CategoricalIndex#
-
class pandas::CategoricalIndex#
Index class for axis labels in pandas data structures.
Example#
#include <pandas/pandas.h>
using namespace pandas;
// Create CategoricalIndex
CategoricalIndex<int64_t> idx({1, 2, 3}, "my_index");
size_t len = idx.size();
Constructors#
Signature |
Location |
Example |
|---|---|---|
|
pd_categorical_index.h:62 |
|
|
pd_categorical_index.h:72 |
|
|
pd_categorical_index.h:83 |
|
|
pd_categorical_index.h:96 |
|
|
pd_categorical_index.h:106 |
|
|
pd_categorical_index.h:113 |
Construction#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
static CategoricalIndex |
pd_categorical_index.h:153 |
Indexing / Selection#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
pd_categorical_index.h:945 |
||
|
CategoricalIndex |
pd_categorical_index.h:884 |
|
|
CategoricalIndex |
pd_categorical_index.h:900 |
|
|
size_t |
pd_categorical_index.h:1003 |
|
|
std::string |
pd_categorical_index.h:587 |
|
|
CategoricalIndex |
pd_categorical_index.h:1761 |
Data Manipulation#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
CategoricalIndex |
pd_categorical_index.h:868 |
|
|
std::pair<CategoricalIndex, numpy::NDArray<numpy::int64>> |
pd_categorical_index.h:1431 |
|
|
CategoricalIndex |
pd_categorical_index.h:482 |
|
|
CategoricalIndex |
pd_categorical_index.h:262 |
|
|
CategoricalIndex |
pd_categorical_index.h:276 |
|
|
CategoricalIndex |
pd_categorical_index.h:1557 |
Statistics#
Aggregation#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
std::unordered_map<GroupT, std::vector<size_t>> |
pd_categorical_index.h:1040 |
|
|
CategoricalIndex |
pd_categorical_index.h:460 |
Arithmetic#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
CategoricalIndex |
pd_categorical_index.h:308 |
Comparison#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
bool |
pd_categorical_index.h:611 |
Sorting#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
numpy::NDArray<numpy::int64> |
pd_categorical_index.h:1793 |
|
|
size_t |
pd_categorical_index.h:1518 |
|
|
CategoricalIndex |
pd_categorical_index.h:1854 |
Reshaping#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
CategoricalIndex |
pd_categorical_index.h:1741 |
Combining#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
pd_categorical_index.h:1182 |
Time Series#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
std::optional<std::string> |
pd_categorical_index.h:683 |
|
|
numpy::NDArray<numpy::int64> |
pd_categorical_index.h:745 |
|
|
numpy::NDArray<numpy::int64> |
pd_categorical_index.h:824 |
|
|
CategoricalIndex |
pd_categorical_index.h:1569 |
I/O#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
CategoricalIndex |
pd_categorical_index.h:1669 |
|
|
std::vector<std::optional<std::string>> |
pd_categorical_index.h:1704 |
|
|
std::vector<std::string> |
pd_categorical_index.h:1678 |
|
|
std::string |
pd_categorical_index.h:517 |
|
|
std::vector<std::optional<std::string>> |
pd_categorical_index.h:1700 |
Conversion#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
std::vector<std::string> |
pd_categorical_index.h:781 |
|
|
CategoricalIndex |
pd_categorical_index.h:473 |
|
|
CategoricalIndex |
pd_categorical_index.h:1139 |
|
|
std::vector<std::optional<std::string>> |
pd_categorical_index.h:1750 |
Type Checking#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
bool |
pd_categorical_index.h:1149 |
|
|
bool |
pd_categorical_index.h:1071 |
|
|
bool |
pd_categorical_index.h:1083 |
|
|
bool |
pd_categorical_index.h:1091 |
|
|
bool |
pd_categorical_index.h:1099 |
|
|
bool |
pd_categorical_index.h:1107 |
|
|
bool |
pd_categorical_index.h:1115 |
|
|
bool |
pd_categorical_index.h:1123 |
Other Methods#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
bool |
pd_categorical_index.h:645 |
|
|
bool |
pd_categorical_index.h:661 |
|
|
std::optional<size_t> |
pd_categorical_index.h:440 |
|
|
std::optional<size_t> |
pd_categorical_index.h:427 |
|
|
CategoricalIndex |
pd_categorical_index.h:369 |
|
|
CategoricalIndex |
pd_categorical_index.h:382 |
|
|
const std::vector<std::string>& |
pd_categorical_index.h:202 |
|
|
std::unique_ptr<IndexBase> |
pd_categorical_index.h:502 |
|
|
const numpy::NDArray<numpy::int32>& |
pd_categorical_index.h:214 |
|
|
std::vector<std::string> |
pd_categorical_index.h:920 |
|
|
bool |
pd_categorical_index.h:243 |
|
|
bool |
pd_categorical_index.h:1063 |
|
|
bool |
pd_categorical_index.h:630 |
|
|
std::string |
pd_categorical_index.h:494 |
|
|
std::string |
pd_categorical_index.h:1158 |
|
|
size_t |
pd_categorical_index.h:1354 |
|
|
size_t |
pd_categorical_index.h:234 |
|
|
bool |
pd_categorical_index.h:226 |
|
|
CategoricalIndex joined_index(joined_values, merged_cats, |
pd_categorical_index.h:1320 |
|
|
CategoricalIndex |
pd_categorical_index.h:1386 |
|
|
std::vector<std::optional<std::string>> |
pd_categorical_index.h:1414 |
|
|
CategoricalIndex |
pd_categorical_index.h:322 |
|
|
CategoricalIndex |
pd_categorical_index.h:334 |
|
|
CategoricalIndex |
pd_categorical_index.h:293 |
|
|
CategoricalIndex |
pd_categorical_index.h:1480 |
|
|
std::string |
pd_categorical_index.h:578 |
|
|
CategoricalIndex |
pd_categorical_index.h:1504 |
|
|
CategoricalIndex |
pd_categorical_index.h:350 |
|
|
std::pair<size_t, size_t> |
pd_categorical_index.h:1604 |
|
|
std::pair<size_t, size_t> |
pd_categorical_index.h:1627 |
|
|
CategoricalIndex |
pd_categorical_index.h:1873 |
|
|
std::pair<CategoricalIndex, numpy::NDArray<numpy::int64>> |
pd_categorical_index.h:1640 |
|
|
IndexTypeId |
pd_categorical_index.h:506 |
Internal Methods#
2 internal methods (prefixed with underscore)
Code Examples#
The following examples are extracted from the test suite.
CategoricalIndex (pd_test_2_all.cpp:20850)
20840 auto sgb = data.groupby(by);
20841 sgb.set_categorical_categories({"A", "B", "C"});
20842 sgb.set_index_name("cat_key");
20843
20844 pandas::Series<numpy::float64> result(values);
20845 std::vector<std::string> idx_labels = {"A", "B"};
20846 result.set_index(std::make_unique<pandas::Index<std::string>>(idx_labels));
20847
20848 sgb.apply_result_index(result);
20849
20850 // Should have CategoricalIndex (dtype_name() returns "category")
20851 check(result.index().dtype_name() == "category", "is_categorical_index");
20852}
20853
20854// =====================================================================
20855// Per-group expanding tests
20856// =====================================================================
20857
20858void test_series_groupby_expanding_sum() {
20859 std::cout << " -- test_series_groupby_expanding_sum --" << std::endl;
CategoricalIndex (pd_test_2_all.cpp:20850)
20840 auto sgb = data.groupby(by);
20841 sgb.set_categorical_categories({"A", "B", "C"});
20842 sgb.set_index_name("cat_key");
20843
20844 pandas::Series<numpy::float64> result(values);
20845 std::vector<std::string> idx_labels = {"A", "B"};
20846 result.set_index(std::make_unique<pandas::Index<std::string>>(idx_labels));
20847
20848 sgb.apply_result_index(result);
20849
20850 // Should have CategoricalIndex (dtype_name() returns "category")
20851 check(result.index().dtype_name() == "category", "is_categorical_index");
20852}
20853
20854// =====================================================================
20855// Per-group expanding tests
20856// =====================================================================
20857
20858void test_series_groupby_expanding_sum() {
20859 std::cout << " -- test_series_groupby_expanding_sum --" << std::endl;
CategoricalIndex (pd_test_2_all.cpp:20850)
20840 auto sgb = data.groupby(by);
20841 sgb.set_categorical_categories({"A", "B", "C"});
20842 sgb.set_index_name("cat_key");
20843
20844 pandas::Series<numpy::float64> result(values);
20845 std::vector<std::string> idx_labels = {"A", "B"};
20846 result.set_index(std::make_unique<pandas::Index<std::string>>(idx_labels));
20847
20848 sgb.apply_result_index(result);
20849
20850 // Should have CategoricalIndex (dtype_name() returns "category")
20851 check(result.index().dtype_name() == "category", "is_categorical_index");
20852}
20853
20854// =====================================================================
20855// Per-group expanding tests
20856// =====================================================================
20857
20858void test_series_groupby_expanding_sum() {
20859 std::cout << " -- test_series_groupby_expanding_sum --" << std::endl;
CategoricalIndex (pd_test_2_all.cpp:20850)
20840 auto sgb = data.groupby(by);
20841 sgb.set_categorical_categories({"A", "B", "C"});
20842 sgb.set_index_name("cat_key");
20843
20844 pandas::Series<numpy::float64> result(values);
20845 std::vector<std::string> idx_labels = {"A", "B"};
20846 result.set_index(std::make_unique<pandas::Index<std::string>>(idx_labels));
20847
20848 sgb.apply_result_index(result);
20849
20850 // Should have CategoricalIndex (dtype_name() returns "category")
20851 check(result.index().dtype_name() == "category", "is_categorical_index");
20852}
20853
20854// =====================================================================
20855// Per-group expanding tests
20856// =====================================================================
20857
20858void test_series_groupby_expanding_sum() {
20859 std::cout << " -- test_series_groupby_expanding_sum --" << std::endl;
CategoricalIndex (pd_test_2_all.cpp:20850)
20840 auto sgb = data.groupby(by);
20841 sgb.set_categorical_categories({"A", "B", "C"});
20842 sgb.set_index_name("cat_key");
20843
20844 pandas::Series<numpy::float64> result(values);
20845 std::vector<std::string> idx_labels = {"A", "B"};
20846 result.set_index(std::make_unique<pandas::Index<std::string>>(idx_labels));
20847
20848 sgb.apply_result_index(result);
20849
20850 // Should have CategoricalIndex (dtype_name() returns "category")
20851 check(result.index().dtype_name() == "category", "is_categorical_index");
20852}
20853
20854// =====================================================================
20855// Per-group expanding tests
20856// =====================================================================
20857
20858void test_series_groupby_expanding_sum() {
20859 std::cout << " -- test_series_groupby_expanding_sum --" << std::endl;
CategoricalIndex (pd_test_2_all.cpp:20850)
20840 auto sgb = data.groupby(by);
20841 sgb.set_categorical_categories({"A", "B", "C"});
20842 sgb.set_index_name("cat_key");
20843
20844 pandas::Series<numpy::float64> result(values);
20845 std::vector<std::string> idx_labels = {"A", "B"};
20846 result.set_index(std::make_unique<pandas::Index<std::string>>(idx_labels));
20847
20848 sgb.apply_result_index(result);
20849
20850 // Should have CategoricalIndex (dtype_name() returns "category")
20851 check(result.index().dtype_name() == "category", "is_categorical_index");
20852}
20853
20854// =====================================================================
20855// Per-group expanding tests
20856// =====================================================================
20857
20858void test_series_groupby_expanding_sum() {
20859 std::cout << " -- test_series_groupby_expanding_sum --" << std::endl;
from_codes (pd_test_1_all.cpp:403)
393 std::cout << " -> tests passed" << std::endl;
394 }
395
396 void pd_test_categorical_array_from_codes() {
397 std::cout << "========= CategoricalArray: from_codes ======================= ";
398
399 std::vector<std::string> cats = {"a", "b", "c"};
400 std::vector<numpy::int32> codes = {0, 1, 2, 0, 1, -1}; // -1 is NA
401
402 pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats, false);
403
404 if (arr.size() != 6) {
405 std::cout << " [FAIL] : in pd_test_categorical_array_from_codes() : size != 6" << std::endl;
406 throw std::runtime_error("pd_test_categorical_array_from_codes failed: size != 6");
407 }
408
409 // Check that code=-1 creates NA
410 if (!arr.is_na(5)) {
411 std::cout << " [FAIL] : in pd_test_categorical_array_from_codes() : code -1 should be NA" << std::endl;
412 throw std::runtime_error("pd_test_categorical_array_from_codes failed: code -1 should be NA");
get_indexer_non_unique (pd_test_3_all.cpp:739)
729 if (indexer.getElementAt({1}) != 3) {
730 std::cout << " [FAIL] : in pd_test_3_all_index_indexers() : 'd' should be at index 3" << std::endl;
731 throw std::runtime_error("pd_test_3_all_index_indexers failed: 'd' index");
732 }
733 // "f" doesn't exist -> -1
734 if (indexer.getElementAt({2}) != -1) {
735 std::cout << " [FAIL] : in pd_test_3_all_index_indexers() : 'f' should be -1" << std::endl;
736 throw std::runtime_error("pd_test_3_all_index_indexers failed: 'f' index");
737 }
738
739 // Test get_indexer_non_unique()
740 std::vector<std::string> target2 = {"a", "c", "z"}; // "z" doesn't exist
741 pandas::Index<std::string> target_idx(target2);
742 auto [indexer2, missing] = idx.get_indexer_non_unique(target_idx);
743
744 if (indexer2.getSize() < 2) {
745 std::cout << " [FAIL] : in pd_test_3_all_index_indexers() : get_indexer_non_unique size too small" << std::endl;
746 throw std::runtime_error("pd_test_3_all_index_indexers failed: get_indexer_non_unique size");
747 }
748
749 // Test slice_indexer()
get_level_values (pd_test_3_all.cpp:4524)
4514 }
4515
4516 std::cout << " -> tests passed" << std::endl;
4517}
4518
4519void pd_test_3_all_interval_index_get_level_values_droplevel() {
4520 std::cout << "========= IntervalIndex.get_level_values/droplevel() ";
4521
4522 pandas::IntervalIndex64 idx = pandas::IntervalIndex64::from_breaks({0, 10, 20, 30});
4523
4524 // get_level_values(0) should work
4525 pandas::IntervalIndex64 level_vals = idx.get_level_values(0);
4526 if (level_vals.size() != idx.size()) {
4527 throw std::runtime_error("get_level_values(0) size mismatch");
4528 }
4529
4530 // get_level_values(1) should throw
4531 bool threw = false;
4532 try {
4533 idx.get_level_values(1);
4534 } catch (const std::out_of_range&) {
get_level_values (pd_test_3_all.cpp:4524)
4514 }
4515
4516 std::cout << " -> tests passed" << std::endl;
4517}
4518
4519void pd_test_3_all_interval_index_get_level_values_droplevel() {
4520 std::cout << "========= IntervalIndex.get_level_values/droplevel() ";
4521
4522 pandas::IntervalIndex64 idx = pandas::IntervalIndex64::from_breaks({0, 10, 20, 30});
4523
4524 // get_level_values(0) should work
4525 pandas::IntervalIndex64 level_vals = idx.get_level_values(0);
4526 if (level_vals.size() != idx.size()) {
4527 throw std::runtime_error("get_level_values(0) size mismatch");
4528 }
4529
4530 // get_level_values(1) should throw
4531 bool threw = false;
4532 try {
4533 idx.get_level_values(1);
4534 } catch (const std::out_of_range&) {
get_slice_bound (pd_test_3_all.cpp:3644)
3634 formatted = idx.format(custom_formatter);
3635
3636 if (formatted[0] != "val:1") {
3637 throw std::runtime_error("custom formatter failed");
3638 }
3639
3640 std::cout << " -> tests passed" << std::endl;
3641}
3642
3643void pd_test_3_all_index_get_slice_bound() {
3644 std::cout << "========= Index.get_slice_bound() ==================";
3645
3646 pandas::Index<numpy::int64> idx({10, 20, 30, 40, 50});
3647
3648 // Exact match, left side
3649 size_t bound = idx.get_slice_bound(30, "left");
3650 if (bound != 2) {
3651 throw std::runtime_error("left bound for 30 should be 2");
3652 }
3653
3654 // Exact match, right side
get_value_str (pd_test_1_all.cpp:4665)
4655 auto corr_df = df.corr();
4656
4657 // Check dimensions
4658 bool passed = corr_df.nrows() == 2 && corr_df.ncols() == 2;
4659 if (!passed) {
4660 std::cout << " [FAIL] : in pd_test_aggregation_dataframe_corr() : corr should be 2x2" << std::endl;
4661 throw std::runtime_error("pd_test_aggregation_dataframe_corr failed: corr should be 2x2");
4662 }
4663
4664 // Diagonal should be 1.0
4665 std::string aa = corr_df["A"].get_value_str(0);
4666 passed = std::abs(std::stod(aa) - 1.0) < 0.001;
4667 if (!passed) {
4668 std::cout << " [FAIL] : in pd_test_aggregation_dataframe_corr() : diagonal should be 1.0" << std::endl;
4669 throw std::runtime_error("pd_test_aggregation_dataframe_corr failed: diagonal should be 1.0");
4670 }
4671
4672 // A-B correlation should be 1.0 (perfect correlation)
4673 std::string ab = corr_df["B"].get_value_str(0);
4674 passed = std::abs(std::stod(ab) - 1.0) < 0.001;
4675 if (!passed) {
where (pd_test_1_all.cpp:22018)
22008 data["B"] = {5.0, 6.0, 7.0, 8.0};
22009 pandas::DataFrame df(data);
22010
22011 // Create condition DataFrame (values > 2)
22012 std::map<std::string, std::vector<numpy::bool_>> cond_data;
22013 cond_data["A"] = {false, false, true, true}; // 1<=2, 2<=2, 3>2, 4>2
22014 cond_data["B"] = {true, true, true, true}; // all >2
22015 pandas::DataFrame cond(cond_data);
22016
22017 // Apply where with replacement value -1
22018 pandas::DataFrame result = df.where(cond, -1.0);
22019
22020 // Get column index for A - it's sorted alphabetically in std::map
22021 size_t col_a_idx = df.get_column_index("A");
22022 size_t col_b_idx = df.get_column_index("B");
22023
22024 bool passed = true;
22025 std::string error_msg;
22026
22027 // Check A column values
22028 std::string a0 = result.iat<double>(0, col_a_idx) == -1.0 ? "ok" : "fail";
droplevel (pd_test_1_all.cpp:14428)
14418 void pd_test_multiindex_droplevel() {
14419 std::cout << "========= droplevel =================================== ";
14420
14421 std::vector<std::vector<std::string>> arrays = {
14422 {"a", "a", "b"},
14423 {"x", "y", "z"},
14424 {"1", "2", "3"}
14425 };
14426
14427 pandas::MultiIndex mi = pandas::MultiIndex::from_arrays<std::string>(arrays);
14428 pandas::MultiIndex dropped = mi.droplevel(1);
14429
14430 bool passed = true;
14431
14432 if (dropped.nlevels() != 2) {
14433 std::cout << " [FAIL] : nlevels should be 2 after drop" << std::endl;
14434 passed = false;
14435 }
14436
14437 // Check remaining levels
14438 auto tup = dropped[0];
reindex (pd_test_1_all.cpp:6708)
6698 }
6699 }
6700
6701 // Test reindex rows
6702 {
6703 std::map<std::string, std::vector<double>> data;
6704 data["A"] = {1.0, 2.0, 3.0};
6705 pandas::DataFrame df(data);
6706 df = df.set_axis({"x", "y", "z"}, 0);
6707
6708 auto reindexed = df.reindex({"x", "z", "w"}, 0);
6709 if (reindexed.nrows() != 3) {
6710 std::cout << " [FAIL] : in pd_test_dataframe_index_ops() : reindex wrong nrows" << std::endl;
6711 throw std::runtime_error("pd_test_dataframe_index_ops failed: reindex nrows");
6712 }
6713 // 'w' should have NaN
6714 std::string val = reindexed["A"].get_value_str(2);
6715 if (!std::isnan(std::stod(val))) {
6716 std::cout << " [FAIL] : in pd_test_dataframe_index_ops() : missing label should be NaN" << std::endl;
6717 throw std::runtime_error("pd_test_dataframe_index_ops failed: reindex NaN");
6718 }
rename (pd_test_1_all.cpp:5816)
5806 std::cout << " -> tests passed" << std::endl;
5807}
5808
5809void pd_test_categorical_index_rename() {
5810 std::cout << "========= rename ======================================";
5811
5812 pandas::CategoricalArray arr({"x", "y"});
5813 pandas::CategoricalIndex idx(arr, "old_name");
5814
5815 pandas::CategoricalIndex renamed = idx.rename("new_name");
5816
5817 bool passed = (renamed.name().has_value() && *renamed.name() == "new_name" &&
5818 renamed.size() == idx.size() && renamed.categories() == idx.categories());
5819 if (!passed) {
5820 std::cout << " [FAIL] : in pd_test_categorical_index_rename()" << std::endl;
5821 throw std::runtime_error("pd_test_categorical_index_rename failed");
5822 }
5823
5824 std::cout << " -> tests passed" << std::endl;
5825}
rename_categories (pd_test_1_all.cpp:655)
645 void pd_test_categorical_array_rename_categories() {
646 std::cout << "========= CategoricalArray: rename_categories ======================= ";
647
648 std::vector<std::string> cats = {"a", "b"};
649 std::vector<numpy::int32> codes = {0, 1, 0}; // a, b, a
650 pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);
651
652 // Rename categories
653 std::vector<std::string> new_names = {"alpha", "beta"};
654 pandas::CategoricalArray result = arr.rename_categories(new_names);
655
656 // Check categories are renamed
657 const std::vector<std::string>& result_cats = result.categories();
658 if (result_cats[0] != "alpha" || result_cats[1] != "beta") {
659 std::cout << " [FAIL] : in pd_test_categorical_array_rename_categories() : categories not renamed" << std::endl;
660 throw std::runtime_error("pd_test_categorical_array_rename_categories failed: categories not renamed");
661 }
662
663 // Values should now be renamed
664 std::optional<std::string> val = result[0];
rename_categories (pd_test_1_all.cpp:655)
645 void pd_test_categorical_array_rename_categories() {
646 std::cout << "========= CategoricalArray: rename_categories ======================= ";
647
648 std::vector<std::string> cats = {"a", "b"};
649 std::vector<numpy::int32> codes = {0, 1, 0}; // a, b, a
650 pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);
651
652 // Rename categories
653 std::vector<std::string> new_names = {"alpha", "beta"};
654 pandas::CategoricalArray result = arr.rename_categories(new_names);
655
656 // Check categories are renamed
657 const std::vector<std::string>& result_cats = result.categories();
658 if (result_cats[0] != "alpha" || result_cats[1] != "beta") {
659 std::cout << " [FAIL] : in pd_test_categorical_array_rename_categories() : categories not renamed" << std::endl;
660 throw std::runtime_error("pd_test_categorical_array_rename_categories failed: categories not renamed");
661 }
662
663 // Values should now be renamed
664 std::optional<std::string> val = result[0];
set_names (pd_test_1_all.cpp:14519)
14509 std::cout << "-> tests passed" << std::endl;
14510 }
14511
14512 void pd_test_multiindex_set_names() {
14513 std::cout << "========= set_names =================================== ";
14514
14515 std::vector<std::vector<std::string>> arrays = {{"a", "b"}, {"x", "y"}};
14516 pandas::MultiIndex mi = pandas::MultiIndex::from_arrays<std::string>(arrays);
14517
14518 std::vector<std::optional<std::string>> new_names = {"level_a", "level_b"};
14519 pandas::MultiIndex named = mi.set_names(new_names);
14520
14521 bool passed = (named.names()[0] == "level_a" && named.names()[1] == "level_b");
14522
14523 if (!passed) {
14524 std::cout << " [FAIL] : names not set correctly" << std::endl;
14525 throw std::runtime_error("pd_test_multiindex_set_names failed");
14526 }
14527
14528 std::cout << "-> tests passed" << std::endl;
14529 }
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'");
groupby (pd_test_1_all.cpp:11495)
11485 std::cout << "========= GroupBy basic =========================";
11486
11487 // Create DataFrame with category column
11488 std::map<std::string, std::vector<double>> data = {
11489 {"category", {1.0, 1.0, 2.0, 2.0, 2.0}},
11490 {"value", {10.0, 20.0, 30.0, 40.0, 50.0}}
11491 };
11492 pandas::DataFrame df(data);
11493
11494 // Test groupby
11495 auto grouped = df.groupby("category");
11496
11497 bool passed = grouped.ngroups() == 2;
11498 if (!passed) {
11499 std::cout << " [FAIL] : in pd_test_groupby_basic() : ngroups should be 2" << std::endl;
11500 throw std::runtime_error("pd_test_groupby_basic failed: ngroups should be 2");
11501 }
11502
11503 std::cout << " -> tests passed" << std::endl;
11504 }
map (pd_test_1_all.cpp:5839)
5829// Map Tests
5830// ============================================================================
5831
5832void pd_test_categorical_index_map() {
5833 std::cout << "========= map =========================================";
5834
5835 pandas::CategoricalArray arr({"yes", "no", "yes"});
5836 pandas::CategoricalIndex idx(arr);
5837
5838 std::unordered_map<std::string, std::string> mapping = {{"yes", "1"}, {"no", "0"}};
5839 pandas::CategoricalIndex mapped = idx.map(mapping);
5840
5841 bool passed = (mapped.has_category("1") && mapped.has_category("0") &&
5842 !mapped.has_category("yes") && !mapped.has_category("no"));
5843 if (!passed) {
5844 std::cout << " [FAIL] : in pd_test_categorical_index_map()" << std::endl;
5845 throw std::runtime_error("pd_test_categorical_index_map failed");
5846 }
5847
5848 std::cout << " -> tests passed" << std::endl;
5849}
add_categories (pd_test_1_all.cpp:555)
545 }
546
547 void pd_test_categorical_array_add_categories() {
548 std::cout << "========= CategoricalArray: add_categories ======================= ";
549
550 std::vector<std::string> cats = {"a", "b"};
551 std::vector<numpy::int32> codes = {0, 1, 0};
552 pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);
553
554 // Add new categories
555 pandas::CategoricalArray result = arr.add_categories({"c", "d"});
556 if (result.categories().size() != 4) {
557 std::cout << " [FAIL] : in pd_test_categorical_array_add_categories() : new categories size != 4" << std::endl;
558 throw std::runtime_error("pd_test_categorical_array_add_categories failed: new categories size != 4");
559 }
560
561 // Original values should be preserved
562 std::optional<std::string> val = result[0];
563 if (!val.has_value() || *val != "a") {
564 std::cout << " [FAIL] : in pd_test_categorical_array_add_categories() : value not preserved" << std::endl;
565 throw std::runtime_error("pd_test_categorical_array_add_categories failed: value not preserved");
equals (pd_test_1_all.cpp:5866)
5856 std::cout << "========= equals ======================================";
5857
5858 pandas::CategoricalArray arr1({"a", "b", "a"});
5859 pandas::CategoricalArray arr2({"a", "b", "a"});
5860 pandas::CategoricalArray arr3({"a", "b", "c"});
5861
5862 pandas::CategoricalIndex idx1(arr1);
5863 pandas::CategoricalIndex idx2(arr2);
5864 pandas::CategoricalIndex idx3(arr3);
5865
5866 bool passed = (idx1.equals(idx2) && !idx1.equals(idx3));
5867 if (!passed) {
5868 std::cout << " [FAIL] : in pd_test_categorical_index_equals()" << std::endl;
5869 throw std::runtime_error("pd_test_categorical_index_equals failed");
5870 }
5871
5872 std::cout << " -> tests passed" << std::endl;
5873}
5874
5875void pd_test_categorical_index_identical() {
5876 std::cout << "========= identical ===================================";
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 }
searchsorted (pd_test_1_all.cpp:18958)
18948 // =========================================================================
18949 // Search Tests
18950 // =========================================================================
18951
18952 void pd_test_range_index_searchsorted() {
18953 std::cout << "========= searchsorted ================================ ";
18954
18955 pandas::RangeIndex ri(0, 10, 2); // [0, 2, 4, 6, 8]
18956
18957 bool passed = (ri.searchsorted(4, "left") == 2 &&
18958 ri.searchsorted(4, "right") == 3 &&
18959 ri.searchsorted(3, "left") == 2 && // 3 would go between 2 and 4
18960 ri.searchsorted(-1, "left") == 0 && // Before all
18961 ri.searchsorted(10, "left") == 5); // After all
18962
18963 if (!passed) {
18964 std::cout << " [FAIL] : searchsorted" << std::endl;
18965 throw std::runtime_error("pd_test_range_index_searchsorted failed");
18966 }
sort_values (pd_test_1_all.cpp:6408)
6398 void pd_test_dataframe_sorting() {
6399 std::cout << "========= sorting ==========================";
6400
6401 std::map<std::string, std::vector<numpy::float64>> data;
6402 data["A"] = {3.0, 1.0, 4.0, 1.0, 5.0};
6403 data["B"] = {9.0, 2.0, 6.0, 5.0, 3.0};
6404
6405 pandas::DataFrame df(data);
6406
6407 // Test sort_values ascending
6408 auto sorted_asc = df.sort_values("A", true);
6409 // First value should be smallest (1.0)
6410 std::string first_val = sorted_asc["A"].get_value_str(0);
6411 if (std::stod(first_val) != 1.0) {
6412 std::cout << " [FAIL] : in pd_test_dataframe_sorting() : sort_values asc first != 1" << std::endl;
6413 throw std::runtime_error("pd_test_dataframe_sorting failed: sort_values asc first != 1");
6414 }
6415
6416 // Test sort_values descending
6417 auto sorted_desc = df.sort_values("A", false);
6418 first_val = sorted_desc["A"].get_value_str(0);
transpose (pd_test_1_all.cpp:16648)
16638 std::cout << " [FAIL] : in pd_test_ndframe_transpose() : T_() size" << std::endl;
16639 throw std::runtime_error("pd_test_ndframe_transpose failed: T_() size");
16640 }
16641
16642 passed = transposed[0] == 1 && transposed[1] == 2 && transposed[2] == 3;
16643 if (!passed) {
16644 std::cout << " [FAIL] : in pd_test_ndframe_transpose() : T_() values" << std::endl;
16645 throw std::runtime_error("pd_test_ndframe_transpose failed: T_() values");
16646 }
16647
16648 // Test transpose() alias
16649 auto transposed2 = s.transpose();
16650 passed = transposed2.size() == s.size();
16651 if (!passed) {
16652 std::cout << " [FAIL] : in pd_test_ndframe_transpose() : transpose() size" << std::endl;
16653 throw std::runtime_error("pd_test_ndframe_transpose failed: transpose() size");
16654 }
16655
16656 std::cout << " -> tests passed" << std::endl;
16657 }
join (pd_test_1_all.cpp:12353)
12343 std::cout << " -> tests passed" << std::endl;
12344 }
12345
12346 void pd_test_index_join() {
12347 std::cout << "========= join ========================================";
12348
12349 pandas::Index<numpy::int64> idx1{1, 2, 3};
12350 pandas::Index<numpy::int64> idx2{2, 3, 4};
12351
12352 auto [inner_joined, left_idx, right_idx] = idx1.join(idx2, "inner");
12353 bool passed = (inner_joined.size() == 2); // {2, 3}
12354
12355 auto [outer_joined, ol_idx, or_idx] = idx1.join(idx2, "outer");
12356 passed = passed && (outer_joined.size() == 4); // {1, 2, 3, 4}
12357
12358 if (!passed) {
12359 std::cout << " [FAIL] : in pd_test_index_join() : join failed" << std::endl;
12360 throw std::runtime_error("pd_test_index_join failed");
12361 }
asof (pd_test_2_all.cpp:366)
356 std::cout << "====================================== [OK] pd_test_add_prefix test suite ========================== " << std::endl;
357 return 0;
358 }
359
360} // namespace dataframe_tests
361// ------------------- pd_test_add_prefix.cpp (end) -----------------------------
362
363// ------------------- pd_test_asof.cpp (start) -----------------------------
364// dataframe_tests/pd_test_asof.cpp
365// Test for DataFrame.asof() method
366
367#include <iostream>
368#include <cmath>
369#include <stdexcept>
370#include <limits>
371#include "../pandas/pd_dataframe.h"
372
373// CRITICAL: No using namespace directives
374
375namespace dataframe_tests {
asof_locs (pd_test_3_all.cpp:3557)
3547 throw std::runtime_error("all() should be true for empty index");
3548 }
3549 if (empty_idx.any()) {
3550 throw std::runtime_error("any() should be false for empty index");
3551 }
3552
3553 std::cout << " -> tests passed" << std::endl;
3554}
3555
3556void pd_test_3_all_index_asof() {
3557 std::cout << "========= Index.asof()/asof_locs() =================";
3558
3559 // Test with monotonically increasing index
3560 pandas::Index<numpy::int64> idx({10, 20, 30, 40, 50});
3561
3562 // Exact match
3563 auto result = idx.asof(30);
3564 if (!result.has_value() || result.value() != 30) {
3565 throw std::runtime_error("asof() exact match should return 30");
3566 }
diff (pd_test_1_all.cpp:5171)
5161 }
5162
5163 void pd_test_arithmetic_dataframe_diff_shift() {
5164 std::cout << "========= DataFrame diff/shift ==================";
5165
5166 std::map<std::string, std::vector<double>> data;
5167 data["A"] = {1.0, 3.0, 6.0, 10.0};
5168 pandas::DataFrame df(data);
5169
5170 // diff: [NaN, 2, 3, 4]
5171 auto d = df.diff();
5172 std::string val = d["A"].get_value_str(1);
5173 bool passed = std::abs(std::stod(val) - 2.0) < 0.001;
5174 if (!passed) {
5175 std::cout << " [FAIL] : in pd_test_arithmetic_dataframe_diff_shift() : diff failed" << std::endl;
5176 throw std::runtime_error("pd_test_arithmetic_dataframe_diff_shift failed: diff failed");
5177 }
5178
5179 // First element should be NaN
5180 val = d["A"].get_value_str(0);
5181 passed = std::isnan(std::stod(val));
shift (pd_test_1_all.cpp:5188)
5178 // First element should be NaN
5179 val = d["A"].get_value_str(0);
5180 passed = std::isnan(std::stod(val));
5181 if (!passed) {
5182 std::cout << " [FAIL] : in pd_test_arithmetic_dataframe_diff_shift() : diff NaN failed" << std::endl;
5183 throw std::runtime_error("pd_test_arithmetic_dataframe_diff_shift failed: diff NaN failed");
5184 }
5185
5186 // shift: [NaN, 1, 3, 6]
5187 auto s = df.shift();
5188 val = s["A"].get_value_str(1);
5189 passed = std::abs(std::stod(val) - 1.0) < 0.001;
5190 if (!passed) {
5191 std::cout << " [FAIL] : in pd_test_arithmetic_dataframe_diff_shift() : shift failed" << std::endl;
5192 throw std::runtime_error("pd_test_arithmetic_dataframe_diff_shift failed: shift failed");
5193 }
5194
5195 std::cout << " -> tests passed" << std::endl;
5196 }
to_flat_index (pd_test_1_all.cpp:14733)
14723 void pd_test_multiindex_to_flat_index() {
14724 std::cout << "========= to_flat_index =============================== ";
14725
14726 std::vector<std::vector<std::string>> arrays = {
14727 {"a", "b"},
14728 {"x", "y"}
14729 };
14730
14731 pandas::MultiIndex mi = pandas::MultiIndex::from_arrays<std::string>(arrays);
14732 auto flat = mi.to_flat_index();
14733
14734 bool passed = (flat.size() == 2 &&
14735 flat[0][0] == "a" && flat[0][1] == "x" &&
14736 flat[1][0] == "b" && flat[1][1] == "y");
14737
14738 if (!passed) {
14739 std::cout << " [FAIL] : to_flat_index incorrect" << std::endl;
14740 throw std::runtime_error("pd_test_multiindex_to_flat_index failed");
14741 }
to_list (pd_test_1_all.cpp:10247)
10237 std::cout << " -> tests passed" << std::endl;
10238}
10239
10240void pd_test_extension_index_to_list() {
10241 std::cout << "========= to_list =========================";
10242
10243 pandas::CategoricalArray arr({"x", "y", "z"});
10244 pandas::CategoricalIndex idx(arr);
10245
10246 auto list = idx.to_list();
10247
10248 bool passed = (list.size() == 3 &&
10249 list[0].has_value() && *list[0] == "x" &&
10250 list[1].has_value() && *list[1] == "y" &&
10251 list[2].has_value() && *list[2] == "z");
10252 if (!passed) {
10253 std::cout << " [FAIL] : in pd_test_extension_index_to_list() : to_list check failed" << std::endl;
10254 throw std::runtime_error("pd_test_extension_index_to_list failed");
10255 }
to_numpy (pd_test_1_all.cpp:16764)
16754 // =====================================================================
16755 // to_numpy Tests
16756 // =====================================================================
16757
16758 void pd_test_ndframe_to_numpy() {
16759 std::cout << "========= to_numpy =============================================" << std::endl;
16760
16761 pandas::Series<int> s({10, 20, 30});
16762
16763 auto arr = s.to_numpy();
16764
16765 bool passed = arr.getSize() == 3;
16766 if (!passed) {
16767 std::cout << " [FAIL] : in pd_test_ndframe_to_numpy() : size" << std::endl;
16768 throw std::runtime_error("pd_test_ndframe_to_numpy failed: size");
16769 }
16770
16771 passed = arr.getElementAt({0}) == 10 && arr.getElementAt({1}) == 20 && arr.getElementAt({2}) == 30;
16772 if (!passed) {
16773 std::cout << " [FAIL] : in pd_test_ndframe_to_numpy() : values" << std::endl;
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];
tolist (pd_test_3_all.cpp:2300)
2290 threw = true;
2291 }
2292 if (!threw) {
2293 throw std::runtime_error("swapaxes should throw for invalid axes");
2294 }
2295
2296 std::cout << " -> tests passed" << std::endl;
2297}
2298
2299void pd_test_3_all_categorical_to_list() {
2300 std::cout << "========= CategoricalArray.to_list()/tolist() =========";
2301
2302 std::vector<std::optional<std::string>> values = {"a", "b", std::nullopt, "c"};
2303 pandas::CategoricalArray arr(values);
2304
2305 auto list = arr.to_list();
2306 if (list.size() != 4 || *list[0] != "a" || *list[1] != "b" ||
2307 list[2].has_value() || *list[3] != "c") {
2308 throw std::runtime_error("to_list failed");
2309 }
astype (pd_test_1_all.cpp:21292)
21282 std::cout << "========= astype all columns to float64 =============";
21283
21284 // Create DataFrame with int64 columns
21285 std::map<std::string, std::vector<numpy::int64>> data;
21286 data["A"] = {1, 2, 3, 4, 5};
21287 data["B"] = {10, 20, 30, 40, 50};
21288
21289 pandas::DataFrame df(data);
21290
21291 // Convert all columns to float64
21292 pandas::DataFrame df_float = df.astype("float64");
21293
21294 // Verify dtype changed
21295 pandas::Series<std::string> dtypes = df_float.dtypes();
21296
21297 bool passed = true;
21298 if (dtypes[static_cast<size_t>(0)] != "float64") {
21299 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;
21300 passed = false;
21301 }
21302 if (dtypes[static_cast<size_t>(1)] != "float64") {
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}
infer_objects (pd_test_1_all.cpp:27595)
27585 // Create DataFrame with string column containing integers
27586 std::map<std::string, std::vector<std::string>> data;
27587 data["A"] = {"1", "2", "3", "4", "5"};
27588
27589 pandas::DataFrame df(data);
27590
27591 // Before inference, dtype should be string/object
27592 std::string before_dtype = df["A"].dtype_name();
27593
27594 // Apply infer_objects
27595 pandas::DataFrame result = df.infer_objects();
27596
27597 // After inference, dtype should be int64
27598 std::string after_dtype = result["A"].dtype_name();
27599
27600 bool passed = (after_dtype == "int64");
27601 if (!passed) {
27602 std::cout << " [FAIL] : in pd_test_infer_objects_integer_column() : expected int64, got " << after_dtype << std::endl;
27603 throw std::runtime_error("pd_test_infer_objects_integer_column failed");
27604 }
view (pd_test_3_all.cpp:2147)
2137 throw std::runtime_error("memory_usage shallow too small");
2138 }
2139 if (deep < shallow) {
2140 throw std::runtime_error("memory_usage deep should be >= shallow");
2141 }
2142
2143 std::cout << " -> tests passed" << std::endl;
2144}
2145
2146void pd_test_3_all_categorical_ravel_view() {
2147 std::cout << "========= CategoricalArray.ravel()/view() =============";
2148
2149 std::vector<std::optional<std::string>> values = {"a", "b", "c"};
2150 pandas::CategoricalArray arr(values);
2151
2152 auto raveled = arr.ravel();
2153 if (raveled.size() != 3 || !raveled.equals(arr)) {
2154 throw std::runtime_error("ravel failed");
2155 }
2156
2157 auto viewed = arr.view();
is_ (pd_test_3_all.cpp:3972)
3962 // For typed Index, this is a no-op
3963 if (result.size() != 5) {
3964 throw std::runtime_error("infer_objects size should be 5");
3965 }
3966
3967 std::cout << " -> tests passed" << std::endl;
3968}
3969
3970void pd_test_3_all_index_is_() {
3971 std::cout << "========= Index.is_() ==============================";
3972
3973 pandas::Index<numpy::int64> idx1({1, 2, 3, 4, 5});
3974 pandas::Index<numpy::int64> idx2({1, 2, 3, 4, 5}); // Different object
3975
3976 // Different objects should not be the same
3977 if (idx1.is_(idx2)) {
3978 throw std::runtime_error("different objects should not be is_() equal");
3979 }
3980
3981 // Same object should be the same
is_boolean (pd_test_3_all.cpp:3290)
3280 std::cout << " -> tests passed" << std::endl;
3281}
3282
3283void pd_test_3_all_datetime_index_type_checks() {
3284 std::cout << "========= DatetimeIndex type checks ======================";
3285
3286 pandas::DatetimeIndex idx = pandas::date_range("2024-01-01", "2024-01-05", std::nullopt, "D");
3287
3288 // Type check methods
3289 if (idx.is_boolean()) {
3290 throw std::runtime_error("is_boolean() should be false");
3291 }
3292 if (idx.is_categorical()) {
3293 throw std::runtime_error("is_categorical() should be false");
3294 }
3295 if (idx.is_floating()) {
3296 throw std::runtime_error("is_floating() should be false");
3297 }
3298 if (idx.is_integer()) {
3299 throw std::runtime_error("is_integer() should be false");
is_categorical (pd_test_3_all.cpp:3293)
3283void pd_test_3_all_datetime_index_type_checks() {
3284 std::cout << "========= DatetimeIndex type checks ======================";
3285
3286 pandas::DatetimeIndex idx = pandas::date_range("2024-01-01", "2024-01-05", std::nullopt, "D");
3287
3288 // Type check methods
3289 if (idx.is_boolean()) {
3290 throw std::runtime_error("is_boolean() should be false");
3291 }
3292 if (idx.is_categorical()) {
3293 throw std::runtime_error("is_categorical() should be false");
3294 }
3295 if (idx.is_floating()) {
3296 throw std::runtime_error("is_floating() should be false");
3297 }
3298 if (idx.is_integer()) {
3299 throw std::runtime_error("is_integer() should be false");
3300 }
3301 if (idx.is_interval()) {
3302 throw std::runtime_error("is_interval() should be false");
is_floating (pd_test_3_all.cpp:622)
612 // Test with integer index
613 pandas::IndexDtype<numpy::int64> int_dtype;
614 if (!int_dtype.is_numeric()) {
615 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should be numeric" << std::endl;
616 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_numeric");
617 }
618 if (!int_dtype.is_integer()) {
619 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should be integer" << std::endl;
620 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_integer");
621 }
622 if (int_dtype.is_floating()) {
623 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should not be floating" << std::endl;
624 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_floating");
625 }
626 if (int_dtype.is_object()) {
627 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should not be object" << std::endl;
628 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_object");
629 }
630
631 // Test with float index
632 pandas::IndexDtype<double> float_dtype;
is_integer (pd_test_3_all.cpp:618)
608void pd_test_3_all_index_dtype_checks() {
609 std::cout << "========= IndexDtype.is_numeric/integer/floating/object() ";
610
611 // Test with integer index
612 pandas::IndexDtype<numpy::int64> int_dtype;
613 if (!int_dtype.is_numeric()) {
614 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should be numeric" << std::endl;
615 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_numeric");
616 }
617 if (!int_dtype.is_integer()) {
618 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should be integer" << std::endl;
619 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_integer");
620 }
621 if (int_dtype.is_floating()) {
622 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should not be floating" << std::endl;
623 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_floating");
624 }
625 if (int_dtype.is_object()) {
626 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should not be object" << std::endl;
627 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_object");
is_interval (pd_test_3_all.cpp:3302)
3292 }
3293 if (idx.is_categorical()) {
3294 throw std::runtime_error("is_categorical() should be false");
3295 }
3296 if (idx.is_floating()) {
3297 throw std::runtime_error("is_floating() should be false");
3298 }
3299 if (idx.is_integer()) {
3300 throw std::runtime_error("is_integer() should be false");
3301 }
3302 if (idx.is_interval()) {
3303 throw std::runtime_error("is_interval() should be false");
3304 }
3305 if (idx.is_numeric()) {
3306 throw std::runtime_error("is_numeric() should be false");
3307 }
3308 if (idx.is_object()) {
3309 throw std::runtime_error("is_object() should be false");
3310 }
3311 if (idx.holds_integer()) {
3312 throw std::runtime_error("holds_integer() should be false");
is_numeric (pd_test_3_all.cpp:614)
604// ============================================================================
605// Category 4: Index Type Checking (IndexDtype)
606// ============================================================================
607
608void pd_test_3_all_index_dtype_checks() {
609 std::cout << "========= IndexDtype.is_numeric/integer/floating/object() ";
610
611 // Test with integer index
612 pandas::IndexDtype<numpy::int64> int_dtype;
613 if (!int_dtype.is_numeric()) {
614 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should be numeric" << std::endl;
615 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_numeric");
616 }
617 if (!int_dtype.is_integer()) {
618 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should be integer" << std::endl;
619 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_integer");
620 }
621 if (int_dtype.is_floating()) {
622 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should not be floating" << std::endl;
623 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_floating");
is_object (pd_test_3_all.cpp:626)
616 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_numeric");
617 }
618 if (!int_dtype.is_integer()) {
619 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should be integer" << std::endl;
620 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_integer");
621 }
622 if (int_dtype.is_floating()) {
623 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should not be floating" << std::endl;
624 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_floating");
625 }
626 if (int_dtype.is_object()) {
627 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : int should not be object" << std::endl;
628 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: int is_object");
629 }
630
631 // Test with float index
632 pandas::IndexDtype<double> float_dtype;
633 if (!float_dtype.is_numeric()) {
634 std::cout << " [FAIL] : in pd_test_3_all_index_dtype_checks() : float should be numeric" << std::endl;
635 throw std::runtime_error("pd_test_3_all_index_dtype_checks failed: float is_numeric");
636 }
all (pd_test_1_all.cpp:247)
237 pandas::BooleanArray has_true({
238 std::optional<bool>(false),
239 std::optional<bool>(true)
240 });
241 any_result = has_true.any();
242 if (!any_result.has_value() || !any_result.value()) {
243 std::cout << " [FAIL] : in pd_test_boolean_array_reductions() : any() with True" << std::endl;
244 throw std::runtime_error("pd_test_boolean_array_reductions failed: any() with True");
245 }
246
247 // Test all()
248 pandas::BooleanArray all_true({
249 std::optional<bool>(true),
250 std::optional<bool>(true)
251 });
252 auto all_result = all_true.all();
253 if (!all_result.has_value() || !all_result.value()) {
254 std::cout << " [FAIL] : in pd_test_boolean_array_reductions() : all() of all True" << std::endl;
255 throw std::runtime_error("pd_test_boolean_array_reductions failed: all() all True");
256 }
any (pd_test_1_all.cpp:226)
216 std::cout << " [FAIL] : in pd_test_boolean_array_kleene_not() : ~NA should be NA" << std::endl;
217 throw std::runtime_error("pd_test_boolean_array_kleene_not failed: ~NA");
218 }
219
220 std::cout << " -> tests passed" << std::endl;
221 }
222
223 void pd_test_boolean_array_reductions() {
224 std::cout << "========= BooleanArray: reductions ======================= ";
225
226 // Test any()
227 pandas::BooleanArray all_false({
228 std::optional<bool>(false),
229 std::optional<bool>(false)
230 });
231 auto any_result = all_false.any();
232 if (!any_result.has_value() || any_result.value()) {
233 std::cout << " [FAIL] : in pd_test_boolean_array_reductions() : any() of all False" << std::endl;
234 throw std::runtime_error("pd_test_boolean_array_reductions failed: any() all False");
235 }
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");
as_ordered (pd_test_1_all.cpp:791)
781 unordered.min();
782 } catch (const std::exception&) {
783 threw = true;
784 }
785 if (!threw) {
786 std::cout << " [FAIL] : in pd_test_categorical_array_ordered_operations() : unordered min should throw" << std::endl;
787 throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: unordered min should throw");
788 }
789
790 // Test as_ordered / as_unordered
791 pandas::CategoricalArray reordered = unordered.as_ordered();
792 if (!reordered.ordered()) {
793 std::cout << " [FAIL] : in pd_test_categorical_array_ordered_operations() : as_ordered failed" << std::endl;
794 throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: as_ordered failed");
795 }
796
797 std::cout << " -> tests passed" << std::endl;
798 }
799
800 void pd_test_categorical_array_comparisons() {
801 std::cout << "========= CategoricalArray: comparisons ======================= ";
as_unordered (pd_test_1_all.cpp:778)
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();
782 } catch (const std::exception&) {
783 threw = true;
784 }
785 if (!threw) {
786 std::cout << " [FAIL] : in pd_test_categorical_array_ordered_operations() : unordered min should throw" << std::endl;
787 throw std::runtime_error("pd_test_categorical_array_ordered_operations failed: unordered min should throw");
788 }
categories (pd_test_1_all.cpp:389)
379 std::vector<std::optional<std::string>> vals = {
380 std::optional<std::string>("low"),
381 std::optional<std::string>("high"),
382 std::optional<std::string>("medium")
383 };
384 pandas::CategoricalArray arr3(vals, cats, true); // ordered
385 if (!arr3.ordered()) {
386 std::cout << " [FAIL] : in pd_test_categorical_array_constructors() : should be ordered" << std::endl;
387 throw std::runtime_error("pd_test_categorical_array_constructors failed: should be ordered");
388 }
389 if (arr3.categories().size() != 3) {
390 std::cout << " [FAIL] : in pd_test_categorical_array_constructors() : categories size != 3" << std::endl;
391 throw std::runtime_error("pd_test_categorical_array_constructors failed: categories size != 3");
392 }
393
394 std::cout << " -> tests passed" << std::endl;
395 }
396
397 void pd_test_categorical_array_from_codes() {
398 std::cout << "========= CategoricalArray: from_codes ======================= ";
clone (pd_test_1_all.cpp:5776)
5766 std::cout << " -> tests passed" << std::endl;
5767}
5768
5769void pd_test_categorical_index_clone() {
5770 std::cout << "========= clone =======================================";
5771
5772 pandas::CategoricalArray arr({"p", "q", "r"});
5773 pandas::CategoricalIndex idx(arr, "original");
5774
5775 std::unique_ptr<pandas::IndexBase> cloned = idx.clone();
5776
5777 bool passed = (cloned != nullptr && cloned->size() == idx.size() &&
5778 cloned->name() == idx.name());
5779 if (!passed) {
5780 std::cout << " [FAIL] : in pd_test_categorical_index_clone()" << std::endl;
5781 throw std::runtime_error("pd_test_categorical_index_clone failed");
5782 }
5783
5784 std::cout << " -> tests passed" << std::endl;
5785}
codes (pd_test_1_all.cpp:473)
463 std::cout << " -> tests passed" << std::endl;
464 }
465
466 void pd_test_categorical_array_codes_property() {
467 std::cout << "========= CategoricalArray: codes property ======================= ";
468
469 std::vector<std::string> cats = {"x", "y", "z"};
470 std::vector<numpy::int32> codes = {0, 1, 2, 1, 0};
471 pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);
472
473 numpy::NDArray<numpy::int32> arr_codes = arr.codes();
474 if (arr_codes.getSize() != 5) {
475 std::cout << " [FAIL] : in pd_test_categorical_array_codes_property() : codes size != 5" << std::endl;
476 throw std::runtime_error("pd_test_categorical_array_codes_property failed: codes size != 5");
477 }
478
479 // Check codes match
480 for (size_t i = 0; i < codes.size(); ++i) {
481 if (arr_codes.getElementAt({i}) != codes[i]) {
482 std::cout << " [FAIL] : in pd_test_categorical_array_codes_property() : code mismatch at " << i << std::endl;
483 throw std::runtime_error("pd_test_categorical_array_codes_property failed: code mismatch");
format (main.cpp:20)
10int main() {
11 // Automatically log all output to temp/pd_test_output.log
12 numpy::TestLogger logger("temp/pd_test_output.log");
13
14 int res = 0;
15 int res1 = 0;
16 std::string resS = "";
17
18 // call all the tests
19 res1 = dataframe_tests::pd_test_main();
20 resS += std::format(" pd_test_main: {} errors\n", res1);
21 res += res1;
22
23 std::cout << "\n------------------------- main --------------------------------------------\n";
24 std::cout << std::endl << "All tests completed. Nb errors = " << res << std::endl;
25 std::cout << "Details: \n" << resS;
26 std::cout << "\n---------------------------------------------------------------------------\n";
27 return res;
28}
has_category (pd_test_1_all.cpp:5303)
5293}
5294
5295void pd_test_categorical_index_values_with_categories_constructor() {
5296 std::cout << "========= values with categories constructor ==========";
5297
5298 std::vector<std::optional<std::string>> values = {"a", "b", "a"};
5299 std::vector<std::string> categories = {"a", "b", "c", "d"};
5300 pandas::CategoricalIndex idx(values, categories, true, "explicit_cats");
5301
5302 bool passed = (idx.size() == 3 && idx.num_categories() == 4 &&
5303 idx.ordered() && idx.has_category("c") && idx.has_category("d"));
5304 if (!passed) {
5305 std::cout << " [FAIL] : in pd_test_categorical_index_values_with_categories_constructor()" << std::endl;
5306 throw std::runtime_error("pd_test_categorical_index_values_with_categories_constructor failed");
5307 }
5308
5309 std::cout << " -> tests passed" << std::endl;
5310}
5311
5312void pd_test_categorical_index_copy_constructor() {
5313 std::cout << "========= copy constructor ============================";
holds_integer (pd_test_3_all.cpp:3311)
3301 }
3302 if (idx.is_interval()) {
3303 throw std::runtime_error("is_interval() should be false");
3304 }
3305 if (idx.is_numeric()) {
3306 throw std::runtime_error("is_numeric() should be false");
3307 }
3308 if (idx.is_object()) {
3309 throw std::runtime_error("is_object() should be false");
3310 }
3311 if (idx.holds_integer()) {
3312 throw std::runtime_error("holds_integer() should be false");
3313 }
3314
3315 std::cout << " -> tests passed" << std::endl;
3316}
3317
3318void pd_test_3_all_datetime_index_sort() {
3319 std::cout << "========= DatetimeIndex.sort_values() ====================";
3320
3321 pandas::DatetimeIndex idx = pandas::date_range("2024-01-01", "2024-01-05", std::nullopt, "D");
identical (pd_test_1_all.cpp:5883)
5873}
5874
5875void pd_test_categorical_index_identical() {
5876 std::cout << "========= identical ===================================";
5877
5878 pandas::CategoricalArray arr({"a", "b"});
5879 pandas::CategoricalIndex idx1(arr, "same_name");
5880 pandas::CategoricalIndex idx2(arr, "same_name");
5881 pandas::CategoricalIndex idx3(arr, "diff_name");
5882
5883 bool passed = (idx1.identical(idx2) && !idx1.identical(idx3));
5884 if (!passed) {
5885 std::cout << " [FAIL] : in pd_test_categorical_index_identical()" << std::endl;
5886 throw std::runtime_error("pd_test_categorical_index_identical failed");
5887 }
5888
5889 std::cout << " -> tests passed" << std::endl;
5890}
5891
5892// ============================================================================
5893// Inherited Operations Tests
inferred_type (pd_test_1_all.cpp:5270)
5260}
5261
5262void pd_test_categorical_index_array_constructor() {
5263 std::cout << "========= array constructor ===========================";
5264
5265 pandas::CategoricalArray arr({"apple", "banana", "apple", "cherry"});
5266 pandas::CategoricalIndex idx(arr, "fruits");
5267
5268 bool passed = (idx.size() == 4 && !idx.empty() &&
5269 idx.name().has_value() && *idx.name() == "fruits" &&
5270 idx.inferred_type() == "categorical");
5271 if (!passed) {
5272 std::cout << " [FAIL] : in pd_test_categorical_index_array_constructor()" << std::endl;
5273 throw std::runtime_error("pd_test_categorical_index_array_constructor failed");
5274 }
5275
5276 std::cout << " -> tests passed" << std::endl;
5277}
5278
5279void pd_test_categorical_index_values_constructor() {
5280 std::cout << "========= values constructor ==========================";
item (pd_test_3_all.cpp:3712)
3702 // Test is_interval (always false for base Index)
3703 if (int_idx.is_interval()) {
3704 throw std::runtime_error("base Index should not be interval");
3705 }
3706
3707 std::cout << " -> tests passed" << std::endl;
3708}
3709
3710void pd_test_3_all_index_item() {
3711 std::cout << "========= Index.item() =============================";
3712
3713 pandas::Index<numpy::int64> idx1({42});
3714 numpy::int64 val = idx1.item();
3715
3716 if (val != 42) {
3717 throw std::runtime_error("item() should return 42");
3718 }
3719
3720 // Test error for size != 1
3721 pandas::Index<numpy::int64> idx2({1, 2, 3});
memory_usage (pd_test_1_all.cpp:27063)
27053 }
27054
27055 std::cout << "====================================== [OK] pd_test_value_counts test suite ========================== " << std::endl;
27056 return 0;
27057 }
27058
27059} // namespace dataframe_tests
27060// ------------------- pd_test_value_counts.cpp (end) -----------------------------
27061
27062// ------------------- pd_test_memory_usage.cpp (start) -----------------------------
27063// Tests for DataFrame.memory_usage() - pandas-compatible memory usage reporting
27064
27065namespace dataframe_tests {
27066 namespace dataframe_tests_memory_usage {
27067
27068 void pd_test_memory_usage_basic() {
27069 std::cout << "========= basic memory_usage =======================";
27070
27071 // Create a simple DataFrame with multiple columns
27072 std::map<std::string, std::vector<double>> data;
27073 data["A"] = {1.0, 2.0, 3.0, 4.0, 5.0};
num_categories (pd_test_1_all.cpp:5285)
5275 std::cout << " -> tests passed" << std::endl;
5276}
5277
5278void pd_test_categorical_index_values_constructor() {
5279 std::cout << "========= values constructor ==========================";
5280
5281 std::vector<std::optional<std::string>> values = {"a", "b", "a", std::nullopt, "c"};
5282 pandas::CategoricalIndex idx(values, std::optional<std::string>("letters"), false);
5283
5284 bool passed = (idx.size() == 5 && idx.num_categories() == 3 &&
5285 !idx.ordered() && idx.name().has_value() && *idx.name() == "letters");
5286 if (!passed) {
5287 std::cout << " [FAIL] : in pd_test_categorical_index_values_constructor()" << std::endl;
5288 throw std::runtime_error("pd_test_categorical_index_values_constructor failed");
5289 }
5290
5291 std::cout << " -> tests passed" << std::endl;
5292}
5293
5294void pd_test_categorical_index_values_with_categories_constructor() {
ordered (pd_test_1_all.cpp:359)
349 void pd_test_categorical_array_constructors() {
350 std::cout << "========= CategoricalArray: constructors ======================= ";
351
352 // Default constructor
353 pandas::CategoricalArray arr1;
354 if (arr1.size() != 0) {
355 std::cout << " [FAIL] : in pd_test_categorical_array_constructors() : default constructor size != 0" << std::endl;
356 throw std::runtime_error("pd_test_categorical_array_constructors failed: default constructor size != 0");
357 }
358 if (arr1.ordered()) {
359 std::cout << " [FAIL] : in pd_test_categorical_array_constructors() : default should be unordered" << std::endl;
360 throw std::runtime_error("pd_test_categorical_array_constructors failed: default should be unordered");
361 }
362
363 // Constructor from values (infer categories)
364 std::vector<std::optional<std::string>> values = {
365 std::optional<std::string>("a"),
366 std::optional<std::string>("b"),
367 std::optional<std::string>("a"),
368 std::optional<std::string>("c")
ordered (pd_test_1_all.cpp:359)
349 void pd_test_categorical_array_constructors() {
350 std::cout << "========= CategoricalArray: constructors ======================= ";
351
352 // Default constructor
353 pandas::CategoricalArray arr1;
354 if (arr1.size() != 0) {
355 std::cout << " [FAIL] : in pd_test_categorical_array_constructors() : default constructor size != 0" << std::endl;
356 throw std::runtime_error("pd_test_categorical_array_constructors failed: default constructor size != 0");
357 }
358 if (arr1.ordered()) {
359 std::cout << " [FAIL] : in pd_test_categorical_array_constructors() : default should be unordered" << std::endl;
360 throw std::runtime_error("pd_test_categorical_array_constructors failed: default should be unordered");
361 }
362
363 // Constructor from values (infer categories)
364 std::vector<std::optional<std::string>> values = {
365 std::optional<std::string>("a"),
366 std::optional<std::string>("b"),
367 std::optional<std::string>("a"),
368 std::optional<std::string>("c")
putmask (pd_test_3_all.cpp:3752)
3742 // Should be at least sizeof index + 5 * sizeof(int64)
3743 if (usage < 5 * sizeof(numpy::int64)) {
3744 throw std::runtime_error("memory_usage too small");
3745 }
3746
3747 std::cout << " -> tests passed" << std::endl;
3748}
3749
3750void pd_test_3_all_index_putmask() {
3751 std::cout << "========= Index.putmask() ==========================";
3752
3753 pandas::Index<numpy::int64> idx({1, 2, 3, 4, 5});
3754 numpy::NDArray<numpy::bool_> mask(std::vector<size_t>{5});
3755 mask.setElementAt({0}, numpy::bool_(true));
3756 mask.setElementAt({1}, numpy::bool_(false));
3757 mask.setElementAt({2}, numpy::bool_(true));
3758 mask.setElementAt({3}, numpy::bool_(false));
3759 mask.setElementAt({4}, numpy::bool_(true));
3760
3761 auto result = idx.putmask(mask, numpy::int64(99));
ravel (pd_test_3_all.cpp:2147)
2137 throw std::runtime_error("memory_usage shallow too small");
2138 }
2139 if (deep < shallow) {
2140 throw std::runtime_error("memory_usage deep should be >= shallow");
2141 }
2142
2143 std::cout << " -> tests passed" << std::endl;
2144}
2145
2146void pd_test_3_all_categorical_ravel_view() {
2147 std::cout << "========= CategoricalArray.ravel()/view() =============";
2148
2149 std::vector<std::optional<std::string>> values = {"a", "b", "c"};
2150 pandas::CategoricalArray arr(values);
2151
2152 auto raveled = arr.ravel();
2153 if (raveled.size() != 3 || !raveled.equals(arr)) {
2154 throw std::runtime_error("ravel failed");
2155 }
2156
2157 auto viewed = arr.view();
remove_categories (pd_test_1_all.cpp:591)
581 }
582
583 void pd_test_categorical_array_remove_categories() {
584 std::cout << "========= CategoricalArray: remove_categories ======================= ";
585
586 std::vector<std::string> cats = {"a", "b", "c"};
587 std::vector<numpy::int32> codes = {0, 1, 2, 1}; // a, b, c, b
588 pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);
589
590 // Remove 'c' - values with 'c' become NA
591 pandas::CategoricalArray result = arr.remove_categories({"c"});
592
593 if (result.categories().size() != 2) {
594 std::cout << " [FAIL] : in pd_test_categorical_array_remove_categories() : categories size != 2" << std::endl;
595 throw std::runtime_error("pd_test_categorical_array_remove_categories failed: categories size != 2");
596 }
597
598 // Element at index 2 should now be NA (was 'c')
599 if (!result.is_na(2)) {
600 std::cout << " [FAIL] : in pd_test_categorical_array_remove_categories() : removed category should be NA" << std::endl;
601 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)
727 std::cout << " -> tests passed" << std::endl;
728 }
729
730 void pd_test_categorical_array_remove_unused_categories() {
731 std::cout << "========= CategoricalArray: remove_unused_categories ======================= ";
732
733 std::vector<std::string> cats = {"a", "b", "c", "d"};
734 std::vector<numpy::int32> codes = {0, 0, 2}; // a, a, c (b and d unused)
735 pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);
736
737 pandas::CategoricalArray result = arr.remove_unused_categories();
738
739 // Only 'a' and 'c' should remain
740 if (result.categories().size() != 2) {
741 std::cout << " [FAIL] : in pd_test_categorical_array_remove_unused_categories() : categories size != 2" << std::endl;
742 throw std::runtime_error("pd_test_categorical_array_remove_unused_categories failed: categories size != 2");
743 }
744
745 // Values should be preserved
746 std::optional<std::string> val0 = result[0];
747 std::optional<std::string> val2 = result[2];
reorder_categories (pd_test_1_all.cpp:695)
685 void pd_test_categorical_array_reorder_categories() {
686 std::cout << "========= CategoricalArray: reorder_categories ======================= ";
687
688 std::vector<std::string> cats = {"a", "b", "c"};
689 std::vector<numpy::int32> codes = {0, 1, 2}; // a, b, c
690 pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);
691
692 // Reorder categories
693 std::vector<std::string> new_order = {"c", "b", "a"};
694 pandas::CategoricalArray result = arr.reorder_categories(new_order);
695
696 // Check categories are reordered
697 const std::vector<std::string>& result_cats = result.categories();
698 if (result_cats[0] != "c" || result_cats[1] != "b" || result_cats[2] != "a") {
699 std::cout << " [FAIL] : in pd_test_categorical_array_reorder_categories() : categories not reordered" << std::endl;
700 throw std::runtime_error("pd_test_categorical_array_reorder_categories failed: categories not reordered");
701 }
702
703 // Values should be preserved
704 std::optional<std::string> val0 = result[0];
repeat (pd_test_3_all.cpp:2166)
2156 auto viewed = arr.view();
2157 if (viewed.size() != 3 || !viewed.equals(arr)) {
2158 throw std::runtime_error("view failed");
2159 }
2160
2161 std::cout << " -> tests passed" << std::endl;
2162}
2163
2164void pd_test_3_all_categorical_repeat() {
2165 std::cout << "========= CategoricalArray.repeat() ===================";
2166
2167 std::vector<std::optional<std::string>> values = {"a", "b"};
2168 pandas::CategoricalArray arr(values);
2169
2170 auto result = arr.repeat(3);
2171 if (result.size() != 6 || *result[0] != "a" || *result[2] != "a" ||
2172 *result[3] != "b" || *result[5] != "b") {
2173 throw std::runtime_error("repeat scalar failed");
2174 }
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}
round (pd_test_1_all.cpp:1688)
1678 void pd_test_floating_array_rounding() {
1679 std::cout << "========= FloatingArray: rounding ======================= ";
1680
1681 pandas::FloatingArray<double> arr({
1682 std::optional<double>(1.234),
1683 std::optional<double>(2.567),
1684 std::nullopt
1685 });
1686
1687 auto rounded = arr.round(2);
1688 if (std::abs(rounded[0].value() - 1.23) > 0.001 ||
1689 std::abs(rounded[1].value() - 2.57) > 0.001) {
1690 std::cout << " [FAIL] : in pd_test_floating_array_rounding() : round(2)" << std::endl;
1691 throw std::runtime_error("pd_test_floating_array_rounding failed: round(2)");
1692 }
1693
1694 if (!rounded.is_na(2)) {
1695 std::cout << " [FAIL] : in pd_test_floating_array_rounding() : round should preserve NA" << std::endl;
1696 throw std::runtime_error("pd_test_floating_array_rounding failed: NA preservation");
1697 }
set_categories (pd_test_1_all.cpp:623)
613 void pd_test_categorical_array_set_categories() {
614 std::cout << "========= CategoricalArray: set_categories ======================= ";
615
616 std::vector<std::string> cats = {"a", "b"};
617 std::vector<numpy::int32> codes = {0, 1, 0}; // a, b, a
618 pandas::CategoricalArray arr = pandas::CategoricalArray::from_codes(codes, cats);
619
620 // Set new categories (values not in new categories become NA)
621 std::vector<std::string> new_cats = {"a", "c"}; // 'b' removed, 'c' added
622 pandas::CategoricalArray result = arr.set_categories(new_cats);
623
624 if (result.categories().size() != 2) {
625 std::cout << " [FAIL] : in pd_test_categorical_array_set_categories() : categories size != 2" << std::endl;
626 throw std::runtime_error("pd_test_categorical_array_set_categories failed: categories size != 2");
627 }
628
629 // Element at index 1 should be NA (was 'b', now not in categories)
630 if (!result.is_na(1)) {
631 std::cout << " [FAIL] : in pd_test_categorical_array_set_categories() : 'b' value should be NA" << std::endl;
632 throw std::runtime_error("pd_test_categorical_array_set_categories failed: 'b' value should be NA");
slice_indexer (pd_test_3_all.cpp:711)
701 }
702
703 std::cout << " -> tests passed" << std::endl;
704}
705
706// ============================================================================
707// Category 6: Index Indexer Methods
708// ============================================================================
709
710void pd_test_3_all_index_indexers() {
711 std::cout << "========= Index.get_indexer_for/non_unique/slice_indexer() ";
712
713 std::vector<std::string> vals = {"a", "b", "c", "d", "e"};
714 pandas::Index<std::string> idx(vals);
715
716 // Test get_indexer_for()
717 std::vector<std::string> target = {"b", "d", "f"}; // "f" doesn't exist
718 numpy::NDArray<numpy::int64> indexer = idx.get_indexer_for(target);
719 if (indexer.getSize() != 3) {
720 std::cout << " [FAIL] : in pd_test_3_all_index_indexers() : get_indexer_for size mismatch" << std::endl;
721 throw std::runtime_error("pd_test_3_all_index_indexers failed: get_indexer_for size");
slice_locs (pd_test_1_all.cpp:18275)
18265 }
18266
18267 std::cout << "-> tests passed" << std::endl;
18268 }
18269
18270 void pd_test_range_index_slice_locs() {
18271 std::cout << "========= slice_locs ================================== ";
18272
18273 pandas::RangeIndex ri(0, 10); // [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
18274
18275 auto [start_idx, stop_idx] = ri.slice_locs(3, 7);
18276
18277 bool passed = (start_idx == 3 && stop_idx == 8);
18278
18279 if (!passed) {
18280 std::cout << " [FAIL] : slice_locs" << std::endl;
18281 throw std::runtime_error("pd_test_range_index_slice_locs failed");
18282 }
18283
18284 std::cout << "-> tests passed" << std::endl;
18285 }
sort (pd_test_3_all.cpp:3869)
3859 throw std::runtime_error("last 2 positions should be NaN");
3860 }
3861 if (std::abs(result[0] - 3.0) > 0.001) {
3862 throw std::runtime_error("shift(-2) [0] should be 3.0");
3863 }
3864
3865 std::cout << " -> tests passed" << std::endl;
3866}
3867
3868void pd_test_3_all_index_sort() {
3869 std::cout << "========= Index.sort() =============================";
3870
3871 pandas::Index<numpy::int64> idx({3, 1, 4, 1, 5, 9, 2, 6});
3872 auto result = idx.sort();
3873
3874 if (result[0] != 1 || result[1] != 1 || result[7] != 9) {
3875 throw std::runtime_error("sort() not working correctly");
3876 }
3877
3878 // Test descending
3879 result = idx.sort(false);
sortlevel (pd_test_1_all.cpp:14676)
14666 void pd_test_multiindex_sortlevel() {
14667 std::cout << "========= sortlevel =================================== ";
14668
14669 std::vector<std::vector<std::string>> arrays = {
14670 {"b", "a", "c"},
14671 {"2", "1", "3"}
14672 };
14673
14674 pandas::MultiIndex mi = pandas::MultiIndex::from_arrays<std::string>(arrays);
14675 auto [sorted, indices] = mi.sortlevel(0);
14676
14677 bool passed = true;
14678
14679 // After sorting by level 0: a, b, c
14680 if (sorted[0][0] != "a" || sorted[1][0] != "b" || sorted[2][0] != "c") {
14681 std::cout << " [FAIL] : not sorted correctly by level 0" << std::endl;
14682 passed = false;
14683 }
14684
14685 if (!passed) {
type_id (pd_test_3_all.cpp:25592)
25582// ------------------- pd_test_value_classify (end) ------------------
25583
25584// ------------------- pd_test_index_type_id (start) ------------------
25585namespace dataframe_tests_index_type_id {
25586
25587void pd_test_index_type_id_dispatch() {
25588 std::cout << "========= IndexTypeId dispatch =======================";
25589
25590 // RangeIndex
25591 ::pandas::RangeIndex ri(0, 5);
25592 if (ri.type_id() != ::pandas::IndexTypeId::RangeIndex)
25593 throw std::runtime_error("RangeIndex type_id failed");
25594
25595 // Index<string>
25596 ::pandas::Index<std::string> si(std::vector<std::string>{"a", "b", "c"});
25597 if (si.type_id() != ::pandas::IndexTypeId::IndexString)
25598 throw std::runtime_error("Index<string> type_id failed");
25599
25600 // Index<int64>
25601 ::pandas::Index<numpy::int64> ii(std::vector<numpy::int64>{1, 2, 3});
25602 if (ii.type_id() != ::pandas::IndexTypeId::IndexInt64)