Skip to content

Latest commit

 

History

History
170 lines (165 loc) · 46.4 KB

File metadata and controls

170 lines (165 loc) · 46.4 KB

text_tokenizer

Source: emel/text/tokenizer/sm.hpp

Mermaid

stateDiagram-v2
  direction TB
  [*] --> uninitialized
  uninitialized --> binding_preprocessor : bind_runtime [can_bind_] / begin_bind_
  uninitialized --> errored : bind_runtime [always] / reject_bind_
  uninitialized --> errored : tokenize_runtime [always] / reject_invalid_
  idle --> binding_preprocessor : bind_runtime [can_bind_] / begin_bind_
  idle --> errored : bind_runtime [always] / reject_bind_
  idle --> preprocessing : tokenize_runtime [can_tokenize_] / begin_tokenize_
  idle --> errored : tokenize_runtime [always] / reject_invalid_
  done --> binding_preprocessor : bind_runtime [can_bind_] / begin_bind_
  done --> errored : bind_runtime [always] / reject_bind_
  done --> preprocessing : tokenize_runtime [can_tokenize_] / begin_tokenize_
  done --> errored : tokenize_runtime [always] / reject_invalid_
  errored --> binding_preprocessor : bind_runtime [can_bind_] / begin_bind_
  errored --> errored : bind_runtime [always] / reject_bind_
  errored --> preprocessing : tokenize_runtime [can_tokenize_] / begin_tokenize_
  errored --> errored : tokenize_runtime [always] / reject_invalid_
  unexpected --> binding_preprocessor : bind_runtime [can_bind_] / begin_bind_
  unexpected --> unexpected : bind_runtime [always] / reject_bind_
  unexpected --> preprocessing : tokenize_runtime [can_tokenize_] / begin_tokenize_
  unexpected --> unexpected : tokenize_runtime [always] / reject_invalid_
  binding_preprocessor --> binding_preprocessor_decision : completion_bind_runtime_ [always] / bind_preprocessor_
  binding_preprocessor_decision --> binding_encoder : completion_bind_runtime_ [bind_preprocessor_error_none_] / none
  binding_preprocessor_decision --> errored : completion_bind_runtime_ [bind_preprocessor_error_invalid_request_] / none
  binding_preprocessor_decision --> errored : completion_bind_runtime_ [bind_preprocessor_error_model_invalid_] / none
  binding_preprocessor_decision --> errored : completion_bind_runtime_ [bind_preprocessor_error_backend_error_] / none
  binding_preprocessor_decision --> errored : completion_bind_runtime_ [bind_preprocessor_error_unknown_] / none
  binding_encoder --> binding_encoder_decision : completion_bind_runtime_ [always] / bind_encoder_
  binding_encoder_decision --> idle : completion_bind_runtime_ [bind_encoder_error_none_] / mark_bind_success_
  binding_encoder_decision --> errored : completion_bind_runtime_ [bind_encoder_error_invalid_request_] / none
  binding_encoder_decision --> errored : completion_bind_runtime_ [bind_encoder_error_model_invalid_] / none
  binding_encoder_decision --> errored : completion_bind_runtime_ [bind_encoder_error_backend_error_] / none
  binding_encoder_decision --> errored : completion_bind_runtime_ [bind_encoder_error_unknown_] / none
  preprocessing --> preprocess_decision : completion_tokenize_runtime_ [always] / dispatch_preprocess_
  preprocess_decision --> errored : completion_tokenize_runtime_ [preprocess_rejected_no_error_] / set_backend_error_
  preprocess_decision --> errored : completion_tokenize_runtime_ [preprocess_reported_error_] / set_error_from_preprocess_
  preprocess_decision --> errored : completion_tokenize_runtime_ [preprocess_fragment_count_invalid_] / set_invalid_request_error_
  preprocess_decision --> prefix_decision : completion_tokenize_runtime_ [preprocess_success_] / none
  prefix_decision --> encoding_ready : completion_tokenize_runtime_ [bos_ready_] / append_bos_
  prefix_decision --> errored : completion_tokenize_runtime_ [bos_no_capacity_] / set_invalid_request_error_
  prefix_decision --> errored : completion_tokenize_runtime_ [bos_invalid_id_] / set_invalid_id_error_
  prefix_decision --> encoding_ready : completion_tokenize_runtime_ [no_prefix_] / none
  encoding_ready --> suffix_decision : completion_tokenize_runtime_ [no_more_fragments_] / none
  encoding_ready --> errored : completion_tokenize_runtime_ [more_fragments_no_capacity_] / set_invalid_request_error_
  encoding_ready --> errored : completion_tokenize_runtime_ [more_fragments_token_invalid_] / set_invalid_request_error_
  encoding_ready --> encoding_token_fragment : completion_tokenize_runtime_ [more_fragments_token_valid_] / none
  encoding_ready --> encoding_raw_fragment : completion_tokenize_runtime_ [more_fragments_raw_] / none
  encoding_token_fragment --> encoding_ready : completion_tokenize_runtime_ [always] / append_fragment_token_
  encoding_raw_fragment --> encoding_raw_decision : completion_tokenize_runtime_ [always] / dispatch_encode_raw_fragment_
  encoding_raw_decision --> errored : completion_tokenize_runtime_ [encode_rejected_no_error_] / set_invalid_id_error_
  encoding_raw_decision --> errored : completion_tokenize_runtime_ [encode_reported_error_] / set_error_from_encode_
  encoding_raw_decision --> errored : completion_tokenize_runtime_ [encode_count_invalid_] / set_invalid_request_error_
  encoding_raw_decision --> encoding_ready : completion_tokenize_runtime_ [encode_success_] / commit_encoded_fragment_
  suffix_decision --> finalizing : completion_tokenize_runtime_ [sep_ready_] / append_sep_
  suffix_decision --> errored : completion_tokenize_runtime_ [sep_no_capacity_] / set_invalid_request_error_
  suffix_decision --> errored : completion_tokenize_runtime_ [sep_invalid_id_] / set_invalid_id_error_
  suffix_decision --> finalizing : completion_tokenize_runtime_ [eos_ready_] / append_eos_
  suffix_decision --> errored : completion_tokenize_runtime_ [eos_no_capacity_] / set_invalid_request_error_
  suffix_decision --> errored : completion_tokenize_runtime_ [eos_invalid_id_] / set_invalid_id_error_
  suffix_decision --> finalizing : completion_tokenize_runtime_ [no_suffix_] / none
  finalizing --> done : completion_tokenize_runtime_ [always] / finalize_
  uninitialized --> unexpected : _ [always] / on_unexpected_
  binding_preprocessor --> unexpected : _ [always] / on_unexpected_
  binding_preprocessor_decision --> unexpected : _ [always] / on_unexpected_
  binding_encoder --> unexpected : _ [always] / on_unexpected_
  binding_encoder_decision --> unexpected : _ [always] / on_unexpected_
  idle --> unexpected : _ [always] / on_unexpected_
  preprocessing --> unexpected : _ [always] / on_unexpected_
  preprocess_decision --> unexpected : _ [always] / on_unexpected_
  prefix_decision --> unexpected : _ [always] / on_unexpected_
  encoding_ready --> unexpected : _ [always] / on_unexpected_
  encoding_token_fragment --> unexpected : _ [always] / on_unexpected_
  encoding_raw_fragment --> unexpected : _ [always] / on_unexpected_
  encoding_raw_decision --> unexpected : _ [always] / on_unexpected_
  suffix_decision --> unexpected : _ [always] / on_unexpected_
  finalizing --> unexpected : _ [always] / on_unexpected_
  done --> unexpected : _ [always] / on_unexpected_
  errored --> unexpected : _ [always] / on_unexpected_
  unexpected --> unexpected : _ [always] / on_unexpected_
Loading

Transitions

Source Event Guard Action Target
uninitialized bind_runtime can_bind> begin_bind> binding_preprocessor
uninitialized bind_runtime always reject_bind> errored
uninitialized tokenize_runtime always reject_invalid> errored
idle bind_runtime can_bind> begin_bind> binding_preprocessor
idle bind_runtime always reject_bind> errored
idle tokenize_runtime can_tokenize> begin_tokenize> preprocessing
idle tokenize_runtime always reject_invalid> errored
done bind_runtime can_bind> begin_bind> binding_preprocessor
done bind_runtime always reject_bind> errored
done tokenize_runtime can_tokenize> begin_tokenize> preprocessing
done tokenize_runtime always reject_invalid> errored
errored bind_runtime can_bind> begin_bind> binding_preprocessor
errored bind_runtime always reject_bind> errored
errored tokenize_runtime can_tokenize> begin_tokenize> preprocessing
errored tokenize_runtime always reject_invalid> errored
unexpected bind_runtime can_bind> begin_bind> binding_preprocessor
unexpected bind_runtime always reject_bind> unexpected
unexpected tokenize_runtime can_tokenize> begin_tokenize> preprocessing
unexpected tokenize_runtime always reject_invalid> unexpected
binding_preprocessor completion<bind_runtime> always bind_preprocessor> binding_preprocessor_decision
binding_preprocessor_decision completion<bind_runtime> bind_preprocessor_error_none> none binding_encoder
binding_preprocessor_decision completion<bind_runtime> bind_preprocessor_error_invalid_request> none errored
binding_preprocessor_decision completion<bind_runtime> bind_preprocessor_error_model_invalid> none errored
binding_preprocessor_decision completion<bind_runtime> bind_preprocessor_error_backend_error> none errored
binding_preprocessor_decision completion<bind_runtime> bind_preprocessor_error_unknown> none errored
binding_encoder completion<bind_runtime> always bind_encoder> binding_encoder_decision
binding_encoder_decision completion<bind_runtime> bind_encoder_error_none> mark_bind_success> idle
binding_encoder_decision completion<bind_runtime> bind_encoder_error_invalid_request> none errored
binding_encoder_decision completion<bind_runtime> bind_encoder_error_model_invalid> none errored
binding_encoder_decision completion<bind_runtime> bind_encoder_error_backend_error> none errored
binding_encoder_decision completion<bind_runtime> bind_encoder_error_unknown> none errored
preprocessing completion<tokenize_runtime> always dispatch_preprocess> preprocess_decision
preprocess_decision completion<tokenize_runtime> preprocess_rejected_no_error> set_backend_error> errored
preprocess_decision completion<tokenize_runtime> preprocess_reported_error> set_error_from_preprocess> errored
preprocess_decision completion<tokenize_runtime> preprocess_fragment_count_invalid> set_invalid_request_error> errored
preprocess_decision completion<tokenize_runtime> preprocess_success> none prefix_decision
prefix_decision completion<tokenize_runtime> bos_ready> append_bos> encoding_ready
prefix_decision completion<tokenize_runtime> bos_no_capacity> set_invalid_request_error> errored
prefix_decision completion<tokenize_runtime> bos_invalid_id> set_invalid_id_error> errored
prefix_decision completion<tokenize_runtime> no_prefix> none encoding_ready
encoding_ready completion<tokenize_runtime> no_more_fragments> none suffix_decision
encoding_ready completion<tokenize_runtime> more_fragments_no_capacity> set_invalid_request_error> errored
encoding_ready completion<tokenize_runtime> more_fragments_token_invalid> set_invalid_request_error> errored
encoding_ready completion<tokenize_runtime> more_fragments_token_valid> none encoding_token_fragment
encoding_ready completion<tokenize_runtime> more_fragments_raw> none encoding_raw_fragment
encoding_token_fragment completion<tokenize_runtime> always append_fragment_token> encoding_ready
encoding_raw_fragment completion<tokenize_runtime> always dispatch_encode_raw_fragment> encoding_raw_decision
encoding_raw_decision completion<tokenize_runtime> encode_rejected_no_error> set_invalid_id_error> errored
encoding_raw_decision completion<tokenize_runtime> encode_reported_error> set_error_from_encode> errored
encoding_raw_decision completion<tokenize_runtime> encode_count_invalid> set_invalid_request_error> errored
encoding_raw_decision completion<tokenize_runtime> encode_success> commit_encoded_fragment> encoding_ready
suffix_decision completion<tokenize_runtime> sep_ready> append_sep> finalizing
suffix_decision completion<tokenize_runtime> sep_no_capacity> set_invalid_request_error> errored
suffix_decision completion<tokenize_runtime> sep_invalid_id> set_invalid_id_error> errored
suffix_decision completion<tokenize_runtime> eos_ready> append_eos> finalizing
suffix_decision completion<tokenize_runtime> eos_no_capacity> set_invalid_request_error> errored
suffix_decision completion<tokenize_runtime> eos_invalid_id> set_invalid_id_error> errored
suffix_decision completion<tokenize_runtime> no_suffix> none finalizing
finalizing completion<tokenize_runtime> always finalize> done
uninitialized _ always on_unexpected> unexpected
binding_preprocessor _ always on_unexpected> unexpected
binding_preprocessor_decision _ always on_unexpected> unexpected
binding_encoder _ always on_unexpected> unexpected
binding_encoder_decision _ always on_unexpected> unexpected
idle _ always on_unexpected> unexpected
preprocessing _ always on_unexpected> unexpected
preprocess_decision _ always on_unexpected> unexpected
prefix_decision _ always on_unexpected> unexpected
encoding_ready _ always on_unexpected> unexpected
encoding_token_fragment _ always on_unexpected> unexpected
encoding_raw_fragment _ always on_unexpected> unexpected
encoding_raw_decision _ always on_unexpected> unexpected
suffix_decision _ always on_unexpected> unexpected
finalizing _ always on_unexpected> unexpected
done _ always on_unexpected> unexpected
errored _ always on_unexpected> unexpected
unexpected _ always on_unexpected> unexpected