dasmith / stanford-corenlp-python

Python wrapper for Stanford CoreNLP tools v3.4.1
GNU General Public License v2.0
610 stars 229 forks source link

parse returning as a string rather than a dictionary. #40

Closed jjm0022 closed 8 years ago

jjm0022 commented 8 years ago

I'm trying to follow the instructions:

from corenlp import * corenlp = StanfordCoreNLP()
corenlp.parse("This is a test.")

When I do this it returns something like this: '{"coref": [[[["This", 0, 0, 0, 1], ["a test", 0, 3, 2, 4]]]], "sentences": [{"parsetree": "(ROOT (S (NP (DT This)) (VP (VBZ is) (NP (DT a) (NN test))) (. .)))", "text": "This is a test.", "dependencies": [["root", "ROOT", "test"], ["nsubj", "test", "This"], ["cop", "test", "is"], ["det", "test", "a"]], "words": [["This", {"NamedEntityTag": "O", "CharacterOffsetEnd": "4", "Lemma": "this", "PartOfSpeech": "DT", "CharacterOffsetBegin": "0"}], ["is", {"NamedEntityTag": "O", "CharacterOffsetEnd": "7", "Lemma": "be", "PartOfSpeech": "VBZ", "CharacterOffsetBegin": "5"}], ["a", {"NamedEntityTag": "O", "CharacterOffsetEnd": "9", "Lemma": "a", "PartOfSpeech": "DT", "CharacterOffsetBegin": "8"}], ["test", {"NamedEntityTag": "O", "CharacterOffsetEnd": "14", "Lemma": "test", "PartOfSpeech": "NN", "CharacterOffsetBegin": "10"}], [".", {"NamedEntityTag": "O", "CharacterOffsetEnd": "15", "Lemma": ".", "PartOfSpeech": ".", "CharacterOffsetBegin": "14"}]]}]}'

Where it is a dictionary wrapped in quotes making it a string. I'm not sure what I'm doing wrong...

jjm0022 commented 8 years ago

Closing the issue. I found a way to convert the string to a dictionary from this.