def integrate_gloss(word):
cs = word["contents"]
if has(cs, "大意"):
glossy = hqueryx(cs, "大意")
if has(cs, "gloss"):
gloss = hqueryx(cs, "gloss")
if glossy["text"] not in gloss["text"].split(", "):
gloss["text"] += ", " + glossy["text"]
else:
cs.append(otm.content("gloss", glossy["text"]))
del cs[hqueryi(cs, "大意")]
return word
def splitnotes(word):
cs = word["contents"]
if has(cs, "notes"):
notes = hqueryx(cs, "notes")["text"]
return splitnotes_(notes)
else:
return []
def delete_emptynotes(word):
'''
空白記号のみの notes を消去。
'''
cs = word["contents"]
if has(cs, "notes") and re.search(r'^\s*$', hqueryx(cs, "notes")["text"]):
del cs[hqueryi(cs, "notes")]
return word
def add_relations_for_multi(task_words, entry_dict):
'''
r"{[a-zA-Z']}" に該当する単語のうち、エントリーのあるものだけを relations に加える。
"ja" の場合「・関連語」も対象にする。
'''
regex = r"\{[a-zA-Z']+\}"
#entry_dict = entry_dict.copy()
for word in task_words:
cs = word["contents"]
_ls = []
if has(cs, "notes"):
_ls = re.findall(regex, hqueryx(cs, "notes")["text"])
if has(cs, "関連語"):
_ls += re.findall(regex, hqueryx(cs, "関連語")["text"])
_ls = [re.sub(r'^[^a-zA-Z]+', '', re.sub(r'[^a-zA-Z]+$', '', e)) for e in _ls]
relations = []
for rel in _ls:
for entry in entry_dict[rel[0]]:
if entry["form"] == rel:
relations.append(otm.relation("", entry))
break
word["relations"] = relations
return task_words
def goodnotes(word):
cs = word["contents"]
if has(cs, "notes"):
after = splitnotes(word)
hqueryx(cs, "notes")["text"] = ""
for elem in after:
if isinstance(elem, dict):
keys = ["大意", "読み方", "語呂合わせ", "関連語"]
for k in keys:
if k in elem:
cs.append({"title":k, "text":elem[k]})
else:
hqueryx(cs, "notes")["text"] += elem + " "
example_extract(word)
return word