Created
January 31, 2014 21:21
-
-
Save vanatteveldt/8743409 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
from amcatclient.api import AmcatAPI | |
from amcat.nlp.naf import NAF_Article | |
from amcat.nlp.syntaxtree import _naf_to_rdf, SyntaxTree | |
from amcat.tools.pysoh.pysoh import SOHServer | |
from collections import defaultdict | |
import re | |
aid = 3958 | |
api = AmcatAPI("http://amcat.vu.nl", "wva", "geheim!") | |
soh = SOHServer("http://localhost:3030/x") | |
ruleset = api.request("rulesets/5", page_size=999) | |
rules = api.request("rulesets/5/rules", page_size=999)['results'] | |
labels = api.request("label", code__codebook_codes__codebook__id=ruleset['lexicon_codebook'], page_size=9999)['results'] | |
codes = defaultdict(dict) | |
for label in labels: | |
codes[int(label['code'])][int(label['language'])] = label['label'] | |
lexicon = defaultdict(set) # {lemma : {lexclass, ... }} | |
lexlang = int(ruleset['lexicon_language']) | |
for code, labels in codes.iteritems(): | |
if lexlang not in labels: continue | |
lexclass = labels[min(labels)] | |
for lemma in set(re.findall("[\w*]+", labels[lexlang])): | |
lexicon[lemma].add(lexclass) | |
naf = api.request("projects/49/articlesets/4635/articles/{aid}/xtas/parse_alpino/".format(**locals()))['results'] | |
naf = NAF_Article.from_dict(naf) | |
rdf = _naf_to_rdf(naf, 73) | |
t = SyntaxTree(soh, rdf) | |
t.apply_lexicon(lexicon) | |
rules = sorted(rules, key=lambda r: int(r['order'])) | |
for rule in rules: | |
print "Applying rule {id}: {label}".format(**rule) | |
t.apply_rule(rule['where'], rule['insert'], rule['remove']) | |
g = t.visualise() | |
open("/tmp/bla.html", "w").write(g.getHTMLObject()) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment