ValueError: invalid width -106 (must be > 0)

[hei-my]

Rasa version: 2.4.3

Rasa SDK version (if used & relevant):

Rasa X version (if used & relevant):

Python version:3.8.8

Operating system (windows, osx, ...):

Issue: ValueError: invalid width -106 (must be > 0)

Error (including full traceback):

Bot loaded. Visualisation at http://localhost:5006/visualization.html .
Type a message and press enter (press 'Ctrl-c' to exit).
? Your input -> 放射性肺炎怎么医治哈
? Your NLU model classified '放射性肺炎怎么医治哈' with intent 'search_treat' and there are no entities, is this correct?  (Y/n)

Cancelled by user

? Do you want to stop?  Undo Last
------
Chat History

 #    Bot                You
────────────────────────────────────
 1    action_listen


Current slots:
        disease: None, symptom: None, dept: None, drug: None, food: None, sure: None, pre_disease: None, session_started_metadata: None

------
? Your input -> 牙龈出血有哪些病导致
? Your NLU model classified '牙龈出血有哪些病导致' with intent 'search_disease' and there are no entities, is this correct?  No
? What intent is it?  1.00 search_disease
? Please mark the entities using [value](type) notation [牙龈出血](symptom)有哪些病导致
------
Chat History

 #    Bot                                                   You
───────────────────────────────────────────────────────────────────────
 1    action_listen
───────────────────────────────────────────────────────────────────────
 2                                     [牙龈出血](symptom)有哪些病导致
                                           intent: search_disease 1.00
───────────────────────────────────────────────────────────────────────
 3    slot{"symptom": "牙龈出血"}


Current slots:
        disease: None, symptom: 牙龈出血, dept: None, drug: None, food: None, sure: None, pre_disease: None, session_started_metadata: None

------
? The bot wants to run 'action_search_disease', correct?  Yes
------
Chat History

 #    Bot                                                                                                                                                                                                                                                                                                            You
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
 1    action_listen                                                                                                                                                                          
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
 2                                                                                                                                                                                                                                                                                              [牙龈出血](symptom)有哪些病导致
                                                                                                                                                                                                                                                                                                    intent: search_disease 1.00
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
 3    slot{"symptom": "牙龈出血"}                                                                                                                                                            
      action_search_disease 1.00                                                                                                                                                             
      牙龈出血的症状对应得疾病可能有：肥大性龈炎、骨髓增生异常综合征、口腔疾病、小儿出血性疾病、边缘性龈炎、获得性维生素K依赖性凝血因子异常、小儿白血病、回归热、坏死性龈口炎、青春期功能失调性子宫出血、郎-奥韦综合征、急性淋巴细胞白血病、感染性血小板减少性紫癜、单纯性牙周炎、单核细胞
      白血病、δ-贮存池病、小儿特发性血小板减少性紫癜、老年人真性红细胞增多症、急性根尖牙周炎、慢性根尖牙周炎、创伤性口炎、青少年牙周炎、慢性牙周炎                                          


Current slots:
        disease: None, symptom: 牙龈出血, dept: None, drug: None, food: None, sure: None, pre_disease: None, session_started_metadata: None

------
? The bot wants to run 'action_listen', correct?  Yes
? Your input -> 口腔疾病怎么医治
? Your NLU model classified '口腔疾病怎么医治' with intent 'search_treat' and there are no entities, is this correct?  No
? What intent is it?  1.00 search_treat
? Please mark the entities using [value](type) notation [口腔疾病](disease)怎么医治
2021-03-31 00:06:45 ERROR    rasa.core.training.interactive  - An exception occurred while recording messages.
Traceback (most recent call last):
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 1479, in record_messages
    await _predict_till_next_listen(
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 964, in _predict_till_next_listen
    await _print_history(conversation_id, endpoint)
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 467, in _print_history
    table = _chat_history_table(events)
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 547, in _chat_history_table
    msg = format_user_msg(event, user_width(table))
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 500, in format_user_msg
    colored(wrap(_md, max_width), "hired"),
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 488, in wrap
    return "\n".join(textwrap.wrap(txt, max_width, replace_whitespace=False))
  File "D:\python\anaconda\install\envs\rasa\lib\textwrap.py", line 379, in wrap
    return w.wrap(text)
  File "D:\python\anaconda\install\envs\rasa\lib\textwrap.py", line 354, in wrap
    return self._wrap_chunks(chunks)
  File "D:\python\anaconda\install\envs\rasa\lib\textwrap.py", line 248, in _wrap_chunks
    raise ValueError("invalid width %r (must be > 0)" % self.width)
ValueError: invalid width -106 (must be > 0)
2021-03-31 00:06:45 ERROR    asyncio  - Task exception was never retrieved
future: <Task finished name='Task-3' coro=<_serve_application.<locals>.run_interactive_io() done, defined at D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py:1569> exception=ValueError('invalid width -106 (must be > 0)')>
Traceback (most recent call last):
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 1572, in run_interactive_io
    await record_messages(
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 1479, in record_messages
    await _predict_till_next_listen(
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 964, in _predict_till_next_listen
    await _print_history(conversation_id, endpoint)
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 467, in _print_history
    table = _chat_history_table(events)
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 547, in _chat_history_table
    msg = format_user_msg(event, user_width(table))
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 500, in format_user_msg
    colored(wrap(_md, max_width), "hired"),
  File "D:\python\anaconda\install\envs\rasa\lib\site-packages\rasa\core\training\interactive.py", line 488, in wrap
    return "\n".join(textwrap.wrap(txt, max_width, replace_whitespace=False))
  File "D:\python\anaconda\install\envs\rasa\lib\textwrap.py", line 379, in wrap
    return w.wrap(text)
  File "D:\python\anaconda\install\envs\rasa\lib\textwrap.py", line 354, in wrap
    return self._wrap_chunks(chunks)
  File "D:\python\anaconda\install\envs\rasa\lib\textwrap.py", line 248, in _wrap_chunks
    raise ValueError("invalid width %r (must be > 0)" % self.width)
ValueError: invalid width -106 (must be > 0)

Command or request that led to error:

python -m rasa interactive -m models/nlu/20210330-231450.tar.gz --endpoints config/endpoints.yml --config config/config_pretrained_embeddings_mitie_zh.yml

Content of configuration file (config.yml) (if relevant):

language: ch
pipeline:
  - name: HFTransformersNLP
    model_name: "bert"
    model_weights: "bert-base-chinese"
  - name: LanguageModelTokenizer
  - name: LanguageModelFeaturizer
  - name: LexicalSyntacticFeaturizer
  - name: CRFEntityExtractor
  - name: EntitySynonymMapper
  - name: DIETClassifier
    epochs: 200

policies:
  - name: AugmentedMemoizationPolicy

  - name: TEDPolicy
    epochs: 100
    featurizer:
      - name: MaxHistoryTrackerFeaturizer
        max_history: 5
        state_featurizer:
          - name: BinarySingleStateFeaturizer


  - name: FallbackPolicy
    fallback_action_name: 'action_default_fallback'
    nlu_threshold: 0.5
    core_threshold: 0.3
    ambiguity_threshold: 0.1

  - name: MappingPolicy

Content of domain file (domain.yml) (if relevant):

session_config:
  session_expiration_time: 60  # value in minutes, 0 means infinitely long
  carry_over_slots_to_new_session: true  # set to false to forget slots between sessions

slots:
  disease:
    type: text
  symptom:
    type: text
  dept:
    type: text
  drug:
    type: text
  food:
    type: text
  sure:
    type: any
  pre_disease:
    type: any


intents:
  - first: {triggers: action_first}
  - greet
  - bye
  - affirmative
  - search_treat
  - search_food
  - search_symptom
  - search_disease
  - search_cause
  - search_neopathy
  - search_drug
  - search_prevention
  - search_drug_func
  - search_disease_treat_time
  - search_easy_get
  - search_disease_dept

entities:
  - disease
  - symptom
  - dept
  - drug
  - food

responses:
  utter_first:
    - text: "您好，我是您的医疗助手Friende，我是个机器人，请问有什么可以帮您？"
  utter_greet:
    - text: "您好～"
    - text: "您好呀～"
  utter_goodbye:
    - text: "再见，祝您身体健康～"
    - text: "拜拜，希望我有帮到您～"
  utter_howcanhelp:
    - text: "您可以这样向我提问: 头痛怎么办/
                              什么人容易头痛/
                              头痛吃什么药/
                              头痛能治吗/
                              头痛属于什么科/
                              头孢地尼分散片用途/
                              如何防止头痛/
                              头痛要治多久/
                              糖尿病有什么并发症/
                              糖尿病有什么症状"
  utter_donknow:
    - text: "啊噢，我没有理解您说的话，我的理解力还需要更多的提升>_<。"

actions:
  - utter_first
  - utter_donknow
  - action_first
  - action_donknow
  - action_echo
  - action_search_treat
  - action_search_food
  - action_search_symptom
  - action_search_cause
  - action_search_disease
  - action_search_neopathy
  - action_search_drug
  - action_search_prevention
  - action_search_drug_func
  - action_search_disease_treat_time
  - action_search_easy_get
  - action_search_disease_dept
  - utter_greet
  - utter_howcanhelp
  - utter_goodbye

[sara-tagger]

Thanks for the issue, @b-quachtran will get back to you about it soon!

You may find help in the docs and the forum, too 🤗

[twerkmeister]

Seems the problem is with the package terminaltables that we are using in the interactive training mode. The problem boils down to non-ascii characters potentially having a different width than ascii characters leading to some internal overflow in the table dimension calculations.

For example 210 character long chinese text from the issue creator:
"牙龈出血的症状对应得疾病可能有：肥大性龈炎、骨髓增生异常综合征、口腔疾病、小儿出血性疾病、边缘性龈炎、获得性维生素K依赖性凝血因子异常、小儿白血病、回归热、坏死性龈口炎、青春期功能失调性子宫出血、郎-奥韦综合征、急性淋巴细胞白血病、感染性血小板减少性紫癜、单纯性牙周炎、单核细胞白血病、δ-贮存池病、小儿特发性血小板减少性紫癜、老年人真性红细胞增多症、急性根尖牙周炎、慢性根尖牙周炎、创伤性口炎、青少年牙周炎、慢性牙周炎"

210 character long ascii text (in monospaced font):
'abcdefgaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'

The problem only arises after the texts have a certain length. For example, for an answer of 50 characters it still works. Crashing will happen somewhere between 50 and 100 characters