import nltk

nltk.download('punkt')  # 初回のみ必要

def split_into_sentences_nltk(text):
    sentences = nltk.sent_tokenize(text)
    return sentences

# テスト
text = "これはサンプルテキストです。これは別の文です。そして、最後の文です。"
sentences = split_into_sentences_nltk(text)
print(sentences)

import spacy

nlp = spacy.load('ja_core_news_sm')  # 日本語モデルを使用

def split_into_sentences_spacy(text):
    doc = nlp(text)
    sentences = [sent.text for sent in doc.sents]
    return sentences

# テスト
text = "これはサンプルテキストです。これは別の文です。そして、最後の文です。"
sentences = split_into_sentences_spacy(text)
print(sentences)

クエリのベクター = vectorize_text(好きな作品のベクター)
作品一覧のベクター = vectorize_text(作品一覧)

作品の関連度 = cosine_similarity(クエリのベクター, 作品一覧)

sorted_作品の関連度.argsort()[0][::-1]

# This snippet has been automatically generated and should be regarded as a
# code template only.
# It will require modifications to work:
# - It may require correct/in-range values for request initialization.
# - It may require specifying regional endpoints when creating the service
#   client as shown in:
#   https://googleapis.dev/python/google-api-core/latest/client_options.html
from google.cloud import discoveryengine_v1beta


def sample_recommend():
    # Create a client
    client = discoveryengine_v1beta.RecommendationServiceClient()

    # Initialize request argument(s)
    user_event = discoveryengine_v1beta.UserEvent()
    user_event.event_type = "event_type_value"
    user_event.user_pseudo_id = "user_pseudo_id_value"

    request = discoveryengine_v1beta.RecommendRequest(
        serving_config="serving_config_value",
        user_event=user_event,
    )

    # Make the request
    response = client.recommend(request=request)

    # Handle the response
    print(response)

from google.cloud import discoveryengine_v1beta

client = discoveryengine_v1beta.DocumentServiceClient()

parent = "projects/xxx/locations/global/collections/default_collection/dataStores/xxx/branches/0"

req = discoveryengine_v1beta.ImportDocumentsRequest({
    "parent": parent,
    "inline_source": {
        "documents": [
            discoveryengine_v1beta.Document({
                "id": "test-1",
                "schema_id": "default_schema",
                "json_data": str({
                    "title": "test-1",
                    "uri": "https://minshumi.app/art/test-art-1",
                    "available_time": "2022-08-26T23:00:17Z",
                }),
            }),
            discoveryengine_v1beta.Document({
                "id": "test-1",
                "schema_id": "default_schema",
                "json_data": str({
                    "title": "test-1",
                    "uri": "https://minshumi.app/art/test-art-1",
                    "available_time": "2022-08-26T23:00:17Z",
                }),
            }),
        ],
    },
})

res = client.import_documents(req)

print(res, res.done())


res = client.list_documents(discoveryengine_v1beta.ListDocumentsRequest({
    "parent": parent,
}))

print("documents", res)

for page in res.pages:
    print(page.documents)

<google.api_core.operation.Operation object at 0x1068c3450> True
documents ListDocumentsPager<>
[]

PROJECT_ID=xxx
DATA_STORE_ID=xxx
DISPLAY_NAME=$DATA_STORE_ID

curl -X POST \
    -H "Authorization: Bearer $(gcloud auth print-access-token)" \
    -H "Content-Type: application/json" \
    -H "X-Goog-User-Project: $PROJECT_ID" \
    "https://discoveryengine.googleapis.com/v1alpha/projects/$PROJECT_ID/locations/global/collections/default_collection/dataStores?dataStoreId=$DATA_STORE_ID" \
    -d "{
    'displayName': '$DISPLAY_NAME',
    'industryVertical': 'GENERIC',
    'solutionTypes': ['SOLUTION_TYPE_SEARCH'],
    "solutionTypes": ['SOLUTION_TYPE_RECOMMENDATION']
    }"

# output
# {
#   "name": "projects/xxx/locations/global/collections/default_collection/operations/create-data-store-11726871623769991452",
#   "done": true
# }

PROJECT_ID=xxx
DATA_STORE_ID=xxx
DISPLAY_NAME=$DATA_STORE_ID

DOCUMENT_ID=test-data-1

curl -X POST \
    -H "Authorization: Bearer $(gcloud auth print-access-token)" \
    -H "Content-Type: application/json" \
    "https://discoveryengine.googleapis.com/v1beta/projects/$PROJECT_ID/locations/global/collections/default_collection/dataStores/$DATA_STORE_ID/branches/0/documents?documentId=$DOCUMENT_ID" \
    -d '{
    "jsonData": "{\"persons\": [{\"role\": \"author\", \"name\": \"つーばーさ\"}], \"country_of_origin\": \"JP\", \"cotent_index\": 0, \"in_languages\": [\"ja-JP\"], \"uri\": \"https://minshumi.app/art/test-art-1\", \"title\": \"つーばーさのすごい小説\", \"available_time\": \"2023-12-01 18:41:02\", \"media_type\": \"novel\", \"language_code\": \"ja-JP\", \"has_tags\": [\"コメディ\", \"感動\", \"ファンタジー\"]}"
    }'

import json

data = {
    "persons": [
        {
            "role": "author",
            "name": "つーばーさ"
        }
    ],
    "country_of_origin": "JP",
    "cotent_index": 0,
    "in_languages": [
        "ja-JP"
    ],
    "uri": "https://minshumi.app/art/test-art-1",
    "title": "つーばーさのすごい小説",
    "available_time": "2023-12-01 18:41:02",
    "media_type": "novel",
    "language_code": "ja-JP",
    "has_tags": [
        "コメディ",
        "感動",
        "ファンタジー"
    ]
}

output = {"jsonData": str(data)}

print(json.dumps(output, ensure_ascii=False))

PROJECT_ID=xxx
DATA_STORE_ID=xxx
DISPLAY_NAME=$DATA_STORE_ID

DOCUMENT_ID=test-data-2
DOCUMENT_DATA="$(python test-data-2.py)"

echo $DOCUMENT_ID
echo $DOCUMENT_DATA

curl -X POST \
    -H "Authorization: Bearer $(gcloud auth print-access-token)" \
    -H "Content-Type: application/json" \
    "https://discoveryengine.googleapis.com/v1beta/projects/$PROJECT_ID/locations/global/collections/default_collection/dataStores/$DATA_STORE_ID/branches/0/documents?documentId=$DOCUMENT_ID" \
    -d "$DOCUMENT_DATA"
    # -d "{
    # 'jsonData': '{\"persons\": [{\"role\": \"author\", \"name\": \"つーばーさ\"}], \"country_of_origin\": \"JP\", \"cotent_index\": 0, \"in_languages\": [\"ja-JP\"], \"uri\": \"https://minshumi.app/art/test-art-1\", \"title\": \"つーばーさのすごい小説\", \"available_time\": \"2023-12-01 18:41:02\", \"media_type\": \"novel\", \"language_code\": \"ja-JP\", \"has_tags\": [\"コメディ\", \"感動\", \"ファンタジー\"]}'
    # }"

# output
# {
#   "name": "projects/xxx/locations/global/collections/default_collection/dataStores/xxx/branches/0/documents/test-data-1",
#   "id": "test-data-1",
#   "schemaId": "default_schema",
#   "jsonData": "{\"language_code\":\"ja-JP\",\"country_of_origin\":\"JP\",\"has_tags\":[\"コメディ\",\"感動\",\"ファンタジー\"],\"in_languages\":[\"ja-JP\"],\"title\":\"つーばーさのすごい小説\",\"cotent_index\":0,\"persons\":[{\"role\":\"author\",\"name\":\"つーばーさ\"}],\"available_time\":\"2023-12-01 18:41:02\",\"media_type\":\"novel\",\"uri\":\"https://minshumi.app/art/test-art-1\"}",
#   "parentDocumentId": "test-data-1"
# }

curl -v -X POST -H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
https://discoveryengine.googleapis.com/v1beta/projects/PROJECT_ID/locations/global/collections/default_collection/dataStores/DATASTORE_ID/servingConfigs/SERVICE_CONFIG:recommend \
-d '{ "userEvent": { "eventType":"view-item", "userPseudoId":"USER_ID", "documents":[{"id":"DOCUMENT_ID"}]}}'

// カスタム属性の例
  attributes: {
    user_age: {text: ["teen", "young adult"]},
    user_location: {text: ["CA"]},
    user_zip: {numbers: [90210]}
  }

import time
from os import path

import gensim

from settings import DIMENSIONS

sentences = [
    ["ファンキー", "な", "直感", "で", "今日", "の", "歌", "だって", "変わってく", "なら"] * 10,
    ["もっと", "仕草", "に", "揺れて", "抑え", "きれない", "ほど", "に"] * 10,
]

FAST_TEXT_MODEL_PATH = "models/fast_text.model"
WORD2VEC_MODEL_PATH = "models/word2vec.model"


def init_fast_text_model():
    fast_text_model = gensim.models.FastText(
        sentences=sentences,
        vector_size=DIMENSIONS,
        window=5,
        min_count=1,
        workers=4,
    )
    fast_text_model.save(FAST_TEXT_MODEL_PATH)
    fast_text_model.train(
        sentences,
        total_examples=2,
        epochs=1,
    )
    return fast_text_model


def init_word2vec_model():
    word2vec_model = gensim.models.Word2Vec(
        sentences=sentences,
        vector_size=DIMENSIONS,
        window=5,
        min_count=1,
        workers=4,
    )
    word2vec_model.save(WORD2VEC_MODEL_PATH)
    word2vec_model.train(
        sentences,
        total_examples=2,
        epochs=1,
    )
    return word2vec_model


def load_fast_text_model():
    if not path.exists(FAST_TEXT_MODEL_PATH):
        return init_fast_text_model()
    return gensim.models.FastText.load(FAST_TEXT_MODEL_PATH)


def load_word2vec_model():
    if not path.exists(WORD2VEC_MODEL_PATH):
        return init_word2vec_model()
    return gensim.models.Word2Vec.load(WORD2VEC_MODEL_PATH)


def to_vec_by_model(text: str, model):
    if model is None:
        raise NotImplementedError("please init_model")
    if text not in model.wv:
        return None
    vector = model.wv[text]
    return vector


word2vec_model = None
fast_text_model = None


def init_models():
    global word2vec_model
    global fast_text_model
    word2vec_model = init_word2vec_model()
    fast_text_model = init_fast_text_model()
    return word2vec_model, fast_text_model


def to_vec(text: str):
    vec_word2vec = to_vec_by_model(text, word2vec_model)
    if vec_word2vec is None:
        vec_fast_text = to_vec_by_model(text, fast_text_model)
        return vec_fast_text, "fast_text"
    return vec_word2vec, "word2vec"


if __name__ == "__main__":
    def test_word(word: str):
        print("test:", word)
        t1 = time.time()
        vec = to_vec(word)
        t2 = time.time()
        print("  vec", vec)
        print("  time", int((t2-t1)*1_000_000) / 1_000, "ms")

    t1 = time.time()
    init_models()
    t2 = time.time()
    print("init models", t2-t1)

    test_word("な")
    test_word("ファンキー")
    test_word("仕草")
    test_word("NewWord")

pip install mecab-python3
pip install unidic-lite

import MeCab

wakati = MeCab.Tagger("-Owakati")


def split_text(text: str):
    return wakati.parse(text).split()