Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 3035e140 authored by Adam Tauber's avatar Adam Tauber
Browse files

Merge pull request #320 from dalf/currency

[enh] currency_convert engine : "1 dollars in euros"
parents c5135f69 5525625d
Loading
Loading
Loading
Loading
+7655 −0

File added.

Preview size limit exceeded, changes collapsed.

+47 −3
Original line number Diff line number Diff line
from datetime import datetime
import re
import os
import json
import unicodedata


categories = []
url = 'https://download.finance.yahoo.com/d/quotes.csv?e=.csv&f=sl1d1t1&s={query}=X'
weight = 100

parser_re = re.compile(r'^\W*(\d+(?:\.\d+)?)\W*([a-z]{3})\W*(?:in)?\W*([a-z]{3})\W*$', re.I)  # noqa
parser_re = re.compile(r'^\W*(\d+(?:\.\d+)?)\W*([^.0-9].+)\W*in?\W*([^\.]+)\W*$', re.I)  # noqa

db = 1


def normalize_name(name):
    name = name.lower().replace('-', ' ')
    name = re.sub(' +', ' ', name)
    return unicodedata.normalize('NFKD', u"" + name).lower()


def name_to_iso4217(name):
    global db

    name = normalize_name(name)
    currencies = db['names'].get(name, [name])
    return currencies[0]


def iso4217_to_name(iso4217, language):
    global db

    return db['iso4217'].get(iso4217, {}).get(language, iso4217)


def request(query, params):
@@ -16,6 +42,8 @@ def request(query, params):

    ammount, from_currency, to_currency = m.groups()
    ammount = float(ammount)
    from_currency = name_to_iso4217(from_currency.strip())
    to_currency = name_to_iso4217(to_currency.strip())

    q = (from_currency + to_currency).upper()

@@ -23,6 +51,8 @@ def request(query, params):
    params['ammount'] = ammount
    params['from'] = from_currency
    params['to'] = to_currency
    params['from_name'] = iso4217_to_name(from_currency, 'en')
    params['to_name'] = iso4217_to_name(to_currency, 'en')

    return params

@@ -35,12 +65,14 @@ def response(resp):
    except:
        return results

    answer = '{0} {1} = {2} {3} (1 {1} = {4} {3})'.format(
    answer = '{0} {1} = {2} {3}, 1 {1} ({5}) = {4} {3} ({6})'.format(
        resp.search_params['ammount'],
        resp.search_params['from'],
        resp.search_params['ammount'] * conversion_rate,
        resp.search_params['to'],
        conversion_rate
        conversion_rate,
        resp.search_params['from_name'],
        resp.search_params['to_name'],
    )

    now_date = datetime.now().strftime('%Y%m%d')
@@ -55,3 +87,15 @@ def response(resp):
    results.append({'answer': answer, 'url': url})

    return results


def load():
    global db

    current_dir = os.path.dirname(os.path.realpath(__file__))
    json_data = open(current_dir + "/../data/currencies.json").read()

    db = json.loads(json_data)


load()
+5 −3
Original line number Diff line number Diff line
@@ -27,9 +27,11 @@ class TestCurrencyConvertEngine(SearxTestCase):

    def test_response(self):
        dicto = defaultdict(dict)
        dicto['ammount'] = 10
        dicto['ammount'] = float(10)
        dicto['from'] = "EUR"
        dicto['to'] = "USD"
        dicto['from_name'] = "euro"
        dicto['to_name'] = "United States dollar"
        response = mock.Mock(text='a,b,c,d', search_params=dicto)
        self.assertEqual(currency_convert.response(response), [])

@@ -38,7 +40,7 @@ class TestCurrencyConvertEngine(SearxTestCase):
        results = currency_convert.response(response)
        self.assertEqual(type(results), list)
        self.assertEqual(len(results), 1)
        self.assertEqual(results[0]['answer'], '10 EUR = 5.0 USD (1 EUR = 0.5 USD)')
        self.assertEqual(results[0]['answer'], '10.0 EUR = 5.0 USD, 1 EUR (euro) = 0.5 USD (United States dollar)')
        now_date = datetime.now().strftime('%Y%m%d')
        self.assertEqual(results[0]['url'], 'https://finance.yahoo.com/currency/converter-results/' +
                                            now_date + '/10-eur-to-usd.html')
                                            now_date + '/10.0-eur-to-usd.html')
+161 −0
Original line number Diff line number Diff line
# -*- coding: utf-8 -*-
import json
import re
import unicodedata
import string
from urllib import urlencode
from requests import get
 
languages = {'de', 'en', 'es', 'fr', 'hu', 'it', 'nl', 'jp'}
 
url_template = 'https://www.wikidata.org/w/api.php?action=wbgetentities&format=json&{query}&props=labels%7Cdatatype%7Cclaims%7Caliases&languages=' + '|'.join(languages)
url_wmflabs_template = 'http://wdq.wmflabs.org/api?q=' 
url_wikidata_search_template='http://www.wikidata.org/w/api.php?action=query&list=search&format=json&srnamespace=0&srprop=sectiontitle&{query}'
 
wmflabs_queries = [ 
    'CLAIM[31:8142]', # all devise
]
 
db = {
    'iso4217' : {
        },
    'names' : {
        }
}


def remove_accents(data):
    return unicodedata.normalize('NFKD', data).lower()
        

def normalize_name(name):
    return re.sub(' +',' ', remove_accents(name.lower()).replace('-', ' '))


def add_currency_name(name, iso4217):
    global db

    db_names = db['names']


    if not isinstance(iso4217, basestring):
        print "problem", name, iso4217
        return

    name = normalize_name(name)

    if name == '':
        print "name empty", iso4217
        return

    iso4217_set = db_names.get(name, None)
    if iso4217_set is not None and iso4217 not in iso4217_set:
        db_names[name].append(iso4217)
    else:
        db_names[name] = [ iso4217 ]


def add_currency_label(label, iso4217, language):
    global db

    db['iso4217'][iso4217] = db['iso4217'].get(iso4217, {})
    db['iso4217'][iso4217][language] = label


def get_property_value(data, name):
    prop = data.get('claims', {}).get(name, {})
    if len(prop) == 0:
        return None
    
    value = prop[0].get('mainsnak', {}).get('datavalue', {}).get('value', '') 
    if value == '':
        return None

    return value
    

def parse_currency(data):
    iso4217 = get_property_value(data, 'P498')
        
    if iso4217 is not None:
        unit = get_property_value(data, 'P558')
        if unit is not None:
            add_currency_name(unit, iso4217)
                
        labels = data.get('labels', {})
        for language in languages:
            name = labels.get(language, {}).get('value', None)
            if name != None:
                add_currency_name(name, iso4217)
                add_currency_label(name, iso4217, language)

        aliases = data.get('aliases', {})
        for language in aliases:
            for i in range(0, len(aliases[language])):
                alias = aliases[language][i].get('value', None)
                add_currency_name(alias, iso4217)

 
def fetch_data(wikidata_ids):
    url = url_template.format(query=urlencode({'ids' : '|'.join(wikidata_ids)}))
    htmlresponse = get(url)
    jsonresponse = json.loads(htmlresponse.content)
    entities = jsonresponse.get('entities', {})
 
    for pname in entities:
        pvalue = entities.get(pname)
        parse_currency(pvalue)
 
 
def add_q(i):
    return "Q" + str(i)
 
 
def fetch_data_batch(wikidata_ids):
    while len(wikidata_ids) > 0:
        if len(wikidata_ids) > 50:
            fetch_data(wikidata_ids[0:49])
            wikidata_ids = wikidata_ids[50:]
        else:
            fetch_data(wikidata_ids)
            wikidata_ids = []
    
 
def wdq_query(query):
    url = url_wmflabs_template + query
    htmlresponse = get(url)
    jsonresponse = json.loads(htmlresponse.content)
    qlist = map(add_q, jsonresponse.get('items', {}))
    error = jsonresponse.get('status', {}).get('error', None)
    if error != None and error != 'OK':
        print "error for query '" + query + "' :" + error

    fetch_data_batch(qlist)
 
 
def wd_query(query, offset=0):
    qlist = []
 
    url = url_wikidata_search_template.format(query=urlencode({'srsearch': query, 'srlimit': 50, 'sroffset': offset}))
    htmlresponse = get(url)
    jsonresponse = json.loads(htmlresponse.content)
    for r in jsonresponse.get('query', {}).get('search', {}):
        qlist.append(r.get('title', ''))
    fetch_data_batch(qlist)
 
## fetch ##
for q in wmflabs_queries:
    wdq_query(q)

# static 
add_currency_name(u"euro", 'EUR')
add_currency_name(u"euros", 'EUR')
add_currency_name(u"dollar", 'USD')
add_currency_name(u"dollars", 'USD')
add_currency_name(u"peso", 'MXN')
add_currency_name(u"pesos", 'MXN')

# write
f = open("currencies.json", "wb")
json.dump(db, f, indent=4, encoding="utf-8")
f.close()