated() {
    std::cout << "========= duplicated =========================";

    pandas::CategoricalArray arr({"a", "b", "a", "c", "a"});
    pandas::CategoricalIndex idx(arr);

    auto dup_mask = idx.duplicated("first");

    bool passed = (dup_mask.getElementAt({0}) == false &&
                   dup_mask.getElementAt({1}) == false &&
                   dup_mask.getElementAt({2}) == true &&
                   dup_mask.getElementAt({3}) == false &&
                   dup_mask.getElementAt({4}) == true);
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_extension_index_duplicated() : duplicated check failed" << std::endl;
        throw std::runtime_error("pd_test_extension_index_duplicated failed");
    }

isin (pd_test_1_all.cpp:5938)

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_categorical_index_isin() {
    std::cout << "========= inherited isin ==============================";

    pandas::CategoricalArray arr({"a", "b", "c", "d"});
    pandas::CategoricalIndex idx(arr);

    std::vector<std::string> values = {"a", "c"};
    numpy::NDArray<numpy::bool_> mask = idx.isin(values);

    bool passed = (mask.getSize() == 4 &&
                   mask.getElementAt({0}) == true &&   // a
                   mask.getElementAt({1}) == false &&  // b
                   mask.getElementAt({2}) == true &&   // c
                   mask.getElementAt({3}) == false);   // d
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_categorical_index_isin()" << std::endl;
        throw std::runtime_error("pd_test_categorical_index_isin failed");
    }

unique (pd_test_1_all.cpp:1345)

        pandas::DatetimeArray arr(std::vector<std::string>{
            "2023-01-01",
            "2023-06-15",
            "2023-01-01",
            "NaT",
            "2023-06-15",
            "NaT"
        });

        // unique
        auto uniq = arr.unique();
        // Should have: NaT, 2023-01-01, 2023-06-15 (3 unique values)
        if (uniq.size() != 3) {
            std::cout << "  [FAIL] : unique size should be 3, got " << uniq.size() << std::endl;
            throw std::runtime_error("pd_test_datetime_array_unique failed: size");
        }

        // factorize
        auto [codes, uniques] = arr.factorize();
        // Codes for NaT should be -1
        if (codes.getElementAt({3}) != -1) {

is_na (pd_test_1_all.cpp:51)

    void pd_test_boolean_array_na_handling() {
        std::cout << "========= BooleanArray: NA handling ======================= ";

        pandas::BooleanArray arr({
            std::optional<bool>(true),
            std::nullopt,  // NA at index 1
            std::optional<bool>(false)
        });

        if (!arr.is_na(1)) {
            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : is_na(1) should be true" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(1) should be true");
        }

        if (arr.is_na(0)) {
            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : is_na(0) should be false" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(0) should be false");
        }

        if (!arr.has_na()) {

argmax (pd_test_1_all.cpp:1323)

        }

        // argmin
        auto min_idx = arr.argmin();
        if (!min_idx.has_value() || min_idx.value() != 2) {
            std::cout << "  [FAIL] : argmin should be 2 (2023-01-01)" << std::endl;
            throw std::runtime_error("pd_test_datetime_array_sorting failed: argmin");
        }

        // argmax
        auto max_idx = arr.argmax();
        if (!max_idx.has_value() || max_idx.value() != 3) {
            std::cout << "  [FAIL] : argmax should be 3 (2023-12-31)" << std::endl;
            throw std::runtime_error("pd_test_datetime_array_sorting failed: argmax");
        }

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_datetime_array_unique() {
        std::cout << "========= DatetimeArray: unique/factorize ======================= ";

argmin (pd_test_1_all.cpp:1316)

        if (indices.getElementAt({0}) != 2) {
            std::cout << "  [FAIL] : argsort: first should be index 2 (2023-01-01)" << std::endl;
            throw std::runtime_error("pd_test_datetime_array_sorting failed: argsort first");
        }
        if (indices.getElementAt({3}) != 1) {
            std::cout << "  [FAIL] : argsort: last should be index 1 (NaT)" << std::endl;
            throw std::runtime_error("pd_test_datetime_array_sorting failed: NaT position");
        }

        // argmin
        auto min_idx = arr.argmin();
        if (!min_idx.has_value() || min_idx.value() != 2) {
            std::cout << "  [FAIL] : argmin should be 2 (2023-01-01)" << std::endl;
            throw std::runtime_error("pd_test_datetime_array_sorting failed: argmin");
        }

        // argmax
        auto max_idx = arr.argmax();
        if (!max_idx.has_value() || max_idx.value() != 3) {
            std::cout << "  [FAIL] : argmax should be 3 (2023-12-31)" << std::endl;
            throw std::runtime_error("pd_test_datetime_array_sorting failed: argmax");

as_ordered (pd_test_1_all.cpp:791)

            unordered.min();
        } catch (const std::exception&) {
            threw = true;
        }
        if (!threw) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : unordered min should throw" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: unordered min should throw");
        }

        // Test as_ordered / as_unordered
        pandas::CategoricalArray reordered = unordered.as_ordered();
        if (!reordered.ordered()) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : as_ordered failed" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: as_ordered failed");
        }

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_categorical_array_comparisons() {
        std::cout << "========= CategoricalArray: comparisons ======================= ";

as_unordered (pd_test_1_all.cpp:778)

        }

        // Test max
        std::optional<std::string> max_val = arr.max();
        if (!max_val.has_value() || *max_val != "high") {
            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : max != 'high'" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: max != 'high'");
        }

        // Test unordered throws for min/max
        pandas::CategoricalArray unordered = arr.as_unordered();
        bool threw = false;
        try {
            unordered.min();
        } catch (const std::exception&) {
            threw = true;
        }
        if (!threw) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_ordered_operations() : unordered min should throw" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: unordered min should throw");
        }

categories (pd_test_1_all.cpp:389)

        std::vector<std::optional<std::string>> vals = {
            std::optional<std::string>("low"),
            std::optional<std::string>("high"),
            std::optional<std::string>("medium")
        };
        pandas::CategoricalArray arr3(vals, cats, true);  // ordered
        if (!arr3.ordered()) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_constructors() : should be ordered" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_constructors failed: should be ordered");
        }
        if (arr3.categories().size() != 3) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_constructors() : categories size != 3" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_constructors failed: categories size != 3");
        }

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_categorical_array_from_codes() {
        std::cout << "========= CategoricalArray: from_codes ======================= ";

check_for_ordered (pd_test_3_all.cpp:1831)

    auto codes = arr.astype_codes();
    if (codes.getSize() != 5) {
        throw std::runtime_error("astype_codes failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_check_ordered() {
    std::cout << "========= CategoricalArray.check_for_ordered() ========";

    std::vector<std::optional<std::string>> values = {"a", "b", "c"};
    pandas::CategoricalArray unordered_arr(values, false);
    pandas::CategoricalArray ordered_arr(values, {"a", "b", "c"}, true);

    bool threw = false;
    try {
        unordered_arr.check_for_ordered("test_op");
    } catch (const std::exception&) {
        threw = true;

check_for_ordered (pd_test_3_all.cpp:1831)

    auto codes = arr.astype_codes();
    if (codes.getSize() != 5) {
        throw std::runtime_error("astype_codes failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_check_ordered() {
    std::cout << "========= CategoricalArray.check_for_ordered() ========";

    std::vector<std::optional<std::string>> values = {"a", "b", "c"};
    pandas::CategoricalArray unordered_arr(values, false);
    pandas::CategoricalArray ordered_arr(values, {"a", "b", "c"}, true);

    bool threw = false;
    try {
        unordered_arr.check_for_ordered("test_op");
    } catch (const std::exception&) {
        threw = true;

check_for_ordered (pd_test_3_all.cpp:1831)

    auto codes = arr.astype_codes();
    if (codes.getSize() != 5) {
        throw std::runtime_error("astype_codes failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_check_ordered() {
    std::cout << "========= CategoricalArray.check_for_ordered() ========";

    std::vector<std::optional<std::string>> values = {"a", "b", "c"};
    pandas::CategoricalArray unordered_arr(values, false);
    pandas::CategoricalArray ordered_arr(values, {"a", "b", "c"}, true);

    bool threw = false;
    try {
        unordered_arr.check_for_ordered("test_op");
    } catch (const std::exception&) {
        threw = true;

codes (pd_test_1_all.cpp:473)

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_categorical_array_codes_property() {
        std::cout << "========= CategoricalArray: codes property ======================= ";

        std::vector<std::string> cats = {"x", "y", "z"};
        std::vector<numpy::int32> codes = {0, 1, 2, 1, 0};
        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);

        numpy::NDArray<numpy::int32> arr_codes = arr.codes();
        if (arr_codes.getSize() != 5) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_codes_property() : codes size != 5" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_codes_property failed: codes size != 5");
        }

        // Check codes match
        for (size_t i = 0; i < codes.size(); ++i) {
            if (arr_codes.getElementAt({i}) != codes[i]) {
                std::cout << "  [FAIL] : in pd_test_categorical_array_codes_property() : code mismatch at " << i << std::endl;
                throw std::runtime_error("pd_test_categorical_array_codes_property failed: code mismatch");

delete_ (pd_test_1_all.cpp:10501)

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_extension_index_delete() {
    std::cout << "========= delete_ =========================";

    pandas::CategoricalArray arr({"a", "b", "c", "d"});
    pandas::CategoricalIndex idx(arr);

    auto deleted = idx.delete_(1);
    auto v0 = deleted[0];
    auto v1 = deleted[1];
    auto v2 = deleted[2];

    bool passed = (deleted.size() == 3 &&
                   v0.has_value() && *v0 == "a" &&
                   v1.has_value() && *v1 == "c" &&
                   v2.has_value() && *v2 == "d");
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_extension_index_delete() : delete_ check failed" << std::endl;

delete_ (pd_test_1_all.cpp:10501)

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_extension_index_delete() {
    std::cout << "========= delete_ =========================";

    pandas::CategoricalArray arr({"a", "b", "c", "d"});
    pandas::CategoricalIndex idx(arr);

    auto deleted = idx.delete_(1);
    auto v0 = deleted[0];
    auto v1 = deleted[1];
    auto v2 = deleted[2];

    bool passed = (deleted.size() == 3 &&
                   v0.has_value() && *v0 == "a" &&
                   v1.has_value() && *v1 == "c" &&
                   v2.has_value() && *v2 == "d");
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_extension_index_delete() : delete_ check failed" << std::endl;

dtype (pd_test_1_all.cpp:295)

            throw std::runtime_error("pd_test_boolean_array_reductions failed: mean");
        }

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_boolean_array_dtype() {
        std::cout << "========= BooleanArray: dtype ======================= ";

        pandas::BooleanArray arr;
        if (arr.dtype().name() != "boolean") {
            std::cout << "  [FAIL] : in pd_test_boolean_array_dtype() : dtype name should be 'boolean'" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_dtype failed: dtype name");
        }

        if (arr.dtype().kind() != "b") {
            std::cout << "  [FAIL] : in pd_test_boolean_array_dtype() : dtype kind should be 'b'" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_dtype failed: dtype kind");
        }

        std::cout << " -> tests passed" << std::endl;

empty (pd_test_1_all.cpp:941)

#include "../pandas/pd_config.h"

namespace dataframe_tests {

namespace dataframe_tests_config {

    void pd_test_config_version() {
        std::cout << "========= df_config: version info ======================= ";
        const char* version = pandas::DataFrameInfo::version();
        if (version == nullptr || std::string(version).empty()) {
            std::cout << "[FAIL] : in pd_test_config_version() : version is null or empty" << std::endl;
            throw std::runtime_error("pd_test_config_version failed: version is null or empty");
        }
        std::cout << "-> tests passed" << std::endl;
    }

    void pd_test_config_na_repr() {
        std::cout << "========= df_config: NA representation ======================= ";
        const char* na_repr = pandas::DataFrameConfig::get_na_repr();
        if (na_repr == nullptr) {

factorize (pd_test_1_all.cpp:1353)

        // unique
        auto uniq = arr.unique();
        // Should have: NaT, 2023-01-01, 2023-06-15 (3 unique values)
        if (uniq.size() != 3) {
            std::cout << "  [FAIL] : unique size should be 3, got " << uniq.size() << std::endl;
            throw std::runtime_error("pd_test_datetime_array_unique failed: size");
        }

        // factorize
        auto [codes, uniques] = arr.factorize();
        // Codes for NaT should be -1
        if (codes.getElementAt({3}) != -1) {
            std::cout << "  [FAIL] : factorize: NaT code should be -1" << std::endl;
            throw std::runtime_error("pd_test_datetime_array_unique failed: NaT code");
        }
        // Same values should have same codes
        if (codes.getElementAt({0}) != codes.getElementAt({2})) {
            std::cout << "  [FAIL] : factorize: 2023-01-01 values should have same code" << std::endl;
            throw std::runtime_error("pd_test_datetime_array_unique failed: same code");
        }

has_na (pd_test_1_all.cpp:61)

        if (!arr.is_na(1)) {
            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : is_na(1) should be true" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(1) should be true");
        }

        if (arr.is_na(0)) {
            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : is_na(0) should be false" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(0) should be false");
        }

        if (!arr.has_na()) {
            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : has_na() should be true" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_na_handling failed: has_na() should be true");
        }

        if (arr.count() != 2) {
            std::cout << "  [FAIL] : in pd_test_boolean_array_na_handling() : count() should be 2" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_na_handling failed: count() should be 2");
        }

        std::cout << " -> tests passed" << std::endl;

memory_usage (pd_test_1_all.cpp:27063)

        }

        std::cout << "====================================== [OK] pd_test_value_counts test suite ========================== " << std::endl;
        return 0;
    }

} // namespace dataframe_tests
// ------------------- pd_test_value_counts.cpp (end) -----------------------------

// ------------------- pd_test_memory_usage.cpp (start) -----------------------------
// Tests for DataFrame.memory_usage() - pandas-compatible memory usage reporting

namespace dataframe_tests {
    namespace dataframe_tests_memory_usage {

        void pd_test_memory_usage_basic() {
            std::cout << "========= basic memory_usage =======================";

            // Create a simple DataFrame with multiple columns
            std::map<std::string, std::vector<double>> data;
            data["A"] = {1.0, 2.0, 3.0, 4.0, 5.0};

name (pd_test_1_all.cpp:295)

            throw std::runtime_error("pd_test_boolean_array_reductions failed: mean");
        }

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_boolean_array_dtype() {
        std::cout << "========= BooleanArray: dtype ======================= ";

        pandas::BooleanArray arr;
        if (arr.dtype().name() != "boolean") {
            std::cout << "  [FAIL] : in pd_test_boolean_array_dtype() : dtype name should be 'boolean'" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_dtype failed: dtype name");
        }

        if (arr.dtype().kind() != "b") {
            std::cout << "  [FAIL] : in pd_test_boolean_array_dtype() : dtype kind should be 'b'" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_dtype failed: dtype kind");
        }

        std::cout << " -> tests passed" << std::endl;

nbytes (pd_test_1_all.cpp:6214)

            }

            // Test empty DataFrame
            pandas::DataFrame empty_df;
            if (!empty_df.empty()) {
                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : should be empty" << std::endl;
                throw std::runtime_error("pd_test_dataframe_properties failed: should be empty");
            }

            // Test nbytes > 0 for non-empty
            if (df.nbytes() == 0) {
                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : nbytes should be > 0" << std::endl;
                throw std::runtime_error("pd_test_dataframe_properties failed: nbytes should be > 0");
            }

            // Test columns index
            if (df.columns().size() != 3) {
                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : columns size != 3" << std::endl;
                throw std::runtime_error("pd_test_dataframe_properties failed: columns size != 3");
            }

ndim (pd_test_1_all.cpp:6195)

            pandas::DataFrame df(data);

            // Test shape
            auto shape = df.shape();
            if (shape.size() != 2 || shape[0] != 4 || shape[1] != 3) {
                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : shape mismatch" << std::endl;
                throw std::runtime_error("pd_test_dataframe_properties failed: shape mismatch");
            }

            // Test ndim
            if (df.ndim() != 2) {
                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : ndim != 2" << std::endl;
                throw std::runtime_error("pd_test_dataframe_properties failed: ndim != 2");
            }

            // Test empty
            if (df.empty()) {
                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : should not be empty" << std::endl;
                throw std::runtime_error("pd_test_dataframe_properties failed: should not be empty");
            }

ordered (pd_test_1_all.cpp:359)

    void pd_test_categorical_array_constructors() {
        std::cout << "========= CategoricalArray: constructors ======================= ";

        // Default constructor
        pandas::CategoricalArray arr1;
        if (arr1.size() != 0) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_constructors() : default constructor size != 0" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_constructors failed: default constructor size != 0");
        }
        if (arr1.ordered()) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_constructors() : default should be unordered" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_constructors failed: default should be unordered");
        }

        // Constructor from values (infer categories)
        std::vector<std::optional<std::string>> values = {
            std::optional<std::string>("a"),
            std::optional<std::string>("b"),
            std::optional<std::string>("a"),
            std::optional<std::string>("c")

ravel (pd_test_3_all.cpp:2147)

        throw std::runtime_error("memory_usage shallow too small");
    }
    if (deep < shallow) {
        throw std::runtime_error("memory_usage deep should be >= shallow");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_ravel_view() {
    std::cout << "========= CategoricalArray.ravel()/view() =============";

    std::vector<std::optional<std::string>> values = {"a", "b", "c"};
    pandas::CategoricalArray arr(values);

    auto raveled = arr.ravel();
    if (raveled.size() != 3 || !raveled.equals(arr)) {
        throw std::runtime_error("ravel failed");
    }

    auto viewed = arr.view();

remove_categories (pd_test_1_all.cpp:591)

    }

    void pd_test_categorical_array_remove_categories() {
        std::cout << "========= CategoricalArray: remove_categories ======================= ";

        std::vector<std::string> cats = {"a", "b", "c"};
        std::vector<numpy::int32> codes = {0, 1, 2, 1};  // a, b, c, b
        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);

        // Remove 'c' - values with 'c' become NA
        pandas::CategoricalArray result = arr.remove_categories({"c"});

        if (result.categories().size() != 2) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_remove_categories() : categories size != 2" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_remove_categories failed: categories size != 2");
        }

        // Element at index 2 should now be NA (was 'c')
        if (!result.is_na(2)) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_remove_categories() : removed category should be NA" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_remove_categories failed: removed category should be NA");

remove_unused_categories (pd_test_1_all.cpp:737)

        std::cout << " -> tests passed" << std::endl;
    }

    void pd_test_categorical_array_remove_unused_categories() {
        std::cout << "========= CategoricalArray: remove_unused_categories ======================= ";

        std::vector<std::string> cats = {"a", "b", "c", "d"};
        std::vector<numpy::int32> codes = {0, 0, 2};  // a, a, c (b and d unused)
        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);

        pandas::CategoricalArray result = arr.remove_unused_categories();

        // Only 'a' and 'c' should remain
        if (result.categories().size() != 2) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_remove_unused_categories() : categories size != 2" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_remove_unused_categories failed: categories size != 2");
        }

        // Values should be preserved
        std::optional<std::string> val0 = result[0];
        std::optional<std::string> val2 = result[2];

reorder_categories (pd_test_1_all.cpp:695)

    void pd_test_categorical_array_reorder_categories() {
        std::cout << "========= CategoricalArray: reorder_categories ======================= ";

        std::vector<std::string> cats = {"a", "b", "c"};
        std::vector<numpy::int32> codes = {0, 1, 2};  // a, b, c
        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);

        // Reorder categories
        std::vector<std::string> new_order = {"c", "b", "a"};
        pandas::CategoricalArray result = arr.reorder_categories(new_order);

        // Check categories are reordered
        const std::vector<std::string>& result_cats = result.categories();
        if (result_cats[0] != "c" || result_cats[1] != "b" || result_cats[2] != "a") {
            std::cout << "  [FAIL] : in pd_test_categorical_array_reorder_categories() : categories not reordered" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_reorder_categories failed: categories not reordered");
        }

        // Values should be preserved
        std::optional<std::string> val0 = result[0];

repeat (pd_test_3_all.cpp:2166)

    auto viewed = arr.view();
    if (viewed.size() != 3 || !viewed.equals(arr)) {
        throw std::runtime_error("view failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_repeat() {
    std::cout << "========= CategoricalArray.repeat() ===================";

    std::vector<std::optional<std::string>> values = {"a", "b"};
    pandas::CategoricalArray arr(values);

    auto result = arr.repeat(3);
    if (result.size() != 6 || *result[0] != "a" || *result[2] != "a" ||
        *result[3] != "b" || *result[5] != "b") {
        throw std::runtime_error("repeat scalar failed");
    }

repeat (pd_test_3_all.cpp:2166)

    auto viewed = arr.view();
    if (viewed.size() != 3 || !viewed.equals(arr)) {
        throw std::runtime_error("view failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_repeat() {
    std::cout << "========= CategoricalArray.repeat() ===================";

    std::vector<std::optional<std::string>> values = {"a", "b"};
    pandas::CategoricalArray arr(values);

    auto result = arr.repeat(3);
    if (result.size() != 6 || *result[0] != "a" || *result[2] != "a" ||
        *result[3] != "b" || *result[5] != "b") {
        throw std::runtime_error("repeat scalar failed");
    }

repr (pd_test_1_all.cpp:10906)

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_extension_index_repr() {
    std::cout << "========= repr =========================";

    pandas::CategoricalArray arr({"a", "b", "c"});
    // Use ExtensionIndex<CategoricalArray> directly to test base class repr
    pandas::ExtensionIndex<pandas::CategoricalArray> idx(arr, "test");

    std::string repr_str = idx.repr();

    bool passed = (!repr_str.empty() && repr_str.find("ExtensionIndex") != std::string::npos);
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_extension_index_repr() : repr check failed" << std::endl;
        throw std::runtime_error("pd_test_extension_index_repr failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

reshape (pd_test_3_all.cpp:2186)

    auto result2 = arr.repeat({1, 2});
    if (result2.size() != 3 || *result2[0] != "a" || *result2[1] != "b" || *result2[2] != "b") {
        throw std::runtime_error("repeat array failed");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_reshape() {
    std::cout << "========= CategoricalArray.reshape() ==================";

    std::vector<std::optional<std::string>> values = {"a", "b", "c", "d"};
    pandas::CategoricalArray arr(values);

    auto result = arr.reshape({4});
    if (result.size() != 4) {
        throw std::runtime_error("reshape failed");
    }

    bool threw = false;

set_categories (pd_test_1_all.cpp:623)

    void pd_test_categorical_array_set_categories() {
        std::cout << "========= CategoricalArray: set_categories ======================= ";

        std::vector<std::string> cats = {"a", "b"};
        std::vector<numpy::int32> codes = {0, 1, 0};  // a, b, a
        pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);

        // Set new categories (values not in new categories become NA)
        std::vector<std::string> new_cats = {"a", "c"};  // 'b' removed, 'c' added
        pandas::CategoricalArray result = arr.set_categories(new_cats);

        if (result.categories().size() != 2) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_set_categories() : categories size != 2" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_set_categories failed: categories size != 2");
        }

        // Element at index 1 should be NA (was 'b', now not in categories)
        if (!result.is_na(1)) {
            std::cout << "  [FAIL] : in pd_test_categorical_array_set_categories() : 'b' value should be NA" << std::endl;
            throw std::runtime_error("pd_test_categorical_array_set_categories failed: 'b' value should be NA");

set_name (pd_test_1_all.cpp:11798)

                throw std::runtime_error("pd_test_index_vector_constructor failed");
            }

            std::cout << " -> tests passed" << std::endl;
        }

        void pd_test_index_copy_constructor() {
            std::cout << "========= copy constructor ============================";

            pandas::Index<numpy::int64> idx1{1, 2, 3};
            idx1.set_name("original");

            pandas::Index<numpy::int64> idx2(idx1);

            bool passed = (idx2.size() == 3);
            passed = passed && (idx2.name().value() == "original");
            passed = passed && idx2.equals(idx1);

            if (!passed) {
                std::cout << "  [FAIL] : in pd_test_index_copy_constructor() : copy failed" << std::endl;
                throw std::runtime_error("pd_test_index_copy_constructor failed");

set_ordered (pd_test_3_all.cpp:2210)

        threw = true;
    }
    if (!threw) {
        throw std::runtime_error("reshape should throw for incompatible shape");
    }

    std::cout << " -> tests passed" << std::endl;
}

void pd_test_3_all_categorical_set_ordered() {
    std::cout << "========= CategoricalArray.set_ordered() ==============";

    std::vector<std::optional<std::string>> values = {"a", "b", "c"};
    pandas::CategoricalArray arr(values, false);

    if (arr.ordered()) {
        throw std::runtime_error("initial should be unordered");
    }

    auto ordered = arr.set_ordered(true);
    if (!ordered.ordered()) {

shape (pd_test_1_all.cpp:6188)

            std::cout << "========= properties =======================";

            std::map<std::string, std::vector<numpy::float64>> data;
            data["A"] = {1.0, 2.0, 3.0, 4.0};
            data["B"] = {5.0, 6.0, 7.0, 8.0};
            data["C"] = {9.0, 10.0, 11.0, 12.0};

            pandas::DataFrame df(data);

            // Test shape
            auto shape = df.shape();
            if (shape.size() != 2 || shape[0] != 4 || shape[1] != 3) {
                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : shape mismatch" << std::endl;
                throw std::runtime_error("pd_test_dataframe_properties failed: shape mismatch");
            }

            // Test ndim
            if (df.ndim() != 2) {
                std::cout << "  [FAIL] : in pd_test_dataframe_properties() : ndim != 2" << std::endl;
                throw std::runtime_error("pd_test_dataframe_properties failed: ndim != 2");
            }

size (pd_test_1_all.cpp:22)

#include "../pandas/pd_boolean_array.h"

namespace dataframe_tests {

namespace dataframe_tests_boolean_array {
    void pd_test_boolean_array_constructors() {
        std::cout << "========= BooleanArray: constructors ======================= ";

        // Default constructor
        pandas::BooleanArray arr1;
        if (arr1.size() != 0) {
            std::cout << "  [FAIL] : in pd_test_boolean_array_constructors() : default constructor size != 0" << std::endl;
            throw std::runtime_error("pd_test_boolean_array_constructors failed: default constructor size != 0");
        }

        // Initializer list constructor
        pandas::BooleanArray arr2({
            std::optional<bool>(true),
            std::optional<bool>(false),
            std::nullopt,
            std::optional<bool>(true)

slice (pd_test_1_all.cpp:17546)

// ============================================================================
// Slicing / Indexing Tests
// ============================================================================

void pd_test_period_index_slice() {
    std::cout << "========= slice method ================================";

    std::vector<int64_t> ordinals = {0, 1, 2, 3, 4};
    pandas::PeriodIndex idx(ordinals, "D");

    pandas::PeriodIndex sliced = idx.slice(1, 4);

    bool passed = (sliced.size() == 3 &&
                   sliced[0].has_value() && *sliced[0] == 1);
    if (!passed) {
        std::cout << "  [FAIL] : in pd_test_period_index_slice()" << std::endl;
        throw std::runtime_error("pd_test_period_index_slice failed");
    }

    std::cout << " -> tests passed" << std::endl;
}