pandas-dev/pandas · error · ValueError

searchsorted requires array to be sorted, which is impossibl

Error message

searchsorted requires array to be sorted, which is impossible with NAs present.

What it means

StringArray.searchsorted requires a sorted array, but missing values (pd.NA or np.nan) have no defined sort order, making a correct result impossible. The method checks only the first and last positions (or sorter[0]/sorter[-1] when a sorter is given) for NA, since NA must be confined to the ends in sorted data, and raises ValueError if found.

Source

Thrown at pandas/core/arrays/string_.py:1198

        """

        # GH#65837: avoid O(n) scan; NA confined to array ends in sorted data.
        # When sorter is given, the sorted order is ndarray[sorter], so check
        # the first/last positions via sorter instead of raw ndarray positions.
        ndarray = self._ndarray
        if len(ndarray):
            if sorter is None:
                has_na = libmissing.checknull(ndarray[0]) or libmissing.checknull(
                    ndarray[-1]
                )
            else:
                has_na = libmissing.checknull(
                    ndarray[sorter[0]]
                ) or libmissing.checknull(ndarray[sorter[-1]])
        else:
            has_na = False
        if has_na:
            raise ValueError(
                "searchsorted requires array to be sorted, which is impossible "
                "with NAs present."
            )
        return super().searchsorted(value=value, side=side, sorter=sorter)

    def _cmp_method(self, other, op):
        from pandas.arrays import (
            ArrowExtensionArray,
            BooleanArray,
        )

        if (
            isinstance(other, BaseStringArray)
            and self.dtype.na_value is not libmissing.NA
            and other.dtype.na_value is libmissing.NA
        ):
            # NA has priority of NaN semantics
            return op(self.astype(other.dtype, copy=False), other)

View on GitHub (pinned to 71959b8cb9)

Solutions

1. Drop NAs before searching: clean = arr[~arr.isna()].
2. Fill missing values with a concrete string if a sentinel is acceptable.
3. Sort and verify no NAs at the ends before calling searchsorted.

Example fix

// before
string_array.searchsorted('x')

// after
clean = string_array[~string_array.isna()]
clean.searchsorted('x')

clean = string_array[~string_array.isna()]
result = clean.searchsorted('x')

import pandas as pd

def has_no_na(arr) -> bool:
    return not bool(pd.isna(arr).any())

When it happens

Trigger: Calling string_array.searchsorted('x') or searchsorted(['a','z']) on a StringArray that contains pd.NA/np.nan at the leading or trailing positions, or passing a sorter whose endpoints point at NA values.

Common situations: Binary-search lookups on string data loaded from CSVs with missing fields; using searchsorted on an uncleaned column; sorted indexes that still carry nulls.

Related errors

• searchsorted requires array to be sorted, which is impossibl
• searchsorted requires array to be sorted, which is impossibl
• invalid na_position: {na_position!r}
• missing values must be missing in the same location both lef
• cannot convert to '{dtype}'-dtype NumPy array with missing v

AI-assisted analysis of pandas-dev/pandas@71959b8cb9 (2026-08-07). Data as JSON: /api/errors/f9d465f22ec115b2. Report an issue: GitHub.