def __setstate__(self, state):
# we do this import here so that we can import 'document' without
# causing a recursive import of 'repository'
from uplib.repository import TheRepository
self.id = state[0]
self.repo = TheRepository
self.__metadata = None
self.__folder = self.repo.doc_location(self.id)
self.__addtime = id_to_time(self.id)
self.__pdforig = None
def __init__ (self, repo, id):
self.repo = repo
self.id = id
self.__metadata = None
self.__folder = repo.doc_location(id)
self.__addtime = id_to_time(id)
self.__pdforig = None
self.__category_strings = None
self.__citation = None
self.__date = None
self.__bboxes = {}
self.__links = {}
self.__removed_links = {}
self.__touch_time = None
self.__pagenumbering = None
self.__icon_size = None
def read_document (doc, categories, collections, authors):
def figure_date(datestring):
d2 = parse_date(datestring)
if (not d2) or (sum(d2) == 0):
return 0
return d2[0] * (13 * 32) + d2[1] * 13 + d2[2]
docdata = {'id': doc.id, 'rloc': 0}
mdata = doc.get_metadata()
docdata['title'] = mdata.get('title', "")
docdata['page-count'] = int(mdata.get('page-count', 1))
date = mdata.get('date')
if date:
docdata['date'] = figure_date(date);
else:
docdata['date'] = 0
docdata['addtime'] = int(id_to_time(doc.id))
# we don't really know the reftime (FIXME) but we'll use the document add time as an approximation
docdata['reftime'] = docdata['addtime']
docdata['categories'] = []
cstring = mdata.get('categories', "")
if cstring:
for category in split_categories_string(cstring):
if not category in categories:
categories[category] = { 'rloc': 0, 'docs': [ doc.id, ], 'name': category }
else:
categories[category]['docs'].append(doc.id)
docdata['categories'].append(category)
docdata['authors'] = []
auths = mdata.get('authors', "").split(" and ")
for auth in auths:
if auth:
authname = figure_author_name(auth)
if not authname in authors:
authors[authname] = { 'rloc': 0, 'docs': [ doc.id, ], 'name': authname }
else:
authors[authname]['docs'].append(doc.id)
docdata['authors'].append(authname)
return docdata
def build_html_abstract_display (self, doc, icon_cid):
fp = StringIO()
dict = doc.get_metadata()
pubdate = dict.get("date")
date = re.sub(" 0|^0", " ",
time.strftime("%d %b %Y, %I:%M %p",
time.localtime(id_to_time(doc.id))))
name = doc.id
page_count = dict.get('page-count')
summary = '<i>(No summary available.)</i>'
if dict:
if dict.has_key('title'):
name = dict.get('title')
elif dict.has_key('name'):
name = '[' + dict.get('name') + ']'
fp.write(u'<table border=0><tr><td>')
fp.write(u'<center>')
fp.write(u'<a href="https://%s:%d/action/basic/dv_show?doc_id=%s" border=0>' % (self.ip, doc.repo.secure_port(), doc.id))
fp.write(u'<img src="cid:%s">' % icon_cid)
fp.write(u'</a><p><small><font color="%s">(%s)</font></small></center></td><td> </td>'
% (STANDARD_DARK_COLOR, date))
fp.write(u'<td valign=top><h3>%s</h3>' % htmlescape(name))
if dict.has_key(u'authors') or pubdate:
fp.write(u'<p><small>')
if dict.has_key('authors'):
fp.write(u'<b> %s</b>'
% (re.sub(' and ', ', ', dict['authors'])))
if pubdate:
formatted_date = format_date(pubdate, True)
fp.write(u' <i><font color="%s">%s</font></i>' % (STANDARD_DARK_COLOR,
formatted_date))
fp.write(u'</small>\n')
if dict.has_key('comment'):
summary = htmlescape(dict.get('comment', ''))
elif dict.has_key('abstract'):
summary = "<i>" + htmlescape(dict.get('abstract', '')) + '</i>'
elif dict.has_key('summary'):
summary = '<font color="%s">' % STANDARD_DARK_COLOR + htmlescape(dict.get('summary')) + '</font>'
fp.write(u'<P>%s' % summary)
if page_count:
fp.write(u'<small><i><font color="%s"> · (%s page%s)'
% (STANDARD_DARK_COLOR, page_count, ((int(page_count) != 1) and "s") or ""))
fp.write(u'</font></i></small>\n')
cstrings = doc.get_category_strings()
fp.write(u'<p>Categories: ')
if cstrings:
fp.write(string.join([htmlescape(s) for s in cstrings], u' · '))
else:
fp.write('(none)')
typ = doc.get_metadata("apparent-mime-type")
if typ:
mtype = ' · <small>%s</small>' % typ
else:
mtype = ''
fp.write(u'<p><a href="https://%s:%s/action/externalAPI/fetch_original?doc_id=%s&browser=true"><font color="%s">(Original%s)</font></a>'
% (self.ip, doc.repo.secure_port(), doc.id, STANDARD_DARK_COLOR, mtype))
fp.write(u' · <a href="https://%s:%s/action/basic/doc_pdf?doc_id=%s"><font color="%s">(PDF)</font></a>'
% (self.ip, doc.repo.secure_port(), doc.id, STANDARD_DARK_COLOR))
if not mtype.lower().startswith("text/html"):
fp.write(u' · <a href="https://%s:%s/action/basic/doc_html?doc_id=%s"><font color="%s">(HTML)</font></a>'
% (self.ip, doc.repo.secure_port(), doc.id, STANDARD_DARK_COLOR))
fp.write(u'</td></tr></table>')
d = fp.getvalue()
fp.close()
return d, name
def get_msg_as_email (self, doc):
try:
mime_type = doc.get_metadata("apparent-mime-type")
if mime_type == "message/rfc822":
f = os.path.join(doc.folder(), "originals")
filepath = os.path.join(f, os.listdir(f)[0])
fp = open(filepath, 'r')
s = fp.read()
fp.close()
msg = message_from_string(s)
else:
def make_header(name, value):
try:
v = value.encode("US-ASCII")
charset = "US-ASCII"
except:
v = value.encode("UTF-8")
charset = "UTF-8"
return name, email.Header.Header(v, charset, 77, name).encode()
def build_icon(doc):
icon = doc.document_icon()
img_part = email.Message.Message()
img_part.set_type("image/png")
cid = "%s.%s.%s.icon" % (self.ip, doc.repo.secure_port(), doc.id)
img_part.add_header("Content-ID", cid)
img_part.add_header("Content-Transfer-Encoding", "base64")
img_part.set_payload(base64.encodestring(icon))
return img_part
def build_description(doc, display):
desc_part = email.Message.Message()
desc_part.set_type("text/html")
desc_part.add_header("Content-Transfer-Encoding", "quoted-printable")
desc_part.set_payload(quopri.encodestring('<html><body bgcolor="%s">' % STANDARD_BACKGROUND_COLOR +
display.encode('UTF-8') + "</body></html>\n"), "UTF-8")
return desc_part
icon_payload = build_icon(doc)
display, name = self.build_html_abstract_display(doc, icon_payload.get("Content-ID"))
msg = email.Message.Message()
msg.set_type("multipart/related;boundary=%s%s%s%s" % (self.ip, doc.repo.secure_port(), doc.id, long(time.time())))
msg.add_header(*make_header("Message-ID", "%s:%s:%s" % (self.ip, doc.repo.secure_port(), doc.id)))
d = doc.get_date()
if d:
try:
d = email.Utils.formatdate(time.mktime((d[0], (d[1] or 1), (d[2] or 1), 0, 0, 0, 0, 1, -1,)))
except:
d = email.Utils.formatdate(id_to_time(doc.id))
else:
d = email.Utils.formatdate(id_to_time(doc.id))
msg.add_header(*make_header("Date", d))
msg.add_header(*make_header("Subject", name))
authors = doc.get_metadata("authors")
if authors:
authors = authors.replace(" and ", ", ").replace('"', '\\"').replace('\r', '\\\r').replace('\\', '\\\\')
msg.add_header(*make_header("From", '"' + authors + '"'))
body_payload = build_description(doc, display)
msg.attach(body_payload)
msg.attach(icon_payload)
# note("msg is:\n%s", str(msg))
return msg
except:
note("Exception getting document %s as email:\n%s", doc.id, string.join(traceback.format_exception(*sys.exc_info())))
return None
