nestor/core/nlp.py

import en_core_web_lg
import fr_core_news_md  # @UnusedImport
from core import logging_


logger = logging_.getLogger("nlp")

logger.info("Nestor se réveille...")
# nlp = fr_core_news_md.load()
nlp = en_core_web_lg.load()
logger.info("> Ok")

def submit(command):
    logger.info("""Message reçu: "%s" """, command)
    doc = nlp(command)

    for token in doc:
        logger.debug("\tJeton: %s", token.text)
        logger.debug("\t\t* Lemma: %s", token.lemma_)
        logger.debug("\t\t* Position: %s", token.pos_)
        logger.debug("\t\t* Tag: %s", token.tag_)
        logger.debug("\t\t* Dependance: %s", token.dep_)
        logger.debug("\t\t* Forme: %s", token.shape_)
        logger.debug("\t\t* Alphanum.: %s", token.is_alpha)
        logger.debug("\t\t* Stop: %s", token.is_stop)

    for ent in doc.ents:
        logger.debug("\tEntité: %s", ent.text)
        logger.debug("\t\t* Start: %s", ent.start_char)
        logger.debug("\t\t* End: %s", ent.end_char)
        logger.debug("\t\t* Label: %s", ent.label_)

    unused = []
    for i in range(len(doc)):
        token = doc[i]
        if token.lemma_ == "copy":
            for j in range(i, len(doc)):
                token = doc[j]
                if token.lemma_ in ["from"]:
                    from_ = doc[j + 1].text
                if token.lemma_ in ["to", "in"]:
                    to_ = doc[j + 1].text

            logger.info("Vous voulez copier %s vers %s", from_, to_)
            break

        if token.lemma_ == "delete":
            what = doc[i + 1]
            logger.info("Vous voulez supprimer %s", what)
            break

        if token.lemma_ == "rename":
            logger.info("Vous voulez renommer")

        if token.lemma_ == "move":
            logger.info("Vous voulez déplacer")

        if token.lemma_ == "open":
            logger.info("Vous voulez ouvrir")

        if token.lemma_ == "new":
            logger.info("Vous voulez créer")

        logger.debug("Connais pas: %s", token.text)
        unused.append(token)



def analyse(message):
    doc = nlp(message)

    for token in doc:
        print(token.text, token.lemma_, token.pos_, token.tag_, token.dep_,
              token.shape_, token.is_alpha, token.is_stop)

    for ent in doc.ents:
        print(ent.text, ent.start_char, ent.end_char, ent.label_)


if __name__ == "__main__":

    submit(r"Copy from .\test\work\a.txt to .\test\work\b.txt")
    submit(r"Delete .\test\work\a.txt")
    # submit(r"Rename .\test\work\a.txt in .\test\work\b.txt")