For reasons, I am currently trying to make use of minizs

tinfl_decompress(tinfl_decompressor *r, const mz_uint8 *pIn_buf_next, size_t *pIn_buf_size, mz_uint8 *pOut_buf_start, mz_uint8 *pOut_buf_next, size_t *pOut_buf_size, const mz_uint32 decomp_flags);

on an ESP32. As stated in the thin documentation available I initialize the pOut_buf_start to have size TINFL_LZ_DICT_SIZE. Since I am decoding a network stream, the size of the decompressed data is potentially infinite, but I feed a maximum of 512 bytes at once in the input buffer to avoid getting too many decompressed data at once.

Everything works fine until I reach the point where pOut_buf_next gets close to the end of the output buffer:

• I can keep decompressing, but will run into a TINFL_STATUS_HAS_MORE_OUTPUT sooner or later (which, as I understand, leaves me in a unrecoverable state?)
• I can wrap the output buffer (pOut_buf_next = pOut_buf_start and pOut_buf_size=TINFL_LZ_DICT_SIZE), but that seems to corrupt the output of my next tinfl_decompress call

How do I handle this best/correct?
Is TINFL_STATUS_HAS_MORE_OUTPUT really unrecoverable? At what point am I allowed to wrap the output buffer?