Feat/ru friendship skill (#120)

* feat: add language parameters * fix: black codestyle * fix: codestyle * fix: dff friendship ru * fix: dff friendship ru * fix: dff friendship ru * fix: dff friendship ru уокротила реплики * fix: dff friendship tests * fix: dff friendship tests * fix: languAGE FOR WIKI * fix: language default value * fix: language default value * fix: language env var * fix: use templates by language * fix: ru templates * fix: no lang env var in common * fix: lang to ackn * fizx: codestyle * feat: default lang value * fix: dummy for russian * fix: no en acks * fix: how are you ru * fix: logs for response functions * fix: logs for condition functions * fix: ru version if what to talk about * feat: ru tests * fix: codestyle * fix: ru condition to resp selector * fix: ru condition to resp selector * fix: logging level and configuration * fix: ascii in tests
deeppavlov · Jun 28, 2022 · 986d5ce · 986d5ce
1 parent 1da2d6d
commit 986d5ce
Show file tree

Hide file tree

Showing 45 changed files with 1,611 additions and 234 deletions.
diff --git a/annotators/BadlistedWordsDetector/Dockerfile b/annotators/BadlistedWordsDetector/Dockerfile
@@ -7,7 +7,7 @@ RUN pip install -r /src/requirements.txt
 RUN spacy download en_core_web_sm
 RUN spacy download ru_core_news_sm
 
-ARG LANGUAGE
+ARG LANGUAGE=EN
 ENV LANGUAGE ${LANGUAGE}
 
 COPY . /src/

diff --git a/annotators/BadlistedWordsDetector/server.py b/annotators/BadlistedWordsDetector/server.py
@@ -15,7 +15,7 @@
 
 
 sentry_sdk.init(getenv("SENTRY_DSN"))
-LANGUAGE = getenv("LANGUAGE", "ENGLISH")
+LANGUAGE = getenv("LANGUAGE", "EN")
 
 logging.basicConfig(format="%(asctime)s - %(name)s - %(levelname)s - %(message)s", level=logging.INFO)
 logger = logging.getLogger(__name__)
@@ -144,7 +144,7 @@ def collect_ngrams(utterance: Doc, max_ngram: int):
     return all_ngrams
 
 
-if LANGUAGE == "RUSSIAN":
+if LANGUAGE == "RU":
     nlp_pipeline = spacy.load("ru_core_news_sm", exclude=["senter", "ner"])
     badlists_dir = Path("./badlists")
     badlists_files = [f for f in badlists_dir.iterdir() if f.is_file() and "_ru" in f.name]

diff --git a/annotators/BadlistedWordsDetector/test.py b/annotators/BadlistedWordsDetector/test.py
@@ -5,8 +5,8 @@
 def main():
     url = "http://0.0.0.0:8018/badlisted_words"
 
-    LANGUAGE = getenv("LANGUAGE", "ENGLISH")
-    if LANGUAGE == "RUSSIAN":
+    LANGUAGE = getenv("LANGUAGE", "EN")
+    if LANGUAGE == "RU":
         request_data = {
             "sentences": ["пизда настала", "застрахуйте уже его", "пошел нахер!"],
         }

diff --git a/annotators/wiki_parser/Dockerfile b/annotators/wiki_parser/Dockerfile
@@ -4,7 +4,7 @@ ARG CONFIG
 ARG COMMIT
 ARG PORT
 ARG SRC_DIR
-ARG LANGUAGE
+ARG LANGUAGE=EN
 
 ENV CONFIG=$CONFIG
 ENV PORT=$PORT

diff --git a/assistant_dists/dream/docker-compose.override.yml b/assistant_dists/dream/docker-compose.override.yml
@@ -62,7 +62,7 @@ services:
       args:
         SERVICE_PORT: 8008
         SERVICE_NAME: dff_program_y_skill
-        LANGUAGE: ENGLISH
+        LANGUAGE: EN
       context: .
       dockerfile: ./skills/dff_program_y_skill/Dockerfile
     command: gunicorn --workers=1 server:app -b 0.0.0.0:8008 --reload

diff --git a/assistant_dists/dream_russian/docker-compose.override.yml b/assistant_dists/dream_russian/docker-compose.override.yml
@@ -16,7 +16,7 @@ services:
       args:
         SERVICE_PORT: 8008
         SERVICE_NAME: dff_program_y_skill
-        LANGUAGE: RUSSIAN
+        LANGUAGE: RU
       context: .
       dockerfile: ./skills/dff_program_y_skill/Dockerfile
     command: gunicorn --workers=1 server:app -b 0.0.0.0:8008 --reload
@@ -32,6 +32,8 @@ services:
     build:
       context: .
       dockerfile: ./response_selectors/convers_evaluation_based_selector/Dockerfile
+      args:
+        LANGUAGE: RU
     command: flask run -h 0.0.0.0 -p 8009
     environment:
       - FLASK_APP=server
@@ -82,7 +84,7 @@ services:
     env_file: [.env]
     build:
       args:
-        LANGUAGE: RUSSIAN
+        LANGUAGE: RU
       context: annotators/BadlistedWordsDetector/
     command: flask run -h 0.0.0.0 -p 8018
     environment:
@@ -219,6 +221,7 @@ services:
       args:
         SERVICE_PORT: 8086
         SERVICE_NAME: dff_friendship_skill # has to be the same with skill dir name
+        LANGUAGE: RU
       context: .
       dockerfile: ./skills/dff_friendship_skill/Dockerfile
     command: gunicorn --workers=1 server:app -b 0.0.0.0:8086
@@ -239,6 +242,7 @@ services:
         SERVICE_PORT: 8111
         SERVICE_NAME: dff_wiki_skill
         COMMIT: 5b99ac3392e8e178e2bb4f9b218d4ddb2ec2e242
+        LANGUAGE: RU
       context: ./
       dockerfile: ./skills/dff_wiki_skill/Dockerfile
     command: gunicorn --workers=1 server:app -b 0.0.0.0:8111

diff --git a/common/acknowledgements.py b/common/acknowledgements.py
@@ -1,5 +1,12 @@
 GENERAL_ACKNOWLEDGEMENTS = {
-    "positive": ["Sounds cool! ", "Great! ", "Wonderful! ", "Cool!", "Nice!"],
-    "neutral": ["Okay. ", "Oh. ", "Huh. ", "Well. ", "Gotcha. ", "Aha. "],
-    "negative": ["Hmm... ", "I see.", "That's okay.", "Okay."],
+    "EN": {
+        "positive": ["Sounds cool! ", "Great! ", "Wonderful! ", "Cool!", "Nice!"],
+        "neutral": ["Okay. ", "Oh. ", "Huh. ", "Well. ", "Gotcha. ", "Aha. "],
+        "negative": ["Hmm... ", "I see.", "That's okay.", "Okay."],
+    },
+    "RU": {
+        "positive": ["Звучит круто! ", "Классно! ", "Великолепно! ", "Отлично!", "Замечательно!"],
+        "neutral": ["Окей. ", "Хм. ", "Ага. ", "Что ж. ", "Понятно. "],
+        "negative": ["Хмм... ", "Понимаю.", "Окей.", "Понятно."],
+    },
 }
diff --git a/common/dff/integration/condition.py b/common/dff/integration/condition.py
@@ -237,7 +237,7 @@ def is_passive_user(ctx: Context, actor: Actor, history_len=2) -> bool:
     return False
 
 
-def get_not_used_and_save_sentiment_acknowledgement(ctx: Context, actor: Actor, sentiment=None):
+def get_not_used_and_save_sentiment_acknowledgement(ctx: Context, actor: Actor, sentiment=None, lang="EN"):
     if sentiment is None:
         sentiment = int_ctx.get_human_sentiment(ctx, actor)
         if is_yes_vars(ctx, actor) or is_no_vars(ctx, actor):
@@ -247,7 +247,7 @@ def get_not_used_and_save_sentiment_acknowledgement(ctx: Context, actor: Actor,
     last_acknowledgements = shared_memory.get("last_acknowledgements", [])
 
     ack = common_utils.get_not_used_template(
-        used_templates=last_acknowledgements, all_templates=GENERAL_ACKNOWLEDGEMENTS[sentiment]
+        used_templates=last_acknowledgements, all_templates=GENERAL_ACKNOWLEDGEMENTS[lang][sentiment]
     )
 
     used_acks = last_acknowledgements + [ack]

diff --git a/common/dialogflow_framework/utils/condition.py b/common/dialogflow_framework/utils/condition.py
@@ -250,7 +250,7 @@ def is_passive_user(vars, history_len=2):
     return False
 
 
-def get_not_used_and_save_sentiment_acknowledgement(vars):
+def get_not_used_and_save_sentiment_acknowledgement(vars, lang="EN"):
     sentiment = state_utils.get_human_sentiment(vars)
     if is_yes_vars(vars) or is_no_vars(vars):
         sentiment = "neutral"
@@ -259,7 +259,7 @@ def get_not_used_and_save_sentiment_acknowledgement(vars):
     last_acknowledgements = shared_memory.get("last_acknowledgements", [])
 
     ack = common_utils.get_not_used_template(
-        used_templates=last_acknowledgements, all_templates=GENERAL_ACKNOWLEDGEMENTS[sentiment]
+        used_templates=last_acknowledgements, all_templates=GENERAL_ACKNOWLEDGEMENTS[lang][sentiment]
     )
 
     used_acks = last_acknowledgements + [ack]

diff --git a/common/emotion.py b/common/emotion.py
@@ -1,8 +1,10 @@
 import re
+
 from common.greeting import HOW_ARE_YOU_RESPONSES
 from common.utils import get_emotions
 from common.universal_templates import if_chat_about_particular_topic
 
+
 POSITIVE_EMOTIONS = set(
     [
         "interest",
@@ -117,7 +119,7 @@
 NOT_LONELINESS_TEMPLATE = re.compile(rf"{NOT_PATTERN} {ALONE_PATTERN}", re.IGNORECASE)
 SAD_TEMPLATE = re.compile(rf"({SAD_PATTERN}|{POOR_ASR_PATTERN})", re.IGNORECASE)
 NOT_SAD_TEMPLATE = re.compile(rf"{NOT_PATTERN} {SAD_PATTERN}", re.IGNORECASE)
-BORING_TEMPLATE = re.compile(rf"(boring|bored)", re.IGNORECASE)
+BORING_TEMPLATE = re.compile(r"(boring|bored)", re.IGNORECASE)
 NOT_BORING_TEMPLATE = re.compile(rf"{NOT_PATTERN} (boring|bored)", re.IGNORECASE)
 JOKE_REQUEST_TEMPLATE = re.compile(rf"{JOKE_PATTERN}", re.IGNORECASE)
 NOT_JOKE_REQUEST_TEMPLATE = re.compile(rf"{NOT_PATTERN} {JOKE_PATTERN}", re.IGNORECASE)
@@ -179,7 +181,7 @@ def emo_advice_requested(uttr):
 
 
 def skill_trigger_phrases():
-    return [HOW_DO_YOU_FEEL] + HOW_ARE_YOU_RESPONSES
+    return [HOW_DO_YOU_FEEL] + sum([HOW_ARE_YOU_RESPONSES[lang] for lang in ["RU", "EN"]], [])
 
 
 def emotion_from_feel_answer(prev_bot_uttr, user_uttr):
@@ -207,7 +209,7 @@ def if_turn_on_emotion(user_utt, bot_uttr):
     how_are_you = any(
         [
             how_are_you_response.lower() in bot_uttr.get("text", "").lower()
-            for how_are_you_response in HOW_ARE_YOU_RESPONSES
+            for how_are_you_response in sum([HOW_ARE_YOU_RESPONSES[lang] for lang in ["RU", "EN"]], [])
         ]
     )
     joke_request_detected = is_joke_requested(user_utt)