IntervalIndex#
-
class pandas::IntervalIndex#
Index class for axis labels in pandas data structures.
Example#
#include <pandas/pandas.h>
using namespace pandas;
// Create IntervalIndex
IntervalIndex<int64_t> idx({1, 2, 3}, "my_index");
size_t len = idx.size();
Constructors#
Signature |
Location |
Example |
|---|---|---|
|
pd_interval_index.h:144 |
|
|
pd_interval_index.h:153 |
|
|
pd_interval_index.h:162 |
|
|
pd_interval_index.h:171 |
Construction#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
static IntervalIndex |
pd_interval_index.h:203 |
|
|
static IntervalIndex |
pd_interval_index.h:231 |
|
|
static IntervalIndex |
pd_interval_index.h:260 |
|
|
static IntervalIndex |
pd_interval_index.h:286 |
|
|
static IntervalIndex |
pd_interval_index.h:313 |
Indexing / Selection#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
pd_interval_index.h:1441 |
||
|
IntervalIndex |
pd_interval_index.h:1751 |
|
|
size_t |
pd_interval_index.h:1364 |
|
|
std::string |
pd_interval_index.h:735 |
|
|
std::string |
pd_interval_index.h:723 |
|
|
IntervalIndex |
pd_interval_index.h:1643 |
Data Manipulation#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
IntervalIndex |
pd_interval_index.h:979 |
|
|
IntervalIndex |
pd_interval_index.h:1763 |
|
|
IntervalIndex |
pd_interval_index.h:1084 |
|
|
std::pair<IntervalIndex, numpy::NDArray<numpy::int64>> |
pd_interval_index.h:1562 |
|
|
IntervalIndex |
pd_interval_index.h:757 |
|
|
IntervalIndex |
pd_interval_index.h:1698 |
Statistics#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
std::optional<std::pair<T, T>> |
pd_interval_index.h:830 |
|
|
std::optional<std::pair<T, T>> |
pd_interval_index.h:816 |
Aggregation#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
std::unordered_map<GroupKey, std::vector<size_t>> |
pd_interval_index.h:1816 |
|
|
std::unordered_map<GroupT, std::vector<size_t>> |
pd_interval_index.h:1848 |
|
|
IntervalIndex |
pd_interval_index.h:1616 |
Arithmetic#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
const std::string& |
pd_interval_index.h:537 |
Comparison#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
numpy::NDArray<T> |
pd_interval_index.h:344 |
|
|
numpy::NDArray<T> |
pd_interval_index.h:366 |
|
|
IntervalArray<T> |
pd_interval_index.h:1939 |
|
|
IntervalArray<T> |
pd_interval_index.h:1959 |
|
|
IntervalArray<T> |
pd_interval_index.h:2028 |
|
|
IntervalArray<T> |
pd_interval_index.h:1100 |
|
|
IntervalArray<T> |
pd_interval_index.h:1189 |
|
|
IntervalArray<T> |
pd_interval_index.h:1212 |
|
|
IntervalArray<T> |
pd_interval_index.h:1545 |
|
|
IntervalArray<T> |
pd_interval_index.h:1601 |
|
|
IntervalArray<T> |
pd_interval_index.h:1633 |
|
|
IntervalArray<T> |
pd_interval_index.h:1660 |
|
|
IntervalArray<T> |
pd_interval_index.h:1687 |
|
|
IntervalArray<T> |
pd_interval_index.h:1741 |
Sorting#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
size_t |
pd_interval_index.h:1477 |
Reshaping#
Combining#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
IntervalIndex |
pd_interval_index.h:1890 |
Time Series#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
std::optional<std::pair<T, T>> |
pd_interval_index.h:1314 |
|
|
numpy::NDArray<numpy::int64> |
pd_interval_index.h:1336 |
|
|
numpy::NDArray<T> |
pd_interval_index.h:1109 |
|
|
IntervalIndex |
pd_interval_index.h:1159 |
I/O#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
IntervalIndex |
pd_interval_index.h:1243 |
|
|
std::vector<std::pair<T, T>> |
pd_interval_index.h:1255 |
|
|
SeriesData |
pd_interval_index.h:2141 |
|
|
std::string |
pd_interval_index.h:649 |
|
|
std::vector<std::optional<std::pair<T, T>>> |
pd_interval_index.h:519 |
|
|
std::vector<std::optional<std::pair<T, T>>> |
pd_interval_index.h:1274 |
Conversion#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
IntervalIndex |
pd_interval_index.h:1782 |
|
|
IntervalIndex |
pd_interval_index.h:770 |
|
|
IntervalIndex |
pd_interval_index.h:1712 |
|
|
IntervalIndex |
pd_interval_index.h:1228 |
Set Operations#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
std::vector<bool> |
pd_interval_index.h:1038 |
|
|
IntervalIndex |
pd_interval_index.h:1973 |
|
|
IntervalIndex |
pd_interval_index.h:960 |
Type Checking#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
bool |
pd_interval_index.h:1772 |
|
|
bool |
pd_interval_index.h:873 |
|
|
bool |
pd_interval_index.h:880 |
|
|
BooleanArray |
pd_interval_index.h:412 |
|
|
bool |
pd_interval_index.h:887 |
|
|
bool |
pd_interval_index.h:894 |
|
|
bool |
pd_interval_index.h:866 |
|
|
bool |
pd_interval_index.h:391 |
|
|
bool |
pd_interval_index.h:439 |
|
|
bool |
pd_interval_index.h:901 |
|
|
bool |
pd_interval_index.h:908 |
|
|
bool |
pd_interval_index.h:424 |
|
|
bool |
pd_interval_index.h:398 |
Other Methods#
Signature |
Return Type |
Location |
Example |
|---|---|---|---|
|
bool |
pd_interval_index.h:783 |
|
|
static bool |
pd_interval_index.h:63 |
|
|
bool |
pd_interval_index.h:801 |
|
|
numpy::int64 |
pd_interval_index.h:848 |
|
|
numpy::int64 |
pd_interval_index.h:841 |
|
|
IntervalArray<T> |
pd_interval_index.h:972 |
|
|
IntervalArray<T> |
pd_interval_index.h:1032 |
|
|
std::unique_ptr<IndexBase> |
pd_interval_index.h:561 |
|
|
IntervalClosed |
pd_interval_index.h:377 |
|
|
std::string |
pd_interval_index.h:384 |
|
|
void |
pd_interval_index.h:74 |
|
|
BooleanArray |
pd_interval_index.h:456 |
|
|
BooleanArray |
pd_interval_index.h:466 |
|
|
IntervalIndex |
pd_interval_index.h:1924 |
|
|
IntervalIndex |
pd_interval_index.h:1948 |
|
|
std::string |
pd_interval_index.h:704 |
|
|
std::string |
pd_interval_index.h:544 |
|
|
oss << |
pd_interval_index.h:601 |
|
|
oss << |
pd_interval_index.h:639 |
|
|
std::vector<std::string> |
pd_interval_index.h:1286 |
|
|
std::string |
pd_interval_index.h:575 |
|
|
bool |
pd_interval_index.h:859 |
|
|
std::string |
pd_interval_index.h:655 |
|
|
std::string |
pd_interval_index.h:554 |
|
|
std::pair<T, T> |
pd_interval_index.h:930 |
|
|
size_t |
pd_interval_index.h:921 |
|
|
FloatingArray<numpy::float64> |
pd_interval_index.h:359 |
|
|
BooleanArray |
pd_interval_index.h:477 |
|
|
BooleanArray |
pd_interval_index.h:489 |
|
|
IntervalIndex |
pd_interval_index.h:1670 |
|
|
IntervalIndex |
pd_interval_index.h:1220 |
|
|
IntervalIndex |
pd_interval_index.h:1199 |
|
|
std::string |
pd_interval_index.h:716 |
|
|
IntervalIndex |
pd_interval_index.h:762 |
|
|
numpy::NDArray<T> |
pd_interval_index.h:351 |
|
|
IntervalIndex |
pd_interval_index.h:1721 |
|
|
IntervalIndex |
pd_interval_index.h:503 |
|
|
IntervalIndex |
pd_interval_index.h:511 |
|
|
void |
pd_interval_index.h:530 |
|
|
std::vector<size_t> |
pd_interval_index.h:1420 |
|
|
std::pair<size_t, size_t> |
pd_interval_index.h:1385 |
|
|
IntervalIndex |
pd_interval_index.h:1509 |
|
|
std::pair<IntervalIndex, numpy::NDArray<numpy::int64>> |
pd_interval_index.h:1522 |
|
|
StringMethods<IntervalIndex<T>> |
pd_interval_index.h:743 |
|
|
IndexTypeId |
pd_interval_index.h:565 |
Code Examples#
The following examples are extracted from the test suite.
IntervalIndex (pd_test_5_all.cpp:1350)
1340 std::cout << " -> tests passed" << std::endl;
1341}
1342
1343
1344// --- cpp_f_test_pandas_advanced_indexing_compare_full_1053.cpp ---
1345void f_test_pandas_advanced_indexing_compare_full_1053() {
1346 std::cout << "========= f_test_pandas_advanced_indexing_compare_full_1053 =======";
1347 int local_fail = 0;
1348 // Expected from pandas: str(pd.cut(range(4), bins=2).categories)
1349 std::string expected =
1350 "IntervalIndex([(-0.003, 1.5], (1.5, 3.0]], dtype='interval[float64, right]')";
1351
1352 // Placeholder: cut() + categories accessor needed in C++ API
1353 // pandas_tests::check(result_categories.to_string() == expected, "binning.cut_categories.str");
1354 pandas_tests::check(true, "binning.cut_categories.str (expected value captured)", local_fail);
1355 if (local_fail > 0) {
1356 std::cout << " [FAIL] : in f_test_pandas_advanced_indexing_compare_full_1053() : " << local_fail << " checks failed" << std::endl;
1357 throw std::runtime_error("f_test_pandas_advanced_indexing_compare_full_1053 failed");
1358 }
1359 std::cout << " -> tests passed" << std::endl;
1360}
IntervalIndex (pd_test_5_all.cpp:1350)
1340 std::cout << " -> tests passed" << std::endl;
1341}
1342
1343
1344// --- cpp_f_test_pandas_advanced_indexing_compare_full_1053.cpp ---
1345void f_test_pandas_advanced_indexing_compare_full_1053() {
1346 std::cout << "========= f_test_pandas_advanced_indexing_compare_full_1053 =======";
1347 int local_fail = 0;
1348 // Expected from pandas: str(pd.cut(range(4), bins=2).categories)
1349 std::string expected =
1350 "IntervalIndex([(-0.003, 1.5], (1.5, 3.0]], dtype='interval[float64, right]')";
1351
1352 // Placeholder: cut() + categories accessor needed in C++ API
1353 // pandas_tests::check(result_categories.to_string() == expected, "binning.cut_categories.str");
1354 pandas_tests::check(true, "binning.cut_categories.str (expected value captured)", local_fail);
1355 if (local_fail > 0) {
1356 std::cout << " [FAIL] : in f_test_pandas_advanced_indexing_compare_full_1053() : " << local_fail << " checks failed" << std::endl;
1357 throw std::runtime_error("f_test_pandas_advanced_indexing_compare_full_1053 failed");
1358 }
1359 std::cout << " -> tests passed" << std::endl;
1360}
IntervalIndex (pd_test_5_all.cpp:1350)
1340 std::cout << " -> tests passed" << std::endl;
1341}
1342
1343
1344// --- cpp_f_test_pandas_advanced_indexing_compare_full_1053.cpp ---
1345void f_test_pandas_advanced_indexing_compare_full_1053() {
1346 std::cout << "========= f_test_pandas_advanced_indexing_compare_full_1053 =======";
1347 int local_fail = 0;
1348 // Expected from pandas: str(pd.cut(range(4), bins=2).categories)
1349 std::string expected =
1350 "IntervalIndex([(-0.003, 1.5], (1.5, 3.0]], dtype='interval[float64, right]')";
1351
1352 // Placeholder: cut() + categories accessor needed in C++ API
1353 // pandas_tests::check(result_categories.to_string() == expected, "binning.cut_categories.str");
1354 pandas_tests::check(true, "binning.cut_categories.str (expected value captured)", local_fail);
1355 if (local_fail > 0) {
1356 std::cout << " [FAIL] : in f_test_pandas_advanced_indexing_compare_full_1053() : " << local_fail << " checks failed" << std::endl;
1357 throw std::runtime_error("f_test_pandas_advanced_indexing_compare_full_1053 failed");
1358 }
1359 std::cout << " -> tests passed" << std::endl;
1360}
IntervalIndex (pd_test_5_all.cpp:1350)
1340 std::cout << " -> tests passed" << std::endl;
1341}
1342
1343
1344// --- cpp_f_test_pandas_advanced_indexing_compare_full_1053.cpp ---
1345void f_test_pandas_advanced_indexing_compare_full_1053() {
1346 std::cout << "========= f_test_pandas_advanced_indexing_compare_full_1053 =======";
1347 int local_fail = 0;
1348 // Expected from pandas: str(pd.cut(range(4), bins=2).categories)
1349 std::string expected =
1350 "IntervalIndex([(-0.003, 1.5], (1.5, 3.0]], dtype='interval[float64, right]')";
1351
1352 // Placeholder: cut() + categories accessor needed in C++ API
1353 // pandas_tests::check(result_categories.to_string() == expected, "binning.cut_categories.str");
1354 pandas_tests::check(true, "binning.cut_categories.str (expected value captured)", local_fail);
1355 if (local_fail > 0) {
1356 std::cout << " [FAIL] : in f_test_pandas_advanced_indexing_compare_full_1053() : " << local_fail << " checks failed" << std::endl;
1357 throw std::runtime_error("f_test_pandas_advanced_indexing_compare_full_1053 failed");
1358 }
1359 std::cout << " -> tests passed" << std::endl;
1360}
from_arrays (pd_test_1_all.cpp:1994)
1984// ============================================================================
1985// Test: from_arrays factory method
1986// ============================================================================
1987void test_from_arrays() {
1988 std::cout << "========= IntervalArray: from_arrays ======================= ";
1989
1990 std::vector<numpy::int64> left_vec = {0, 10, 20};
1991 std::vector<numpy::int64> right_vec = {5, 15, 25};
1992
1993 auto arr = pandas::IntervalArrayInt64::from_arrays(left_vec, right_vec);
1994
1995 if (arr.size() != 3) {
1996 std::cout << "[FAIL] : in test_from_arrays() : size" << std::endl;
1997 return;
1998 }
1999
2000 auto interval1 = arr[1];
2001 if (!interval1.has_value() || interval1->first != 10 || interval1->second != 15) {
2002 std::cout << "[FAIL] : in test_from_arrays() : interval values" << std::endl;
2003 return;
from_arrays (pd_test_1_all.cpp:1994)
1984// ============================================================================
1985// Test: from_arrays factory method
1986// ============================================================================
1987void test_from_arrays() {
1988 std::cout << "========= IntervalArray: from_arrays ======================= ";
1989
1990 std::vector<numpy::int64> left_vec = {0, 10, 20};
1991 std::vector<numpy::int64> right_vec = {5, 15, 25};
1992
1993 auto arr = pandas::IntervalArrayInt64::from_arrays(left_vec, right_vec);
1994
1995 if (arr.size() != 3) {
1996 std::cout << "[FAIL] : in test_from_arrays() : size" << std::endl;
1997 return;
1998 }
1999
2000 auto interval1 = arr[1];
2001 if (!interval1.has_value() || interval1->first != 10 || interval1->second != 15) {
2002 std::cout << "[FAIL] : in test_from_arrays() : interval values" << std::endl;
2003 return;
from_breaks (pd_test_1_all.cpp:1955)
1945}
1946
1947// ============================================================================
1948// Test: from_breaks factory method
1949// ============================================================================
1950void test_from_breaks() {
1951 std::cout << "========= IntervalArray: from_breaks ======================= ";
1952
1953 // Create from breaks
1954 std::vector<numpy::float64> breaks = {0.0, 1.0, 2.0, 3.0, 4.0};
1955 auto arr = pandas::IntervalArrayFloat64::from_breaks(breaks);
1956
1957 if (arr.size() != 4) {
1958 std::cout << "[FAIL] : in test_from_breaks() : size should be n-1" << std::endl;
1959 return;
1960 }
1961
1962 // Check intervals
1963 auto interval0 = arr[0];
1964 if (!interval0.has_value() || interval0->first != 0.0 || interval0->second != 1.0) {
1965 std::cout << "[FAIL] : in test_from_breaks() : first interval" << std::endl;
from_breaks (pd_test_1_all.cpp:1955)
1945}
1946
1947// ============================================================================
1948// Test: from_breaks factory method
1949// ============================================================================
1950void test_from_breaks() {
1951 std::cout << "========= IntervalArray: from_breaks ======================= ";
1952
1953 // Create from breaks
1954 std::vector<numpy::float64> breaks = {0.0, 1.0, 2.0, 3.0, 4.0};
1955 auto arr = pandas::IntervalArrayFloat64::from_breaks(breaks);
1956
1957 if (arr.size() != 4) {
1958 std::cout << "[FAIL] : in test_from_breaks() : size should be n-1" << std::endl;
1959 return;
1960 }
1961
1962 // Check intervals
1963 auto interval0 = arr[0];
1964 if (!interval0.has_value() || interval0->first != 0.0 || interval0->second != 1.0) {
1965 std::cout << "[FAIL] : in test_from_breaks() : first interval" << std::endl;
from_tuples (pd_test_1_all.cpp:2022)
2012// ============================================================================
2013void test_from_tuples() {
2014 std::cout << "========= IntervalArray: from_tuples ======================= ";
2015
2016 std::vector<std::pair<numpy::float64, numpy::float64>> tuples = {
2017 {0.0, 1.5},
2018 {1.5, 3.0},
2019 {3.0, 4.5}
2020 };
2021
2022 auto arr = pandas::IntervalArrayFloat64::from_tuples(tuples);
2023
2024 if (arr.size() != 3) {
2025 std::cout << "[FAIL] : in test_from_tuples() : size" << std::endl;
2026 return;
2027 }
2028
2029 auto interval2 = arr[2];
2030 if (!interval2.has_value() || interval2->first != 3.0 || interval2->second != 4.5) {
2031 std::cout << "[FAIL] : in test_from_tuples() : interval values" << std::endl;
2032 return;
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_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_string (pd_test_3_all.cpp:27746)
27736 }
27737 }
27738
27739 pandas::Series<numpy::int64> si({10, 20, 30}, "ints");
27740 auto result2 = si.astype("str");
27741 auto* str_s2 = dynamic_cast<pandas::Series<std::string>*>(result2.get());
27742 if (!str_s2) {
27743 std::cout << " FAIL: expected Series<string> from int" << std::endl;
27744 fail++;
27745 } else {
27746 if (str_s2->get_string(0) != "10") {
27747 std::cout << " FAIL: expected '10', got '" << str_s2->get_string(0) << "'" << std::endl;
27748 fail++;
27749 }
27750 }
27751
27752 if (fail == 0) std::cout << " OK" << std::endl;
27753}
27754
27755void pd_test_astype_datetime_to_string() {
27756 std::cout << " -- pd_test_astype_datetime_to_string --" << std::endl;
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";
drop_duplicates (pd_test_1_all.cpp:6639)
6629 }
6630 }
6631
6632 // Test drop_duplicates
6633 {
6634 std::map<std::string, std::vector<numpy::int64>> dup_data;
6635 dup_data["A"] = {1, 1, 2, 2};
6636 dup_data["B"] = {1, 1, 2, 3};
6637 pandas::DataFrame df_dup(dup_data);
6638
6639 auto deduped = df_dup.drop_duplicates();
6640 // Rows 0 and 1 are duplicates (A=1, B=1), so should have 3 rows
6641 if (deduped.nrows() != 3) {
6642 std::cout << " [FAIL] : in pd_test_dataframe_manipulation() : drop_duplicates nrows != 3, got " << deduped.nrows() << std::endl;
6643 throw std::runtime_error("pd_test_dataframe_manipulation failed: drop_duplicates");
6644 }
6645 }
6646
6647 // Test assign
6648 {
6649 std::map<std::string, std::vector<numpy::int64>> assign_data;
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];
insert (pd_test_1_all.cpp:12028)
12018 }
12019
12020 std::cout << " -> tests passed" << std::endl;
12021 }
12022
12023 void pd_test_index_insert_delete() {
12024 std::cout << "========= insert and delete ===========================";
12025
12026 pandas::Index<numpy::int64> idx{1, 2, 4, 5};
12027
12028 auto inserted = idx.insert(2, 3);
12029 bool passed = (inserted.size() == 5);
12030 passed = passed && (inserted[2] == 3);
12031
12032 auto deleted = inserted.delete_(2);
12033 passed = passed && (deleted.size() == 4);
12034 passed = passed && deleted.equals(idx);
12035
12036 if (!passed) {
12037 std::cout << " [FAIL] : in pd_test_index_insert_delete() : insert/delete failed" << std::endl;
12038 throw std::runtime_error("pd_test_index_insert_delete failed");
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}
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 }
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}
subtype_override (pd_test_3_all.cpp:24889)
24879 return dataframe_tests_bdate_timedelta_range::pd_test_bdate_timedelta_range_main();
24880}
24881// ------------------- pd_test_bdate_timedelta_range (end) ---------------------------
24882
24883// ------------------- pd_test_interval_type_inference (begin) ---------------------------
24884namespace dataframe_tests_interval_type_inference {
24885
24886void pd_test_interval_type_inference_breaks_int() {
24887 std::cout << "========= interval_type_inference_breaks_int ======================= ";
24888 auto idx = pandas::IntervalIndex<double>::from_breaks({0.0, 1.0, 2.0, 3.0});
24889 if (idx.subtype_override() != "int64")
24890 throw std::runtime_error("expected subtype_override 'int64', got '" + idx.subtype_override() + "'");
24891 std::string dtype = idx.dtype_name();
24892 if (dtype.find("int64") == std::string::npos)
24893 throw std::runtime_error("expected dtype containing 'int64', got '" + dtype + "'");
24894 std::string fmt = idx.format_interval(0);
24895 if (fmt.find('.') != std::string::npos)
24896 throw std::runtime_error("expected integer format without decimal, got '" + fmt + "'");
24897 std::cout << " -> tests passed" << std::endl;
24898}
left (pd_test_1_all.cpp:1909)
1899 if (empty.size() != 0) {
1900 std::cout << "[FAIL] : in test_constructors() : default constructor size" << std::endl;
1901 return;
1902 }
1903 if (empty.closed() != pandas::IntervalClosed::Right) {
1904 std::cout << "[FAIL] : in test_constructors() : default closure" << std::endl;
1905 return;
1906 }
1907
1908 // Constructor from left/right arrays
1909 numpy::NDArray<numpy::float64> left(std::vector<size_t>{3});
1910 numpy::NDArray<numpy::float64> right(std::vector<size_t>{3});
1911 left.setElementAt({0}, 0.0); right.setElementAt({0}, 1.0);
1912 left.setElementAt({1}, 1.0); right.setElementAt({1}, 2.0);
1913 left.setElementAt({2}, 2.0); right.setElementAt({2}, 3.0);
1914
1915 pandas::IntervalArrayFloat64 arr1(left, right);
1916 if (arr1.size() != 3) {
1917 std::cout << "[FAIL] : in test_constructors() : array size" << std::endl;
1918 return;
1919 }
length (pd_test_1_all.cpp:2137)
2127 auto mid0 = mid_arr[0];
2128 auto mid1 = mid_arr[1];
2129 auto mid2 = mid_arr[2];
2130 if (!mid0.has_value() || std::abs(mid0.value() - 1.0) > 1e-10 ||
2131 !mid1.has_value() || std::abs(mid1.value() - 3.5) > 1e-10 ||
2132 !mid2.has_value() || std::abs(mid2.value() - 7.5) > 1e-10) {
2133 std::cout << "[FAIL] : in test_left_right_mid_length() : mid()" << std::endl;
2134 return;
2135 }
2136
2137 // Test length()
2138 auto len_arr = arr.length();
2139 if (len_arr.getElementAt({0}) != 2.0 ||
2140 len_arr.getElementAt({1}) != 3.0 ||
2141 len_arr.getElementAt({2}) != 5.0) {
2142 std::cout << "[FAIL] : in test_left_right_mid_length() : length()" << std::endl;
2143 return;
2144 }
2145
2146 std::cout << "-> tests passed" << std::endl;
2147}
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 }
to_frame (pd_test_3_all.cpp:4931)
4921 size_t usage = mi.memory_usage(true);
4922 if (usage == 0) {
4923 throw std::runtime_error("memory_usage() should return > 0");
4924 }
4925
4926 std::cout << " -> tests passed" << std::endl;
4927}
4928
4929void pd_test_3_all_multiindex_to_frame() {
4930 std::cout << "========= MultiIndex.to_frame() =======================";
4931
4932 std::vector<std::vector<std::string>> arrays = {{"a", "b"}, {"x", "y"}};
4933 std::vector<std::optional<std::string>> names = {"first", "second"};
4934 pandas::MultiIndex mi = pandas::MultiIndex::from_arrays<std::string>(arrays, names);
4935
4936 auto frame = mi.to_frame();
4937 if (frame.find("first") == frame.end() || frame.find("second") == frame.end()) {
4938 throw std::runtime_error("to_frame() missing columns");
4939 }
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_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_series (pd_test_3_all.cpp:5788)
5778 throw std::runtime_error("to_frame use_index should be false when index=false");
5779 }
5780 if (frame3.column_name != "0") {
5781 throw std::runtime_error("to_frame column_name should be '0' when no name");
5782 }
5783
5784 std::cout << " -> tests passed" << std::endl;
5785}
5786
5787void pd_test_3_all_period_index_to_series() {
5788 std::cout << "========= PeriodIndex.to_series() =====================";
5789
5790 pandas::PeriodIndex idx = make_period_index({1, 2, 3}, "M").rename("periods");
5791
5792 // Test to_series() with default parameters
5793 pandas::PeriodIndex::SeriesData series = idx.to_series();
5794
5795 // values should have same size
5796 if (series.values.size() != 3) {
5797 throw std::runtime_error("to_series values size should be 3");
5798 }
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];
to_tuples (pd_test_1_all.cpp:13037)
13027 }
13028
13029 std::cout << " -> tests passed" << std::endl;
13030}
13031
13032void pd_test_interval_index_to_tuples() {
13033 std::cout << "========= to_tuples =========================";
13034
13035 auto idx = pandas::IntervalIndex64::from_breaks({0, 1, 2, 3});
13036
13037 auto tuples = idx.to_tuples();
13038
13039 bool passed = (tuples.size() == 3 &&
13040 tuples[0].has_value() && tuples[0]->first == 0 && tuples[0]->second == 1 &&
13041 tuples[1].has_value() && tuples[1]->first == 1 && tuples[1]->second == 2 &&
13042 tuples[2].has_value() && tuples[2]->first == 2 && tuples[2]->second == 3);
13043 if (!passed) {
13044 std::cout << " [FAIL] : in pd_test_interval_index_to_tuples() : check failed" << std::endl;
13045 throw std::runtime_error("pd_test_interval_index_to_tuples failed");
13046 }
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();
duplicated (pd_test_1_all.cpp:10583)
10573 std::cout << " -> tests passed" << std::endl;
10574}
10575
10576void pd_test_extension_index_duplicated() {
10577 std::cout << "========= duplicated =========================";
10578
10579 pandas::CategoricalArray arr({"a", "b", "a", "c", "a"});
10580 pandas::CategoricalIndex idx(arr);
10581
10582 auto dup_mask = idx.duplicated("first");
10583
10584 bool passed = (dup_mask.getElementAt({0}) == false &&
10585 dup_mask.getElementAt({1}) == false &&
10586 dup_mask.getElementAt({2}) == true &&
10587 dup_mask.getElementAt({3}) == false &&
10588 dup_mask.getElementAt({4}) == true);
10589 if (!passed) {
10590 std::cout << " [FAIL] : in pd_test_extension_index_duplicated() : duplicated check failed" << std::endl;
10591 throw std::runtime_error("pd_test_extension_index_duplicated failed");
10592 }
union_ (pd_test_1_all.cpp:10694)
10684 std::cout << "========= union =========================";
10685
10686 // Use same categories for both arrays
10687 std::vector<std::string> cats = {"a", "b", "c", "d", "e"};
10688 pandas::CategoricalArray arr1({"a", "b", "c"}, cats);
10689 pandas::CategoricalIndex idx1(arr1);
10690
10691 pandas::CategoricalArray arr2({"b", "c", "d", "e"}, cats);
10692 pandas::CategoricalIndex idx2(arr2);
10693
10694 auto uni = idx1.union_(idx2);
10695
10696 bool passed = (uni.size() == 5 &&
10697 uni.contains("a") && uni.contains("b") && uni.contains("c") &&
10698 uni.contains("d") && uni.contains("e"));
10699 if (!passed) {
10700 std::cout << " [FAIL] : in pd_test_extension_index_union() : union check failed" << std::endl;
10701 throw std::runtime_error("pd_test_extension_index_union failed");
10702 }
10703
10704 std::cout << " -> tests passed" << std::endl;
unique (pd_test_1_all.cpp:1345)
1335 pandas::DatetimeArray arr(std::vector<std::string>{
1336 "2023-01-01",
1337 "2023-06-15",
1338 "2023-01-01",
1339 "NaT",
1340 "2023-06-15",
1341 "NaT"
1342 });
1343
1344 // unique
1345 auto uniq = arr.unique();
1346 // Should have: NaT, 2023-01-01, 2023-06-15 (3 unique values)
1347 if (uniq.size() != 3) {
1348 std::cout << " [FAIL] : unique size should be 3, got " << uniq.size() << std::endl;
1349 throw std::runtime_error("pd_test_datetime_array_unique failed: size");
1350 }
1351
1352 // factorize
1353 auto [codes, uniques] = arr.factorize();
1354 // Codes for NaT should be -1
1355 if (codes.getElementAt({3}) != -1) {
is_ (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_empty (pd_test_1_all.cpp:2164)
2154 // Test with right-closed intervals (a, b]
2155 std::vector<std::pair<numpy::float64, numpy::float64>> tuples = {
2156 {0.0, 1.0}, // Not empty
2157 {1.0, 1.0}, // Empty (1, 1] has no points
2158 {2.0, 2.0}, // Empty
2159 {2.0, 3.0} // Not empty
2160 };
2161
2162 auto arr_right = pandas::IntervalArrayFloat64::from_tuples(tuples, pandas::IntervalClosed::Right);
2163 auto empty_right = arr_right.is_empty();
2164
2165 if (empty_right[0].value_or(true) != false ||
2166 empty_right[1].value_or(false) != true ||
2167 empty_right[2].value_or(false) != true ||
2168 empty_right[3].value_or(true) != false) {
2169 std::cout << "[FAIL] : in test_is_empty() : right-closed" << std::endl;
2170 return;
2171 }
2172
2173 // Test with both-closed intervals [a, b] - [1, 1] is NOT empty
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_left_closed (pd_test_1_all.cpp:12830)
12820}
12821
12822void pd_test_interval_index_is_left_right_closed() {
12823 std::cout << "========= is_left_closed/is_right_closed =========================";
12824
12825 auto idx_right = pandas::IntervalIndex64::from_breaks({0, 1, 2}, pandas::IntervalClosed::Right);
12826 auto idx_left = pandas::IntervalIndex64::from_breaks({0, 1, 2}, pandas::IntervalClosed::Left);
12827 auto idx_both = pandas::IntervalIndex64::from_breaks({0, 1, 2}, pandas::IntervalClosed::Both);
12828 auto idx_neither = pandas::IntervalIndex64::from_breaks({0, 1, 2}, pandas::IntervalClosed::Neither);
12829
12830 bool passed = (!idx_right.is_left_closed() && idx_right.is_right_closed() &&
12831 idx_left.is_left_closed() && !idx_left.is_right_closed() &&
12832 idx_both.is_left_closed() && idx_both.is_right_closed() &&
12833 !idx_neither.is_left_closed() && !idx_neither.is_right_closed());
12834 if (!passed) {
12835 std::cout << " [FAIL] : in pd_test_interval_index_is_left_right_closed() : check failed" << std::endl;
12836 throw std::runtime_error("pd_test_interval_index_is_left_right_closed failed");
12837 }
12838
12839 std::cout << " -> tests passed" << std::endl;
12840}
is_non_overlapping_monotonic (pd_test_1_all.cpp:2457)
2447// ============================================================================
2448// Test: is_non_overlapping_monotonic
2449// ============================================================================
2450void test_is_non_overlapping_monotonic() {
2451 std::cout << "========= IntervalArray: is_non_overlapping_monotonic ======================= ";
2452
2453 // Monotonic, non-overlapping
2454 std::vector<numpy::float64> breaks1 = {0.0, 1.0, 2.0, 3.0};
2455 auto arr1 = pandas::IntervalArrayFloat64::from_breaks(breaks1, pandas::IntervalClosed::Right);
2456 if (!arr1.is_non_overlapping_monotonic()) {
2457 std::cout << "[FAIL] : in test_is_non_overlapping_monotonic() : should be true for breaks" << std::endl;
2458 return;
2459 }
2460
2461 // Non-monotonic (out of order)
2462 std::vector<std::pair<numpy::float64, numpy::float64>> tuples2 = {
2463 {2.0, 3.0}, {0.0, 1.0}, {1.0, 2.0}
2464 };
2465 auto arr2 = pandas::IntervalArrayFloat64::from_tuples(tuples2);
2466 if (arr2.is_non_overlapping_monotonic()) {
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 }
is_overlapping (pd_test_1_all.cpp:12891)
12881 std::cout << "========= is_overlapping =========================";
12882
12883 // Non-overlapping intervals
12884 auto idx1 = pandas::IntervalIndex64::from_breaks({0, 1, 2, 3});
12885
12886 // Overlapping intervals: [0,5], [3,8]
12887 std::vector<numpy::int64> left_vals = {0, 3};
12888 std::vector<numpy::int64> right_vals = {5, 8};
12889 auto idx2 = pandas::IntervalIndex64::from_arrays(left_vals, right_vals, pandas::IntervalClosed::Both);
12890
12891 bool passed = (!idx1.is_overlapping() && idx2.is_overlapping());
12892 if (!passed) {
12893 std::cout << " [FAIL] : in pd_test_interval_index_is_overlapping() : check failed" << std::endl;
12894 throw std::runtime_error("pd_test_interval_index_is_overlapping failed");
12895 }
12896
12897 std::cout << " -> tests passed" << std::endl;
12898}
12899
12900void pd_test_interval_index_is_non_overlapping_monotonic() {
12901 std::cout << "========= is_non_overlapping_monotonic =========================";
is_right_closed (pd_test_1_all.cpp:12830)
12820}
12821
12822void pd_test_interval_index_is_left_right_closed() {
12823 std::cout << "========= is_left_closed/is_right_closed =========================";
12824
12825 auto idx_right = pandas::IntervalIndex64::from_breaks({0, 1, 2}, pandas::IntervalClosed::Right);
12826 auto idx_left = pandas::IntervalIndex64::from_breaks({0, 1, 2}, pandas::IntervalClosed::Left);
12827 auto idx_both = pandas::IntervalIndex64::from_breaks({0, 1, 2}, pandas::IntervalClosed::Both);
12828 auto idx_neither = pandas::IntervalIndex64::from_breaks({0, 1, 2}, pandas::IntervalClosed::Neither);
12829
12830 bool passed = (!idx_right.is_left_closed() && idx_right.is_right_closed() &&
12831 idx_left.is_left_closed() && !idx_left.is_right_closed() &&
12832 idx_both.is_left_closed() && idx_both.is_right_closed() &&
12833 !idx_neither.is_left_closed() && !idx_neither.is_right_closed());
12834 if (!passed) {
12835 std::cout << " [FAIL] : in pd_test_interval_index_is_left_right_closed() : check failed" << std::endl;
12836 throw std::runtime_error("pd_test_interval_index_is_left_right_closed failed");
12837 }
12838
12839 std::cout << " -> tests passed" << std::endl;
12840}
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");
arr (pd_test_1_all.cpp:45)
35 std::cout << " [FAIL] : in pd_test_boolean_array_constructors() : initializer_list size != 4" << std::endl;
36 throw std::runtime_error("pd_test_boolean_array_constructors failed: initializer_list size != 4");
37 }
38
39 std::cout << " -> tests passed" << std::endl;
40 }
41
42 void pd_test_boolean_array_na_handling() {
43 std::cout << "========= BooleanArray: NA handling ======================= ";
44
45 pandas::BooleanArray arr({
46 std::optional<bool>(true),
47 std::nullopt, // NA at index 1
48 std::optional<bool>(false)
49 });
50
51 if (!arr.is_na(1)) {
52 std::cout << " [FAIL] : in pd_test_boolean_array_na_handling() : is_na(1) should be true" << std::endl;
53 throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(1) should be true");
54 }
arr (pd_test_1_all.cpp:45)
35 std::cout << " [FAIL] : in pd_test_boolean_array_constructors() : initializer_list size != 4" << std::endl;
36 throw std::runtime_error("pd_test_boolean_array_constructors failed: initializer_list size != 4");
37 }
38
39 std::cout << " -> tests passed" << std::endl;
40 }
41
42 void pd_test_boolean_array_na_handling() {
43 std::cout << "========= BooleanArray: NA handling ======================= ";
44
45 pandas::BooleanArray arr({
46 std::optional<bool>(true),
47 std::nullopt, // NA at index 1
48 std::optional<bool>(false)
49 });
50
51 if (!arr.is_na(1)) {
52 std::cout << " [FAIL] : in pd_test_boolean_array_na_handling() : is_na(1) should be true" << std::endl;
53 throw std::runtime_error("pd_test_boolean_array_na_handling failed: is_na(1) should be true");
54 }
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}
closed (pd_test_1_all.cpp:1903)
1893// ============================================================================
1894void test_constructors() {
1895 std::cout << "========= IntervalArray: constructors ======================= ";
1896
1897 // Default constructor
1898 pandas::IntervalArrayFloat64 empty;
1899 if (empty.size() != 0) {
1900 std::cout << "[FAIL] : in test_constructors() : default constructor size" << std::endl;
1901 return;
1902 }
1903 if (empty.closed() != pandas::IntervalClosed::Right) {
1904 std::cout << "[FAIL] : in test_constructors() : default closure" << std::endl;
1905 return;
1906 }
1907
1908 // Constructor from left/right arrays
1909 numpy::NDArray<numpy::float64> left(std::vector<size_t>{3});
1910 numpy::NDArray<numpy::float64> right(std::vector<size_t>{3});
1911 left.setElementAt({0}, 0.0); right.setElementAt({0}, 1.0);
1912 left.setElementAt({1}, 1.0); right.setElementAt({1}, 2.0);
1913 left.setElementAt({2}, 2.0); right.setElementAt({2}, 3.0);
closed_string (pd_test_1_all.cpp:12813)
12803 std::cout << " -> tests passed" << std::endl;
12804}
12805
12806void pd_test_interval_index_closed_string() {
12807 std::cout << "========= closed_string =========================";
12808
12809 auto idx_right = pandas::IntervalIndex64::from_breaks({0, 1, 2}, pandas::IntervalClosed::Right);
12810 auto idx_left = pandas::IntervalIndex64::from_breaks({0, 1, 2}, pandas::IntervalClosed::Left);
12811
12812 bool passed = (idx_right.closed_string() == "right" && idx_left.closed_string() == "left");
12813 if (!passed) {
12814 std::cout << " [FAIL] : in pd_test_interval_index_closed_string() : closed_string check failed" << std::endl;
12815 throw std::runtime_error("pd_test_interval_index_closed_string failed");
12816 }
12817
12818 std::cout << " -> tests passed" << std::endl;
12819}
12820
12821void pd_test_interval_index_is_left_right_closed() {
12822 std::cout << "========= is_left_closed/is_right_closed =========================";
contains (pd_test_1_all.cpp:2200)
2190// Test: contains method
2191// ============================================================================
2192void test_contains() {
2193 std::cout << "========= IntervalArray: contains ======================= ";
2194
2195 std::vector<numpy::float64> breaks = {0.0, 1.0, 2.0, 3.0};
2196
2197 // Right-closed intervals: (0, 1], (1, 2], (2, 3]
2198 auto arr_right = pandas::IntervalArrayFloat64::from_breaks(breaks, pandas::IntervalClosed::Right);
2199
2200 // Test contains(1.0) - should be in interval 0 but not 1 (since 1 is exclusive on left of interval 1)
2201 auto contains_1 = arr_right.contains(1.0);
2202 // (0, 1] contains 1: yes, (1, 2] contains 1: no (open on left), (2, 3] contains 1: no
2203 if (contains_1[0].value_or(false) != true ||
2204 contains_1[1].value_or(true) != false ||
2205 contains_1[2].value_or(true) != false) {
2206 std::cout << "[FAIL] : in test_contains() : right-closed contains 1.0" << std::endl;
2207 return;
2208 }
2209
2210 // Left-closed intervals: [0, 1), [1, 2), [2, 3)
contains (pd_test_1_all.cpp:2200)
2190// Test: contains method
2191// ============================================================================
2192void test_contains() {
2193 std::cout << "========= IntervalArray: contains ======================= ";
2194
2195 std::vector<numpy::float64> breaks = {0.0, 1.0, 2.0, 3.0};
2196
2197 // Right-closed intervals: (0, 1], (1, 2], (2, 3]
2198 auto arr_right = pandas::IntervalArrayFloat64::from_breaks(breaks, pandas::IntervalClosed::Right);
2199
2200 // Test contains(1.0) - should be in interval 0 but not 1 (since 1 is exclusive on left of interval 1)
2201 auto contains_1 = arr_right.contains(1.0);
2202 // (0, 1] contains 1: yes, (1, 2] contains 1: no (open on left), (2, 3] contains 1: no
2203 if (contains_1[0].value_or(false) != true ||
2204 contains_1[1].value_or(true) != false ||
2205 contains_1[2].value_or(true) != false) {
2206 std::cout << "[FAIL] : in test_contains() : right-closed contains 1.0" << std::endl;
2207 return;
2208 }
2209
2210 // Left-closed intervals: [0, 1), [1, 2), [2, 3)
delete_ (pd_test_1_all.cpp:10501)
10491 std::cout << " -> tests passed" << std::endl;
10492}
10493
10494void pd_test_extension_index_delete() {
10495 std::cout << "========= delete_ =========================";
10496
10497 pandas::CategoricalArray arr({"a", "b", "c", "d"});
10498 pandas::CategoricalIndex idx(arr);
10499
10500 auto deleted = idx.delete_(1);
10501 auto v0 = deleted[0];
10502 auto v1 = deleted[1];
10503 auto v2 = deleted[2];
10504
10505 bool passed = (deleted.size() == 3 &&
10506 v0.has_value() && *v0 == "a" &&
10507 v1.has_value() && *v1 == "c" &&
10508 v2.has_value() && *v2 == "d");
10509 if (!passed) {
10510 std::cout << " [FAIL] : in pd_test_extension_index_delete() : delete_ check failed" << std::endl;
delete_ (pd_test_1_all.cpp:10501)
10491 std::cout << " -> tests passed" << std::endl;
10492}
10493
10494void pd_test_extension_index_delete() {
10495 std::cout << "========= delete_ =========================";
10496
10497 pandas::CategoricalArray arr({"a", "b", "c", "d"});
10498 pandas::CategoricalIndex idx(arr);
10499
10500 auto deleted = idx.delete_(1);
10501 auto v0 = deleted[0];
10502 auto v1 = deleted[1];
10503 auto v2 = deleted[2];
10504
10505 bool passed = (deleted.size() == 3 &&
10506 v0.has_value() && *v0 == "a" &&
10507 v1.has_value() && *v1 == "c" &&
10508 v2.has_value() && *v2 == "d");
10509 if (!passed) {
10510 std::cout << " [FAIL] : in pd_test_extension_index_delete() : delete_ check failed" << std::endl;
dtype_name (pd_test_1_all.cpp:10104)
10094}
10095
10096void pd_test_extension_index_array_constructor() {
10097 std::cout << "========= array constructor =========================";
10098
10099 pandas::CategoricalArray arr({"apple", "banana", "apple", "cherry"});
10100 pandas::CategoricalIndex idx(arr, "fruits");
10101
10102 bool passed = (idx.size() == 4 && !idx.empty() &&
10103 idx.name().has_value() && *idx.name() == "fruits" &&
10104 idx.dtype_name() == "category");
10105 if (!passed) {
10106 std::cout << " [FAIL] : in pd_test_extension_index_array_constructor() : array constructor check failed" << std::endl;
10107 throw std::runtime_error("pd_test_extension_index_array_constructor failed");
10108 }
10109
10110 std::cout << " -> tests passed" << std::endl;
10111}
10112
10113void pd_test_extension_index_copy_constructor() {
10114 std::cout << "========= copy constructor =========================";
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}
format_interval (pd_test_3_all.cpp:24894)
24884namespace dataframe_tests_interval_type_inference {
24885
24886void pd_test_interval_type_inference_breaks_int() {
24887 std::cout << "========= interval_type_inference_breaks_int ======================= ";
24888 auto idx = pandas::IntervalIndex<double>::from_breaks({0.0, 1.0, 2.0, 3.0});
24889 if (idx.subtype_override() != "int64")
24890 throw std::runtime_error("expected subtype_override 'int64', got '" + idx.subtype_override() + "'");
24891 std::string dtype = idx.dtype_name();
24892 if (dtype.find("int64") == std::string::npos)
24893 throw std::runtime_error("expected dtype containing 'int64', got '" + dtype + "'");
24894 std::string fmt = idx.format_interval(0);
24895 if (fmt.find('.') != std::string::npos)
24896 throw std::runtime_error("expected integer format without decimal, got '" + fmt + "'");
24897 std::cout << " -> tests passed" << std::endl;
24898}
24899
24900void pd_test_interval_type_inference_breaks_float() {
24901 std::cout << "========= interval_type_inference_breaks_float ===================== ";
24902 auto idx = pandas::IntervalIndex<double>::from_breaks({0.0, 1.5, 3.0});
24903 if (!idx.subtype_override().empty())
24904 throw std::runtime_error("expected empty subtype_override, got '" + idx.subtype_override() + "'");
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");
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};
mid (pd_test_1_all.cpp:2124)
2114 // Test right()
2115 auto right_arr = arr.right();
2116 if (right_arr.getElementAt({0}) != 2.0 ||
2117 right_arr.getElementAt({1}) != 5.0 ||
2118 right_arr.getElementAt({2}) != 10.0) {
2119 std::cout << "[FAIL] : in test_left_right_mid_length() : right()" << std::endl;
2120 return;
2121 }
2122
2123 // Test mid()
2124 auto mid_arr = arr.mid();
2125 // (0+2)/2=1, (2+5)/2=3.5, (5+10)/2=7.5
2126 auto mid0 = mid_arr[0];
2127 auto mid1 = mid_arr[1];
2128 auto mid2 = mid_arr[2];
2129 if (!mid0.has_value() || std::abs(mid0.value() - 1.0) > 1e-10 ||
2130 !mid1.has_value() || std::abs(mid1.value() - 3.5) > 1e-10 ||
2131 !mid2.has_value() || std::abs(mid2.value() - 7.5) > 1e-10) {
2132 std::cout << "[FAIL] : in test_left_right_mid_length() : mid()" << std::endl;
2133 return;
overlaps (pd_test_1_all.cpp:2244)
2234// Test: overlaps method
2235// ============================================================================
2236void test_overlaps() {
2237 std::cout << "========= IntervalArray: overlaps ======================= ";
2238
2239 std::vector<numpy::float64> breaks = {0.0, 2.0, 4.0, 6.0};
2240 // Right-closed: (0, 2], (2, 4], (4, 6]
2241 auto arr = pandas::IntervalArrayFloat64::from_breaks(breaks, pandas::IntervalClosed::Right);
2242
2243 // Check overlap with (1, 3]
2244 auto overlap_1_3 = arr.overlaps(1.0, 3.0);
2245 // (0, 2] overlaps (1, 3]? Yes (share 1-2)
2246 // (2, 4] overlaps (1, 3]? Yes (share 2-3)
2247 // (4, 6] overlaps (1, 3]? No
2248 if (overlap_1_3[0].value_or(false) != true ||
2249 overlap_1_3[1].value_or(false) != true ||
2250 overlap_1_3[2].value_or(true) != false) {
2251 std::cout << "[FAIL] : in test_overlaps() : overlaps (1, 3]" << std::endl;
2252 return;
2253 }
overlaps (pd_test_1_all.cpp:2244)
2234// Test: overlaps method
2235// ============================================================================
2236void test_overlaps() {
2237 std::cout << "========= IntervalArray: overlaps ======================= ";
2238
2239 std::vector<numpy::float64> breaks = {0.0, 2.0, 4.0, 6.0};
2240 // Right-closed: (0, 2], (2, 4], (4, 6]
2241 auto arr = pandas::IntervalArrayFloat64::from_breaks(breaks, pandas::IntervalClosed::Right);
2242
2243 // Check overlap with (1, 3]
2244 auto overlap_1_3 = arr.overlaps(1.0, 3.0);
2245 // (0, 2] overlaps (1, 3]? Yes (share 1-2)
2246 // (2, 4] overlaps (1, 3]? Yes (share 2-3)
2247 // (4, 6] overlaps (1, 3]? No
2248 if (overlap_1_3[0].value_or(false) != true ||
2249 overlap_1_3[1].value_or(false) != true ||
2250 overlap_1_3[2].value_or(true) != false) {
2251 std::cout << "[FAIL] : in test_overlaps() : overlaps (1, 3]" << std::endl;
2252 return;
2253 }
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();
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}
result (pd_test_1_all.cpp:15406)
15396 data.setElementAt({0}, numpy::datetime64(100LL, numpy::DateTimeUnit::Nanosecond));
15397 data.setElementAt({1}, numpy::datetime64(200LL, numpy::DateTimeUnit::Nanosecond));
15398
15399 numpy::NDArray<numpy::bool_> mask(std::vector<size_t>{2});
15400 mask.setElementAt({0}, numpy::bool_(false));
15401 mask.setElementAt({1}, numpy::bool_(false));
15402
15403 pandas::DatetimeArray arr(data, mask);
15404 pandas::DatetimeIndexBase idx(arr, "original");
15405
15406 // Create join result (int64 values)
15407 numpy::NDArray<numpy::int64> join_result(std::vector<size_t>{3});
15408 join_result.setElementAt({0}, numpy::int64(500LL));
15409 join_result.setElementAt({1}, numpy::int64(600LL));
15410 join_result.setElementAt({2}, numpy::int64(700LL));
15411
15412 auto new_idx = idx._from_join_target(join_result);
15413
15414 bool passed = (new_idx.size() == 3 &&
15415 new_idx.name().has_value() && *new_idx.name() == "original");
15416 if (!passed) {
right (pd_test_1_all.cpp:1910)
1900 std::cout << "[FAIL] : in test_constructors() : default constructor size" << std::endl;
1901 return;
1902 }
1903 if (empty.closed() != pandas::IntervalClosed::Right) {
1904 std::cout << "[FAIL] : in test_constructors() : default closure" << std::endl;
1905 return;
1906 }
1907
1908 // Constructor from left/right arrays
1909 numpy::NDArray<numpy::float64> left(std::vector<size_t>{3});
1910 numpy::NDArray<numpy::float64> right(std::vector<size_t>{3});
1911 left.setElementAt({0}, 0.0); right.setElementAt({0}, 1.0);
1912 left.setElementAt({1}, 1.0); right.setElementAt({1}, 2.0);
1913 left.setElementAt({2}, 2.0); right.setElementAt({2}, 3.0);
1914
1915 pandas::IntervalArrayFloat64 arr1(left, right);
1916 if (arr1.size() != 3) {
1917 std::cout << "[FAIL] : in test_constructors() : array size" << std::endl;
1918 return;
1919 }
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_closed (pd_test_1_all.cpp:2285)
2275 std::vector<numpy::float64> breaks = {0.0, 1.0, 2.0};
2276 auto arr = pandas::IntervalArrayFloat64::from_breaks(breaks, pandas::IntervalClosed::Right);
2277
2278 if (arr.closed() != pandas::IntervalClosed::Right) {
2279 std::cout << "[FAIL] : in test_set_closed() : initial closure" << std::endl;
2280 return;
2281 }
2282
2283 // Change to left-closed
2284 auto arr_left = arr.set_closed(pandas::IntervalClosed::Left);
2285 if (arr_left.closed() != pandas::IntervalClosed::Left) {
2286 std::cout << "[FAIL] : in test_set_closed() : set to Left" << std::endl;
2287 return;
2288 }
2289
2290 // Original should be unchanged
2291 if (arr.closed() != pandas::IntervalClosed::Right) {
2292 std::cout << "[FAIL] : in test_set_closed() : original changed" << std::endl;
2293 return;
2294 }
set_closed (pd_test_1_all.cpp:2285)
2275 std::vector<numpy::float64> breaks = {0.0, 1.0, 2.0};
2276 auto arr = pandas::IntervalArrayFloat64::from_breaks(breaks, pandas::IntervalClosed::Right);
2277
2278 if (arr.closed() != pandas::IntervalClosed::Right) {
2279 std::cout << "[FAIL] : in test_set_closed() : initial closure" << std::endl;
2280 return;
2281 }
2282
2283 // Change to left-closed
2284 auto arr_left = arr.set_closed(pandas::IntervalClosed::Left);
2285 if (arr_left.closed() != pandas::IntervalClosed::Left) {
2286 std::cout << "[FAIL] : in test_set_closed() : set to Left" << std::endl;
2287 return;
2288 }
2289
2290 // Original should be unchanged
2291 if (arr.closed() != pandas::IntervalClosed::Right) {
2292 std::cout << "[FAIL] : in test_set_closed() : original changed" << std::endl;
2293 return;
2294 }
set_subtype_override (pd_test_3_all.cpp:24977)
24967 std::cout << "========= Interval repr float bounds ====================";
24968 pandas::Interval<double> iv(0.0, 1.5);
24969 if (iv.repr() != "Interval(0.0, 1.5, closed='right')")
24970 throw std::runtime_error("repr mismatch: " + iv.repr());
24971 std::cout << " -> tests passed" << std::endl;
24972}
24973
24974void pd_test_interval_repr_timedelta() {
24975 std::cout << "========= Interval repr timedelta subtype ===============";
24976 pandas::Interval<double> iv(0.0, 86400000000000.0); // 1 day in nanos
24977 iv.set_subtype_override("timedelta64[ns]");
24978 std::string r = iv.repr();
24979 if (r.find("Timedelta") == std::string::npos)
24980 throw std::runtime_error("expected Timedelta in repr: " + r);
24981 if (r.find("1 days") == std::string::npos)
24982 throw std::runtime_error("expected '1 days' in repr: " + r);
24983 std::cout << " -> tests passed" << std::endl;
24984}
24985
24986void pd_test_interval_str_integer() {
24987 std::cout << "========= Interval to_string integer bounds =============";
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) {
str (pd_test_1_all.cpp:7137)
7127 // Test basic info() with stringstream
7128 std::map<std::string, std::vector<int>> data = {
7129 {"A", {1, 2, 3, 4, 5}},
7130 {"B", {10, 20, 30, 40, 50}},
7131 {"C", {100, 200, 300, 400, 500}}
7132 };
7133 pandas::DataFrame df(data);
7134
7135 std::ostringstream oss;
7136 df.info(oss);
7137 std::string output = oss.str();
7138
7139 // Verify key components
7140 if (output.find("<class 'pandas.core.frame.DataFrame'>") == std::string::npos) {
7141 std::cout << " [FAIL] : info missing class name" << std::endl;
7142 throw std::runtime_error("pd_test_dataframe_info failed: missing class name");
7143 }
7144 if (output.find("RangeIndex:") == std::string::npos) {
7145 std::cout << " [FAIL] : info missing RangeIndex" << std::endl;
7146 throw std::runtime_error("pd_test_dataframe_info failed: missing RangeIndex");
7147 }
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)