def query(query):
db = database.root
query = query.replace('+', ' ')
results = db.query(query)
if not results['entry'] and len(results['matches']) == 1:
return redirect(url_for('query', query=results['matches'].pop()))
sourcemetaphor = []
unknownaffixes = None
similar = None
if not results['matches']:
sourcemetaphor = [
e for a in compound2affixes(query) if len(a) != 1
for e in db.entries.itervalues() if a in e.searchaffixes
]
unknownaffixes = len(sourcemetaphor) != \
len([a for a in compound2affixes(query)
if len(a) != 1])
similar = [
e.word for e in db.entries.itervalues()
if dameraulevenshtein(query, e.word) == 1
]
similar += [
g.gloss for g in db.glosses if g.gloss not in similar
and dameraulevenshtein(query, g.gloss) == 1
]
results.update(locals())
return render_response('query.html', results)
def query(query):
db = database.root
query = query.replace('+', ' ')
results = db.query(query)
if not results['entry'] and len(results['matches']) == 1:
return redirect(url_for('query', query=results['matches'].pop()))
sourcemetaphor = []
unknownaffixes = None
similar = None
if not results['matches']:
sourcemetaphor = [e for a in compound2affixes(query)
if len(a) != 1
for e in db.entries.itervalues()
if a in e.searchaffixes]
unknownaffixes = len(sourcemetaphor) != \
len([a for a in compound2affixes(query)
if len(a) != 1])
similar = [e.word for e in db.entries.itervalues()
if dameraulevenshtein(query, e.word) == 1]
similar += [g.gloss for g in db.glosses
if g.gloss not in similar
and dameraulevenshtein(query, g.gloss) == 1]
results.update(locals())
return render_response('query.html', results)
def compound2affixes_splits_compounds():
"""The compound2affixes util splits compounds into affixes"""
same(compound2affixes('jbobau'), ['jbo', 'bau'])
same(compound2affixes('lobybau'), ['lob', 'y', 'bau'])
same(compound2affixes('jbobangu'), ['jbo', 'bangu'])
same(compound2affixes('lojbybau'), ['lojb', 'y', 'bau'])
same(compound2affixes('lobybangu'), ['lob', 'y', 'bangu'])
same(compound2affixes('lojbybangu'), ['lojb', 'y', 'bangu'])
same(compound2affixes("ro'inre'o"), ["ro'i", 'n', "re'o"])
def compound2affixes_splits_compounds():
"""The compound2affixes util splits compounds into affixes"""
same(compound2affixes('jbobau'), ['jbo', 'bau'])
same(compound2affixes('lobybau'), ['lob', 'y', 'bau'])
same(compound2affixes('jbobangu'), ['jbo', 'bangu'])
same(compound2affixes('lojbybau'), ['lojb', 'y', 'bau'])
same(compound2affixes('lobybangu'), ['lob', 'y', 'bangu'])
same(compound2affixes('lojbybangu'), ['lojb', 'y', 'bangu'])
same(compound2affixes("ro'inre'o"), ["ro'i", 'n', "re'o"])
def query(self, target, query):
fields = 'affix|class|type|notes|cll|url|components'
if query == 'help!':
self.msg(target, '<query http://tiny.cc/query-format > '
'[(%s)]' % fields)
return
field = 'definition'
match = re.search(r'\s\((?P<field>%s)\)$' % fields, query)
if match:
field = match.group('field')
query = re.sub(r'\s\(.+?\)$', '', query)
url = 'http://vlasisku.lojban.org/%s' % url_quote_plus(query)
results = database.root.query(query)
entry = results['entry']
if not entry and len(results['matches']) == 1:
entry = results['matches'].pop()
if entry or field == 'components':
case = lambda x: field == x
if case('definition'):
data = entry.textdefinition.encode('utf-8')
elif case('affix'):
data = ', '.join('-%s-' % i for i in entry.affixes)
elif case('class'):
data = entry.grammarclass
elif case('type'):
data = entry.type
elif case('notes'):
data = entry.textnotes.encode('utf-8')
elif case('cll'):
data = ' '.join(link for (chap, link) in entry.cll)
elif case('url'):
data = url
elif case('components'):
entry = query
data = ' '.join(e.word for a in compound2affixes(query)
if len(a) != 1
for e in database.root.entries.itervalues()
if a in e.searchaffixes)
data = data or '(none)'
if field == 'definition':
format = '%s = %s'
self.msg(target, format % (entry, data))
else:
format = '%s (%s) = %s'
self.msg(target, format % (entry, field, data))
class WordBot(BotBase):
nickname = 'valsi'
def query(self, target, query):
fields = 'affix|class|type|notes|cll|url|components|lujvo'
if query == 'help!':
self.msg(target, '<query http://tiny.cc/query-format > '
'[(%s)]' % fields)
return
elif query == 'update!':
def do_update():
from contextlib import closing
import urllib2
import xml.etree.cElementTree as etree
import os
self.msg(target, 'downloading...')
opener = urllib2.build_opener()
opener.addheaders.append(('Cookie', 'jbovlastesessionid=%s' % self.factory.app.config['BOT_KEY']))
url = 'http://jbovlaste.lojban.org/export/xml-export.html?lang=en'
with closing(opener.open(url)) as data:
xml = etree.parse(data)
assert xml.getroot().tag == 'dictionary'
with open('vlasisku/data/jbovlaste.xml', 'w') as file:
xml.write(file, 'utf-8')
self.msg(target, 'updating...')
database.init_app(database.app, True)
self.msg(target, 'done!')
import threading
threading.Thread(target=do_update).start()
return
field = 'definition'
match = re.search(r'\s\((?P<field>%s)\)$' % fields, query)
if match:
field = match.group('field')
query = re.sub(r'\s\(.+?\)$', '', query)
if field == 'lujvo':
try:
lujvos = jvocuhadju(query)
tanru = query
query = lujvos[0]
except ValueError, e:
self.msg(target, 'error: %s' % e)
return
url = 'http://%s/%s' % (self.factory.app.config['WEBSITE'], url_quote_plus(query))
results = database.root.query(query)
entry = results['entry']
if not entry and len(results['matches']) == 1:
entry = results['matches'].pop()
if entry or field in ['components', 'lujvo']:
case = lambda x: field == x
if case('definition'):
data = entry.textdefinition.encode('utf-8')
elif case('affix'):
data = ', '.join('-%s-' % i for i in entry.affixes)
elif case('class'):
data = entry.grammarclass
elif case('type'):
data = entry.type
elif case('notes'):
data = entry.textnotes.encode('utf-8')
elif case('cll'):
data = ' '.join(link for (chap, link) in entry.cll)
elif case('url'):
data = url
elif case('components'):
entry = query
data = ' '.join(e.word for a in compound2affixes(query)
if len(a) != 1
for e in database.root.entries.itervalues()
if a in e.searchaffixes)
elif case('lujvo'):
data = ', '.join(lujvos)
if entry:
data += ' (defined as %s = %s)' % (query, entry.textdefinition.encode('utf-8'))
entry = tanru
data = data or '(none)'
if field == 'definition':
format = '%s = %s'
self.msg(target, format % (entry, data))
else:
format = '%s (%s) = %s'
self.msg(target, format % (entry, field, data))
+results['glosses']
+results['affix']
+results['classes']
+results['types']
+results['definitions']
+results['notes'])
data = ', '.join(map(str, matches[:10]))
if len(results['matches']) > 10:
data += '…'
self.msg(target, format % (len(results['matches']),
's' if len(results['matches']) != 1
else '',
data))
else:
if field not in ['components', 'lujvo']:
rafsi = compound2affixes(query)
try:
if len(rafsi) > 0:
tanru = [e.word for a in rafsi
if len(a) != 1
for e in database.root.entries.itervalues()
if a in e.searchaffixes]
lujvo = jvocuhadju(' '.join(tanru))[0]
if lujvo != query:
if field != 'definition':
return self.query(target, '%s (%s)' % (lujvo, field))
else:
return self.query(target, lujvo)
except:
pass
self.msg(target, 'no results. %s' % url)
def query(self, target, query):
fields = 'affix|class|type|notes|cll|url|components'
if query == 'help!':
self.msg(target, '<query http://tiny.cc/query-format > '
'[(%s)]' % fields)
return
field = 'definition'
match = re.search(r'\s\((?P<field>%s)\)$' % fields, query)
if match:
field = match.group('field')
query = re.sub(r'\s\(.+?\)$', '', query)
url = 'http://vlasisku.lojban.org/%s' % url_quote_plus(query)
results = database.root.query(query)
entry = results['entry']
if not entry and len(results['matches']) == 1:
entry = results['matches'].pop()
if entry or field == 'components':
case = lambda x: field == x
if case('definition'):
data = entry.textdefinition.encode('utf-8')
elif case('affix'):
data = ', '.join('-%s-' % i for i in entry.affixes)
elif case('class'):
data = entry.grammarclass
elif case('type'):
data = entry.type
elif case('notes'):
data = entry.textnotes.encode('utf-8')
elif case('cll'):
data = ' '.join(link for (chap, link) in entry.cll)
elif case('url'):
data = url
elif case('components'):
entry = query
data = ' '.join(e.word for a in compound2affixes(query)
if len(a) != 1
for e in database.root.entries.itervalues()
if a in e.searchaffixes)
data = data or '(none)'
if field == 'definition':
format = '%s = %s'
self.msg(target, format % (entry, data))
else:
format = '%s (%s) = %s'
self.msg(target, format % (entry, field, data))
elif results['matches']:
format = '%d result%s: %s'
matches = (results['words']
+results['glosses']
+results['affix']
+results['classes']
+results['types']
+results['definitions']
+results['notes'])
data = ', '.join(map(str, matches[:10]))
if len(results['matches']) > 10:
data += '…'
self.msg(target, format % (len(results['matches']),
's' if len(results['matches']) != 1
else '',
data))
else:
self.msg(target, 'no results. %s' % url)
