Function Strs_sample

python/stringzilla.c:6447–6532

}

static PyObject *Strs_sample(Strs *self, PyObject *const *args, Py_ssize_t positional_args_count,
                             PyObject *args_names_tuple) {
    PyObject *sample_size_obj = NULL;
    PyObject *seed_obj = NULL;

    // Check for positional arguments
    if (positional_args_count > 1) {
        PyErr_SetString(PyExc_TypeError, "sample() takes 1 positional argument and 1 keyword argument");
        return NULL;
    }
    else if (positional_args_count == 1) { sample_size_obj = args[0]; }

    // Parse keyword arguments
    if (args_names_tuple) {
        Py_ssize_t args_names_count = PyTuple_GET_SIZE(args_names_tuple);
        for (Py_ssize_t i = 0; i < args_names_count; ++i) {
            PyObject *key = PyTuple_GET_ITEM(args_names_tuple, i);
            PyObject *value = args[positional_args_count + i];
            if (PyUnicode_CompareWithASCIIString(key, "seed") == 0 && !seed_obj) { seed_obj = value; }
            else if (PyErr_Format(PyExc_TypeError, "Got an unexpected keyword argument '%U'", key)) { return NULL; }
        }
    }

    // Translate the seed and the sample size to C types
    sz_size_t sample_size = 0;
    if (sample_size_obj) {
        if (!PyLong_Check(sample_size_obj)) {
            PyErr_SetString(PyExc_TypeError, "The sample size must be an integer");
            return NULL;
        }
        sample_size = PyLong_AsSize_t(sample_size_obj);
    }
    unsigned int seed = (unsigned int)time(NULL); // Default seed
    if (seed_obj) {
        if (!PyLong_Check(seed_obj)) {
            PyErr_SetString(PyExc_TypeError, "The seed must be an integer");
            return NULL;
        }
        seed = PyLong_AsUnsignedLong(seed_obj);
    }

    // Create a new `Strs` object
    Strs *result = (Strs *)StrsType.tp_alloc(&StrsType, 0);
    if (result == NULL && PyErr_NoMemory()) return NULL;

    // Initialize the memory allocator with default malloc wrapper
    sz_memory_allocator_init_default(&result->data.fragmented.allocator);

    result->layout = STRS_FRAGMENTED;
    result->data.fragmented.count = 0;
    result->data.fragmented.spans = NULL;
    result->data.fragmented.parent = NULL;
    if (sample_size == 0) { return (PyObject *)result; }

    // Now create a new Strs object with the sampled strings
    sz_string_view_t *result_spans = malloc(sample_size * sizeof(sz_string_view_t));
    if (!result_spans) {
        PyErr_SetString(PyExc_MemoryError, "Failed to allocate memory for the sample");