def _friend_item(users, user, item):
"""Parses a new friend item."""
# Sometimes API returns an empty new friend item:
# { "date": 1390235880, "source_id": 1334862, "type": "friend" }
if "friends" not in item:
return None
html = ""
rows = []
for friend in item["friends"]["items"]:
friend = users[friend["uid"]]
friend_url = _get_user_url(friend["id"])
rows.append([
_link(friend_url, _image(friend["photo"])),
_link(friend_url, friend["name"]),
])
html += _table(rows, column_spacing=7)
if item["friends"]["count"] > len(item["friends"]["items"]):
html += _block("[показаны не все новые друзья]")
return {
"title": user["name"] + ": новые друзья",
"text": html,
"url": _vk_url("friends?id={}§ion=all".format(user["id"])),
}
def _friend_item(users, user, item):
"""Parses a new friend item."""
# Sometimes API returns an empty new friend item:
# { "date": 1390235880, "source_id": 1334862, "type": "friend" }
if "friends" not in item:
return None
html = ""
friends = item["friends"][1:]
rows = []
for friend in friends:
friend = users[friend["uid"]]
friend_url = _get_user_url(friend["id"])
rows.append([
_link(friend_url, _image(friend["photo"])),
_link(friend_url, friend["name"]),
])
html += _table(rows, column_spacing=7)
if item["friends"][0] > len(friends):
html += _block("[показаны не все новые друзья]")
return {
"title": user["name"] + ": новые друзья",
"text": html,
"url": _vk_url("friends?id={}§ion=all".format(user["id"])),
}
def _parse_text(html):
"""Parses a post text."""
html = _TEXT_URL_RE.sub(r"\1" + _link(r"\2", r"\2") + r"\3", html)
html = _DOMAIN_ONLY_TEXT_URL_RE.sub(r"\1" + _link(r"http://\2", r"\2") + r"\3", html)
html = _USER_LINK_RE.sub(_em(_link(_vk_url(r"\1"), r"\2")), html)
return html.strip()
def _parse_text(html):
"""Parses a post text."""
html = _TEXT_URL_RE.sub(r"\1" + _link(r"\2", r"\2") + r"\3", html)
html = _DOMAIN_ONLY_TEXT_URL_RE.sub(
r"\1" + _link(r"http://\2", r"\2") + r"\3", html)
html = _USER_LINK_RE.sub(_em(_link(_vk_url(r"\1"), r"\2")), html)
return html.strip()
def _photo_item(users, user, api_item):
"""Parses a photo item."""
if api_item["type"] == "photo":
title = "новые фотографии"
photos = api_item["photos"]
get_photo_url = lambda photo: _vk_url("feed?" + urlencode({
"section": "photos",
"z": "photo{owner_id}_{photo_id}/feed1_{source_id}_{timestamp}".format(
owner_id=photo["owner_id"], photo_id=photo["pid"],
source_id=api_item["source_id"], timestamp=api_item["date"])}))
elif api_item["type"] == "photo_tag":
title = "новые отметки на фотографиях"
photos = api_item["photo_tags"]
get_photo_url = lambda photo: _vk_url("feed?" + urlencode({
"z": "photo{owner_id}_{photo_id}/feed3_{source_id}_{timestamp}".format(
owner_id=photo["owner_id"], photo_id=photo["pid"],
source_id=api_item["source_id"], timestamp=api_item["date"])}))
else:
raise Error("Logical error.")
item = {
"title": user["name"] + ": " + title,
"text": "",
}
for photo in photos[1:]:
url = get_photo_url(photo)
item.setdefault("url", url)
item["text"] += _block(_link(url, _image(photo["src_big"])))
if photos[0] > len(photos) - 1:
item["text"] += _block("[показаны не все фотографии]")
return item
def _parse_text(text, tweet_entities):
"""Parses a tweet text."""
sorted_entities = []
for entity_type, entities in tweet_entities.items():
for entity in entities:
sorted_entities.append((entity_type, entity))
sorted_entities.sort(
key=lambda entity_tuple: entity_tuple[1]["indices"][0], reverse=True)
html = ""
media_html = ""
pos = len(text)
for entity_type, entity in sorted_entities:
start, end = entity["indices"]
if end < pos:
html = text[end:pos] + html
if entity_type == "urls":
html = _link(entity["expanded_url"], entity["display_url"]) + html
elif entity_type == "user_mentions":
html = _link(_twitter_user_url(entity["screen_name"]),
entity["name"]) + html
elif entity_type == "hashtags":
html = _link(
_TWITTER_URL + "search?" + urlencode(
{"q": "#" + entity["full_text"]}), text[start:end]) + html
elif entity_type == "media":
html = _link(entity["expanded_url"], entity["display_url"]) + html
media_html += _block(
_link(entity["expanded_url"],
_image(entity["media_url_https"])))
else:
LOG.error("Unknown tweet entity:\n%s", pprint.pformat(entity))
html = text[start:end] + html
pos = start
if pos:
html = text[:pos] + html
return _block(html) + media_html
def _parse_text(text, tweet_entities):
"""Parses a tweet text."""
sorted_entities = []
for entity_type, entities in tweet_entities.items():
for entity in entities:
sorted_entities.append(( entity_type, entity ))
sorted_entities.sort(
key=lambda entity_tuple: entity_tuple[1]["indices"][0], reverse=True)
html = ""
media_html = ""
pos = len(text)
for entity_type, entity in sorted_entities:
start, end = entity["indices"]
if end < pos:
html = text[end:pos] + html
if entity_type == "urls":
html = _link(entity["expanded_url"], entity["display_url"]) + html
elif entity_type == "user_mentions":
html = _link(_twitter_user_url(entity["screen_name"]), entity["name"]) + html
elif entity_type == "hashtags":
html = _link(_TWITTER_URL + "search?" + urlencode({ "q": "#" + entity["text"] }), text[start:end]) + html
elif entity_type == "media":
html = _link(entity["expanded_url"], entity["display_url"]) + html
media_html += _block(_link(entity["expanded_url"], _image(entity["media_url_https"])))
else:
LOG.error("Unknown tweet entity:\n%s", pprint.pformat(entity))
html = text[start:end] + html
pos = start
if pos:
html = text[:pos] + html
return _block(html) + media_html
def _photo_item(users, user, api_item):
"""Parses a photo item."""
if api_item["type"] == "photo":
title = "новые фотографии"
photos = api_item["photos"]
get_photo_url = lambda photo: _vk_url("feed?" + urlencode({
"section":
"photos",
"z":
"photo{owner_id}_{photo_id}/feed1_{source_id}_{timestamp}".format(
owner_id=photo["owner_id"],
photo_id=photo["id"],
source_id=api_item["source_id"],
timestamp=api_item["date"])
}))
elif api_item["type"] == "photo_tag":
title = "новые отметки на фотографиях"
photos = api_item["photo_tags"]
get_photo_url = lambda photo: _vk_url("feed?" + urlencode({
"z":
"photo{owner_id}_{photo_id}/feed3_{source_id}_{timestamp}".format(
owner_id=photo["owner_id"],
photo_id=photo["id"],
source_id=api_item["source_id"],
timestamp=api_item["date"])
}))
else:
raise Error("Logical error.")
item = {
"title": user["name"] + ": " + title,
"text": "",
}
for photo in photos["items"]:
url = get_photo_url(photo)
item.setdefault("url", url)
item["text"] += _block(_link(url, _image(photo["photo_604"])))
if photos["count"] > len(photos["items"]):
item["text"] += _block("[показаны не все фотографии]")
return item
def _post_item(users, user, item):
"""Parses a wall post item."""
attachment_order = (
"doc",
"note",
"page",
"poll",
"album",
"posted_photo",
"photo",
"graffiti",
"app",
"video",
"link",
"audio",
)
def attachment_sort_key(attachment):
try:
return attachment_order.index(attachment["type"])
except ValueError:
return len(attachment_order)
top_html = ""
bottom_html = ""
categories = set()
unknown_attachments = set()
attachments = sorted(
item.get("attachments", []), key=attachment_sort_key)
if not item["text"] and not attachments and "geo" in item:
LOG.debug("Skip check-in item from %s from %s.", user["name"], item["date"])
return
if (
"attachment" in item and
item["text"] == item["attachment"][item["attachment"]["type"]].get("title")
):
main_html = ""
else:
main_html = _parse_text(item["text"])
photo_count = functools.reduce(
lambda count, attachment:
count + ( attachment["type"] in ("app", "graffiti", "photo", "posted_photo") ),
attachments, 0)
big_image = photo_count == 1
for attachment in attachments:
attachment_category = attachment["type"]
# Notice: attachment object is not always stored in
# attachment[attachment["type"]] - sometimes it's stored under a
# different key, so we can't obtain it here for all attachment types.
if attachment["type"] == "app":
info = attachment[attachment["type"]]
top_html += _block(
_vk_link(_vk_id("app", info["app_id"]),
_image(info["src_big" if big_image else "src"])))
elif attachment["type"] == "graffiti":
info = attachment[attachment["type"]]
top_html += _block(
_vk_link(_vk_id("graffiti", info["gid"]),
_image(info["src_big" if big_image else "src"])))
elif attachment["type"] == "link":
info = attachment[attachment["type"]]
link_block = _em("Ссылка: " + _link(info["url"], info["title"]))
link_description = _parse_text(info["description"]) or info["title"]
if "image_src" in info:
if link_description:
link_block += _image_block(info["url"], info["image_src"], link_description)
else:
link_block += _block(_link(info["url"], _image(info["image_src"])))
elif link_description:
link_block += _block(link_description)
top_html += _block(link_block)
elif attachment["type"] == "album":
info = attachment[attachment["type"]]
top_html += _image_block(
_vk_url("album", info["owner_id"], info["aid"]), info["thumb"]["src"],
"Альбом: {description} ({size} фото)".format(description=info["description"].strip(), size=info["size"]))
elif attachment["type"] == "photo":
top_html += _photo(attachment[attachment["type"]], big_image)
attachment_category = "posted_photo"
elif attachment["type"] == "posted_photo":
top_html += _photo(attachment[attachment["type"]], big_image)
elif attachment["type"] == "photos_list":
# It seems like photos_list always duplicates photo attachments
attachment_category = "posted_photo"
elif attachment["type"] == "audio":
info = attachment[attachment["type"]]
bottom_html += _block(_em(
"Аудиозапись: " +
_vk_link(
"search?" + urlencode({
"c[q]": info["performer"] + " - " + info["title"],
"c[section]": "audio"
}),
"{} - {} ({})".format(info["performer"], info["title"],
_duration(info["duration"])))))
elif attachment["type"] == "doc":
info = attachment[attachment["type"]]
if "url" in info and "thumb" in info:
bottom_html += _block(_image_block(
info["url"], info["thumb"],
_link(info["url"], info["title"])))
else:
bottom_html += _block(_em("Документ: {}".format(info["title"])))
elif attachment["type"] == "video":
info = attachment[attachment["type"]]
top_html += _block(
_image(info["image"]) +
_block(_em("{} ({})".format(info["title"], _duration(info["duration"])))))
elif attachment["type"] == "note":
top_html += _block(_em("Заметка: {}".format(
attachment[attachment["type"]]["title"])))
elif attachment["type"] == "page":
top_html += _block(_em("Страница: {}".format(
attachment[attachment["type"]]["title"])))
elif attachment["type"] == "poll":
top_html += _block(_em("Опрос: {}".format(
attachment[attachment["type"]]["question"])))
else:
unknown_attachments.add(attachment["type"])
categories.add(_CATEGORY_TYPE + attachment_category)
if unknown_attachments:
LOG.error("Got a post with unknown attachment type (%s):\n%s",
", ".join(unknown_attachments), pprint.pformat(item))
html = top_html + main_html + bottom_html
if "copy_owner_id" in item and "copy_post_id" in item:
html = _block(
_em(_link(
_get_user_url(item["copy_owner_id"]),
users[item["copy_owner_id"]]["name"]
)) + " пишет:"
) + html
if "copy_text" in item:
html = _quote_block(item["copy_text"], html)
categories.add(_CATEGORY_TYPE_REPOST)
return {
"title": user["name"] + ": запись на стене",
"text": html,
"url": _vk_url("wall", user["id"], item["post_id"]),
"categories": categories,
}
def _vk_link(target, html):
"""Renders a VK link."""
return _link(_vk_url(target), html)
def _post_item(users, user, item):
"""Parses a wall post item."""
attachment_order = (
"doc",
"note",
"page",
"poll",
"album",
"posted_photo",
"photo",
"graffiti",
"app",
"video",
"link",
"audio",
)
def attachment_sort_key(attachment):
try:
return attachment_order.index(attachment["type"])
except ValueError:
return len(attachment_order)
top_html = ""
bottom_html = ""
categories = set()
unknown_attachments = set()
attachments = sorted(item.get("attachments", []), key=attachment_sort_key)
if not item["text"] and not attachments and "geo" in item:
LOG.debug("Skip check-in item from %s from %s.", user["name"],
item["date"])
return
if ("attachment" in item and item["text"]
== item["attachment"][item["attachment"]["type"]].get("title")):
main_html = ""
else:
main_html = _parse_text(item["text"])
photo_count = functools.reduce(
lambda count, attachment: count +
(attachment["type"] in
("app", "graffiti", "photo", "posted_photo")), attachments, 0)
big_image = photo_count == 1
for attachment in attachments:
attachment_category = attachment["type"]
# Notice: attachment object is not always stored in
# attachment[attachment["type"]] - sometimes it's stored under a
# different key, so we can't obtain it here for all attachment types.
if attachment["type"] == "app":
info = attachment[attachment["type"]]
top_html += _block(
_vk_link(
_vk_id("app", info["id"]),
_image(info["photo_604" if big_image else "photo_130"])))
elif attachment["type"] == "graffiti":
info = attachment[attachment["type"]]
top_html += _block(
_vk_link(
_vk_id("graffiti", info["id"]),
_image(info["photo_604" if big_image else "photo_130"])))
elif attachment["type"] == "link":
info = attachment[attachment["type"]]
link_block = _em("Ссылка: " + _link(info["url"], info["title"]))
link_description = _parse_text(
info["description"]) or info["title"]
if "photo" in info:
if link_description:
link_block += _image_block(info["url"],
info["photo"]["photo_130"],
link_description)
else:
link_block += _block(
_link(info["url"], _image(info["photo"]["photo_130"])))
elif link_description:
link_block += _block(link_description)
top_html += _block(link_block)
elif attachment["type"] == "album":
info = attachment[attachment["type"]]
top_html += _image_block(
_vk_url("album", info["owner_id"],
info["id"]), info["thumb"]["photo_130"],
"Альбом: {description} ({size} фото)".format(
description=info["description"].strip(),
size=info["size"]))
elif attachment["type"] == "photo":
top_html += _photo(attachment[attachment["type"]], big_image)
attachment_category = "posted_photo"
elif attachment["type"] == "posted_photo":
top_html += _photo(attachment[attachment["type"]], big_image)
elif attachment["type"] == "photos_list":
# It seems like photos_list always duplicates photo attachments
attachment_category = "posted_photo"
elif attachment["type"] == "audio":
info = attachment[attachment["type"]]
bottom_html += _block(
_em("Аудиозапись: " + _vk_link(
"search?" +
urlencode({
"c[q]": info["artist"] + " - " + info["title"],
"c[section]": "audio"
}), "{} - {} ({})".format(info["artist"], info["title"],
_duration(info["duration"])))))
elif attachment["type"] == "doc":
info = attachment[attachment["type"]]
bottom_html += _block(_em("Документ: {}".format(info["title"])))
elif attachment["type"] == "video":
info = attachment[attachment["type"]]
video_block = _block(
_em("{} ({})".format(info["title"],
_duration(info["duration"]))))
# Restricted videos may appear without image
if "photo_320" in info:
video_block = _block(_image(info["photo_320"]) + video_block)
top_html += video_block
categories.add(_CATEGORY_ATTACHMENT + "video/" + str(info["id"]))
elif attachment["type"] == "note":
top_html += _block(
_em("Заметка: {}".format(
attachment[attachment["type"]]["title"])))
elif attachment["type"] == "page":
top_html += _block(
_em("Страница: {}".format(
attachment[attachment["type"]]["title"])))
elif attachment["type"] == "poll":
top_html += _block(
_em("Опрос: {}".format(
attachment[attachment["type"]]["question"])))
else:
unknown_attachments.add(attachment["type"])
categories.add(_CATEGORY_TYPE + attachment_category)
if unknown_attachments:
LOG.error("Got a post with unknown attachment type (%s):\n%s",
", ".join(unknown_attachments), pprint.pformat(item))
html = top_html + main_html + bottom_html
if "copy_owner_id" in item and "copy_post_id" in item:
html = _block(
_em(
_link(_get_user_url(item["copy_owner_id"]), users[
item["copy_owner_id"]]["name"])) + " пишет:") + html
if "copy_text" in item:
html = _quote_block(item["copy_text"], html)
categories.add(_CATEGORY_TYPE_REPOST)
return {
"title": user["name"] + ": запись на стене",
"text": html,
"url": _vk_url("wall", user["id"], item["post_id"]),
"categories": categories,
}
def _vk_link(target, html):
"""Renders a VK link."""
return _link(_vk_url(target), html)