spacy - 临时

import os
import sys
import re
import spacy
import json
from i18n import *

# pip install spacy
# python -m spacy download en


nlp = spacy.load("en_core_web_sm")



class Word:
    def __init__(self,token) -> None:
        self.token = token

    token = None

    def getPos(self):
        return self.token.pos_

    def toObj(self):
        ret =  {
            "content" : str(self.token),
            "pos1" : self.token.pos_,
            "pos2" : self.getPos(),
            "depen" : dependency_labels_i18n(self.token.dep_,1)

        }
        if self.token.lemma_ != str(self.token):
            ret["lemma"] = self.token.lemma_ # 词性还原

        return ret




astr = '''
this is a dog
'''

# str = str.replace("\n"," ")

astr = astr.strip()
astr = re.sub(r"(\s){2,}"," ",astr)


document = nlp(astr)



# 拆分句子，将每个单词送去分析
data = []
tmp_data = []
for token in document:
    if token.is_sent_start:
        if len(tmp_data)!=0:
            data.append(tmp_data)
            tmp_data = []
    tmp_data.append(Word(token).toObj())

if len(tmp_data)!=0:
    data.append(tmp_data)
    tmp_data = []



# 准备返回的数据
sententce = [str(s) for s in list(document.sents) ]

j = json.dumps({
    "sents" : sententce,
    "count" : len(sententce),
    "data" : data
},ensure_ascii=False)

print(j)


    # data.append({token:token.pos_ for token in sent})
    # sentences.append(data)

# for i in sentences:
#     print(i)



#tags

#all_tags = {w.pos: w.pos_ for w in document}
#for word in list(document.sents)[0]:  
#    print(" %s -- %s " % (word,word.tag_))



# 'DET', 'ADJ', 'NOUN', 'VERB', 'DET', 'ADJ', 'ADJ', 'NOUN', 'PUNCT']
# 对应于中文是 【冠词，形容词，名词，动词，冠词，形容词，形容词，名词，标点】

# https://github.com/clir/clearnlp-guidelines/blob/master/md/specifications/dependency_labels.md

dependency_labels = {
    "ACL" : ["Clausalmodifierofnoun","名词的从句修饰语"],
    "ACOMP" : ["Adjectivalcomplement","形容词补语"],
    "ADVCL" : ["Adverbialclausemodifier","状语从句修饰语"],
    "ADVMOD" : ["Adverbialmodifier","状语修饰语"],
    "AGENT" : ["Agent","代理"],
    "AMOD" : ["Adjectivalmodifier","形容词修饰语"],
    "APPOS" : ["Appositionalmodifier","同位修饰语"],
    "ATTR" : ["Attribute","属性"],
    "AUX" : ["Auxiliary","辅助的"],
    "AUXPASS" : ["Auxiliary(passive)","辅助（被动）"],
    "CASE" : ["Casemarker","格位标记"],
    "CC" : ["Coordinatingconjunction","并列连词"],
    "CCOMP" : ["Clausalcomplement","从句补语"],
    "COMPOUND" : ["Compoundmodifier","复合名词修饰"],
    "CONJ" : ["Conjunct","连词"],
    "CSUBJ" : ["Clausalsubject","主语"],
    "CSUBJPASS" : ["Clausalsubject(passive)","子句主语（被动）"],
    "DATIVE" : ["Dative","与格"],
    "DEP" : ["Unclassifieddependent","未分类从属"],
    "DET" : ["Determiner","限定词"],
    "DOBJ" : ["DirectObject","直接宾语"],
    "EXPL" : ["Expletive","虚词"],
    "INTJ" : ["Interjection","感叹词"],
    "MARK" : ["Marker","标记"],
    "META" : ["Metamodifier","元修饰符"],
    "NEG" : ["Negationmodifier","否定修饰符"],
    "NOUNMOD" : ["Modifierofnominal","名词修饰语"],
    "NPMOD" : ["Nounphraseasadverbialmodifier","名词短语状语修饰语"],
    "NSUBJ" : ["Nominalsubject","名词性主语"],
    "NSUBJPASS" : ["Nominalsubject(passive)","名义主语（被动）"],
    "NUMMOD" : ["Numbermodifier","数字修饰符"],
    "OPRD" : ["Objectpredicate","对象谓词"],
    "PARATAXIS" : ["Parataxis","并列"],
    "PCOMP" : ["Complementofpreposition","介词补语"],
    "POBJ" : ["Objectofpreposition","介词宾语"],
    "POSS" : ["Possessionmodifier","占有修饰符"],
    "PRECONJ" : ["Pre-correlativeconjunction","pre-关联连词"],
    "PREDET" : ["Pre-determiner","前位限定词"],
    "PREP" : ["Prepositionalmodifier","介词修饰语"],
    "PRT" : ["Particle","小品词"],
    "PUNCT" : ["Punctuation","标点"],
    "QUANTMOD" : ["Modifierofquantifier","量词修饰语"],
    "RELCL" : ["Relativeclausemodifier","关系从句修饰符"],
    "ROOT" : ["Root","根节点"],
    "XCOMP" : ["Openclausalcomplement","开句补语"]
}


def dependency_labels_i18n(original_str,index = 0):
    newstr = str.upper(original_str)
    if newstr in dependency_labels:
        ret = dependency_labels[newstr]
        if len(ret) > index:
            return ret[index]
        else:
            return ret[0]
    else:
        return original_str