def variant(x):
# Map the keywords used in var_type to values used
# in the database kw* tables, where they differ.
vmap = {'njecd': 'halpern_njecd', 'oneill': 'oneill_names'}
vt = x.get('var_type')
vt = vmap.get(vt, vt)
if vt == 'ucs': kw = 0
else: kw = KW.CINF[Xml2db.CINF.get(vt, vt)].id
return jdb.Cinf(kw=kw, value=x.text)
def radical(x, c):
cinf = c._cinf
if len(list(x.keys())) != 1: warn('Expected only one rad_value attribute')
rad_attr, rad_type = list(x.items())[0]
if rad_attr != 'rad_type':
warn('Unexpected rad_value attribute: %s', rad_attr)
if rad_type == 'classical': c.bushu = int(x.text)
elif rad_type == 'nelson_c':
cinf.append(jdb.Cinf(kw=KW.CINF_nelson_rad, value=int(x.text)))
else:
warn("Unknown radical attribute value: %s=\"%s\"",
(rad_attr, rad_type))
def codepoint(x, c, chtxt):
cinf = c._cinf
if len(list(x.keys())) != 1: warn('Expected only one cp_value attribute')
cp_attr, cp_type = list(x.items())[0]
if cp_attr != 'cp_type': warn('Unexpected cp_value attribute', cp_attr)
if cp_type == 'ucs':
if int(x.text, 16) != jdb.uord(chtxt):
warn ("xml codepoint ucs value '%s' doesnt match character %s (0x%x)." \
% (x.text, chtxt, jdb.uord (chtxt)))
else:
cinf.append(
jdb.Cinf(kw=KW.CINF[Xml2db.CINF.get(cp_type, cp_type)].id,
value=x.text))
def rmgroup(rmg, langs=None):
rdngs = []
glosses = []
cinf = []
dupchk = {}
for x in rmg.findall('reading'):
rtype = None
rstat = None
cinfrec = None
for aname, aval in list(x.items()):
if aname == 'r_type': rtype = aval
if aname == 'on_type': rtype = aval
if aname == 'r_status': rstat = aval
if rtype in ('pinyin', 'korean_r', 'korean_h', 'vietnam'):
if (rtype, x.text) in dupchk:
warn("Duplicate reading ignored: %s, %s" % (rtype, x.text))
continue
dupchk[(rtype, x.text)] = True
cinf.append(jdb.Cinf(kw=KW.CINF[rtype].id, value=x.text))
elif rtype == 'ja_on' or rtype == 'ja_kun':
if x.text in dupchk:
warn('Duplicate reading ignored: %s' % x.text)
continue
dupchk[x.text] = True
rdng = jdb.Rdng(txt=x.text, _inf=[])
rdng._inf.append(
jdb.Rinf(kw=KW.RINF[Xml2db.RINF.get(aval, aval)].id))
if rstat:
rdng._inf.append(
jdb.Rinf(kw=KW.RINF[Xml2db.RINF.get(rstat, rstat)].id))
rdngs.append(rdng)
else:
raise KeyError('Unkown r_type attribute: %s' % rtype)
dupchk = {}
for x in rmg.findall('meaning'):
lang = x.get('m_lang', 'en')
langkw = KW.LANG[Xml2db.LANG.get(lang, lang)].id
if (lang, x.text) in dupchk:
warn("Duplicate lang,meaning pair ignored: %s:%s" % (lang, x.text))
continue
dupchk[(lang, x.text)] = True
if not langs or langkw in langs:
glosses.append(jdb.Gloss(txt=x.text, lang=langkw, ginf=1))
return rdngs, glosses, cinf
def dicnum(dic_number, cinf):
dupchk = {}
for x in dic_number.findall('dic_ref'):
drtype = x.get('dr_type')
val = x.text
if x.get('m_vol'):
val = "%s.%s.%s" % (x.get('m_vol'), x.get('m_page'), x.text)
key = Xml2db.CINF.get(drtype, drtype)
try:
kw = KW.CINF[key].id
except KeyError:
warn('Unknown CINF keyword: "%s"' % key)
continue
if (kw, val) in dupchk:
warn('Duplicate dr_type,value pair ignored: %s, %s' %
(drtype, val))
continue
dupchk[(kw, val)] = True
cinf.append(jdb.Cinf(kw=kw, value=val))
def qcode(query_code, cinf):
dupchk = {}
saw_misclass = False
saw_skip = False
for x in query_code.findall('q_code'):
qctype = x.get('qc_type')
val = x.text
kw = KW.CINF[Xml2db.CINF.get(qctype, qctype)].id
misclass = x.get('skip_misclass', '')
if (kw, val) in dupchk:
warn('Duplicate qc_type,value pair ignored: %s,%s' % (qctype, val))
continue
dupchk[(kw, val)] = True
if misclass:
if qctype != "skip":
raise KeyError("'skip_misclass' attr on non-skip element")
saw_misclass = True
elif qctype == 'skip':
saw_skip = True
cinf.append(jdb.Cinf(kw=kw, value=val, mctype=misclass))
if saw_misclass and not saw_skip:
warn("Has skip_misclass but no skip")
def strokes(x, n, c):
cinf = c._cinf
if n == 0:
c.strokes = int(x.text)
else:
cinf.append(jdb.Cinf(kw=KW.CINF_strokes, value=int(x.text)))