def comment_to_object(self, comment, photo_id):
"""Convert a Flickr comment json object to an ActivityStreams comment.
Args:
comment: dict, the comment object from Flickr
photo_id: string, the Flickr ID of the photo that this comment belongs to
Returns:
dict, an ActivityStreams object
"""
obj = {
'objectType': 'comment',
'url': comment.get('permalink'),
'id': self.tag_uri(comment.get('id')),
'inReplyTo': [{'id': self.tag_uri(photo_id)}],
'content': comment.get('_content', ''),
'published': util.maybe_timestamp_to_rfc3339(comment.get('datecreate')),
'updated': util.maybe_timestamp_to_rfc3339(comment.get('datecreate')),
'author': {
'objectType': 'person',
'displayName': comment.get('realname') or comment.get('authorname'),
'username': comment.get('authorname'),
'id': self.tag_uri(comment.get('author')),
'url': self.user_url(comment.get('path_alias') or comment.get('author')),
'image': {
'url': self.get_user_image(comment.get('iconfarm'),
comment.get('iconserver'),
comment.get('author')),
},
}
}
self.postprocess_object(obj)
return obj
def comment_to_object(self, comment, photo_id):
"""Convert a Flickr comment json object to an ActivityStreams comment.
Args:
comment: dict, the comment object from Flickr
photo_id: string, the Flickr ID of the photo that this comment belongs to
Returns:
dict, an ActivityStreams object
"""
obj = {
'objectType': 'comment',
'url': comment.get('permalink'),
'id': self.tag_uri(comment.get('id')),
'inReplyTo': [{'id': self.tag_uri(photo_id)}],
'content': comment.get('_content', ''),
'published': util.maybe_timestamp_to_rfc3339(comment.get('datecreate')),
'updated': util.maybe_timestamp_to_rfc3339(comment.get('datecreate')),
'author': {
'objectType': 'person',
'displayName': comment.get('realname') or comment.get('authorname'),
'username': comment.get('authorname'),
'id': self.tag_uri(comment.get('author')),
'url': self.user_url(comment.get('path_alias') or comment.get('author')),
'image': {
'url': self.get_user_image(comment.get('iconfarm'),
comment.get('iconserver'),
comment.get('author')),
},
}
}
self.postprocess_object(obj)
return obj
def like_to_object(self, person, photo_activity):
"""Convert a Flickr favorite into an ActivityStreams like tag.
Args:
person: dict, the person object from Flickr
photo_activity: dict, the ActivityStreams object representing
the photo this like belongs to
Returns:
dict, an ActivityStreams object
"""
return {
'author': {
'objectType': 'person',
'displayName': person.get('realname') or person.get('username'),
'username': person.get('username'),
'id': self.tag_uri(person.get('nsid')),
'image': {
'url': self.get_user_image(person.get('iconfarm'),
person.get('iconserver'),
person.get('nsid')),
},
},
'created': util.maybe_timestamp_to_rfc3339(photo_activity.get('favedate')),
'url': u'{}#liked-by-{}'.format(
photo_activity.get('url'), person.get('nsid')),
'object': {'url': photo_activity.get('url')},
'id': self.tag_uri(u'{}_liked_by_{}'.format(
photo_activity.get('flickr_id'), person.get('nsid'))),
'objectType': 'activity',
'verb': 'like',
}
def comment_to_object(self, comment, media_id, media_url):
"""Converts a comment to an object.
Args:
comment: JSON object retrieved from the Instagram API
media_id: string
media_url: string
Returns:
an ActivityStreams object dict, ready to be JSON-encoded
"""
return self.postprocess_object({
'objectType':
'comment',
'id':
self.tag_uri(comment.get('id')),
'inReplyTo': [{
'id': self.tag_uri(media_id)
}],
'url':
'%s#comment-%s' %
(media_url, comment.get('id')) if media_url else None,
# TODO: add PST time zone
'published':
util.maybe_timestamp_to_rfc3339(comment.get('created_time')),
'content':
comment.get('text'),
'author':
self.user_to_actor(comment.get('from')),
'to': [{
'objectType': 'group',
'alias': '@public'
}],
})
def like_to_object(self, person, photo_activity):
"""Convert a Flickr favorite into an ActivityStreams like tag.
Args:
person: dict, the person object from Flickr
photo_activity: dict, the ActivityStreams object representing
the photo this like belongs to
Returns:
dict, an ActivityStreams object
"""
return {
'author': {
'objectType': 'person',
'displayName': person.get('realname') or person.get('username'),
'username': person.get('username'),
'id': self.tag_uri(person.get('nsid')),
'image': {
'url': self.get_user_image(person.get('iconfarm'),
person.get('iconserver'),
person.get('nsid')),
},
},
'created': util.maybe_timestamp_to_rfc3339(photo_activity.get('favedate')),
'url': u'{}#liked-by-{}'.format(
photo_activity.get('url'), person.get('nsid')),
'object': {'url': photo_activity.get('url')},
'id': self.tag_uri(u'{}_liked_by_{}'.format(
photo_activity.get('flickr_id'), person.get('nsid'))),
'objectType': 'activity',
'verb': 'like',
}
def comment_to_object(self, comment, media_id, media_url):
"""Converts a comment to an object.
Args:
comment: JSON object retrieved from the Instagram API
media_id: string
media_url: string
Returns:
an ActivityStreams object dict, ready to be JSON-encoded
"""
return self.postprocess_object({
'objectType': 'comment',
'id': self.tag_uri(comment.get('id')),
'inReplyTo': [{'id': self.tag_uri(media_id)}],
'url': '%s#comment-%s' % (media_url, comment.get('id')) if media_url else None,
# TODO: add PST time zone
'published': util.maybe_timestamp_to_rfc3339(comment.get('created_time')),
'content': comment.get('text'),
'author': self.user_to_actor(comment.get('from')),
'to': [{'objectType': 'group', 'alias': '@public'}],
})
def media_to_object(self, media):
"""Converts a media to an object.
Args:
media: JSON object retrieved from the Instagram API
Returns:
an ActivityStreams object dict, ready to be JSON-encoded
"""
id = media.get('id')
user = media.get('user', {})
content = xml.sax.saxutils.escape(
media.get('caption', {}).get('text', ''))
object = {
'id':
self.tag_uri(id),
# TODO: detect videos. (the type field is in the JSON respose but not
# propagated into the Media object.)
'objectType':
OBJECT_TYPES.get(media.get('type', 'image'), 'photo'),
'published':
util.maybe_timestamp_to_rfc3339(media.get('created_time')),
'author':
self.user_to_actor(user),
'content':
content,
'url':
media.get('link'),
'to': [{
'objectType':
'group',
'alias':
'@private' if user.get('is_private') else '@public',
}],
'attachments': [{
'objectType':
'video' if 'videos' in media else 'image',
'url':
media.get('link'),
# ActivityStreams 2.0 allows image to be a JSON array.
# http://jasnell.github.io/w3c-socialwg-activitystreams/activitystreams2.html#link
'image':
sorted(
media.get('images', {}).values(),
# sort by size, descending, since atom.py
# uses the first image in the list.
key=operator.itemgetter('width'),
reverse=True),
# video object defined in
# http://activitystrea.ms/head/activity-schema.html#rfc.section.4.18
'stream':
sorted(media.get('videos', {}).values(),
key=operator.itemgetter('width'),
reverse=True),
}],
# comments go in the replies field, according to the "Responses for
# Activity Streams" extension spec:
# http://activitystrea.ms/specs/json/replies/1.0/
'replies': {
'items': [
self.comment_to_object(c, id, media.get('link'))
for c in media.get('comments', {}).get('data', [])
],
'totalItems':
media.get('comments', {}).get('count'),
},
'tags': [
{
'objectType': 'hashtag',
'id': self.tag_uri(tag),
'displayName': tag,
# TODO: url
} for tag in media.get('tags', [])
] + [
self.user_to_actor(u.get('user'))
for u in media.get('users_in_photo', [])
] + [
self.like_to_object(u, id, media.get('link'))
for u in media.get('likes', {}).get('data', [])
] + self._mention_tags_from_content(content)
}
# alt text
# https://instagram-press.com/blog/2018/11/28/creating-a-more-accessible-instagram/
alt = media.get('accessibility_caption')
if alt and not any(
alt.startswith(prefix) for prefix in AUTO_ALT_TEXT_PREFIXES):
for att in object['attachments']:
for img in att.get('image', []):
img['displayName'] = alt
for version in ('standard_resolution', 'low_resolution', 'thumbnail'):
image = media.get('images', {}).get(version)
if image:
object['image'] = {'url': image.get('url')}
break
for version in ('standard_resolution', 'low_resolution',
'low_bandwidth'):
video = media.get('videos', {}).get(version)
if video:
object['stream'] = {'url': video.get('url')}
break
# http://instagram.com/developer/endpoints/locations/
if 'location' in media:
media_loc = media.get('location', {})
object['location'] = {
'id':
self.tag_uri(media_loc.get('id')),
'displayName':
media_loc.get('name'),
'latitude':
media_loc.get('point', {}).get('latitude'),
'longitude':
media_loc.get('point', {}).get('longitude'),
'address': {
'formatted': media_loc.get('street_address')
},
'url': (media_loc.get('id')
and 'https://instagram.com/explore/locations/%s/' %
media_loc.get('id')),
}
return self.postprocess_object(object)
def photo_to_activity(self, photo):
"""Convert a Flickr photo to an ActivityStreams object. Takes either
data in the expanded form returned by flickr.photos.getInfo or the
abbreviated form returned by flickr.people.getPhotos.
Args:
photo: dict response from Flickr
Returns:
dict, an ActivityStreams object
"""
owner = photo.get('owner')
if isinstance(owner, dict):
owner_id = owner.get('nsid')
path_alias = owner.get('path_alias')
else:
owner_id = owner
path_alias = photo.get('pathalias')
created = photo.get('dates', {}).get('taken') or photo.get('datetaken')
published = util.maybe_timestamp_to_rfc3339(
photo.get('dates', {}).get('posted') or photo.get('dateupload'))
# TODO replace owner_id with path_alias?
photo_permalink = self.photo_url(path_alias or owner_id, photo.get('id'))
title = photo.get('title')
if isinstance(title, dict):
title = title.get('_content', '')
public = (photo.get('visibility') or photo).get('ispublic')
activity = {
'id': self.tag_uri(photo.get('id')),
'flickr_id': photo.get('id'),
'url': photo_permalink,
'actor': {
'numeric_id': owner_id,
},
'object': {
'displayName': title,
'url': photo_permalink,
'id': self.tag_uri(photo.get('id')),
'image': {
'url': u'https://farm{}.staticflickr.com/{}/{}_{}_{}.jpg'.format(
photo.get('farm'), photo.get('server'),
photo.get('id'), photo.get('secret'), 'b'),
},
'content': '\n'.join((
title, photo.get('description', {}).get('_content', ''))),
'objectType': 'photo',
'created': created,
'published': published,
'to': [{'objectType': 'group',
'alias': '@public' if public else '@private'}],
},
'verb': 'post',
'created': created,
'published': published,
}
if isinstance(owner, dict):
activity['object']['author'] = {
'objectType': 'person',
'displayName': owner.get('realname') or owner.get('username'),
'username': owner.get('username'),
'id': self.tag_uri(owner.get('username')),
'image': {
'url': self.get_user_image(owner.get('iconfarm'),
owner.get('iconserver'),
owner.get('nsid')),
},
}
if isinstance(photo.get('tags'), dict):
activity['object']['tags'] = [{
'objectType': 'hashtag',
'id': self.tag_uri(tag.get('id')),
'url': u'https://www.flickr.com/search?tags={}'.format(
tag.get('_content')),
'displayName': tag.get('raw'),
} for tag in photo.get('tags', {}).get('tag', [])]
elif isinstance(photo.get('tags'), basestring):
activity['object']['tags'] = [{
'objectType': 'hashtag',
'url': u'https://www.flickr.com/search?tags={}'.format(
tag.strip()),
'displayName': tag.strip(),
} for tag in photo.get('tags').split(' ') if tag.strip()]
self.postprocess_activity(activity)
return activity
def photo_to_activity(self, photo):
"""Convert a Flickr photo to an ActivityStreams object. Takes either
data in the expanded form returned by flickr.photos.getInfo or the
abbreviated form returned by flickr.people.getPhotos.
Args:
photo: dict response from Flickr
Returns:
dict, an ActivityStreams object
"""
owner = photo.get('owner')
if isinstance(owner, dict):
owner_id = owner.get('nsid')
path_alias = owner.get('path_alias')
else:
owner_id = owner
path_alias = photo.get('pathalias')
created = photo.get('dates', {}).get('taken') or photo.get('datetaken')
published = util.maybe_timestamp_to_rfc3339(
photo.get('dates', {}).get('posted') or photo.get('dateupload'))
# TODO replace owner_id with path_alias?
photo_permalink = self.photo_url(path_alias or owner_id,
photo.get('id'))
title = photo.get('title')
if isinstance(title, dict):
title = title.get('_content', '')
public = (photo.get('visibility') or photo).get('ispublic')
activity = {
'id': self.tag_uri(photo.get('id')),
'flickr_id': photo.get('id'),
'url': photo_permalink,
'actor': {
'numeric_id': owner_id,
},
'object': {
'displayName':
title,
'url':
photo_permalink,
'id':
self.tag_uri(photo.get('id')),
'image': {
'url':
u'https://farm{}.staticflickr.com/{}/{}_{}_{}.jpg'.format(
photo.get('farm'), photo.get('server'),
photo.get('id'), photo.get('secret'), 'b'),
},
'content':
'\n'.join((title, photo.get('description',
{}).get('_content', ''))),
'objectType':
'photo',
'created':
created,
'published':
published,
'to': [{
'objectType': 'group',
'alias': '@public' if public else '@private'
}],
},
'verb': 'post',
'created': created,
'published': published,
}
if isinstance(owner, dict):
username = owner.get('username')
# Flickr API is evidently inconsistent in what it puts in realname vs
# username vs path_alias. if username has spaces, it's probably actually
# real name, so use path alias instead.
# https://github.com/snarfed/bridgy/issues/687
# https://www.flickr.com/groups/51035612836@N01/discuss/72157625945600254
if not username or ' ' in username:
username = owner.get('path_alias')
activity['object']['author'] = {
'objectType': 'person',
'displayName': owner.get('realname') or username,
'username': username,
'id': self.tag_uri(username),
'image': {
'url':
self.get_user_image(owner.get('iconfarm'),
owner.get('iconserver'),
owner.get('nsid')),
},
}
if isinstance(photo.get('tags'), dict):
activity['object']['tags'] = [{
'objectType':
'hashtag',
'id':
self.tag_uri(tag.get('id')),
'url':
u'https://www.flickr.com/search?tags={}'.format(
tag.get('_content')),
'displayName':
tag.get('raw'),
} for tag in photo.get('tags', {}).get('tag', [])]
elif isinstance(photo.get('tags'), basestring):
activity['object']['tags'] = [{
'objectType':
'hashtag',
'url':
u'https://www.flickr.com/search?tags={}'.format(tag.strip()),
'displayName':
tag.strip(),
} for tag in photo.get('tags').split(' ') if tag.strip()]
# location is represented differently in a list of photos vs a
# single photo info
lat = photo.get('latitude') or photo.get('location',
{}).get('latitude')
lng = photo.get('longitude') or photo.get('location',
{}).get('longitude')
if lat and lng and float(lat) != 0 and float(lng) != 0:
activity['object']['location'] = {
'objectType': 'place',
'latitude': float(lat),
'longitude': float(lng),
}
self.postprocess_object(activity['object'])
self.postprocess_activity(activity)
return activity
def media_to_object(self, media):
"""Converts a media to an object.
Args:
media: JSON object retrieved from the Instagram API
Returns:
an ActivityStreams object dict, ready to be JSON-encoded
"""
id = media.get('id')
object = {
'id': self.tag_uri(id),
# TODO: detect videos. (the type field is in the JSON respose but not
# propagated into the Media object.)
'objectType': OBJECT_TYPES.get(media.get('type', 'image'), 'photo'),
'published': util.maybe_timestamp_to_rfc3339(media.get('created_time')),
'author': self.user_to_actor(media.get('user')),
'content': xml.sax.saxutils.escape(
media.get('caption', {}).get('text', '')),
'url': media.get('link'),
'to': [{'objectType': 'group', 'alias': '@public'}],
'attachments': [{
'objectType': 'video' if 'videos' in media else 'image',
# ActivityStreams 2.0 allows image to be a JSON array.
# http://jasnell.github.io/w3c-socialwg-activitystreams/activitystreams2.html#link
'image': sorted(
media.get('images', {}).values(),
# sort by size, descending, since atom.py
# uses the first image in the list.
key=operator.itemgetter('width'), reverse=True),
# video object defined in
# http://activitystrea.ms/head/activity-schema.html#rfc.section.4.18
'stream': sorted(
media.get('videos', {}).values(),
key=operator.itemgetter('width'), reverse=True),
}],
# comments go in the replies field, according to the "Responses for
# Activity Streams" extension spec:
# http://activitystrea.ms/specs/json/replies/1.0/
'replies': {
'items': [self.comment_to_object(c, id, media.get('link'))
for c in media.get('comments', {}).get('data', [])],
'totalItems': media.get('comments', {}).get('count'),
},
'tags': [{
'objectType': 'hashtag',
'id': self.tag_uri(tag),
'displayName': tag,
# TODO: url
} for tag in media.get('tags', [])] +
[self.user_to_actor(user.get('user'))
for user in media.get('users_in_photo', [])] +
[self.like_to_object(user, id, media.get('link'))
for user in media.get('likes', {}).get('data', [])],
}
for version in ('standard_resolution', 'low_resolution', 'thumbnail'):
image = media.get('images').get(version)
if image:
object['image'] = {'url': image.get('url')}
break
for version in ('standard_resolution', 'low_resolution', 'low_bandwidth'):
video = media.get('videos', {}).get(version)
if video:
object['stream'] = {'url': video.get('url')}
break
# http://instagram.com/developer/endpoints/locations/
if 'location' in media:
media_loc = media.get('location', {})
object['location'] = {
'id': media_loc.get('id'),
'displayName': media_loc.get('name'),
'latitude': media_loc.get('point', {}).get('latitude'),
'longitude': media_loc.get('point', {}).get('longitude'),
'address': {'formatted': media_loc.get('street_address')},
'url': (media_loc.get('id')
and 'https://instagram.com/explore/locations/%s/'
% media_loc.get('id')),
}
return self.postprocess_object(object)
def photo_to_activity(self, photo):
"""Convert a Flickr photo to an ActivityStreams object. Takes either
data in the expanded form returned by flickr.photos.getInfo or the
abbreviated form returned by flickr.people.getPhotos.
Args:
photo: dict response from Flickr
Returns:
dict, an ActivityStreams object
"""
owner = photo.get('owner')
owner_id = owner.get('nsid') if isinstance(owner, dict) else owner
created = photo.get('dates', {}).get('taken') or photo.get('datetaken')
published = util.maybe_timestamp_to_rfc3339(
photo.get('dates', {}).get('posted') or photo.get('dateupload'))
photo_permalink = 'https://www.flickr.com/photos/{}/{}/'.format(
owner_id, photo.get('id'))
title = photo.get('title')
if isinstance(title, dict):
title = title.get('_content', '')
public = (photo.get('visibility') or photo).get('ispublic')
activity = {
'id': self.tag_uri(photo.get('id')),
'flickr_id': photo.get('id'),
'url': photo_permalink,
'actor': {
'numeric_id': owner_id,
},
'object': {
'displayName':
title,
'url':
photo_permalink,
'id':
self.tag_uri(photo.get('id')),
'image': {
'url':
'https://farm{}.staticflickr.com/{}/{}_{}_{}.jpg'.format(
photo.get('farm'), photo.get('server'),
photo.get('id'), photo.get('secret'), 'b'),
},
'content':
'\n'.join((title, photo.get('description',
{}).get('_content', ''))),
'objectType':
'photo',
'created':
created,
'published':
published,
'to': [{
'objectType': 'group',
'alias': '@public' if public else '@private'
}],
},
'verb': 'post',
'created': created,
'published': published,
}
if isinstance(owner, dict):
activity['object']['author'] = {
'objectType': 'person',
'displayName': owner.get('realname') or owner.get('username'),
'username': owner.get('username'),
'id': self.tag_uri(owner.get('username')),
'image': {
'url':
self.get_user_image(owner.get('iconfarm'),
owner.get('iconserver'),
owner.get('nsid')),
},
}
if isinstance(photo.get('tags'), dict):
activity['object']['tags'] = [{
'objectType':
'hashtag',
'id':
self.tag_uri(tag.get('id')),
'url':
'https://www.flickr.com/search?tags={}'.format(
tag.get('_content')),
'displayName':
tag.get('raw'),
} for tag in photo.get('tags', {}).get('tag', [])]
elif isinstance(photo.get('tags'), basestring):
activity['object']['tags'] = [{
'objectType':
'hashtag',
'url':
'https://www.flickr.com/search?tags={}'.format(tag.strip()),
'displayName':
tag.strip(),
} for tag in photo.get('tags').split(' ') if tag.strip()]
self.postprocess_activity(activity)
return activity
def photo_to_activity(self, photo):
"""Convert a Flickr photo to an ActivityStreams object. Takes either
data in the expanded form returned by flickr.photos.getInfo or the
abbreviated form returned by flickr.people.getPhotos.
Args:
photo: dict response from Flickr
Returns:
dict, an ActivityStreams object
"""
owner = photo.get('owner')
if isinstance(owner, dict):
owner_id = owner.get('nsid')
path_alias = owner.get('path_alias')
else:
owner_id = owner
path_alias = photo.get('pathalias')
created = photo.get('dates', {}).get('taken') or photo.get('datetaken')
published = util.maybe_timestamp_to_rfc3339(
photo.get('dates', {}).get('posted') or photo.get('dateupload'))
# TODO replace owner_id with path_alias?
photo_permalink = self.photo_url(path_alias or owner_id, photo.get('id'))
title = photo.get('title')
if isinstance(title, dict):
title = title.get('_content', '')
public = (photo.get('visibility') or photo).get('ispublic')
activity = {
'id': self.tag_uri(photo.get('id')),
'flickr_id': photo.get('id'),
'url': photo_permalink,
'actor': {
'numeric_id': owner_id,
},
'object': {
'displayName': title,
'url': photo_permalink,
'id': self.tag_uri(photo.get('id')),
'image': {
'url': 'https://farm{}.staticflickr.com/{}/{}_{}_{}.jpg'.format(
photo.get('farm'), photo.get('server'),
photo.get('id'), photo.get('secret'), 'b'),
},
'content': '\n'.join((
title, photo.get('description', {}).get('_content', ''))),
'objectType': 'photo',
'created': created,
'published': published,
'to': [{'objectType': 'group',
'alias': '@public' if public else '@private'}],
},
'verb': 'post',
'created': created,
'published': published,
}
if isinstance(owner, dict):
username = owner.get('username')
# Flickr API is evidently inconsistent in what it puts in realname vs
# username vs path_alias. if username has spaces, it's probably actually
# real name, so use path alias instead.
# https://github.com/snarfed/bridgy/issues/687
# https://www.flickr.com/groups/51035612836@N01/discuss/72157625945600254
if not username or ' ' in username:
username = owner.get('path_alias')
activity['object']['author'] = {
'objectType': 'person',
'displayName': owner.get('realname') or username,
'username': username,
'id': self.tag_uri(username),
'image': {
'url': self.get_user_image(owner.get('iconfarm'),
owner.get('iconserver'),
owner.get('nsid')),
},
}
if isinstance(photo.get('tags'), dict):
activity['object']['tags'] = [{
'objectType': 'hashtag',
'id': self.tag_uri(tag.get('id')),
'url': 'https://www.flickr.com/search?tags={}'.format(
tag.get('_content')),
'displayName': tag.get('raw'),
} for tag in photo.get('tags', {}).get('tag', [])]
elif isinstance(photo.get('tags'), basestring):
activity['object']['tags'] = [{
'objectType': 'hashtag',
'url': 'https://www.flickr.com/search?tags={}'.format(
tag.strip()),
'displayName': tag.strip(),
} for tag in photo.get('tags').split(' ') if tag.strip()]
# location is represented differently in a list of photos vs a
# single photo info
lat = photo.get('latitude') or photo.get('location', {}).get('latitude')
lng = photo.get('longitude') or photo.get('location', {}).get('longitude')
if lat and lng and float(lat) != 0 and float(lng) != 0:
activity['object']['location'] = {
'objectType': 'place',
'latitude': float(lat),
'longitude': float(lng),
}
self.postprocess_object(activity['object'])
self.postprocess_activity(activity)
return activity
