# Purpose
On KODE=1, CBESK computes an N member sequence of complex Bessel functions CY(L)=K(FNU+L-1,Z) for real nonnegative orders FNU+L-1, L=1,...,N and complex Z.NE.0 in the cut plane -pi<arg(Z)<=pi. On KODE=2, CBESJ returns the scaled
# Description
This canonical unsafe binding exposes original SLATEC routine `CBESK`. Its documented behavior is taken from the selected, source-hash-verified provider prologue; the exact implementation source is [CBESK](https://www.netlib.org/slatec/src/cbesk.f).
# Arguments
## `Z`
**Direction:** `input`. **Fortran type:** `COMPLEX`. **Rust ABI type:** `*mut crate::Complex32`. **Shape:** scalar.
Nonzero argument of type COMPLEX.
## `FNU`
**Direction:** `input`. **Fortran type:** `REAL`. **Rust ABI type:** `*mut f32`. **Shape:** scalar.
Initial order of type REAL, FNU>=0.
## `KODE`
**Direction:** `input`. **Fortran type:** `INTEGER`. **Rust ABI type:** `*mut crate::FortranInteger`. **Shape:** scalar.
A parameter to indicate the scaling option 1 returns.
## `N`
**Direction:** `input`. **Fortran type:** `INTEGER`. **Rust ABI type:** `*mut crate::FortranInteger`. **Shape:** scalar.
Number of terms in the sequence, N>=1.
## `CY`
**Direction:** `input-output`. **Fortran type:** `COMPLEX`. **Rust ABI type:** `*mut crate::Complex32`. **Shape:** rank 1; dimensions (N).
exp(Z)*K(FNU+L-1,Z), L=1,. ,N which remove the exponential growth in both the left and right half planes as Z goes to infinity. Definitions and notation are found in the NBS Handbook of Mathematical Functions (Ref. 1). =2 returns K(FNU+L-1,Z)*EXP(Z), L=1,. ,N Result vector of type COMPLEX.
## `NZ`
**Direction:** `status-output`. **Fortran type:** `INTEGER`. **Rust ABI type:** `*mut crate::FortranInteger`. **Shape:** scalar.
Number of underflows set to zero NZ=0 Normal return NZ>0 CY(L)=0 for NZ values of L (if Re(Z)>0 then CY(L)=0 for L=1,. ,NZ; in the complementary half plane the underflows may not be in an uninterrupted sequence).
## `IERR`
**Direction:** `output`. **Fortran type:** `INTEGER`. **Rust ABI type:** `*mut crate::FortranInteger`. **Shape:** scalar.
Error flag 0 Normal return - COMPUTATION COMPLETED 1 Input error - NO COMPUTATION 2 Overflow - NO COMPUTATION (abs(Z) too small and/or FNU+N-1 too large) 3 Precision warning - COMPUTATION COMPLETED (Result has half precision or less because abs(Z) or FNU+N-1 is large) 4 Precision error - NO COMPUTATION (Result has no precision because abs(Z) or FNU+N-1 is too large) 5 Algorithmic error - NO COMPUTATION (Termination condition not met).
# Return value
This is a Fortran subroutine and has no direct return value. Its results, status, and any persistent solver state are communicated through the documented arguments.
# Status and error values
| `NZ` | `0` | Normal return for NZ values of L (if Re(Z)>0 for L=1,...,NZ; in the complementary half plane the underflows may not be in an uninterrupted sequence) |
# Workspace and array requirements
- `CY`: not a workspace argument
# ABI notes
- Canonical Rust path: `slatec_sys::special::complex::cbesk`
- Original SLATEC routine: `CBESK`
- Native symbol: `cbesk_`
- ABI fingerprint: `subroutine:void(mut_complex32,mut_f32,mut_i32,mut_i32,mut_complex32_array_rank1,mut_i32,mut_i32)`
- Exact Netlib source file: [CBESK](https://www.netlib.org/slatec/src/cbesk.f)
# Safety
Every raw pointer must be non-null unless this argument contract expressly permits null, correctly aligned, and valid for its documented readable or writable extent. Preserve Fortran column-major layout, length, leading-dimension, workspace, callback-lifetime, aliasing, and provider-runtime requirements. The native routine does not retain ordinary argument pointers beyond this call.