def loadMails(self):
obj2id = self.obj2id
mails = []
for mail in mailLoaderGen(self.archivedir):
#tags = set(getattr(mail, 'tags', []) + getattr(mail, 'type', []))
if hasattr(mail, 'tags'):
alltags.update(mail.tags)
if hasattr(mail, 'type'):
alltags[mail.type] += 1
#mail.tags = [obj2id.add(Tag(tag)) for tag in mail.tags]
mail.id = obj2id.add(mail)
mails.append(mail)
self.mails = mails
"date",
"tags",
"sfbi_url",
"contact-nom",
"contact-email",
"date-candidature",
"validite",
"duree",
"ville",
"lieu",
"labo",
]
outdir = "archives_SFBI_AnnotationManuelle"
mails = list(mailLoaderGen())
words = Counter()
for mail in mails:
mail.sents = list(iterTokenizedSentences(mail.description))
for sent in mail.sents:
words.update(sent)
stemmer = Stemmer(set(word for (word, n) in words.items() if n > 10))
for m in mails:
outf = outdir + m.mailfile.strip("archives_SFBI")
d = m.__dict__
d["date"] = date.fromtimestamp(d["timestamp"]).strftime("%d %B %Y")
with open(outf, "wt") as f:
d["from"] = d.pop("sender")
