Page MenuHomePhorge
Files F85827587

No OneTemporary
Actions

Authored By
Unknown
Size
112 KB
Referenced Files
None
Subscribers
None

View Options

diff --git a/changelog.d/paradedb-search.add b/changelog.d/paradedb-search.add
index 00a6e2361..f72ebca14 100644
--- a/changelog.d/paradedb-search.add
+++ b/changelog.d/paradedb-search.add
@@ -1 +1 @@
-Add ParadeDB (pg_search) as an optional external search backend.
+Add ParadeDB (pg_search) as an optional external search backend, indexing status content, content warnings, and attachment descriptions.
diff --git a/config/config.exs b/config/config.exs
index 2e405413c..3c308f093 100644
--- a/config/config.exs
+++ b/config/config.exs
@@ -1,985 +1,985 @@
# .i;;;;i.
# iYcviii;vXY:
# .YXi .i1c.
# .YC. . in7.
# .vc. ...... ;1c.
# i7, .. .;1;
# i7, .. ... .Y1i
# ,7v .6MMM@; .YX,
# .7;. ..IMMMMMM1 :t7.
# .;Y. ;$MMMMMM9. :tc.
# vY. .. .nMMM@MMU. ;1v.
# i7i ... .#MM@M@C. .....:71i
# it: .... $MMM@9;.,i;;;i,;tti
# :t7. ..... 0MMMWv.,iii:::,,;St.
# .nC. ..... IMMMQ..,::::::,.,czX.
# .ct: ....... .ZMMMI..,:::::::,,:76Y.
# c2: ......,i..Y$M@t..:::::::,,..inZY
# vov ......:ii..c$MBc..,,,,,,,,,,..iI9i
# i9Y ......iii:..7@MA,..,,,,,,,,,....;AA:
# iIS. ......:ii::..;@MI....,............;Ez.
# .I9. ......:i::::...8M1..................C0z.
# .z9; ......:i::::,.. .i:...................zWX.
# vbv ......,i::::,,. ................. :AQY
# c6Y. .,...,::::,,..:t0@@QY. ................ :8bi
# :6S. ..,,...,:::,,,..EMMMMMMI. ............... .;bZ,
# :6o, .,,,,..:::,,,..i#MMMMMM#v................. YW2.
# .n8i ..,,,,,,,::,,,,.. tMMMMM@C:.................. .1Wn
# 7Uc. .:::,,,,,::,,,,.. i1t;,..................... .UEi
# 7C...::::::::::::,,,,.. .................... vSi.
# ;1;...,,::::::,......... .................. Yz:
# v97,......... .voC.
# izAotX7777777777777777777777777777777777777777Y7n92:
# .;CoIIIIIUAA666666699999ZZZZZZZZZZZZZZZZZZZZ6ov.
#
# !!! ATTENTION !!!
# DO NOT EDIT THIS FILE! THIS FILE CONTAINS THE DEFAULT VALUES FOR THE CON-
# FIGURATION! EDIT YOUR SECRET FILE (either prod.secret.exs, dev.secret.exs).
#
# This file is responsible for configuring your application
# and its dependencies with the aid of the Config module.
#
# This configuration file is loaded before any dependency and
# is restricted to this project.
import Config
# General application configuration
config :pleroma, ecto_repos: [Pleroma.Repo]
config :pleroma, Pleroma.Repo,
telemetry_event: [Pleroma.Repo.Instrumenter],
migration_lock: :pg_advisory_lock
config :pleroma, Pleroma.Captcha,
enabled: true,
seconds_valid: 300,
method: Pleroma.Captcha.Native
config :pleroma, Pleroma.Captcha.Kocaptcha, endpoint: "https://captcha.kotobank.ch"
# Upload configuration
config :pleroma, Pleroma.Upload,
uploader: Pleroma.Uploaders.Local,
filters: [Pleroma.Upload.Filter.Dedupe],
link_name: false,
proxy_remote: false,
filename_display_max_length: 30,
default_description: nil,
base_url: nil,
allowed_mime_types: ["image", "audio", "video"]
config :pleroma, Pleroma.Uploaders.Local, uploads: "uploads"
config :pleroma, Pleroma.Uploaders.S3,
bucket: nil,
bucket_namespace: nil,
truncated_namespace: nil,
streaming_enabled: true
config :ex_aws, :s3,
# host: "s3.wasabisys.com", # required if not Amazon AWS
access_key_id: nil,
secret_access_key: nil,
# region: "us-east-1", # may be required for Amazon AWS
scheme: "https://"
config :pleroma, Pleroma.Uploaders.IPFS,
post_gateway_url: "http://localhost:5001",
get_gateway_url: "http://localhost:8080"
config :pleroma, :emoji,
shortcode_globs: ["/emoji/custom/**/*.png"],
pack_extensions: [".png", ".gif"],
groups: [
Custom: ["/emoji/*.png", "/emoji/**/*.png"]
],
default_manifest: "https://git.pleroma.social/pleroma/emoji-index/raw/master/index.json",
shared_pack_cache_seconds_per_file: 60
config :pleroma, :uri_schemes,
valid_schemes: [
"https",
"http",
"dat",
"dweb",
"gopher",
"hyper",
"ipfs",
"ipns",
"irc",
"ircs",
"magnet",
"mailto",
"mumble",
"ssb",
"xmpp"
]
# Configures the endpoint
config :pleroma, Pleroma.Web.Endpoint,
url: [host: "localhost"],
http: [
ip: {127, 0, 0, 1}
],
protocol: "https",
secret_key_base: "aK4Abxf29xU9TTDKre9coZPUgevcVCFQJe/5xP/7Lt4BEif6idBIbjupVbOrbKxl",
live_view: [signing_salt: "U5ELgdEwTD3n1+D5s0rY0AMg8/y1STxZ3Zvsl3bWh+oBcGrYdil0rXqPMRd3Glcq"],
signing_salt: "CqaoopA2",
render_errors: [view: Pleroma.Web.ErrorView, accepts: ~w(json)],
pubsub_server: Pleroma.PubSub,
secure_cookie_flag: true,
extra_cookie_attrs: [
"SameSite=Lax"
]
# Configures Elixir's Logger
config :logger, backends: [:console]
config :logger, :console,
level: :debug,
format: "\n$time $metadata[$level] $message\n",
metadata: [:actor, :path, :type, :user]
config :logger, :ex_syslogger,
level: :debug,
ident: "pleroma",
format: "$metadata[$level] $message",
metadata: [:actor, :path, :type, :user]
config :mime, :types, %{
"application/xml" => ["xml"],
"application/xrd+xml" => ["xrd+xml"],
"application/jrd+json" => ["jrd+json"],
"application/activity+json" => ["activity+json"],
"application/ld+json" => ["activity+json"],
# Can be removed when bumping MIME past 2.0.5
# see https://akkoma.dev/AkkomaGang/akkoma/issues/657
"image/apng" => ["apng"]
}
config :tesla, adapter: Tesla.Adapter.Hackney
# Configures http settings, upstream proxy etc.
config :pleroma, :http,
proxy_url: nil,
send_user_agent: true,
user_agent: :default,
adapter: []
config :pleroma, :instance,
name: "Pleroma",
email: "example@example.com",
notify_email: "noreply@example.com",
description: "Pleroma: An efficient and flexible fediverse server",
short_description: "",
background_image: "/images/city.jpg",
instance_thumbnail: "/instance/thumbnail.jpeg",
favicon: "/favicon.png",
limit: 5_000,
description_limit: 5_000,
remote_limit: 100_000,
upload_limit: 16_000_000,
avatar_upload_limit: 2_000_000,
background_upload_limit: 4_000_000,
banner_upload_limit: 4_000_000,
poll_limits: %{
max_options: 20,
max_option_chars: 200,
min_expiration: 0,
max_expiration: 365 * 24 * 60 * 60
},
registrations_open: true,
invites_enabled: false,
account_activation_required: false,
account_approval_required: false,
federating: true,
federation_incoming_replies_max_depth: 100,
allow_relay: true,
public: true,
quarantined_instances: [],
rejected_instances: [],
static_dir: "instance/static/",
allowed_post_formats: [
"text/plain",
"text/html",
"text/markdown",
"text/bbcode",
"text/x.misskeymarkdown"
],
autofollowed_nicknames: [],
autofollowing_nicknames: [],
max_pinned_statuses: 1,
attachment_links: false,
max_report_comment_size: 1000,
report_strip_status: true,
safe_dm_mentions: false,
healthcheck: false,
remote_post_retention_days: 90,
skip_thread_containment: true,
limit_to_local_content: :unauthenticated,
user_bio_length: 5000,
user_name_length: 100,
max_account_fields: 10,
max_remote_account_fields: 20,
account_field_name_length: 512,
account_field_value_length: 2048,
registration_reason_length: 500,
external_user_synchronization: true,
extended_nickname_format: true,
cleanup_attachments: false,
multi_factor_authentication: [
totp: [
# digits 6 or 8
digits: 6,
period: 30
],
backup_codes: [
number: 5,
length: 16
]
],
show_reactions: true,
password_reset_token_validity: 60 * 60 * 24,
profile_directory: true,
admin_privileges: [
:users_read,
:users_manage_invites,
:users_manage_activation_state,
:users_manage_tags,
:users_manage_credentials,
:users_delete,
:messages_read,
:messages_delete,
:instances_delete,
:reports_manage_reports,
:moderation_log_read,
:announcements_manage_announcements,
:emoji_manage_emoji,
:statistics_read
],
moderator_privileges: [:messages_delete, :reports_manage_reports],
max_endorsed_users: 20,
birthday_required: false,
birthday_min_age: 0,
max_media_attachments: 1_000
config :pleroma, :welcome,
direct_message: [
enabled: false,
sender_nickname: nil,
message: nil
],
chat_message: [
enabled: false,
sender_nickname: nil,
message: nil
],
email: [
enabled: false,
sender: nil,
subject: "Welcome to <%= instance_name %>",
html: "Welcome to <%= instance_name %>",
text: "Welcome to <%= instance_name %>"
]
config :pleroma, :feed,
post_title: %{
max_length: 100,
omission: "..."
}
config :pleroma, :markup,
# XXX - unfortunately, inline images must be enabled by default right now, because
# of custom emoji. Issue #275 discusses defanging that somehow.
allow_inline_images: true,
allow_headings: false,
allow_tables: false,
allow_fonts: false,
scrub_policy: [
Pleroma.HTML.Scrubber.Default,
Pleroma.HTML.Transform.MediaProxy
]
config :pleroma, :frontend_configurations,
pleroma_fe: %{
alwaysShowSubjectInput: true,
background: "/images/city.jpg",
collapseMessageWithSubject: false,
disableChat: false,
greentext: false,
embeddedToS: true,
hideFilteredStatuses: false,
hideMutedPosts: false,
hidePostStats: false,
hideSitename: false,
hideUserStats: false,
loginMethod: "password",
logo: "/static/logo.svg",
logoMargin: ".1em",
logoMask: true,
minimalScopesMode: false,
noAttachmentLinks: false,
nsfwCensorImage: "",
postContentType: "text/plain",
redirectRootLogin: "/main/friends",
redirectRootNoLogin: "/main/all",
scopeCopy: true,
sidebarRight: false,
showFeaturesPanel: true,
showInstanceSpecificPanel: false,
subjectLineBehavior: "email",
theme: "pleroma-dark",
webPushNotifications: false
}
config :pleroma, :assets,
mascots: [
pleroma_fox_tan: %{
url: "/images/pleroma-fox-tan-smol.png",
mime_type: "image/png"
},
pleroma_fox_tan_shy: %{
url: "/images/pleroma-fox-tan-shy.png",
mime_type: "image/png"
}
],
default_mascot: :pleroma_fox_tan
config :pleroma, :manifest,
icons: [
%{
src: "/static/logo.svg",
sizes: "512x512",
purpose: "any",
type: "image/svg+xml"
}
],
theme_color: "#282c37",
background_color: "#191b22"
config :pleroma, :activitypub,
unfollow_blocked: true,
outgoing_blocks: true,
blockers_visible: true,
follow_handshake_timeout: 500,
note_replies_output_limit: 5,
sign_object_fetches: true,
authorized_fetch_mode: false,
client_api_enabled: false,
anonymize_reporter: false,
anonymize_reporter_local_nickname: ""
config :pleroma, :streamer,
workers: 3,
overflow_workers: 2
config :pleroma, :user, deny_follow_blocked: true
config :pleroma, :mrf_normalize_markup, scrub_policy: Pleroma.HTML.Scrubber.Default
config :pleroma, :mrf_rejectnonpublic,
allow_followersonly: false,
allow_direct: false
config :pleroma, :mrf_hellthread,
delist_threshold: 10,
reject_threshold: 20
config :pleroma, :mrf_simple,
media_removal: [],
media_nsfw: [],
federated_timeline_removal: [],
report_removal: [],
reject: [],
followers_only: [],
accept: [],
avatar_removal: [],
banner_removal: [],
reject_deletes: []
config :pleroma, :mrf_keyword,
reject: [],
federated_timeline_removal: [],
replace: []
config :pleroma, :mrf_emoji,
remove_url: [],
remove_shortcode: [],
federated_timeline_removal_url: [],
federated_timeline_removal_shortcode: []
config :pleroma, :mrf_hashtag,
sensitive: ["nsfw"],
reject: [],
federated_timeline_removal: []
config :pleroma, :mrf_subchain, match_actor: %{}
config :pleroma, :mrf_activity_expiration, days: 365
config :pleroma, :mrf_vocabulary,
accept: [],
reject: []
# threshold of 7 days
config :pleroma, :mrf_object_age,
threshold: 604_800,
actions: [:delist, :strip_followers]
config :pleroma, :mrf_nsfw_api,
url: "http://127.0.0.1:5000/",
threshold: 0.7,
mark_sensitive: true,
unlist: false,
reject: false
config :pleroma, :mrf_follow_bot, follower_nickname: nil
config :pleroma, :mrf_inline_quote, template: "<bdi>RT:</bdi> {url}"
config :pleroma, :mrf_remote_report,
reject_all: false,
reject_anonymous: true,
reject_empty_message: true
config :pleroma, :mrf_force_mention,
mention_parent: true,
mention_quoted: true
config :pleroma, :mrf_antimentionspam, user_age_limit: 30_000
config :pleroma, :rich_media,
enabled: true,
ignore_hosts: [],
ignore_tld: ["local", "localdomain", "lan"],
parsers: [
Pleroma.Web.RichMedia.Parsers.TwitterCard,
Pleroma.Web.RichMedia.Parsers.OEmbed
],
timeout: 5_000,
ttl_setters: [
Pleroma.Web.RichMedia.Parser.TTL.AwsSignedUrl,
Pleroma.Web.RichMedia.Parser.TTL.Opengraph
],
max_body: 5_000_000
config :pleroma, :media_proxy,
enabled: false,
invalidation: [
enabled: false,
provider: Pleroma.Web.MediaProxy.Invalidation.Script
],
proxy_opts: [
redirect_on_failure: false,
max_body_length: 25 * 1_048_576,
# Note: max_read_duration defaults to Pleroma.ReverseProxy.max_read_duration_default/1
max_read_duration: 30_000,
http: [
follow_redirect: true,
pool: :media
]
],
whitelist: []
config :pleroma, Pleroma.Web.MediaProxy.Invalidation.Http,
method: :purge,
headers: [],
options: []
config :pleroma, Pleroma.Web.MediaProxy.Invalidation.Script,
script_path: nil,
url_format: nil
# Note: media preview proxy depends on media proxy to be enabled
config :pleroma, :media_preview_proxy,
enabled: false,
thumbnail_max_width: 600,
thumbnail_max_height: 600,
image_quality: 85,
min_content_length: 100 * 1024
config :pleroma, :shout,
enabled: true,
limit: 5_000
config :phoenix, :format_encoders, json: Jason, "activity+json": Jason
config :phoenix, :json_library, Jason
config :phoenix, :filter_parameters, ["password", "confirm"]
config :pleroma, :gopher,
enabled: false,
ip: {0, 0, 0, 0},
port: 9999
config :pleroma, Pleroma.Web.Metadata,
providers: [
Pleroma.Web.Metadata.Providers.OpenGraph,
Pleroma.Web.Metadata.Providers.TwitterCard
],
unfurl_nsfw: false
config :pleroma, Pleroma.Web.Preload,
providers: [
Pleroma.Web.Preload.Providers.Instance
]
config :pleroma, :http_security,
enabled: true,
sts: false,
sts_max_age: 31_536_000,
ct_max_age: 2_592_000,
referrer_policy: "same-origin",
allow_unsafe_eval: false
config :cors_plug,
max_age: 86_400,
methods: ["POST", "PUT", "DELETE", "GET", "PATCH", "OPTIONS"],
expose: [
"Link",
"X-RateLimit-Reset",
"X-RateLimit-Limit",
"X-RateLimit-Remaining",
"X-Request-Id",
"Idempotency-Key"
],
credentials: true,
headers: ["Authorization", "Content-Type", "Idempotency-Key"]
config :pleroma, Pleroma.User,
restricted_nicknames: [
".well-known",
"~",
"about",
"activities",
"api",
"auth",
"check_password",
"dev",
"friend-requests",
"inbox",
"internal",
"main",
"media",
"nodeinfo",
"notice",
"oauth",
"objects",
"ostatus_subscribe",
"pleroma",
"proxy",
"push",
"registration",
"relay",
"settings",
"status",
"tag",
"user-search",
"user_exists",
"users",
"web",
"verify_credentials",
"update_credentials",
"relationships",
"search",
"confirmation_resend",
"mfa"
],
email_blacklist: []
# The Pruner :max_age must be longer than Worker :unique
# value or it cannot enforce uniqueness.
config :pleroma, Oban,
repo: Pleroma.Repo,
notifier: Oban.Notifiers.PG,
log: false,
queues: [
activity_expiration: 10,
federator_incoming: 5,
federator_outgoing: 25,
web_push: 50,
background: 20,
search_indexing: [limit: 10, paused: true],
slow: 5
],
plugins: [Oban.Plugins.Lazarus, {Oban.Plugins.Pruner, max_age: 900}],
crontab: [
{"0 0 * * 0", Pleroma.Workers.Cron.DigestEmailsWorker},
{"0 0 * * *", Pleroma.Workers.Cron.NewUsersDigestWorker},
{"*/10 * * * *", Pleroma.Workers.Cron.AppCleanupWorker}
]
config :pleroma, Pleroma.Formatter,
class: false,
rel: "ugc",
new_window: false,
truncate: false,
strip_prefix: false,
extra: true,
validate_tld: :no_scheme
config :pleroma, :ldap,
enabled: System.get_env("LDAP_ENABLED") == "true",
host: System.get_env("LDAP_HOST", "localhost"),
port: String.to_integer(System.get_env("LDAP_PORT", "389")),
ssl: System.get_env("LDAP_SSL") == "true",
sslopts: [],
tls: System.get_env("LDAP_TLS") == "true",
tlsopts: [],
base: System.get_env("LDAP_BASE", "dc=example,dc=com"),
uid: System.get_env("LDAP_UID", "cn"),
# defaults to CAStore's Mozilla roots
cacertfile: System.get_env("LDAP_CACERTFILE", nil),
mail: System.get_env("LDAP_MAIL", "mail")
oauth_consumer_strategies =
System.get_env("OAUTH_CONSUMER_STRATEGIES")
|> to_string()
|> String.split()
|> Enum.map(&hd(String.split(&1, ":")))
ueberauth_providers =
for strategy <- oauth_consumer_strategies do
strategy_module_name = "Elixir.Ueberauth.Strategy.#{String.capitalize(strategy)}"
strategy_module = String.to_atom(strategy_module_name)
{String.to_atom(strategy), {strategy_module, [callback_params: ["state"]]}}
end
config :ueberauth,
Ueberauth,
base_path: "/oauth",
providers: ueberauth_providers
config :pleroma, :auth, oauth_consumer_strategies: oauth_consumer_strategies
config :pleroma, Pleroma.Emails.Mailer, adapter: Swoosh.Adapters.Sendmail, enabled: false
config :pleroma, Pleroma.Emails.UserEmail,
logo: nil,
styling: %{
link_color: "#d8a070",
background_color: "#2C3645",
content_background_color: "#1B2635",
header_color: "#d8a070",
text_color: "#b9b9ba",
text_muted_color: "#b9b9ba"
}
config :pleroma, Pleroma.Emails.NewUsersDigestEmail, enabled: false
config :pleroma, Pleroma.PromEx,
disabled: false,
manual_metrics_start_delay: :no_delay,
drop_metrics_groups: [],
grafana: [
host: System.get_env("GRAFANA_HOST", "http://localhost:3000"),
auth_token: System.get_env("GRAFANA_TOKEN"),
upload_dashboards_on_start: false,
folder_name: "BEAM",
annotate_app_lifecycle: true
],
metrics_server: [
port: 4021,
path: "/metrics",
protocol: :http,
pool_size: 5,
cowboy_opts: [],
auth_strategy: :none
],
datasource: "Prometheus"
config :pleroma, Pleroma.ScheduledActivity,
daily_user_limit: 25,
total_user_limit: 300,
enabled: true
config :pleroma, :email_notifications,
digest: %{
active: false,
interval: 7,
inactivity_threshold: 7
}
config :pleroma, :oauth2,
token_expires_in: 3600 * 24 * 365 * 100,
issue_new_refresh_token: true,
clean_expired_tokens: false
config :pleroma, :database, rum_enabled: false
config :pleroma, :features, improved_hashtag_timeline: :auto
config :pleroma, :populate_hashtags_table, fault_rate_allowance: 0.01
config :pleroma, :delete_context_objects, fault_rate_allowance: 0.01
config :pleroma, :env, Mix.env()
config :http_signatures,
adapter: Pleroma.Signature
config :pleroma, :rate_limit,
authentication: {60_000, 15},
timeline: {500, 3},
search: [{1000, 10}, {1000, 30}],
app_account_creation: {1_800_000, 25},
oauth_app_creation: {900_000, 5},
relations_actions: {10_000, 10},
relation_id_action: {60_000, 2},
statuses_actions: {10_000, 15},
status_id_action: {60_000, 3},
password_reset: {1_800_000, 5},
account_confirmation_resend: {8_640_000, 5},
ap_routes: {60_000, 15}
config :pleroma, Pleroma.Workers.PurgeExpiredActivity, enabled: true, min_lifetime: 600
config :pleroma, Pleroma.Web.Plugs.RemoteIp,
enabled: true,
headers: ["x-forwarded-for"],
clients: [],
proxies: [],
reserved: [
"127.0.0.0/8",
"::1/128",
"fc00::/7",
"10.0.0.0/8",
"172.16.0.0/12",
"192.168.0.0/16"
]
config :pleroma, :static_fe, enabled: false
# Example of frontend configuration
# This example will make us serve the primary frontend from the
# frontends directory within your `:pleroma, :instance, static_dir`.
# e.g., instance/static/frontends/pleroma/develop/
#
# With no frontend configuration, the bundled files from the `static` directory will
# be used.
#
# config :pleroma, :frontends,
# primary: %{"name" => "pleroma-fe", "ref" => "develop"},
# admin: %{"name" => "admin-fe", "ref" => "stable"},
# available: %{...}
config :pleroma, :frontends,
available: %{
"kenoma" => %{
"name" => "kenoma",
"git" => "https://git.pleroma.social/lambadalambda/kenoma",
"build_url" =>
"https://git.pleroma.social/lambadalambda/kenoma/-/jobs/artifacts/${ref}/download?job=build",
"ref" => "master"
},
"pleroma-fe" => %{
"name" => "pleroma-fe",
"git" => "https://git.pleroma.social/pleroma/pleroma-fe",
"build_url" =>
"https://git.pleroma.social/api/packages/pleroma/generic/pleroma-fe-builds/${ref}/latest.zip",
"ref" => "develop"
},
"fedi-fe" => %{
"name" => "fedi-fe",
"git" => "https://git.pleroma.social/pleroma/fedi-fe",
"build_url" =>
"https://git.pleroma.social/pleroma/fedi-fe/-/jobs/artifacts/${ref}/download?job=build_release",
"ref" => "master",
"custom-http-headers" => [
{"service-worker-allowed", "/"}
]
},
"admin-fe" => %{
"name" => "admin-fe",
"git" => "https://git.pleroma.social/pleroma/admin-fe",
"build_url" =>
"https://git.pleroma.social/pleroma/admin-fe/-/jobs/artifacts/${ref}/download?job=build",
"ref" => "develop"
},
"soapbox" => %{
"name" => "soapbox",
"git" => "https://gitlab.com/soapbox-pub/soapbox",
"build_url" =>
"https://gitlab.com/soapbox-pub/soapbox/-/jobs/artifacts/${ref}/download?job=build-production",
"ref" => "v3.0.0-beta.1",
"build_dir" => "static"
},
"glitch-lily" => %{
"name" => "glitch-lily",
"git" => "https://lily-is.land/infra/glitch-lily",
"build_url" =>
"https://lily-is.land/infra/glitch-lily/-/jobs/artifacts/${ref}/download?job=build",
"ref" => "servant",
"build_dir" => "public"
},
"pl-fe" => %{
"name" => "pl-fe",
"git" => "https://github.com/mkljczk/pl-fe",
"build_url" => "https://pl.mkljczk.pl/pl-fe.zip",
"ref" => "develop",
"build_dir" => "."
}
}
config :pleroma, :web_cache_ttl,
activity_pub: nil,
activity_pub_question: 30_000
config :pleroma, :modules, runtime_dir: "instance/modules"
config :pleroma, configurable_from_database: false
config :pleroma, Pleroma.Repo,
parameters: [gin_fuzzy_search_limit: "500", jit: "off"],
prepare: :unnamed
config :pleroma, :connections_pool,
reclaim_multiplier: 0.1,
connection_acquisition_wait: 250,
connection_acquisition_retries: 5,
max_connections: 250,
max_idle_time: 30_000,
retry: 0,
connect_timeout: 5_000
config :pleroma, :pools,
federation: [
size: 75,
max_waiting: 20,
recv_timeout: 10_000
],
media: [
size: 75,
max_waiting: 20,
recv_timeout: 15_000
],
rich_media: [
size: 25,
max_waiting: 20,
recv_timeout: 15_000
],
upload: [
size: 25,
max_waiting: 20,
recv_timeout: 15_000
],
default: [
size: 50,
max_waiting: 2,
recv_timeout: 5_000
]
config :pleroma, :hackney_pools,
federation: [
max_connections: 50,
timeout: 10_000
],
media: [
max_connections: 50,
timeout: 15_000
],
rich_media: [
max_connections: 50,
timeout: 15_000
],
upload: [
max_connections: 25,
timeout: 15_000
]
config :pleroma, :majic_pool, size: 2
private_instance? = :if_instance_is_private
config :pleroma, :restrict_unauthenticated,
timelines: %{local: private_instance?, federated: private_instance?},
profiles: %{local: private_instance?, remote: private_instance?},
activities: %{local: private_instance?, remote: private_instance?}
config :pleroma, Pleroma.Web.ApiSpec.CastAndValidate, strict: false
config :pleroma, :mrf,
policies: [
Pleroma.Web.ActivityPub.MRF.ObjectAgePolicy,
Pleroma.Web.ActivityPub.MRF.TagPolicy,
Pleroma.Web.ActivityPub.MRF.InlineQuotePolicy
],
transparency: true,
transparency_exclusions: []
config :tzdata, :http_client, Pleroma.HTTP.Tzdata
config :ex_aws, http_client: Pleroma.HTTP.ExAws
config :web_push_encryption, http_client: Pleroma.HTTP.WebPush
config :pleroma, :instances_favicons, enabled: false
config :floki, :html_parser, Floki.HTMLParser.FastHtml
config :pleroma, Pleroma.Web.Auth.Authenticator, Pleroma.Web.Auth.PleromaAuthenticator
config :pleroma, Pleroma.User.Backup,
purge_after_days: 30,
limit_days: 7,
dir: nil,
process_chunk_size: 100,
timeout: :timer.minutes(30)
config :pleroma, ConcurrentLimiter, [
{Pleroma.Search, [max_running: 30, max_waiting: 50]}
]
config :pleroma, Pleroma.Web.WebFinger, domain: nil, update_nickname_on_user_fetch: true
config :pleroma, Pleroma.Search, module: Pleroma.Search.DatabaseSearch
config :pleroma, Pleroma.Search.Meilisearch,
url: "http://127.0.0.1:7700/",
private_key: nil,
initial_indexing_chunk_size: 100_000
config :pleroma, Pleroma.Search.ParadeDB,
url: nil,
table: "pleroma_search_documents",
fuzzy_distance: 0
config :pleroma, Pleroma.Search.ParadeDB.Repo,
- pool_size: 10,
+ pool_size: 2,
prepare: :unnamed
config :pleroma, Pleroma.Application,
background_migrators: true,
internal_fetch: true,
load_custom_modules: true,
max_restarts: 3,
streamer_registry: true
config :pleroma, Pleroma.Uploaders.Uploader, timeout: 30_000
config :pleroma, Pleroma.Search.QdrantSearch,
qdrant_url: "http://127.0.0.1:6333/",
qdrant_api_key: "",
openai_url: "http://127.0.0.1:11345",
# The healthcheck url has to be set to nil when used with the real openai
# API, as it doesn't have a healthcheck endpoint.
openai_healthcheck_url: "http://127.0.0.1:11345/health",
openai_model: "snowflake/snowflake-arctic-embed-xs",
openai_api_key: "",
qdrant_index_configuration: %{
vectors: %{size: 384, distance: "Cosine"}
}
config :pleroma, :database_config_whitelist, [
{:pleroma},
{:cors_plug},
{:ex_aws, :s3},
{:mime},
{:prometheus, Pleroma.Web.Endpoint.MetricsExporter},
{:web_push_encryption, :vapid_details}
]
# Import environment specific config. This must remain at the bottom
# of this file so it overrides the configuration defined above.
import_config "#{Mix.env()}.exs"
diff --git a/docs/configuration/search.md b/docs/configuration/search.md
index c0d3f8462..1d5056ddf 100644
--- a/docs/configuration/search.md
+++ b/docs/configuration/search.md
@@ -1,184 +1,190 @@
# Configuring search
{! backend/administration/CLI_tasks/general_cli_task_info.include !}
## Built-in search
To use built-in search that has no external dependencies, set the search module to `Pleroma.Activity`:
> config :pleroma, Pleroma.Search, module: Pleroma.Search.DatabaseSearch
While it has no external dependencies, it has problems with performance and relevancy.
## QdrantSearch
This uses the vector search engine [Qdrant](https://qdrant.tech) to search the posts in a vector space. This needs a way to generate embeddings and uses the [OpenAI API](https://platform.openai.com/docs/guides/embeddings/what-are-embeddings). This is implemented by several project besides OpenAI itself, including the python-based fastembed-server found in `supplemental/search/fastembed-api`.
The default settings will support a setup where both the fastembed server and Qdrant run on the same system as pleroma. To use it, set the search provider and run the fastembed server, see the README in `supplemental/search/fastembed-api`:
> config :pleroma, Pleroma.Search, module: Pleroma.Search.QdrantSearch
Then, start the Qdrant server, see [here](https://qdrant.tech/documentation/quick-start/) for instructions.
You will also need to create the Qdrant index once by running `mix pleroma.search.indexer create_index`. Running `mix pleroma.search.indexer index` will retroactively index the last 100_000 activities.
### Indexing and model options
To see the available configuration options, check out the QdrantSearch section in `config/config.exs`.
The default indexing option work for the default model (`snowflake-arctic-embed-xs`). To optimize for a low memory footprint, adjust the index configuration as described in the [Qdrant docs](https://qdrant.tech/documentation/guides/optimize/). See also [this blog post](https://qdrant.tech/articles/memory-consumption/) that goes into detail.
Different embedding models will need different vector size settings. You can see a list of the models supported by the fastembed server [here](https://qdrant.github.io/fastembed/examples/Supported_Models), including their vector dimensions. These vector dimensions need to be set in the `qdrant_index_configuration`.
E.g, If you want to use `sentence-transformers/all-MiniLM-L6-v2` as a model, you will not need to adjust things, because it and `snowflake-arctic-embed-xs` are both 384 dimensional models. If you want to use `snowflake/snowflake-arctic-embed-l`, you will need to adjust the `size` parameter in the `qdrant_index_configuration` to 1024, as it has a dimension of 1024.
When using a different model, you will need do drop the index and recreate it (`mix pleroma.search.indexer drop_index` and `mix pleroma.search.indexer create_index`), as the different embeddings are not compatible with each other.
## Meilisearch
Note that it's quite a bit more memory hungry than PostgreSQL (around 4-5G for ~1.2 million
posts while idle and up to 7G while indexing initially). The disk usage for this additional index is also
around 4 gigabytes. Like [RUM](./cheatsheet.md#rum-indexing-for-full-text-search) indexes, it offers considerably
higher performance and ordering by timestamp in a reasonable amount of time.
Additionally, the search results seem to be more accurate.
Due to high memory usage, it may be best to set it up on a different machine, if running pleroma on a low-resource
computer, and use private key authentication to secure the remote search instance.
To use [meilisearch](https://www.meilisearch.com/), set the search module to `Pleroma.Search.Meilisearch`:
> config :pleroma, Pleroma.Search, module: Pleroma.Search.Meilisearch
You then need to set the address of the meilisearch instance, and optionally the private key for authentication. You might
also want to change the `initial_indexing_chunk_size` to be smaller if you're server is not very powerful, but not higher than `100_000`,
because meilisearch will refuse to process it if it's too big. However, in general you want this to be as big as possible, because meilisearch
indexes faster when it can process many posts in a single batch.
> config :pleroma, Pleroma.Search.Meilisearch,
> url: "http://127.0.0.1:7700/",
> private_key: "private key",
> initial_indexing_chunk_size: 100_000
Information about setting up meilisearch can be found in the
[official documentation](https://docs.meilisearch.com/learn/getting_started/installation.html).
You probably want to start it with `MEILI_NO_ANALYTICS=true` environment variable to disable analytics.
At least version 0.25.0 is required, but you are strongly advised to use at least 0.26.0, as it introduces
the `--enable-auto-batching` option which drastically improves performance. Without this option, the search
is hardly usable on a somewhat big instance.
### Private key authentication (optional)
To set the private key, use the `MEILI_MASTER_KEY` environment variable when starting. After setting the _master key_,
you have to get the _private key_, which is actually used for authentication.
=== "OTP"
```sh
./bin/pleroma_ctl search.meilisearch show-keys <your master key here>
```
=== "From Source"
```sh
mix pleroma.search.meilisearch show-keys <your master key here>
```
You will see a "Default Admin API Key", this is the key you actually put into your configuration file.
### Initial indexing
-After setting up the configuration, you'll want to index all of your already existing posts. Only public posts are indexed. You'll only
+After setting up the configuration, you'll want to index all of your already existing posts. Only public/unlisted posts are indexed. You'll only
have to do it one time, but it might take a while, depending on the amount of posts your instance has seen. This is also a fairly RAM
consuming process for `meilisearch`, and it will take a lot of RAM when running if you have a lot of posts (seems to be around 5G for ~1.2
million posts while idle and up to 7G while indexing initially, but your experience may be different).
The sequence of actions is as follows:
1. First, change the configuration to use `Pleroma.Search.Meilisearch` as the search backend
2. Restart your instance, at this point it can be used while the search indexing is running, though search won't return anything
3. Start the initial indexing process (as described below with `index`),
and wait until the task says it sent everything from the database to index
4. Wait until everything is actually indexed (by checking with `stats` as described below),
at this point you don't have to do anything, just wait a while.
To start the initial indexing, run the `index` command:
=== "OTP"
```sh
./bin/pleroma_ctl search.meilisearch index
```
=== "From Source"
```sh
mix pleroma.search.meilisearch index
```
This will show you the total amount of posts to index, and then show you the amount of posts indexed currently, until the numbers eventually
become the same. The posts are indexed in big batches and meilisearch will take some time to actually index them, even after you have
inserted all the posts into it. Depending on the amount of posts, this may be as long as several hours. To get information about the status
of indexing and how many posts have actually been indexed, use the `stats` command:
=== "OTP"
```sh
./bin/pleroma_ctl search.meilisearch stats
```
=== "From Source"
```sh
mix pleroma.search.meilisearch stats
```
### Clearing the index
In case you need to clear the index (for example, to re-index from scratch, if that needs to happen for some reason), you can
use the `clear` command:
=== "OTP"
```sh
./bin/pleroma_ctl search.meilisearch clear
```
=== "From Source"
```sh
mix pleroma.search.meilisearch clear
```
This will clear **all** the posts from the search index. Note, that deleted posts are also removed from index by the instance itself, so
there is no need to actually clear the whole index, unless you want **all** of it gone. That said, the index does not hold any information
that cannot be re-created from the database, it should also generally be a lot smaller than the size of your database. Still, the size
depends on the amount of text in posts.
## ParadeDB
[ParadeDB](https://www.paradedb.com/) is a Postgres extension that provides BM25 full-text search. In Pleroma, it can be used as an
external search backend by pointing Pleroma at a separate Postgres instance with the `pg_search` extension installed.
Pleroma will maintain a small `pleroma_search_documents` table in that database (via the existing search indexing queue) and run search
queries against it.
+The indexed text includes status content, content warnings, and attachment descriptions.
To use it, set the search module to `Pleroma.Search.ParadeDB` and configure the ParadeDB database URL:
+These settings must be in static/runtime config (or `PARADEDB_DATABASE_URL`) so the dedicated ParadeDB Repo is started with the application.
> config :pleroma, Pleroma.Search, module: Pleroma.Search.ParadeDB
>
> config :pleroma, Pleroma.Search.ParadeDB,
> url: System.get_env("PARADEDB_DATABASE_URL") || "postgres://postgres:postgres@127.0.0.1:5432/paradedb",
> table: "pleroma_search_documents",
> fuzzy_distance: 0
+>
+> config :pleroma, Pleroma.Search.ParadeDB.Repo,
+> pool_size: 2,
+> prepare: :unnamed
Then, create the table and BM25 index once:
=== "From Source"
```sh
mix pleroma.search.indexer create_index
```
After that, run an initial indexing pass to backfill existing posts:
=== "From Source"
```sh
mix pleroma.search.indexer index
```
Note: Like the other external search backends, only public/unlisted Notes are indexed.
Set `fuzzy_distance` to `1` or `2` to allow typo-tolerant matching (`2` is the maximum ParadeDB supports).
Indexing runs via the Oban `search_indexing` queue, so ensure that queue is enabled (not paused) in your Oban configuration.
diff --git a/lib/pleroma/search.ex b/lib/pleroma/search.ex
index 9cd2768c4..cd4907026 100644
--- a/lib/pleroma/search.ex
+++ b/lib/pleroma/search.ex
@@ -1,80 +1,98 @@
defmodule Pleroma.Search do
alias Pleroma.Activity
alias Pleroma.Object
alias Pleroma.Web.ActivityPub.Visibility
alias Pleroma.Workers.SearchIndexingWorker
@spec add_to_index(Activity.t()) :: {:ok, Oban.Job.t() | :noop} | {:error, Oban.Job.changeset()}
- def add_to_index(%Activity{id: activity_id, object: %Object{} = object} = activity) do
- with {_, true} <- {:indexable, indexable?(activity)},
- {_, "public"} <- {:visibility, Visibility.get_visibility(object)} do
+ def add_to_index(%Activity{id: activity_id, object: %Object{}} = activity) do
+ if indexable?(activity) do
SearchIndexingWorker.new(%{"op" => "add_to_index", "activity" => activity_id})
|> Oban.insert()
else
- _ -> {:ok, :noop}
+ {:ok, :noop}
end
end
def add_to_index(%Activity{id: activity_id}) do
case Activity.get_by_id_with_object(activity_id) do
%Activity{} = preloaded -> add_to_index(preloaded)
_ -> {:ok, :noop}
end
end
@spec remove_from_index(Object.t()) :: {:ok, Oban.Job.t()} | {:error, Oban.Job.changeset()}
def remove_from_index(%Pleroma.Object{id: object_id}) do
SearchIndexingWorker.new(%{"op" => "remove_from_index", "object" => object_id})
|> Oban.insert()
end
def search(query, options) do
search_module = Pleroma.Config.get([Pleroma.Search, :module])
search_module.search(options[:for_user], query, options)
end
def healthcheck_endpoints do
search_module = Pleroma.Config.get([Pleroma.Search, :module])
search_module.healthcheck_endpoints()
end
def object_to_search_data(%Object{} = object) do
data = object.data
- content_str =
- case data["content"] do
- [nil | rest] -> to_string(rest)
- str -> str
- end
-
content =
- with {:ok, scrubbed} <-
- FastSanitize.Sanitizer.scrub(content_str, Pleroma.HTML.Scrubber.SearchIndexing),
- trimmed <- String.trim(scrubbed) do
- trimmed
- end
-
- # Make sure we have a non-empty string
- if content != "" do
- {:ok, published, _} = DateTime.from_iso8601(data["published"])
+ data
+ |> search_texts()
+ |> Enum.map(&sanitize_search_text/1)
+ |> Enum.reject(&(&1 in ["", "."]))
+ |> Enum.join(" ")
+ with true <- content != "",
+ published when is_binary(published) <- data["published"],
+ {:ok, published, _} <- DateTime.from_iso8601(published) do
%{
id: object.id,
content: content,
ap: data["id"],
published: published |> DateTime.to_unix()
}
+ else
+ _ -> nil
end
end
- defp indexable?(%Activity{
- data: %{"type" => "Create"},
- object: %Object{
- data: %{"content" => content, "published" => published, "type" => "Note"}
- }
- })
- when not is_nil(content) and content not in ["", "."] and not is_nil(published),
- do: true
+ def indexable?(%Activity{
+ data: %{"type" => "Create"},
+ object: %Object{data: %{"published" => published, "type" => "Note"}} = object
+ })
+ when not is_nil(published) do
+ Visibility.get_visibility(object) in ["public", "unlisted"] and
+ not is_nil(object_to_search_data(object))
+ end
+
+ def indexable?(_), do: false
+
+ defp search_texts(data) do
+ [data["content"], data["summary"] | attachment_names(data["attachment"])]
+ end
+
+ defp attachment_names(attachments) when is_list(attachments) do
+ Enum.map(attachments, fn
+ %{"name" => name} -> name
+ _ -> nil
+ end)
+ end
+
+ defp attachment_names(_), do: []
+
+ defp sanitize_search_text([nil | rest]), do: sanitize_search_text(to_string(rest))
+
+ defp sanitize_search_text(text) when is_binary(text) do
+ case FastSanitize.Sanitizer.scrub(text, Pleroma.HTML.Scrubber.SearchIndexing) do
+ {:ok, scrubbed} -> String.trim(scrubbed)
+ _ -> ""
+ end
+ end
- defp indexable?(_), do: false
+ defp sanitize_search_text(_), do: ""
end
diff --git a/lib/pleroma/search/paradedb.ex b/lib/pleroma/search/parade_db.ex
similarity index 96%
rename from lib/pleroma/search/paradedb.ex
rename to lib/pleroma/search/parade_db.ex
index 20be5b6f9..ee00e74af 100644
--- a/lib/pleroma/search/paradedb.ex
+++ b/lib/pleroma/search/parade_db.ex
@@ -1,286 +1,287 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2022 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Search.ParadeDB do
@behaviour Pleroma.Search.SearchBackend
require Logger
require Pleroma.Constants
import Ecto.Query
import Pleroma.Search.DatabaseSearch,
only: [
maybe_fetch: 3,
maybe_restrict_author: 2,
maybe_restrict_blocked: 2,
maybe_restrict_local: 2
]
alias Pleroma.Activity
alias Pleroma.Config.Getting, as: Config
alias Pleroma.Search
@max_limit 40
@default_table "pleroma_search_documents"
defp client_impl do
Config.get([Pleroma.Search.ParadeDB, :client_impl]) ||
Pleroma.Search.ParadeDB.Client.Postgres
end
defp table do
Config.get([Pleroma.Search.ParadeDB, :table], @default_table)
end
defp fuzzy_distance do
case Config.get([Pleroma.Search.ParadeDB, :fuzzy_distance], 0) do
value when is_integer(value) and value in 0..2 ->
value
value when is_binary(value) ->
case Integer.parse(value) do
{parsed, ""} when parsed in 0..2 -> parsed
_ -> 0
end
_ ->
0
end
end
@impl true
def create_index do
with {:ok, _} <- client_impl().query("CREATE EXTENSION IF NOT EXISTS pg_search", []),
{:ok, _} <- client_impl().query(create_table_sql(table()), []),
{:ok, _} <- client_impl().query(create_object_id_index_sql(table()), []),
{:ok, _} <- client_impl().query(create_bm25_index_sql(table()), []) do
:ok
else
{:error, error} ->
Logger.error("ParadeDB create_index failed: #{inspect(error)}")
{:error, error}
end
end
@impl true
def drop_index do
- with {:ok, _} <- client_impl().query("DROP TABLE IF EXISTS #{table()}", []) do
- :ok
- else
+ case client_impl().query("DROP TABLE IF EXISTS #{table()}", []) do
+ {:ok, _} ->
+ :ok
+
{:error, error} ->
Logger.error("ParadeDB drop_index failed: #{inspect(error)}")
{:error, error}
end
end
@impl true
def healthcheck_endpoints, do: nil
@impl true
def add_to_index(activity) do
- maybe_search_data = Search.object_to_search_data(activity.object)
-
- if activity.data["type"] == "Create" and maybe_search_data do
+ with true <- Search.indexable?(activity),
+ maybe_search_data when not is_nil(maybe_search_data) <-
+ Search.object_to_search_data(activity.object) do
dumped_activity_id = dump_activity_id(activity.id)
actor_ap_id = activity.data["actor"]
published_at = DateTime.from_unix!(maybe_search_data.published)
if is_nil(dumped_activity_id) do
Logger.error("ParadeDB add_to_index failed: invalid activity id #{inspect(activity.id)}")
:ok
else
sql =
"""
INSERT INTO #{table()} (id, object_id, object_ap_id, actor_ap_id, content, published_at)
VALUES ($1, $2, $3, $4, $5, $6)
ON CONFLICT (id) DO UPDATE SET
object_id = EXCLUDED.object_id,
object_ap_id = EXCLUDED.object_ap_id,
actor_ap_id = EXCLUDED.actor_ap_id,
content = EXCLUDED.content,
published_at = EXCLUDED.published_at
"""
params = [
dumped_activity_id,
maybe_search_data.id,
maybe_search_data.ap,
actor_ap_id,
maybe_search_data.content,
published_at
]
case client_impl().query(sql, params) do
{:ok, _} ->
:ok
{:error, error} ->
Logger.error("ParadeDB add_to_index failed: #{inspect(error)}")
{:error, error}
end
end
else
- :ok
+ _ -> :ok
end
end
@impl true
def remove_from_index(object) do
sql = "DELETE FROM #{table()} WHERE object_id = $1"
case client_impl().query(sql, [object.id]) do
{:ok, _} ->
:ok
{:error, error} ->
Logger.error("ParadeDB remove_from_index failed: #{inspect(error)}")
{:error, error}
end
end
@impl true
def search(user, query, options \\ []) do
limit = options |> Keyword.get(:limit, @max_limit) |> min(@max_limit)
offset = Keyword.get(options, :offset, 0)
author = Keyword.get(options, :author)
case search_ids(query, limit, offset, author) do
{:ok, []} ->
maybe_fetch([], user, query)
{:ok, ids} ->
ids
|> fetch_activities(user, query, author)
{:error, error} ->
Logger.error("ParadeDB search failed: #{inspect(error)}")
maybe_fetch([], user, query)
end
end
defp fetch_activities(ids, user, query, author) do
dumped_ids =
ids
|> Enum.map(&dump_activity_id/1)
|> Enum.reject(&is_nil/1)
from(a in Activity, where: a.id in ^dumped_ids)
|> Activity.with_preloaded_object()
|> Activity.restrict_deactivated_users()
|> maybe_restrict_public(user)
|> maybe_restrict_local(user)
|> maybe_restrict_author(author)
|> maybe_restrict_blocked(user)
|> order_by(
[a],
fragment(
"array_position(?, ?)",
^dumped_ids,
a.id
)
)
|> Pleroma.Repo.all()
|> maybe_fetch(user, query)
end
defp maybe_restrict_public(query, %Pleroma.User{}) do
intended_recipients = [
Pleroma.Constants.as_public(),
Pleroma.Web.ActivityPub.Utils.as_local_public()
]
from(a in query,
where: fragment("?->>'type' = 'Create'", a.data),
where: fragment("? && ?", ^intended_recipients, a.recipients)
)
end
defp maybe_restrict_public(query, _user) do
from(a in query,
where: fragment("?->>'type' = 'Create'", a.data),
where: ^Pleroma.Constants.as_public() in a.recipients
)
end
defp search_ids(query, limit, offset, author) do
base_sql =
"SELECT id FROM #{table()} WHERE content &&& #{search_query_param_sql(fuzzy_distance())}"
params = [query]
{base_sql, params} =
case author do
%Pleroma.User{ap_id: ap_id} when is_binary(ap_id) ->
{base_sql <> " AND actor_ap_id = $2", params ++ [ap_id]}
_ ->
{base_sql, params}
end
limit_arg = length(params) + 1
offset_arg = length(params) + 2
sql =
base_sql <>
" ORDER BY published_at DESC, id DESC LIMIT $#{limit_arg} OFFSET $#{offset_arg}"
params = params ++ [limit, offset]
with {:ok, %{rows: rows}} <- client_impl().query(sql, params) do
{:ok, Enum.map(rows, fn [id] -> id end)}
end
end
defp search_query_param_sql(0), do: "$1"
defp search_query_param_sql(distance), do: "($1::text)::pdb.fuzzy(#{distance})"
defp create_table_sql(table) do
"""
CREATE TABLE IF NOT EXISTS #{table} (
id uuid PRIMARY KEY,
object_id bigint NOT NULL,
object_ap_id text NOT NULL,
actor_ap_id text,
content text NOT NULL,
published_at timestamptz
)
"""
end
defp create_object_id_index_sql(table) do
"CREATE INDEX IF NOT EXISTS #{table}_object_id_idx ON #{table} (object_id)"
end
defp create_bm25_index_sql(table) do
"""
CREATE INDEX IF NOT EXISTS #{table}_bm25_idx ON #{table}
USING bm25 (id, content, (actor_ap_id::pdb.literal), published_at)
WITH (key_field = 'id')
"""
end
defp dump_activity_id(<<_::binary-size(16)>> = uuid), do: uuid
defp dump_activity_id(id) when is_binary(id) do
case Ecto.UUID.dump(id) do
{:ok, <<_::binary-size(16)>> = uuid} ->
uuid
:error ->
dump_flake_activity_id(id)
end
end
defp dump_activity_id(_), do: nil
defp dump_flake_activity_id(id) do
case FlakeId.Ecto.CompatType.dump(id) do
{:ok, <<_::binary-size(16)>> = uuid} -> uuid
_ -> nil
end
rescue
_ -> nil
end
end
diff --git a/lib/pleroma/search/paradedb/client.ex b/lib/pleroma/search/parade_db/client.ex
similarity index 100%
rename from lib/pleroma/search/paradedb/client.ex
rename to lib/pleroma/search/parade_db/client.ex
diff --git a/lib/pleroma/search/paradedb/repo.ex b/lib/pleroma/search/parade_db/repo.ex
similarity index 100%
rename from lib/pleroma/search/paradedb/repo.ex
rename to lib/pleroma/search/parade_db/repo.ex
diff --git a/lib/pleroma/web/activity_pub/mrf/anti_followbot_policy.ex b/lib/pleroma/web/activity_pub/mrf/anti_followbot_policy.ex
index 8ea61aec2..d0320ce49 100644
--- a/lib/pleroma/web/activity_pub/mrf/anti_followbot_policy.ex
+++ b/lib/pleroma/web/activity_pub/mrf/anti_followbot_policy.ex
@@ -1,83 +1,87 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2022 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Web.ActivityPub.MRF.AntiFollowbotPolicy do
alias Pleroma.User
@moduledoc "Prevent followbots from following with a bit of heuristic"
@behaviour Pleroma.Web.ActivityPub.MRF.Policy
# XXX: this should become User.normalize_by_ap_id() or similar, really.
defp normalize_by_ap_id(%{"id" => id}), do: User.get_cached_by_ap_id(id)
defp normalize_by_ap_id(uri) when is_binary(uri), do: User.get_cached_by_ap_id(uri)
defp normalize_by_ap_id(_), do: nil
defp score_nickname("followbot@" <> _), do: 1.0
defp score_nickname("federationbot@" <> _), do: 1.0
defp score_nickname("federation_bot@" <> _), do: 1.0
defp score_nickname(_), do: 0.0
defp score_displayname("federation bot"), do: 1.0
defp score_displayname("federationbot"), do: 1.0
defp score_displayname("fedibot"), do: 1.0
defp score_displayname(_), do: 0.0
- defp determine_if_followbot(%User{nickname: nickname, name: displayname, actor_type: actor_type}) do
+ defp determine_if_followbot(%User{
+ nickname: nickname,
+ name: displayname,
+ actor_type: actor_type
+ }) do
# nickname will be a binary string except when following a relay
nick_score =
if is_binary(nickname) do
nickname
|> String.downcase()
|> score_nickname()
else
0.0
end
# displayname will either be a binary string or nil, if a displayname isn't set.
name_score =
if is_binary(displayname) do
displayname
|> String.downcase()
|> score_displayname()
else
0.0
end
# actor_type "Service" is a Bot account
actor_type_score =
if actor_type == "Service" do
1.0
else
0.0
end
nick_score + name_score + actor_type_score
end
defp bot_allowed?(%{"object" => target}, bot_actor) do
%User{} = user = normalize_by_ap_id(target)
User.following?(user, bot_actor)
end
@impl true
def filter(%{"type" => "Follow", "actor" => actor_id} = activity) do
%User{} = actor = normalize_by_ap_id(actor_id)
score = determine_if_followbot(actor)
if score < 0.8 || bot_allowed?(activity, actor) do
{:ok, activity}
else
{:reject, "[AntiFollowbotPolicy] Scored #{actor_id} as #{score}"}
end
end
@impl true
def filter(activity), do: {:ok, activity}
@impl true
def describe, do: {:ok, %{}}
end
diff --git a/lib/pleroma/web/activity_pub/mrf/simple_policy.ex b/lib/pleroma/web/activity_pub/mrf/simple_policy.ex
index a97e8db7b..93f0616ec 100644
--- a/lib/pleroma/web/activity_pub/mrf/simple_policy.ex
+++ b/lib/pleroma/web/activity_pub/mrf/simple_policy.ex
@@ -1,376 +1,376 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2022 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Web.ActivityPub.MRF.SimplePolicy do
@moduledoc "Filter activities depending on their origin instance"
@behaviour Pleroma.Web.ActivityPub.MRF.Policy
alias Pleroma.Config
alias Pleroma.FollowingRelationship
alias Pleroma.User
alias Pleroma.Web.ActivityPub.MRF
require Pleroma.Constants
defp check_accept(%{host: actor_host} = _actor_info, activity) do
accepts =
instance_list(:accept)
|> MRF.subdomains_regex()
cond do
accepts == [] -> {:ok, activity}
actor_host == Config.get([Pleroma.Web.Endpoint, :url, :host]) -> {:ok, activity}
MRF.subdomain_match?(accepts, actor_host) -> {:ok, activity}
true -> {:reject, "[SimplePolicy] host not in accept list"}
end
end
defp check_reject(%{host: actor_host} = _actor_info, activity) do
rejects =
instance_list(:reject)
|> MRF.subdomains_regex()
if MRF.subdomain_match?(rejects, actor_host) do
{:reject, "[SimplePolicy] host in reject list"}
else
{:ok, activity}
end
end
defp check_media_removal(
%{host: actor_host} = _actor_info,
%{"type" => type, "object" => %{"attachment" => object_attachment}} = activity
)
when length(object_attachment) > 0 and type in ["Create", "Update"] do
media_removal =
instance_list(:media_removal)
|> MRF.subdomains_regex()
activity =
if MRF.subdomain_match?(media_removal, actor_host) do
object = Map.delete(activity["object"], "attachment")
Map.put(activity, "object", object)
else
activity
end
{:ok, activity}
end
defp check_media_removal(_actor_info, activity), do: {:ok, activity}
defp check_media_nsfw(
%{host: actor_host} = _actor_info,
%{
"type" => type,
"object" => %{} = _object
} = activity
)
when type in ["Create", "Update"] do
media_nsfw =
instance_list(:media_nsfw)
|> MRF.subdomains_regex()
activity =
if MRF.subdomain_match?(media_nsfw, actor_host) do
Kernel.put_in(activity, ["object", "sensitive"], true)
else
activity
end
{:ok, activity}
end
defp check_media_nsfw(_actor_info, activity), do: {:ok, activity}
defp check_ftl_removal(%{host: actor_host} = _actor_info, activity) do
timeline_removal =
instance_list(:federated_timeline_removal)
|> MRF.subdomains_regex()
activity =
with true <- MRF.subdomain_match?(timeline_removal, actor_host),
user <- User.get_cached_by_ap_id(activity["actor"]),
true <- Pleroma.Constants.as_public() in activity["to"] do
to = List.delete(activity["to"], Pleroma.Constants.as_public()) ++ [user.follower_address]
cc = List.delete(activity["cc"], user.follower_address) ++ [Pleroma.Constants.as_public()]
activity
|> Map.put("to", to)
|> Map.put("cc", cc)
else
_ -> activity
end
{:ok, activity}
end
defp intersection(list1, list2) do
- list1 -- list1 -- list2
+ list1 -- (list1 -- list2)
end
defp check_followers_only(%{host: actor_host} = _actor_info, activity) do
followers_only =
instance_list(:followers_only)
|> MRF.subdomains_regex()
activity =
with true <- MRF.subdomain_match?(followers_only, actor_host),
user <- User.get_cached_by_ap_id(activity["actor"]) do
# Don't use Map.get/3 intentionally, these must not be nil
fixed_to = activity["to"] || []
fixed_cc = activity["cc"] || []
to = FollowingRelationship.followers_ap_ids(user, fixed_to)
cc = FollowingRelationship.followers_ap_ids(user, fixed_cc)
activity
|> Map.put("to", intersection([user.follower_address | to], fixed_to))
|> Map.put("cc", intersection([user.follower_address | cc], fixed_cc))
else
_ -> activity
end
{:ok, activity}
end
defp check_report_removal(%{host: actor_host} = _actor_info, %{"type" => "Flag"} = activity) do
report_removal =
instance_list(:report_removal)
|> MRF.subdomains_regex()
if MRF.subdomain_match?(report_removal, actor_host) do
{:reject, "[SimplePolicy] host in report_removal list"}
else
{:ok, activity}
end
end
defp check_report_removal(_actor_info, activity), do: {:ok, activity}
defp check_avatar_removal(%{host: actor_host} = _actor_info, %{"icon" => _icon} = activity) do
avatar_removal =
instance_list(:avatar_removal)
|> MRF.subdomains_regex()
if MRF.subdomain_match?(avatar_removal, actor_host) do
{:ok, Map.delete(activity, "icon")}
else
{:ok, activity}
end
end
defp check_avatar_removal(_actor_info, activity), do: {:ok, activity}
defp check_banner_removal(%{host: actor_host} = _actor_info, %{"image" => _image} = activity) do
banner_removal =
instance_list(:banner_removal)
|> MRF.subdomains_regex()
if MRF.subdomain_match?(banner_removal, actor_host) do
{:ok, Map.delete(activity, "image")}
else
{:ok, activity}
end
end
defp check_banner_removal(_actor_info, activity), do: {:ok, activity}
defp check_object(%{"object" => object} = activity) do
with {:ok, _object} <- filter(object) do
{:ok, activity}
end
end
defp check_object(activity), do: {:ok, activity}
defp instance_list(config_key) do
Config.get([:mrf_simple, config_key])
|> MRF.instance_list_from_tuples()
end
@impl true
def id_filter(id) do
host_info = URI.parse(id)
with {:ok, _} <- check_accept(host_info, %{}),
{:ok, _} <- check_reject(host_info, %{}) do
true
else
_ -> false
end
end
@impl true
def filter(%{"type" => "Delete", "actor" => actor} = activity) do
%{host: actor_host} = URI.parse(actor)
reject_deletes =
instance_list(:reject_deletes)
|> MRF.subdomains_regex()
if MRF.subdomain_match?(reject_deletes, actor_host) do
{:reject, "[SimplePolicy] host in reject_deletes list"}
else
{:ok, activity}
end
end
@impl true
def filter(%{"actor" => actor} = activity) do
actor_info = URI.parse(actor)
with {:ok, activity} <- check_accept(actor_info, activity),
{:ok, activity} <- check_reject(actor_info, activity),
{:ok, activity} <- check_media_removal(actor_info, activity),
{:ok, activity} <- check_media_nsfw(actor_info, activity),
{:ok, activity} <- check_ftl_removal(actor_info, activity),
{:ok, activity} <- check_followers_only(actor_info, activity),
{:ok, activity} <- check_report_removal(actor_info, activity),
{:ok, activity} <- check_object(activity) do
{:ok, activity}
else
{:reject, _} = e -> e
end
end
def filter(%{"id" => actor, "type" => actor_type} = activity)
when actor_type in ["Application", "Group", "Organization", "Person", "Service"] do
actor_info = URI.parse(actor)
with {:ok, activity} <- check_accept(actor_info, activity),
{:ok, activity} <- check_reject(actor_info, activity),
{:ok, activity} <- check_avatar_removal(actor_info, activity),
{:ok, activity} <- check_banner_removal(actor_info, activity) do
{:ok, activity}
else
{:reject, _} = e -> e
end
end
def filter(activity) when is_binary(activity) do
uri = URI.parse(activity)
with {:ok, activity} <- check_accept(uri, activity),
{:ok, activity} <- check_reject(uri, activity) do
{:ok, activity}
else
{:reject, _} = e -> e
end
end
def filter(activity), do: {:ok, activity}
@impl true
def describe do
exclusions = Config.get([:mrf, :transparency_exclusions]) |> MRF.instance_list_from_tuples()
mrf_simple_excluded =
Config.get(:mrf_simple)
|> Enum.map(fn {rule, instances} ->
{rule, Enum.reject(instances, fn {host, _} -> host in exclusions end)}
end)
mrf_simple =
mrf_simple_excluded
|> Enum.map(fn {rule, instances} ->
{rule, Enum.map(instances, fn {host, _} -> host end)}
end)
|> Map.new()
# This is for backwards compatibility. We originally didn't sent
# extra info like a reason why an instance was rejected/quarantined/etc.
# Because we didn't want to break backwards compatibility it was decided
# to add an extra "info" key.
mrf_simple_info =
mrf_simple_excluded
|> Enum.map(fn {rule, instances} ->
{rule, Enum.reject(instances, fn {_, reason} -> reason == "" end)}
end)
|> Enum.reject(fn {_, instances} -> instances == [] end)
|> Enum.map(fn {rule, instances} ->
instances =
instances
|> Enum.map(fn {host, reason} -> {host, %{"reason" => reason}} end)
|> Map.new()
{rule, instances}
end)
|> Map.new()
{:ok, %{mrf_simple: mrf_simple, mrf_simple_info: mrf_simple_info}}
end
@impl true
def config_description do
%{
key: :mrf_simple,
related_policy: "Pleroma.Web.ActivityPub.MRF.SimplePolicy",
label: "MRF Simple",
description: "Simple ingress policies",
children:
[
%{
key: :media_removal,
description:
"List of instances to strip media attachments from and the reason for doing so"
},
%{
key: :media_nsfw,
label: "Media NSFW",
description:
"List of instances to tag all media as NSFW (sensitive) from and the reason for doing so"
},
%{
key: :federated_timeline_removal,
description:
"List of instances to remove from the Federated (aka The Whole Known Network) Timeline and the reason for doing so"
},
%{
key: :reject,
description:
"List of instances to reject activities from (except deletes) and the reason for doing so"
},
%{
key: :accept,
description:
"List of instances to only accept activities from (except deletes) and the reason for doing so"
},
%{
key: :followers_only,
description:
"Force posts from the given instances to be visible by followers only and the reason for doing so"
},
%{
key: :report_removal,
description: "List of instances to reject reports from and the reason for doing so"
},
%{
key: :avatar_removal,
description: "List of instances to strip avatars from and the reason for doing so"
},
%{
key: :banner_removal,
description: "List of instances to strip banners from and the reason for doing so"
},
%{
key: :reject_deletes,
description: "List of instances to reject deletions from and the reason for doing so"
}
]
|> Enum.map(fn setting ->
Map.merge(
setting,
%{
type: {:list, :tuple},
key_placeholder: "instance",
value_placeholder: "reason",
suggestions: [{"example.com", "Some reason"}, {"*.example.com", "Another reason"}]
}
)
end)
}
end
end
diff --git a/lib/pleroma/web/mastodon_api/views/status_view.ex b/lib/pleroma/web/mastodon_api/views/status_view.ex
index 6949c5f2d..5b18261a7 100644
--- a/lib/pleroma/web/mastodon_api/views/status_view.ex
+++ b/lib/pleroma/web/mastodon_api/views/status_view.ex
@@ -1,870 +1,872 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2022 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Web.MastodonAPI.StatusView do
use Pleroma.Web, :view
require Pleroma.Constants
alias Pleroma.Activity
alias Pleroma.HTML
alias Pleroma.Maps
alias Pleroma.Object
alias Pleroma.Repo
alias Pleroma.User
alias Pleroma.UserRelationship
alias Pleroma.Web.CommonAPI
alias Pleroma.Web.CommonAPI.Utils
alias Pleroma.Web.MastodonAPI.AccountView
alias Pleroma.Web.MastodonAPI.PollView
alias Pleroma.Web.MastodonAPI.StatusView
alias Pleroma.Web.MediaProxy
alias Pleroma.Web.PleromaAPI.EmojiReactionController
alias Pleroma.Web.RichMedia.Card
import Pleroma.Web.ActivityPub.Visibility, only: [get_visibility: 1, visible_for_user?: 2]
# This is a naive way to do this, just spawning a process per activity
# to fetch the preview. However it should be fine considering
# pagination is restricted to 40 activities at a time
# Force disable Websockets streaming for backfill jobs,
# otherwise old posts can show up on timelines.
defp fetch_rich_media_for_activities(activities, opts) do
opts = Map.put(opts, :stream, false)
Enum.each(activities, fn activity ->
Card.get_by_activity(activity, opts)
end)
end
# TODO: Add cached version.
defp get_replied_to_activities([]), do: %{}
defp get_replied_to_activities(activities) do
activities
|> Enum.map(fn
%{data: %{"type" => "Create"}} = activity ->
object = Object.normalize(activity, fetch: false)
object && object.data["inReplyTo"] != "" && object.data["inReplyTo"]
_ ->
nil
end)
|> Enum.filter(& &1)
|> Activity.create_by_object_ap_id_with_object()
|> Repo.all()
|> Enum.reduce(%{}, fn activity, acc ->
object = Object.normalize(activity, fetch: false)
if object, do: Map.put(acc, object.data["id"], activity), else: acc
end)
end
defp get_quoted_activities([]), do: %{}
defp get_quoted_activities(activities) do
activities
|> Enum.map(fn
%{data: %{"type" => "Create"}} = activity ->
object = Object.normalize(activity, fetch: false)
object && object.data["quoteUrl"] != "" && object.data["quoteUrl"]
_ ->
nil
end)
|> Enum.filter(& &1)
|> Activity.create_by_object_ap_id_with_object()
|> Repo.all()
|> Enum.reduce(%{}, fn activity, acc ->
object = Object.normalize(activity, fetch: false)
if object, do: Map.put(acc, object.data["id"], activity), else: acc
end)
end
# DEPRECATED This field seems to be a left-over from the StatusNet era.
# If your application uses `pleroma.conversation_id`: this field is deprecated.
# It is currently stubbed instead by doing a CRC32 of the context, and
# clearing the MSB to avoid overflow exceptions with signed integers on the
# different clients using this field (Java/Kotlin code, mostly; see Husky.)
# This should be removed in a future version of Pleroma. Pleroma-FE currently
# depends on this field, as well.
defp get_context_id(%{data: %{"context" => context}}) when is_binary(context) do
import Bitwise
:erlang.crc32(context)
|> band(bnot(0x8000_0000))
end
defp get_context_id(_), do: nil
# Check if the user reblogged this status
defp reblogged?(activity, %User{ap_id: ap_id}) do
with %Object{data: %{"announcements" => announcements}} when is_list(announcements) <-
Object.normalize(activity, fetch: false) do
ap_id in announcements
else
_ -> false
end
end
# False if the user is logged out
defp reblogged?(_activity, _user), do: false
def render("index.json", opts) do
reading_user = opts[:for]
# To do: check AdminAPIControllerTest on the reasons behind nil activities in the list
activities = Enum.filter(opts.activities, & &1)
# Start prefetching rich media before doing anything else
fetch_rich_media_for_activities(activities, opts)
replied_to_activities = get_replied_to_activities(activities)
quoted_activities = get_quoted_activities(activities)
parent_activities =
activities
|> Enum.filter(&(&1.data["type"] == "Announce" && &1.data["object"]))
|> Enum.map(&Object.normalize(&1, fetch: false).data["id"])
|> Activity.create_by_object_ap_id()
|> Activity.with_preloaded_object(:left)
|> Activity.with_preloaded_bookmark(reading_user)
|> Activity.with_set_thread_muted_field(reading_user)
|> Repo.all()
relationships_opt =
cond do
Map.has_key?(opts, :relationships) ->
opts[:relationships]
is_nil(reading_user) ->
UserRelationship.view_relationships_option(nil, [])
true ->
# Note: unresolved users are filtered out
actors =
(activities ++ parent_activities)
|> Enum.map(&CommonAPI.get_user(&1.data["actor"], false))
|> Enum.filter(& &1)
UserRelationship.view_relationships_option(reading_user, actors, subset: :source_mutes)
end
opts =
opts
|> Map.put(:replied_to_activities, replied_to_activities)
|> Map.put(:quoted_activities, quoted_activities)
|> Map.put(:parent_activities, parent_activities)
|> Map.put(:relationships, relationships_opt)
safe_render_many(activities, StatusView, "show.json", opts)
end
def render(
"show.json",
%{activity: %{data: %{"type" => "Announce", "object" => _object}} = activity} = opts
) do
user = CommonAPI.get_user(activity.data["actor"])
created_at = Utils.to_masto_date(activity.data["published"])
object = Object.normalize(activity, fetch: false)
reblogged_parent_activity =
if opts[:parent_activities] do
Activity.Queries.find_by_object_ap_id(
opts[:parent_activities],
object.data["id"]
)
else
Activity.create_by_object_ap_id(object.data["id"])
|> Activity.with_preloaded_bookmark(opts[:for])
|> Activity.with_set_thread_muted_field(opts[:for])
|> Repo.one()
end
reblog_rendering_opts = Map.put(opts, :activity, reblogged_parent_activity)
reblogged = render("show.json", reblog_rendering_opts)
favorited = opts[:for] && opts[:for].ap_id in (object.data["likes"] || [])
bookmark = Activity.get_bookmark(reblogged_parent_activity, opts[:for])
bookmark_folder =
if bookmark != nil do
bookmark.folder_id
else
nil
end
mentions =
activity.recipients
|> Enum.map(fn ap_id -> User.get_cached_by_ap_id(ap_id) end)
|> Enum.filter(& &1)
|> Enum.map(fn user -> AccountView.render("mention.json", %{user: user}) end)
{pinned?, pinned_at} = pin_data(object, user)
%{
id: to_string(activity.id),
uri: object.data["id"],
url: object.data["id"],
account:
AccountView.render("show.json", %{
user: user,
for: opts[:for]
}),
in_reply_to_id: nil,
in_reply_to_account_id: nil,
reblog: reblogged,
content: reblogged[:content] || "",
created_at: created_at,
reblogs_count: 0,
replies_count: 0,
favourites_count: 0,
reblogged: reblogged?(reblogged_parent_activity, opts[:for]),
favourited: present?(favorited),
bookmarked: present?(bookmark),
muted: false,
pinned: pinned?,
sensitive: false,
spoiler_text: "",
visibility: get_visibility(activity),
media_attachments: reblogged[:media_attachments] || [],
mentions: mentions,
tags: reblogged[:tags] || [],
application: build_application(object.data["generator"]),
language: get_language(object),
emojis: [],
pleroma: %{
local: activity.local,
pinned_at: pinned_at,
bookmark_folder: bookmark_folder
}
}
end
def render("show.json", %{activity: %{data: %{"object" => _object}} = activity} = opts) do
object = Object.normalize(activity, fetch: false)
user = CommonAPI.get_user(object.data["actor"])
user_follower_address = user.follower_address
like_count = object.data["like_count"] || 0
announcement_count = object.data["announcement_count"] || 0
hashtags = Object.hashtags(object)
sensitive = object.data["sensitive"] || Enum.member?(hashtags, "nsfw")
tags = Object.tags(object)
tag_mentions =
tags
|> Enum.filter(fn tag -> is_map(tag) and tag["type"] == "Mention" end)
|> Enum.map(fn tag -> tag["href"] end)
mentions =
(object.data["to"] ++ tag_mentions)
|> Enum.uniq()
|> Enum.map(fn
Pleroma.Constants.as_public() -> nil
^user_follower_address -> nil
ap_id -> User.get_cached_by_ap_id(ap_id)
end)
|> Enum.filter(& &1)
|> Enum.map(fn user -> AccountView.render("mention.json", %{user: user}) end)
favorited = opts[:for] && opts[:for].ap_id in (object.data["likes"] || [])
bookmark = Activity.get_bookmark(activity, opts[:for])
bookmark_folder =
if bookmark != nil do
bookmark.folder_id
else
nil
end
client_posted_this_activity = opts[:for] && user.id == opts[:for].id
expires_at =
with true <- client_posted_this_activity,
%Oban.Job{scheduled_at: scheduled_at} <-
Pleroma.Workers.PurgeExpiredActivity.get_expiration(activity.id) do
scheduled_at
else
_ -> nil
end
thread_muted? =
cond do
is_nil(opts[:for]) -> false
is_boolean(activity.thread_muted?) -> activity.thread_muted?
true -> CommonAPI.thread_muted?(activity, opts[:for])
end
attachment_data = object.data["attachment"] || []
attachments = render_many(attachment_data, StatusView, "attachment.json", as: :attachment)
created_at = Utils.to_masto_date(object.data["published"])
edited_at =
with %{"updated" => updated} <- object.data,
date <- Utils.to_masto_date(updated),
true <- date != "" do
date
else
_ ->
nil
end
reply_to = get_reply_to(activity, opts)
reply_to_user = reply_to && CommonAPI.get_user(reply_to.data["actor"])
history_len =
1 +
(Object.Updater.history_for(object.data)
|> Map.get("orderedItems")
|> length())
# See render("history.json", ...) for more details
# Here the implicit index of the current content is 0
chrono_order = history_len - 1
quote_activity = get_quote(activity, opts)
quote_id =
case quote_activity do
%Activity{id: id} -> id
_ -> nil
end
quote_post =
if visible_for_user?(quote_activity, opts[:for]) and opts[:show_quote] != false do
quote_rendering_opts = Map.merge(opts, %{activity: quote_activity, show_quote: false})
render("show.json", quote_rendering_opts)
else
nil
end
content =
object
|> render_content()
content_html =
content
|> Activity.HTML.get_cached_scrubbed_html_for_activity(
User.html_filter_policy(opts[:for]),
activity,
"mastoapi:content:#{chrono_order}"
)
content_plaintext =
content
|> Activity.HTML.get_cached_stripped_html_for_activity(
activity,
"mastoapi:content:#{chrono_order}"
)
summary = object.data["summary"] || ""
# Force disable Websockets streaming for backfill jobs which the below call will create,
# otherwise old posts can show up on timelines.
card =
case Card.get_by_activity(activity, Map.put(opts, :stream, false)) do
%Card{} = result -> render("card.json", result)
_ -> nil
end
url =
if user.local do
Pleroma.Web.Router.Helpers.o_status_url(Pleroma.Web.Endpoint, :notice, activity)
else
object.data["url"] || object.data["external_url"] || object.data["id"]
end
direct_conversation_id =
with {_, nil} <- {:direct_conversation_id, opts[:direct_conversation_id]},
{_, true} <- {:include_id, opts[:with_direct_conversation_id]},
{_, %User{} = for_user} <- {:for_user, opts[:for]} do
Activity.direct_conversation_id(activity, for_user)
else
{:direct_conversation_id, participation_id} when is_integer(participation_id) ->
participation_id
_e ->
nil
end
emoji_reactions =
object
|> Object.get_emoji_reactions()
|> EmojiReactionController.filter_allowed_users(
opts[:for],
Map.get(opts, :with_muted, false)
)
|> Stream.map(fn {emoji, users, url} ->
build_emoji_map(emoji, users, url, opts[:for])
end)
|> Enum.to_list()
# Status muted state (would do 1 request per status unless user mutes are preloaded)
muted =
thread_muted? ||
UserRelationship.exists?(
get_in(opts, [:relationships, :user_relationships]),
:mute,
opts[:for],
user,
fn for_user, user -> User.mutes?(for_user, user) end
)
{pinned?, pinned_at} = pin_data(object, user)
%{
id: to_string(activity.id),
uri: object.data["id"],
url: url,
account:
AccountView.render("show.json", %{
user: user,
for: opts[:for]
}),
in_reply_to_id: reply_to && to_string(reply_to.id),
in_reply_to_account_id: reply_to_user && to_string(reply_to_user.id),
reblog: nil,
card: card,
content: content_html,
text: opts[:with_source] && get_source_text(object.data["source"]),
created_at: created_at,
edited_at: edited_at,
reblogs_count: announcement_count,
replies_count: object.data["repliesCount"] || 0,
favourites_count: like_count,
reblogged: reblogged?(activity, opts[:for]),
favourited: present?(favorited),
bookmarked: present?(bookmark),
muted: muted,
pinned: pinned?,
sensitive: sensitive,
spoiler_text: summary,
visibility: get_visibility(object),
media_attachments: attachments,
poll: render(PollView, "show.json", object: object, for: opts[:for]),
mentions: mentions,
tags: build_tags(tags),
application: build_application(object.data["generator"]),
language: get_language(object),
emojis: build_emojis(object.data["emoji"]),
quotes_count: object.data["quotesCount"] || 0,
pleroma: %{
local: activity.local,
conversation_id: get_context_id(activity),
context: object.data["context"],
in_reply_to_account_acct: reply_to_user && reply_to_user.nickname,
quote: quote_post,
quote_id: quote_id,
quote_url: object.data["quoteUrl"],
quote_visible: visible_for_user?(quote_activity, opts[:for]),
content: %{"text/plain" => content_plaintext},
spoiler_text: %{"text/plain" => summary},
expires_at: expires_at,
direct_conversation_id: direct_conversation_id,
thread_muted: thread_muted?,
emoji_reactions: emoji_reactions,
parent_visible: visible_for_user?(reply_to, opts[:for]),
pinned_at: pinned_at,
quotes_count: object.data["quotesCount"] || 0,
bookmark_folder: bookmark_folder,
list_id: get_list_id(object, client_posted_this_activity)
}
}
end
def render("show.json", _) do
nil
end
def render("history.json", %{activity: %{data: %{"object" => _object}} = activity} = opts) do
object = Object.normalize(activity, fetch: false)
hashtags = Object.hashtags(object)
user = CommonAPI.get_user(activity.data["actor"])
past_history =
Object.Updater.history_for(object.data)
|> Map.get("orderedItems")
|> Enum.map(&Map.put(&1, "id", object.data["id"]))
|> Enum.map(&%Object{data: &1, id: object.id})
history =
[object | past_history]
# Mastodon expects the original to be at the first
|> Enum.reverse()
|> Enum.with_index()
|> Enum.map(fn {object, chrono_order} ->
%{
# The history is prepended every time there is a new edit.
# In chrono_order, the oldest item is always at 0, and so on.
# The chrono_order is an invariant kept between edits.
chrono_order: chrono_order,
object: object
}
end)
individual_opts =
opts
|> Map.put(:as, :item)
|> Map.put(:user, user)
|> Map.put(:hashtags, hashtags)
render_many(history, StatusView, "history_item.json", individual_opts)
end
def render(
"history_item.json",
%{
activity: activity,
user: user,
item: %{object: object, chrono_order: chrono_order},
hashtags: hashtags
} = opts
) do
sensitive = object.data["sensitive"] || Enum.member?(hashtags, "nsfw")
attachment_data = object.data["attachment"] || []
attachments = render_many(attachment_data, StatusView, "attachment.json", as: :attachment)
created_at = Utils.to_masto_date(object.data["updated"] || object.data["published"])
content =
object
|> render_content()
content_html =
content
|> Activity.HTML.get_cached_scrubbed_html_for_activity(
User.html_filter_policy(opts[:for]),
activity,
"mastoapi:content:#{chrono_order}"
)
summary = object.data["summary"] || ""
%{
account:
AccountView.render("show.json", %{
user: user,
for: opts[:for]
}),
content: content_html,
sensitive: sensitive,
spoiler_text: summary,
created_at: created_at,
media_attachments: attachments,
emojis: build_emojis(object.data["emoji"]),
poll: render(PollView, "show.json", object: object, for: opts[:for])
}
end
def render("source.json", %{activity: %{data: %{"object" => _object}} = activity} = _opts) do
object = Object.normalize(activity, fetch: false)
%{
id: activity.id,
text: get_source_text(Map.get(object.data, "source", "")),
spoiler_text: Map.get(object.data, "summary", ""),
content_type: get_source_content_type(object.data["source"])
}
end
def render("card.json", %Card{fields: rich_media}) do
page_url_data = URI.parse(rich_media["url"])
page_url = page_url_data |> to_string
image_url = proxied_url(rich_media["image"], page_url_data)
audio_url = proxied_url(rich_media["audio"], page_url_data)
video_url = proxied_url(rich_media["video"], page_url_data)
%{
type: "link",
provider_name: page_url_data.host,
provider_url: page_url_data.scheme <> "://" <> page_url_data.host,
url: page_url,
image: image_url,
image_description: rich_media["image:alt"] || "",
title: rich_media["title"] || "",
description: rich_media["description"] || "",
pleroma: %{
opengraph:
rich_media
|> Maps.put_if_present("image", image_url)
|> Maps.put_if_present("audio", audio_url)
|> Maps.put_if_present("video", video_url)
}
}
end
def render("card.json", _), do: nil
def render("attachment.json", %{attachment: attachment}) do
[attachment_url | _] = attachment["url"]
media_type = attachment_url["mediaType"] || attachment_url["mimeType"] || "image"
href_remote = attachment_url["href"]
href = href_remote |> MediaProxy.url()
href_preview = attachment_url["href"] |> MediaProxy.preview_url()
meta = render("attachment_meta.json", %{attachment: attachment})
type =
cond do
String.contains?(media_type, "image") -> "image"
String.contains?(media_type, "video") -> "video"
String.contains?(media_type, "audio") -> "audio"
true -> "unknown"
end
attachment_id =
with {_, ap_id} when is_binary(ap_id) <- {:ap_id, attachment["id"]},
{_, %Object{data: _object_data, id: object_id}} <-
{:object, Object.get_by_ap_id(ap_id)} do
to_string(object_id)
else
_ ->
<<hash_id::signed-32, _rest::binary>> = :crypto.hash(:md5, href)
to_string(attachment["id"] || hash_id)
end
description =
if attachment["summary"] do
HTML.strip_tags(attachment["summary"])
else
attachment["name"]
end
name = if attachment["summary"], do: attachment["name"]
pleroma =
%{mime_type: media_type}
|> Maps.put_if_present(:name, name)
%{
id: attachment_id,
url: href,
remote_url: href_remote,
preview_url: href_preview,
text_url: href,
type: type,
description: description,
pleroma: pleroma,
blurhash: attachment["blurhash"]
}
|> Maps.put_if_present(:meta, meta)
end
def render("attachment_meta.json", %{
attachment: %{"url" => [%{"width" => width, "height" => height} | _]}
})
when is_integer(width) and is_integer(height) do
%{
original: %{
width: width,
height: height,
aspect: width / height
}
}
end
def render("attachment_meta.json", _), do: nil
def render("context.json", %{activity: activity, activities: activities, user: user}) do
%{ancestors: ancestors, descendants: descendants} =
activities
|> Enum.reverse()
- |> Enum.group_by(fn %{id: id} -> if id < activity.id, do: :ancestors, else: :descendants end)
+ |> Enum.group_by(fn %{id: id} ->
+ if id < activity.id, do: :ancestors, else: :descendants
+ end)
|> Map.put_new(:ancestors, [])
|> Map.put_new(:descendants, [])
%{
ancestors: render("index.json", for: user, activities: ancestors, as: :activity),
descendants: render("index.json", for: user, activities: descendants, as: :activity)
}
end
def render("translation.json", %{
content: content,
detected_source_language: detected_source_language,
provider: provider
}) do
%{content: content, detected_source_language: detected_source_language, provider: provider}
end
def get_reply_to(activity, %{replied_to_activities: replied_to_activities}) do
object = Object.normalize(activity, fetch: false)
with nil <- replied_to_activities[object.data["inReplyTo"]] do
# If user didn't participate in the thread
Activity.get_in_reply_to_activity(activity)
end
end
def get_reply_to(%{data: %{"object" => _object}} = activity, _) do
object = Object.normalize(activity, fetch: false)
if object.data["inReplyTo"] && object.data["inReplyTo"] != "" do
Activity.get_create_by_object_ap_id(object.data["inReplyTo"])
else
nil
end
end
def get_quote(activity, %{quoted_activities: quoted_activities}) do
object = Object.normalize(activity, fetch: false)
with nil <- quoted_activities[object.data["quoteUrl"]] do
# For when a quote post is inside an Announce
Activity.get_create_by_object_ap_id_with_object(object.data["quoteUrl"])
end
end
def get_quote(%{data: %{"object" => _object}} = activity, _) do
object = Object.normalize(activity, fetch: false)
if object.data["quoteUrl"] && object.data["quoteUrl"] != "" do
Activity.get_create_by_object_ap_id(object.data["quoteUrl"])
else
nil
end
end
def render_content(%{data: %{"name" => name}} = object) when not is_nil(name) and name != "" do
url = object.data["url"] || object.data["id"]
"<p><a href=\"#{url}\">#{name}</a></p>#{object.data["content"]}"
end
def render_content(object), do: object.data["content"] || ""
@doc """
Builds a dictionary tags.
## Examples
iex> Pleroma.Web.MastodonAPI.StatusView.build_tags(["fediverse", "nextcloud"])
[{"name": "fediverse", "url": "/tag/fediverse"},
{"name": "nextcloud", "url": "/tag/nextcloud"}]
"""
@spec build_tags(list(any())) :: list(map())
def build_tags(object_tags) when is_list(object_tags) do
object_tags
|> Enum.filter(&is_binary/1)
|> Enum.map(&%{name: &1, url: "#{Pleroma.Web.Endpoint.url()}/tag/#{URI.encode(&1)}"})
end
def build_tags(_), do: []
@doc """
Builds list emojis.
Arguments: `nil` or list tuple of name and url.
Returns list emojis.
## Examples
iex> Pleroma.Web.MastodonAPI.StatusView.build_emojis([{"2hu", "corndog.png"}])
[%{shortcode: "2hu", static_url: "corndog.png", url: "corndog.png", visible_in_picker: false}]
"""
@spec build_emojis(nil | list(tuple())) :: list(map())
def build_emojis(nil), do: []
def build_emojis(emojis) do
emojis
|> Enum.map(fn {name, url} ->
name = HTML.strip_tags(name)
url =
url
|> HTML.strip_tags()
|> MediaProxy.url()
%{shortcode: name, url: url, static_url: url, visible_in_picker: false}
end)
end
defp present?(nil), do: false
defp present?(false), do: false
defp present?(_), do: true
defp pin_data(%Object{data: %{"id" => object_id}}, %User{pinned_objects: pinned_objects}) do
if pinned_at = pinned_objects[object_id] do
{true, Utils.to_masto_date(pinned_at)}
else
{false, nil}
end
end
defp build_emoji_map(emoji, users, url, current_user) do
%{
name: Pleroma.Web.PleromaAPI.EmojiReactionView.emoji_name(emoji, url),
count: length(users),
url: MediaProxy.url(url),
me: !!(current_user && current_user.ap_id in users),
account_ids: Enum.map(users, fn user -> User.get_cached_by_ap_id(user).id end)
}
end
@spec build_application(map() | nil) :: map() | nil
defp build_application(%{"type" => _type, "name" => name, "url" => url}),
do: %{name: name, website: url}
defp build_application(_), do: nil
@spec build_image_url(URI.t(), URI.t()) :: String.t()
defp build_image_url(%URI{} = image_url_data, %URI{} = page_url_data) do
URI.merge(page_url_data, image_url_data) |> to_string
end
defp get_source_text(%{"content" => content} = _source) do
content
end
defp get_source_text(source) when is_binary(source) do
source
end
defp get_source_text(_) do
""
end
defp get_source_content_type(%{"mediaType" => type} = _source) do
type
end
defp get_source_content_type(_source) do
Utils.get_content_type(nil)
end
defp get_language(%{data: %{"language" => "und"}}), do: nil
defp get_language(object), do: object.data["language"]
defp proxied_url(url, page_url_data) do
if is_binary(url) do
build_image_url(URI.parse(url), page_url_data) |> MediaProxy.url()
else
nil
end
end
defp get_list_id(object, client_posted_this_activity) do
with true <- client_posted_this_activity,
%{data: %{"listMessage" => list_ap_id}} when is_binary(list_ap_id) <- object,
%{id: list_id} <- Pleroma.List.get_by_ap_id(list_ap_id) do
list_id
else
_ -> nil
end
end
end
diff --git a/lib/pleroma/workers/mailer_worker.ex b/lib/pleroma/workers/mailer_worker.ex
index b0259b191..9ef6a80c9 100644
--- a/lib/pleroma/workers/mailer_worker.ex
+++ b/lib/pleroma/workers/mailer_worker.ex
@@ -1,18 +1,20 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2022 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Workers.MailerWorker do
use Oban.Worker, queue: :background
@impl true
- def perform(%Job{args: %{"op" => "email", "encoded_email" => encoded_email, "config" => config}}) do
+ def perform(%Job{
+ args: %{"op" => "email", "encoded_email" => encoded_email, "config" => config}
+ }) do
encoded_email
|> Base.decode64!()
|> :erlang.binary_to_term()
|> Pleroma.Emails.Mailer.deliver(config)
end
@impl true
def timeout(_job), do: :timer.seconds(5)
end
diff --git a/test/pleroma/search/meilisearch_test.exs b/test/pleroma/search/meilisearch_test.exs
index ff32491c5..6e4a42c95 100644
--- a/test/pleroma/search/meilisearch_test.exs
+++ b/test/pleroma/search/meilisearch_test.exs
@@ -1,137 +1,137 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2021 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Search.MeilisearchTest do
require Pleroma.Constants
use Pleroma.DataCase, async: true
use Oban.Testing, repo: Pleroma.Repo
import Pleroma.Factory
import Tesla.Mock
import Mox
alias Pleroma.Search.Meilisearch
alias Pleroma.UnstubbedConfigMock, as: Config
alias Pleroma.Web.CommonAPI
alias Pleroma.Workers.SearchIndexingWorker
describe "meilisearch" do
test "indexes a local post on creation" do
user = insert(:user)
Tesla.Mock.mock(fn
%{
method: :put,
url: "http://127.0.0.1:7700/indexes/objects/documents",
body: body
} ->
assert match?(
[%{"content" => "guys i just don&#39;t wanna leave the swamp"}],
Jason.decode!(body)
)
# To make sure that the worker is called
send(self(), "posted_to_meilisearch")
%{
"enqueuedAt" => "2023-11-12T12:36:46.927517Z",
"indexUid" => "objects",
"status" => "enqueued",
"taskUid" => 6,
"type" => "documentAdditionOrUpdate"
}
|> json()
end)
Config
|> expect(:get, 3, fn
[Pleroma.Search, :module], nil ->
Meilisearch
[Pleroma.Search.Meilisearch, :url], nil ->
"http://127.0.0.1:7700"
[Pleroma.Search.Meilisearch, :private_key], nil ->
"secret"
end)
{:ok, activity} =
CommonAPI.post(user, %{
status: "guys i just don't wanna leave the swamp",
visibility: "public"
})
args = %{"op" => "add_to_index", "activity" => activity.id}
assert_enqueued(
worker: SearchIndexingWorker,
args: args
)
assert :ok = perform_job(SearchIndexingWorker, args)
assert_received("posted_to_meilisearch")
end
test "deletes posts from index when deleted locally" do
user = insert(:user)
Tesla.Mock.mock(fn
%{
method: :put,
url: "http://127.0.0.1:7700/indexes/objects/documents",
body: body
} ->
assert match?(
[%{"content" => "guys i just don&#39;t wanna leave the swamp"}],
Jason.decode!(body)
)
%{
"enqueuedAt" => "2023-11-12T12:36:46.927517Z",
"indexUid" => "objects",
"status" => "enqueued",
"taskUid" => 6,
"type" => "documentAdditionOrUpdate"
}
|> json()
%{method: :delete, url: "http://127.0.0.1:7700/indexes/objects/documents/" <> id} ->
send(self(), "called_delete")
- assert String.length(id) > 1
+ assert id != ""
json(%{})
end)
Config
|> expect(:get, 6, fn
[Pleroma.Search, :module], nil ->
Meilisearch
[Pleroma.Search.Meilisearch, :url], nil ->
"http://127.0.0.1:7700"
[Pleroma.Search.Meilisearch, :private_key], nil ->
"secret"
end)
{:ok, activity} =
CommonAPI.post(user, %{
status: "guys i just don't wanna leave the swamp",
visibility: "public"
})
args = %{"op" => "add_to_index", "activity" => activity.id}
assert_enqueued(worker: SearchIndexingWorker, args: args)
assert :ok = perform_job(SearchIndexingWorker, args)
{:ok, _} = CommonAPI.delete(activity.id, user)
delete_args = %{"op" => "remove_from_index", "object" => activity.object.id}
assert_enqueued(worker: SearchIndexingWorker, args: delete_args)
assert :ok = perform_job(SearchIndexingWorker, delete_args)
assert_received("called_delete")
end
end
end
diff --git a/test/pleroma/search/paradedb_integration_test.exs b/test/pleroma/search/parade_db_integration_test.exs
similarity index 100%
rename from test/pleroma/search/paradedb_integration_test.exs
rename to test/pleroma/search/parade_db_integration_test.exs
diff --git a/test/pleroma/search/paradedb_test.exs b/test/pleroma/search/parade_db_test.exs
similarity index 94%
rename from test/pleroma/search/paradedb_test.exs
rename to test/pleroma/search/parade_db_test.exs
index 3a3b8298f..3133b6b38 100644
--- a/test/pleroma/search/paradedb_test.exs
+++ b/test/pleroma/search/parade_db_test.exs
@@ -1,280 +1,286 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2022 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Search.ParadeDBTest do
use Pleroma.DataCase, async: true
use Oban.Testing, repo: Pleroma.Repo
import Mox
import Pleroma.Factory
alias Pleroma.Search.ParadeDB
alias Pleroma.Search.ParadeDB.ClientMock
alias Pleroma.UnstubbedConfigMock, as: Config
alias Pleroma.Web.CommonAPI
alias Pleroma.Workers.SearchIndexingWorker
describe "ParadeDB" do
test "indexes a public post on creation" do
user = insert(:user)
{:ok, activity} =
CommonAPI.post(user, %{
status: "guys i just don't wanna leave the swamp",
visibility: "public"
})
args = %{"op" => "add_to_index", "activity" => activity.id}
assert_enqueued(worker: SearchIndexingWorker, args: args)
ClientMock
|> expect(:query, fn sql, params ->
assert sql =~ "INSERT INTO pleroma_search_documents"
[activity_id, object_id, object_ap_id, actor_ap_id, content, published_at] = params
{:ok, dumped_activity_id} = FlakeId.Ecto.CompatType.dump(activity.id)
assert activity_id == dumped_activity_id
assert is_integer(object_id)
assert is_binary(object_ap_id)
assert actor_ap_id == activity.data["actor"]
assert content == "guys i just don&#39;t wanna leave the swamp"
assert %DateTime{} = published_at
send(self(), :inserted)
{:ok, %{}}
end)
Config
|> expect(:get, 3, fn
[Pleroma.Search, :module], nil ->
ParadeDB
[Pleroma.Search.ParadeDB, :client_impl], nil ->
ClientMock
[Pleroma.Search.ParadeDB, :table], "pleroma_search_documents" ->
"pleroma_search_documents"
end)
assert :ok = perform_job(SearchIndexingWorker, args)
assert_received(:inserted)
end
test "doesn't index posts that are not public" do
user = insert(:user)
Enum.each(["private", "direct"], fn visibility ->
{:ok, activity} =
CommonAPI.post(user, %{
status: "guys i just don't wanna leave the swamp",
visibility: visibility
})
args = %{"op" => "add_to_index", "activity" => activity.id}
- assert_enqueued(worker: SearchIndexingWorker, args: args)
+ refute_enqueued(worker: SearchIndexingWorker, args: args)
+ end)
+ end
- Config
- |> expect(:get, fn
- [Pleroma.Search, :module], nil ->
- ParadeDB
- end)
+ test "backend skips non-public posts when called directly" do
+ user = insert(:user)
+
+ Enum.each(["private", "direct"], fn visibility ->
+ {:ok, activity} =
+ CommonAPI.post(user, %{
+ status: "guys i just don't wanna leave the swamp",
+ visibility: visibility
+ })
- assert :ok = perform_job(SearchIndexingWorker, args)
+ assert :ok = ParadeDB.add_to_index(activity)
end)
end
test "deletes posts from index when deleted locally" do
user = insert(:user)
ClientMock
|> expect(:query, 2, fn sql, params ->
cond do
String.contains?(sql, "INSERT INTO pleroma_search_documents") ->
send(self(), :inserted)
{:ok, %{}}
String.contains?(sql, "DELETE FROM pleroma_search_documents") ->
assert [object_id] = params
assert is_integer(object_id)
send(self(), :deleted)
{:ok, %{}}
true ->
flunk("Unexpected ParadeDB SQL: #{inspect(sql)}")
end
end)
Config
|> expect(:get, 6, fn
[Pleroma.Search, :module], nil ->
ParadeDB
[Pleroma.Search.ParadeDB, :client_impl], nil ->
ClientMock
[Pleroma.Search.ParadeDB, :table], "pleroma_search_documents" ->
"pleroma_search_documents"
end)
{:ok, activity} =
CommonAPI.post(user, %{
status: "guys i just don't wanna leave the swamp",
visibility: "public"
})
args = %{"op" => "add_to_index", "activity" => activity.id}
assert_enqueued(worker: SearchIndexingWorker, args: args)
assert :ok = perform_job(SearchIndexingWorker, args)
assert_received(:inserted)
{:ok, _} = CommonAPI.delete(activity.id, user)
delete_args = %{"op" => "remove_from_index", "object" => activity.object.id}
assert_enqueued(worker: SearchIndexingWorker, args: delete_args)
assert :ok = perform_job(SearchIndexingWorker, delete_args)
assert_received(:deleted)
end
test "search returns activities in backend order" do
user = insert(:user)
{:ok, activity1} = CommonAPI.post(user, %{status: "first swamp", visibility: "public"})
{:ok, activity2} = CommonAPI.post(user, %{status: "second swamp", visibility: "public"})
activity1_id = activity1.id
activity2_id = activity2.id
{:ok, dumped_activity1_id} = FlakeId.Ecto.CompatType.dump(activity1_id)
{:ok, dumped_activity2_id} = FlakeId.Ecto.CompatType.dump(activity2_id)
ClientMock
|> expect(:query, fn sql, params ->
assert sql =~ "SELECT id FROM pleroma_search_documents"
assert ["swamp", _limit, _offset] = params
{:ok, %{rows: [[dumped_activity2_id], [dumped_activity1_id]]}}
end)
Config
|> expect(:get, 3, fn
[Pleroma.Search.ParadeDB, :client_impl], nil ->
ClientMock
[Pleroma.Search.ParadeDB, :table], "pleroma_search_documents" ->
"pleroma_search_documents"
[Pleroma.Search.ParadeDB, :fuzzy_distance], 0 ->
0
end)
assert [%{id: ^activity2_id}, %{id: ^activity1_id}] =
ParadeDB.search(nil, "swamp", limit: 40, offset: 0)
end
test "search excludes non-public activities from backend hits" do
user = insert(:user)
{:ok, public_activity} =
CommonAPI.post(user, %{status: "public swamp", visibility: "public"})
{:ok, private_activity} =
CommonAPI.post(user, %{status: "private swamp", visibility: "private"})
public_activity_id = public_activity.id
{:ok, dumped_public_id} = FlakeId.Ecto.CompatType.dump(public_activity.id)
{:ok, dumped_private_id} = FlakeId.Ecto.CompatType.dump(private_activity.id)
ClientMock
|> expect(:query, fn sql, params ->
assert sql =~ "SELECT id FROM pleroma_search_documents"
assert ["swamp", _limit, _offset] = params
{:ok, %{rows: [[dumped_private_id], [dumped_public_id]]}}
end)
Config
|> expect(:get, 3, fn
[Pleroma.Search.ParadeDB, :client_impl], nil ->
ClientMock
[Pleroma.Search.ParadeDB, :table], "pleroma_search_documents" ->
"pleroma_search_documents"
[Pleroma.Search.ParadeDB, :fuzzy_distance], 0 ->
0
end)
assert [%{id: ^public_activity_id}] = ParadeDB.search(nil, "swamp", limit: 40, offset: 0)
end
test "search supports backend ids returned as uuid strings" do
user = insert(:user)
{:ok, activity} = CommonAPI.post(user, %{status: "uuid swamp", visibility: "public"})
activity_id = activity.id
{:ok, dumped_id} = FlakeId.Ecto.CompatType.dump(activity_id)
{:ok, uuid_string} = Ecto.UUID.load(dumped_id)
ClientMock
|> expect(:query, fn sql, params ->
assert sql =~ "SELECT id FROM pleroma_search_documents"
assert ["swamp", _limit, _offset] = params
{:ok, %{rows: [[uuid_string]]}}
end)
Config
|> expect(:get, 3, fn
[Pleroma.Search.ParadeDB, :client_impl], nil ->
ClientMock
[Pleroma.Search.ParadeDB, :table], "pleroma_search_documents" ->
"pleroma_search_documents"
[Pleroma.Search.ParadeDB, :fuzzy_distance], 0 ->
0
end)
assert [%{id: ^activity_id}] = ParadeDB.search(nil, "swamp", limit: 40, offset: 0)
end
test "search uses configured fuzzy distance" do
user = insert(:user)
{:ok, activity} = CommonAPI.post(user, %{status: "runing shoes", visibility: "public"})
activity_id = activity.id
{:ok, dumped_id} = FlakeId.Ecto.CompatType.dump(activity_id)
ClientMock
|> expect(:query, fn sql, params ->
assert sql =~ "SELECT id FROM pleroma_search_documents"
assert sql =~ "($1::text)::pdb.fuzzy(1)"
assert ["running shoes", _limit, _offset] = params
{:ok, %{rows: [[dumped_id]]}}
end)
Config
|> expect(:get, 3, fn
[Pleroma.Search.ParadeDB, :client_impl], nil ->
ClientMock
[Pleroma.Search.ParadeDB, :table], "pleroma_search_documents" ->
"pleroma_search_documents"
[Pleroma.Search.ParadeDB, :fuzzy_distance], 0 ->
1
end)
assert [%{id: ^activity_id}] = ParadeDB.search(nil, "running shoes", limit: 40, offset: 0)
end
end
end
diff --git a/test/pleroma/search_test.exs b/test/pleroma/search_test.exs
index d777bcda2..00d8b10ed 100644
--- a/test/pleroma/search_test.exs
+++ b/test/pleroma/search_test.exs
@@ -1,69 +1,147 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2021 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.SearchTest do
+ require Pleroma.Constants
+
use Pleroma.DataCase, async: true
use Oban.Testing, repo: Pleroma.Repo
import Pleroma.Factory
+ alias Pleroma.Object
+ alias Pleroma.Search
alias Pleroma.Web.CommonAPI
alias Pleroma.Workers.SearchIndexingWorker
- test "indexes posts that are public" do
+ test "indexes posts that are public or unlisted" do
user = insert(:user)
- {:ok, activity} =
- CommonAPI.post(user, %{
- status: "Well this is a story all about how my life got flipped turned upside down",
- visibility: "public"
- })
+ Enum.each(["public", "unlisted"], fn visibility ->
+ {:ok, activity} =
+ CommonAPI.post(user, %{
+ status: "Well this is a story all about how my life got flipped turned upside down",
+ visibility: visibility
+ })
- args = %{"op" => "add_to_index", "activity" => activity.id}
+ args = %{"op" => "add_to_index", "activity" => activity.id}
- assert_enqueued(worker: SearchIndexingWorker, args: args)
+ assert_enqueued(worker: SearchIndexingWorker, args: args)
+ end)
end
test "doesn't index posts that are not public" do
user = insert(:user)
Enum.each(["private", "direct"], fn visibility ->
{:ok, activity} =
CommonAPI.post(user, %{
status: "guys i just don't wanna leave the swamp",
visibility: visibility
})
args = %{"op" => "add_to_index", "activity" => activity.id}
refute_enqueued(worker: SearchIndexingWorker, args: args)
end)
end
test "Indexes appropriate activity types" do
user = insert(:user)
{:ok, activity} =
CommonAPI.post(user, %{
status: "I'm my own hype man",
visibility: "public"
})
args = %{"op" => "add_to_index", "activity" => activity.id}
assert_enqueued(worker: SearchIndexingWorker, args: args)
{:ok, fav_activity} = CommonAPI.favorite(activity.id, user)
args = %{"op" => "add_to_index", "activity" => fav_activity.id}
refute_enqueued(worker: SearchIndexingWorker, args: args)
{:ok, repeat_activity} = CommonAPI.repeat(activity.id, user)
args = %{"op" => "add_to_index", "activity" => repeat_activity.id}
refute_enqueued(worker: SearchIndexingWorker, args: args)
end
+
+ test "object_to_search_data includes content warnings and attachment descriptions" do
+ object = %Object{
+ id: 1,
+ data: %{
+ "id" => "https://example.com/objects/1",
+ "type" => "Note",
+ "content" => "body text",
+ "summary" => "subject text",
+ "attachment" => [
+ %{"type" => "Document", "name" => "image description"},
+ %{"type" => "Document", "name" => ""}
+ ],
+ "published" => DateTime.utc_now() |> DateTime.to_iso8601()
+ }
+ }
+
+ assert %{content: content} = Search.object_to_search_data(object)
+ assert content =~ "body text"
+ assert content =~ "subject text"
+ assert content =~ "image description"
+ end
+
+ test "object_to_search_data ignores objects with invalid published dates" do
+ for published <- [nil, 123, "not a date"] do
+ object = %Object{
+ id: 1,
+ data: %{
+ "id" => "https://example.com/objects/1",
+ "type" => "Note",
+ "content" => "body text",
+ "published" => published
+ }
+ }
+
+ refute Search.object_to_search_data(object)
+ end
+ end
+
+ test "indexable? accepts searchable content warnings and attachment descriptions" do
+ public = Pleroma.Constants.as_public()
+ published = DateTime.utc_now() |> DateTime.to_iso8601()
+
+ for data <- [
+ %{"content" => ".", "summary" => "subject text", "attachment" => []},
+ %{
+ "content" => ".",
+ "summary" => "",
+ "attachment" => [%{"type" => "Document", "name" => "image description"}]
+ }
+ ] do
+ object = %Object{
+ id: 1,
+ data:
+ Map.merge(data, %{
+ "id" => "https://example.com/objects/1",
+ "type" => "Note",
+ "to" => [public],
+ "published" => published
+ })
+ }
+
+ activity = %Pleroma.Activity{
+ id: Ecto.UUID.generate(),
+ data: %{"type" => "Create"},
+ object: object,
+ recipients: [public]
+ }
+
+ assert Search.indexable?(activity)
+ end
+ end
end

File Metadata

Mime Type
text/x-diff
Expires
Fri, Oct 30, 9:41 PM (15 h, 37 m)
Storage Engine
blob
Storage Format
Raw Data
Storage Handle
1800577
Default Alt Text
(112 KB)

Event Timeline