Merge remote-tracking branch 'origin/develop' into translate-posts

Signed-off-by: marcin mikołajczak <git@mkljczk.pl>
This commit is contained in:
marcin mikołajczak 2024-06-12 17:13:23 +02:00
commit b07fd324fb
193 changed files with 4639 additions and 784 deletions

View file

@ -1,8 +1,8 @@
image: git.pleroma.social:5050/pleroma/pleroma/ci-base
image: git.pleroma.social:5050/pleroma/pleroma/ci-base:elixir-1.13.4-otp-25
variables: &global_variables
# Only used for the release
ELIXIR_VER: 1.12.3
ELIXIR_VER: 1.13.4
POSTGRES_DB: pleroma_test
POSTGRES_USER: postgres
POSTGRES_PASSWORD: postgres
@ -72,7 +72,7 @@ check-changelog:
tags:
- amd64
build-1.12.3:
build-1.13.4-otp-25:
extends:
- .build_changes_policy
- .using-ci-base
@ -80,12 +80,12 @@ build-1.12.3:
script:
- mix compile --force
build-1.15.7-otp-25:
build-1.15.8-otp-26:
extends:
- .build_changes_policy
- .using-ci-base
stage: build
image: git.pleroma.social:5050/pleroma/pleroma/ci-base:elixir-1.15
image: git.pleroma.social:5050/pleroma/pleroma/ci-base:elixir-1.15.8-otp-26
allow_failure: true
script:
- mix compile --force
@ -121,7 +121,7 @@ benchmark:
- mix ecto.migrate
- mix pleroma.load_testing
unit-testing-1.12.3:
unit-testing-1.13.4-otp-25:
extends:
- .build_changes_policy
- .using-ci-base
@ -144,18 +144,18 @@ unit-testing-1.12.3:
coverage_format: cobertura
path: coverage.xml
unit-testing-1.15.7-otp-25:
unit-testing-1.15.8-otp-26:
extends:
- .build_changes_policy
- .using-ci-base
stage: test
image: git.pleroma.social:5050/pleroma/pleroma/ci-base:elixir-1.15-otp25
image: git.pleroma.social:5050/pleroma/pleroma/ci-base:elixir-1.15.8-otp-26
allow_failure: true
cache: *testing_cache_policy
services: *testing_services
script: *testing_script
unit-testing-1.12-erratic:
unit-testing-1.13.4-otp-25-erratic:
extends:
- .build_changes_policy
- .using-ci-base

View file

@ -1,7 +1,7 @@
ARG ELIXIR_IMG=hexpm/elixir
ARG ELIXIR_VER=1.12.3
ARG ERLANG_VER=24.2.1
ARG ALPINE_VER=3.17.0
ARG ELIXIR_VER=1.13.4
ARG ERLANG_VER=24.3.4.15
ARG ALPINE_VER=3.17.5
FROM ${ELIXIR_IMG}:${ELIXIR_VER}-erlang-${ERLANG_VER}-alpine-${ALPINE_VER} as build

View file

@ -0,0 +1 @@
Fix Emoji object IDs not always being valid

View file

@ -0,0 +1 @@
HTTP Security: By default, don't allow unsafe-eval. The setting needs to be changed to allow Flash emulation.

0
changelog.d/3907.skip Normal file
View file

View file

@ -0,0 +1 @@
Uploader: Add support for uploading attachments using IPFS

View file

@ -0,0 +1 @@
Add NSFW-detecting MRF

View file

@ -0,0 +1 @@
Add DNSRBL MRF

View file

@ -0,0 +1 @@
Add options to the mix prune_objects task

View file

@ -0,0 +1 @@
Add Anti-mention Spam MRF backported from Rebased

View file

View file

@ -0,0 +1 @@
HTTPSignaturePlug: Add :authorized_fetch_mode_exceptions configuration

View file

@ -0,0 +1 @@
Add an option to reject certain domains when authorized fetch is enabled.

View file

@ -0,0 +1 @@
Update Bandit to 1.5.2

View file

@ -0,0 +1 @@
Elixir 1.13 is the minimum required version.

View file

0
changelog.d/cleanup.skip Normal file
View file

View file

@ -0,0 +1 @@
Restore Cowboy's ability to stream MediaProxy responses without Chunked encoding.

View file

@ -0,0 +1,2 @@
Update dependencies held back due to old Elixir version

View file

@ -0,0 +1 @@
Implement `/api/v1/accounts/familiar_followers`

View file

@ -0,0 +1 @@
Fix webfinger spoofing.

View file

@ -0,0 +1 @@
no comment

View file

@ -0,0 +1 @@
Logger metadata is now attached to some logs to help with troubleshooting and analysis

View file

@ -0,0 +1 @@
The query for marking notifications as read has been simplified

View file

@ -0,0 +1 @@
Ensure MediaProxy HTTP requests obey all the defined connection settings

View file

@ -0,0 +1 @@
Add missing indexes on foreign key relationships

View file

@ -0,0 +1 @@
Permit passing --chunk and --step values to the Pleroma.Search.Indexer Mix task

View file

@ -0,0 +1 @@
noop

View file

View file

@ -0,0 +1 @@
Oban queues have refactored to simplify the queue design

View file

@ -0,0 +1 @@
Use proper workers for fetching pins instead of an ad-hoc task, fixing a potential fetch loop

1
changelog.d/pools.change Normal file
View file

@ -0,0 +1 @@
HTTP connection pool adjustments

View file

@ -0,0 +1 @@
Update the documentation for configuring Prometheus metrics.

View file

@ -0,0 +1 @@
PromEx documentation

View file

@ -0,0 +1 @@
Add Qdrant/OpenAI embedding search

View file

@ -0,0 +1 @@
pleroma_ctl: Use realpath(1) instead of readlink(1)

View file

@ -0,0 +1 @@
A 422 error is returned when attempting to reply to a deleted status

View file

View file

@ -0,0 +1 @@
Parsing of RichMedia TTLs for Amazon URLs when query parameters are nil

View file

@ -0,0 +1 @@
Monitoring of search backend health to control the processing of jobs in the search indexing Oban queue

View file

@ -0,0 +1 @@
Display reposted replies with exclude_replies: true

View file

View file

@ -0,0 +1 @@
Add "status" notification type

View file

@ -0,0 +1 @@
End of poll notifications were not streamed over websockets or web push

View file

@ -0,0 +1 @@
Support honk-style attachment summaries as alt-text.

View file

@ -0,0 +1 @@
Video thumbnails were not being generated due to a negative cache lookup logic error

View file

@ -0,0 +1 @@
Fix validate_webfinger when running a different domain for Webfinger

View file

@ -0,0 +1 @@
Render nice web push notifications for polls

View file

@ -1,4 +1,4 @@
FROM elixir:1.15.7-otp-25
FROM elixir:1.13.4-otp-25
# Single RUN statement, otherwise intermediate images are created
# https://docs.docker.com/develop/develop-images/dockerfile_best-practices/#run

View file

@ -0,0 +1 @@
docker buildx build --platform linux/amd64,linux/arm64 -t git.pleroma.social:5050/pleroma/pleroma/ci-base:elixir-1.13.4-otp-25 --push .

View file

@ -1 +0,0 @@
docker buildx build --platform linux/amd64 -t git.pleroma.social:5050/pleroma/pleroma/ci-base:elixir-1.15-otp25 --push .

View file

@ -0,0 +1,8 @@
FROM elixir:1.15.8-otp-26
# Single RUN statement, otherwise intermediate images are created
# https://docs.docker.com/develop/develop-images/dockerfile_best-practices/#run
RUN apt-get update &&\
apt-get install -y libmagic-dev cmake libimage-exiftool-perl ffmpeg &&\
mix local.hex --force &&\
mix local.rebar --force

View file

@ -0,0 +1 @@
docker buildx build --platform linux/amd64,linux/arm64 -t git.pleroma.social:5050/pleroma/pleroma/ci-base:elixir-1.15.8-otp-26 --push .

View file

@ -82,6 +82,10 @@ config :ex_aws, :s3,
# 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"],
@ -131,13 +135,13 @@ config :pleroma, Pleroma.Web.Endpoint,
config :logger, :console,
level: :debug,
format: "\n$time $metadata[$level] $message\n",
metadata: [:request_id]
metadata: [:actor, :path, :type, :user]
config :logger, :ex_syslogger,
level: :debug,
ident: "pleroma",
format: "$metadata[$level] $message",
metadata: [:request_id]
metadata: [:actor, :path, :type, :user]
config :mime, :types, %{
"application/xml" => ["xml"],
@ -188,6 +192,7 @@ config :pleroma, :instance,
allow_relay: true,
public: true,
quarantined_instances: [],
rejected_instances: [],
static_dir: "instance/static/",
allowed_post_formats: [
"text/plain",
@ -406,11 +411,23 @@ config :pleroma, :mrf_vocabulary,
accept: [],
reject: []
config :pleroma, :mrf_dnsrbl,
nameserver: "127.0.0.1",
port: 53,
zone: "bl.pleroma.com"
# 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}"
@ -419,6 +436,8 @@ 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: [],
@ -501,7 +520,8 @@ config :pleroma, :http_security,
sts: false,
sts_max_age: 31_536_000,
ct_max_age: 2_592_000,
referrer_policy: "same-origin"
referrer_policy: "same-origin",
allow_unsafe_eval: false
config :cors_plug,
max_age: 86_400,
@ -563,24 +583,14 @@ config :pleroma, Oban,
log: false,
queues: [
activity_expiration: 10,
token_expiration: 5,
filter_expiration: 1,
backup: 1,
federator_incoming: 5,
federator_outgoing: 5,
ingestion_queue: 50,
web_push: 50,
mailer: 10,
transmogrifier: 20,
scheduled_activities: 10,
poll_notifications: 10,
background: 5,
remote_fetcher: 2,
attachments_cleanup: 1,
new_users_digest: 1,
mute_expire: 5,
search_indexing: 10,
rich_media_expiration: 2
search_indexing: [limit: 10, paused: true],
slow: 1
],
plugins: [Oban.Plugins.Pruner],
crontab: [
@ -818,22 +828,27 @@ config :pleroma, :connections_pool,
config :pleroma, :pools,
federation: [
size: 50,
max_waiting: 10,
size: 75,
max_waiting: 20,
recv_timeout: 10_000
],
media: [
size: 50,
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: 5,
max_waiting: 20,
recv_timeout: 15_000
],
default: [
size: 10,
size: 50,
max_waiting: 2,
recv_timeout: 5_000
]
@ -847,6 +862,10 @@ config :pleroma, :hackney_pools,
max_connections: 50,
timeout: 150_000
],
rich_media: [
max_connections: 50,
timeout: 150_000
],
upload: [
max_connections: 25,
timeout: 300_000
@ -892,8 +911,6 @@ config :pleroma, Pleroma.User.Backup,
process_chunk_size: 100
config :pleroma, ConcurrentLimiter, [
{Pleroma.Web.RichMedia.Helpers, [max_running: 5, max_waiting: 5]},
{Pleroma.Web.ActivityPub.MRF.MediaProxyWarmingPolicy, [max_running: 5, max_waiting: 5]},
{Pleroma.Search, [max_running: 30, max_waiting: 50]}
]
@ -915,6 +932,19 @@ config :pleroma, Pleroma.Application,
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"}
}
# 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"

View file

@ -136,6 +136,31 @@ config :pleroma, :config_description, [
}
]
},
%{
group: :pleroma,
key: Pleroma.Uploaders.IPFS,
type: :group,
description: "IPFS uploader-related settings",
children: [
%{
key: :get_gateway_url,
type: :string,
description: "GET Gateway URL",
suggestions: [
"https://ipfs.mydomain.com/{CID}",
"https://{CID}.ipfs.mydomain.com/"
]
},
%{
key: :post_gateway_url,
type: :string,
description: "POST Gateway URL",
suggestions: [
"http://localhost:5001/"
]
}
]
},
%{
group: :pleroma,
key: Pleroma.Uploaders.S3,
@ -749,6 +774,18 @@ config :pleroma, :config_description, [
{"*.quarantined.com", "Reason"}
]
},
%{
key: :rejected_instances,
type: {:list, :tuple},
key_placeholder: "instance",
value_placeholder: "reason",
description:
"List of ActivityPub instances to reject requests from if authorized_fetch_mode is enabled",
suggestions: [
{"rejected.com", "Reason"},
{"*.rejected.com", "Reason"}
]
},
%{
key: :static_dir,
type: :string,
@ -1791,6 +1828,12 @@ config :pleroma, :config_description, [
type: :boolean,
description: "Require HTTP signatures for AP fetches"
},
%{
key: :authorized_fetch_mode_exceptions,
type: {:list, :string},
description:
"List of IPs (CIDR format accepted) to exempt from HTTP Signatures requirement (for example to allow debugging, you shouldn't otherwise need this)"
},
%{
key: :note_replies_output_limit,
type: :integer,

View file

@ -35,8 +35,8 @@ config :pleroma, Pleroma.Emails.Mailer, adapter: Swoosh.Adapters.Local
# configured to run both http and https servers on
# different ports.
# Do not include metadata nor timestamps in development logs
config :logger, :console, format: "[$level] $message\n"
# Do not include timestamps in development logs
config :logger, :console, format: "$metadata[$level] $message\n"
# Set a higher stacktrace during development. Avoid configuring such
# in production as building large stacktraces may be expensive.

View file

@ -153,6 +153,12 @@ config :pleroma, Pleroma.Uploaders.S3, config_impl: Pleroma.UnstubbedConfigMock
config :pleroma, Pleroma.Upload, config_impl: Pleroma.UnstubbedConfigMock
config :pleroma, Pleroma.ScheduledActivity, config_impl: Pleroma.UnstubbedConfigMock
config :pleroma, Pleroma.Web.RichMedia.Helpers, config_impl: Pleroma.StaticStubbedConfigMock
config :pleroma, Pleroma.Uploaders.IPFS, config_impl: Pleroma.UnstubbedConfigMock
config :pleroma, Pleroma.Web.Plugs.HTTPSecurityPlug, config_impl: Pleroma.StaticStubbedConfigMock
config :pleroma, Pleroma.Web.Plugs.HTTPSignaturePlug, config_impl: Pleroma.StaticStubbedConfigMock
config :pleroma, Pleroma.Web.Plugs.HTTPSignaturePlug,
http_signatures_impl: Pleroma.StubbedHTTPSignaturesMock
peer_module =
if String.to_integer(System.otp_release()) >= 25 do

View file

@ -21,16 +21,18 @@ Replaces embedded objects with references to them in the `objects` table. Only n
mix pleroma.database remove_embedded_objects [option ...]
```
### Options
- `--vacuum` - run `VACUUM FULL` after the embedded objects are replaced with their references
## Prune old remote posts from the database
This will prune remote posts older than 90 days (configurable with [`config :pleroma, :instance, remote_post_retention_days`](../../configuration/cheatsheet.md#instance)) from the database, they will be refetched from source when accessed.
This will prune remote posts older than 90 days (configurable with [`config :pleroma, :instance, remote_post_retention_days`](../../configuration/cheatsheet.md#instance)) from the database. Pruned posts may be refetched in some cases.
!!! note
The disk space will only be reclaimed after a proper vacuum. By default Postgresql does this for you on a regular basis, but if your instance has been running for a long time and there are many rows deleted, it may be advantageous to use `VACUUM FULL` (e.g. by using the `--vacuum` option).
!!! danger
The disk space will only be reclaimed after `VACUUM FULL`. You may run out of disk space during the execution of the task or vacuuming if you don't have about 1/3rds of the database size free.
You may run out of disk space during the execution of the task or vacuuming if you don't have about 1/3rds of the database size free. Vacuum causes a substantial increase in I/O traffic, and may lead to a degraded experience while it is running.
=== "OTP"
@ -45,7 +47,11 @@ This will prune remote posts older than 90 days (configurable with [`config :ple
```
### Options
- `--vacuum` - run `VACUUM FULL` after the objects are pruned
- `--keep-threads` - Don't prune posts when they are part of a thread where at least one post has seen local interaction (e.g. one of the posts is a local post, or is favourited by a local user, or has been repeated by a local user...). It also won't delete posts when at least one of the posts in that thread is kept (e.g. because one of the posts has seen recent activity).
- `--keep-non-public` - Keep non-public posts like DM's and followers-only, even if they are remote.
- `--prune-orphaned-activities` - Also prune orphaned activities afterwards. Activities are things like Like, Create, Announce, Flag (aka reports). They can significantly help reduce the database size. Note: this can take a very long time.
- `--vacuum` - Run `VACUUM FULL` after the objects are pruned. This should not be used on a regular basis, but is useful if your instance has been running for a long time before pruning.
## Create a conversation for all existing DMs
@ -93,6 +99,9 @@ Can be safely re-run
## Vacuum the database
!!! note
By default Postgresql has an autovacuum deamon running. While the tasks described here can help in some cases, they shouldn't be needed on a regular basis. See [the Postgresql docs on vacuuming](https://www.postgresql.org/docs/current/sql-vacuum.html) for more information on this.
### Analyze
Running an `analyze` vacuum job can improve performance by updating statistics used by the query planner. **It is safe to cancel this.**

View file

@ -41,6 +41,7 @@ To add configuration to your config file, you can copy it from the base config.
* `allow_relay`: Permits remote instances to subscribe to all public posts of your instance. This may increase the visibility of your instance.
* `public`: Makes the client API in authenticated mode-only except for user-profiles. Useful for disabling the Local Timeline and The Whole Known Network. Note that there is a dependent setting restricting or allowing unauthenticated access to specific resources, see `restrict_unauthenticated` for more details.
* `quarantined_instances`: ActivityPub instances where private (DMs, followers-only) activities will not be send.
* `rejected_instances`: ActivityPub instances to reject requests from if authorized_fetch_mode is enabled.
* `allowed_post_formats`: MIME-type list of formats allowed to be posted (transformed into HTML).
* `extended_nickname_format`: Set to `true` to use extended local nicknames format (allows underscores/dashes). This will break federation with
older software for theses nicknames.
@ -284,6 +285,7 @@ Notes:
* `deny_follow_blocked`: Whether to disallow following an account that has blocked the user in question
* `sign_object_fetches`: Sign object fetches with HTTP signatures
* `authorized_fetch_mode`: Require HTTP signatures for AP fetches
* `authorized_fetch_mode_exceptions`: List of IPs (CIDR format accepted) to exempt from HTTP Signatures requirement (for example to allow debugging, you shouldn't otherwise need this)
## Pleroma.User
@ -472,6 +474,7 @@ This will make Pleroma listen on `127.0.0.1` port `8080` and generate urls start
* ``ct_max_age``: The maximum age for the `Expect-CT` header if sent.
* ``referrer_policy``: The referrer policy to use, either `"same-origin"` or `"no-referrer"`.
* ``report_uri``: Adds the specified url to `report-uri` and `report-to` group in CSP header.
* `allow_unsafe_eval`: Adds `wasm-unsafe-eval` to the CSP header. Needed for some non-essential frontend features like Flash emulation.
### Pleroma.Web.Plugs.RemoteIp
@ -661,6 +664,19 @@ config :ex_aws, :s3,
host: "s3.eu-central-1.amazonaws.com"
```
#### Pleroma.Uploaders.IPFS
* `post_gateway_url`: URL with port of POST Gateway (unauthenticated)
* `get_gateway_url`: URL of public GET Gateway
Example:
```elixir
config :pleroma, Pleroma.Uploaders.IPFS,
post_gateway_url: "http://localhost:5001",
get_gateway_url: "http://{CID}.ipfs.mydomain.com"
```
### Upload filters
#### Pleroma.Upload.Filter.AnonymizeFilename

View file

@ -10,6 +10,30 @@ To use built-in search that has no external dependencies, set the search module
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

View file

@ -295,9 +295,7 @@ See [Admin-API](admin_api.md)
"id": "9umDrYheeY451cQnEe",
"name": "Read later",
"emoji": "🕓",
"source": {
"emoji": "🕓"
}
"emoji_url": null
}
]
```

View file

@ -1,44 +1,47 @@
# Prometheus Metrics
# Prometheus / OpenTelemetry Metrics
Pleroma includes support for exporting metrics via the [prometheus_ex](https://github.com/deadtrickster/prometheus.ex) library.
Pleroma includes support for exporting metrics via the [prom_ex](https://github.com/akoutmos/prom_ex) library.
The metrics are exposed by a dedicated webserver/port to improve privacy and security.
Config example:
```
config :prometheus, Pleroma.Web.Endpoint.MetricsExporter,
enabled: true,
auth: {:basic, "myusername", "mypassword"},
ip_whitelist: ["127.0.0.1"],
path: "/api/pleroma/app_metrics",
format: :text
```
* `enabled` (Pleroma extension) enables the endpoint
* `ip_whitelist` (Pleroma extension) could be used to restrict access only to specified IPs
* `auth` sets the authentication (`false` for no auth; configurable to HTTP Basic Auth, see [prometheus-plugs](https://github.com/deadtrickster/prometheus-plugs#exporting) documentation)
* `format` sets the output format (`:text` or `:protobuf`)
* `path` sets the path to app metrics page
## `/api/pleroma/app_metrics`
### Exports Prometheus application metrics
* Method: `GET`
* Authentication: not required by default (see configuration options above)
* Params: none
* Response: text
## Grafana
### Config example
The following is a config example to use with [Grafana](https://grafana.com)
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"
```
- job_name: 'beam'
metrics_path: /api/pleroma/app_metrics
scheme: https
PromEx supports the ability to automatically publish dashboards to your Grafana server as well as register Annotations. If you do not wish to configure this capability you must generate the dashboard JSON files and import them directly. You can find the mix commands in the upstream [documentation](https://hexdocs.pm/prom_ex/Mix.Tasks.PromEx.Dashboard.Export.html). You can find the list of modules enabled in Pleroma for which you should generate dashboards for by examining the contents of the `lib/pleroma/prom_ex.ex` module.
## prometheus.yml
The following is a bare minimum config example to use with [Prometheus](https://prometheus.io) or Prometheus-compatible software like [VictoriaMetrics](https://victoriametrics.com).
```
global:
scrape_interval: 15s
scrape_configs:
- job_name: 'pleroma'
scheme: http
static_configs:
- targets: ['pleroma.soykaf.com']
- targets: ['pleroma.soykaf.com:4021']
```

View file

@ -14,7 +14,7 @@ Note: This article is potentially outdated because at this time we may not have
- PostgreSQL 11.0以上 (Ubuntu16.04では9.5しか提供されていないので,[](https://www.postgresql.org/download/linux/ubuntu/)こちらから新しいバージョンを入手してください)
- `postgresql-contrib` 11.0以上 (同上)
- Elixir 1.8 以上 ([Debianのリポジトリからインストールしないこと ここからインストールすること!](https://elixir-lang.org/install.html#unix-and-unix-like)。または [asdf](https://github.com/asdf-vm/asdf) をpleromaユーザーでインストールしてください)
- Elixir 1.13 以上 ([Debianのリポジトリからインストールしないこと ここからインストールすること!](https://elixir-lang.org/install.html#unix-and-unix-like)。または [asdf](https://github.com/asdf-vm/asdf) をpleromaユーザーでインストールしてください)
- `erlang-dev`
- `erlang-nox`
- `git`

View file

@ -1,7 +1,7 @@
## Required dependencies
* PostgreSQL >=11.0
* Elixir >=1.11.0 <1.15
* Elixir >=1.13.0 <1.15
* Erlang OTP >=22.2.0 (supported: <27)
* git
* file / libmagic

View file

@ -0,0 +1,15 @@
[Unit]
Description=NSFW API
After=docker.service
Requires=docker.service
[Service]
TimeoutStartSec=0
Restart=always
ExecStartPre=-/usr/bin/docker stop %n
ExecStartPre=-/usr/bin/docker rm %n
ExecStartPre=/usr/bin/docker pull eugencepoi/nsfw_api:latest
ExecStart=/usr/bin/docker run --rm -p 127.0.0.1:5000:5000/tcp --env PORT=5000 --name %n eugencepoi/nsfw_api:latest
[Install]
WantedBy=multi-user.target

View file

@ -67,43 +67,168 @@ defmodule Mix.Tasks.Pleroma.Database do
OptionParser.parse(
args,
strict: [
vacuum: :boolean
vacuum: :boolean,
keep_threads: :boolean,
keep_non_public: :boolean,
prune_orphaned_activities: :boolean
]
)
start_pleroma()
deadline = Pleroma.Config.get([:instance, :remote_post_retention_days])
time_deadline = NaiveDateTime.utc_now() |> NaiveDateTime.add(-(deadline * 86_400))
Logger.info("Pruning objects older than #{deadline} days")
log_message = "Pruning objects older than #{deadline} days"
time_deadline =
NaiveDateTime.utc_now()
|> NaiveDateTime.add(-(deadline * 86_400))
log_message =
if Keyword.get(options, :keep_non_public) do
log_message <> ", keeping non public posts"
else
log_message
end
from(o in Object,
where:
fragment(
"?->'to' \\? ? OR ?->'cc' \\? ?",
o.data,
^Pleroma.Constants.as_public(),
o.data,
^Pleroma.Constants.as_public()
),
where: o.inserted_at < ^time_deadline,
where:
log_message =
if Keyword.get(options, :keep_threads) do
log_message <> ", keeping threads intact"
else
log_message
end
log_message =
if Keyword.get(options, :prune_orphaned_activities) do
log_message <> ", pruning orphaned activities"
else
log_message
end
log_message =
if Keyword.get(options, :vacuum) do
log_message <>
", doing a full vacuum (you shouldn't do this as a recurring maintanance task)"
else
log_message
end
Logger.info(log_message)
if Keyword.get(options, :keep_threads) do
# We want to delete objects from threads where
# 1. the newest post is still old
# 2. none of the activities is local
# 3. none of the activities is bookmarked
# 4. optionally none of the posts is non-public
deletable_context =
if Keyword.get(options, :keep_non_public) do
Pleroma.Activity
|> join(:left, [a], b in Pleroma.Bookmark, on: a.id == b.activity_id)
|> group_by([a], fragment("? ->> 'context'::text", a.data))
|> having(
[a],
not fragment(
# Posts (checked on Create Activity) is non-public
"bool_or((not(?->'to' \\? ? OR ?->'cc' \\? ?)) and ? ->> 'type' = 'Create')",
a.data,
^Pleroma.Constants.as_public(),
a.data,
^Pleroma.Constants.as_public(),
a.data
)
)
else
Pleroma.Activity
|> join(:left, [a], b in Pleroma.Bookmark, on: a.id == b.activity_id)
|> group_by([a], fragment("? ->> 'context'::text", a.data))
end
|> having([a], max(a.updated_at) < ^time_deadline)
|> having([a], not fragment("bool_or(?)", a.local))
|> having([_, b], fragment("max(?::text) is null", b.id))
|> select([a], fragment("? ->> 'context'::text", a.data))
Pleroma.Object
|> where([o], fragment("? ->> 'context'::text", o.data) in subquery(deletable_context))
else
if Keyword.get(options, :keep_non_public) do
Pleroma.Object
|> where(
[o],
fragment(
"?->'to' \\? ? OR ?->'cc' \\? ?",
o.data,
^Pleroma.Constants.as_public(),
o.data,
^Pleroma.Constants.as_public()
)
)
else
Pleroma.Object
end
|> where([o], o.updated_at < ^time_deadline)
|> where(
[o],
fragment("split_part(?->>'actor', '/', 3) != ?", o.data, ^Pleroma.Web.Endpoint.host())
)
)
end
|> Repo.delete_all(timeout: :infinity)
prune_hashtags_query = """
if !Keyword.get(options, :keep_threads) do
# Without the --keep-threads option, it's possible that bookmarked
# objects have been deleted. We remove the corresponding bookmarks.
"""
delete from public.bookmarks
where id in (
select b.id from public.bookmarks b
left join public.activities a on b.activity_id = a.id
left join public.objects o on a."data" ->> 'object' = o.data ->> 'id'
where o.id is null
)
"""
|> Repo.query([], timeout: :infinity)
end
if Keyword.get(options, :prune_orphaned_activities) do
# Prune activities who link to a single object
"""
delete from public.activities
where id in (
select a.id from public.activities a
left join public.objects o on a.data ->> 'object' = o.data ->> 'id'
left join public.activities a2 on a.data ->> 'object' = a2.data ->> 'id'
left join public.users u on a.data ->> 'object' = u.ap_id
where not a.local
and jsonb_typeof(a."data" -> 'object') = 'string'
and o.id is null
and a2.id is null
and u.id is null
)
"""
|> Repo.query([], timeout: :infinity)
# Prune activities who link to an array of objects
"""
delete from public.activities
where id in (
select a.id from public.activities a
join json_array_elements_text((a."data" -> 'object')::json) as j on jsonb_typeof(a."data" -> 'object') = 'array'
left join public.objects o on j.value = o.data ->> 'id'
left join public.activities a2 on j.value = a2.data ->> 'id'
left join public.users u on j.value = u.ap_id
group by a.id
having max(o.data ->> 'id') is null
and max(a2.data ->> 'id') is null
and max(u.ap_id) is null
)
"""
|> Repo.query([], timeout: :infinity)
end
"""
DELETE FROM hashtags AS ht
WHERE NOT EXISTS (
SELECT 1 FROM hashtags_objects hto
WHERE ht.id = hto.hashtag_id)
"""
Repo.query(prune_hashtags_query)
|> Repo.query()
if Keyword.get(options, :vacuum) do
Maintenance.vacuum("full")

View file

@ -0,0 +1,83 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2021 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Mix.Tasks.Pleroma.Search.Indexer do
import Mix.Pleroma
import Ecto.Query
alias Pleroma.Workers.SearchIndexingWorker
def run(["create_index"]) do
start_pleroma()
with :ok <- Pleroma.Config.get([Pleroma.Search, :module]).create_index() do
IO.puts("Index created")
else
e -> IO.puts("Could not create index: #{inspect(e)}")
end
end
def run(["drop_index"]) do
start_pleroma()
with :ok <- Pleroma.Config.get([Pleroma.Search, :module]).drop_index() do
IO.puts("Index dropped")
else
e -> IO.puts("Could not drop index: #{inspect(e)}")
end
end
def run(["index" | options]) do
{options, [], []} =
OptionParser.parse(
options,
strict: [
chunk: :integer,
limit: :integer,
step: :integer
]
)
start_pleroma()
chunk_size = Keyword.get(options, :chunk, 100)
limit = Keyword.get(options, :limit, 100_000)
per_step = Keyword.get(options, :step, 1000)
chunks = max(div(limit, per_step), 1)
1..chunks
|> Enum.each(fn step ->
q =
from(a in Pleroma.Activity,
limit: ^per_step,
offset: ^per_step * (^step - 1),
select: [:id],
order_by: [desc: :id]
)
{:ok, ids} =
Pleroma.Repo.transaction(fn ->
Pleroma.Repo.stream(q, timeout: :infinity)
|> Enum.map(fn a ->
a.id
end)
end)
IO.puts("Got #{length(ids)} activities, adding to indexer")
ids
|> Enum.chunk_every(chunk_size)
|> Enum.each(fn chunk ->
IO.puts("Adding #{length(chunk)} activities to indexing queue")
chunk
|> Enum.map(fn id ->
SearchIndexingWorker.new(%{"op" => "add_to_index", "activity" => id})
end)
|> Oban.insert_all()
end)
end)
end
end

View file

@ -14,6 +14,7 @@ defmodule Pleroma.Application do
@name Mix.Project.config()[:name]
@version Mix.Project.config()[:version]
@repository Mix.Project.config()[:source_url]
@compile_env Mix.env()
def name, do: @name
def version, do: @version
@ -51,7 +52,11 @@ defmodule Pleroma.Application do
Pleroma.HTML.compile_scrubbers()
Pleroma.Config.Oban.warn()
Config.DeprecationWarnings.warn()
Pleroma.Web.Plugs.HTTPSecurityPlug.warn_if_disabled()
if @compile_env != :test do
Pleroma.Web.Plugs.HTTPSecurityPlug.warn_if_disabled()
end
Pleroma.ApplicationRequirements.verify!()
load_custom_modules()
Pleroma.Docs.JSON.compile()
@ -109,7 +114,8 @@ defmodule Pleroma.Application do
streamer_registry() ++
background_migrators() ++
shout_child(shout_enabled?()) ++
[Pleroma.Gopher.Server]
[Pleroma.Gopher.Server] ++
[Pleroma.Search.Healthcheck]
# See http://elixir-lang.org/docs/stable/elixir/Supervisor.html
# for other strategies and supported options
@ -163,6 +169,7 @@ defmodule Pleroma.Application do
limit: 500_000
),
build_cachex("rel_me", limit: 2500),
build_cachex("host_meta", default_ttl: :timer.minutes(120), limit: 5_000),
build_cachex("translations", default_ttl: :timer.hours(24), limit: 5_000)
]
end
@ -286,8 +293,6 @@ defmodule Pleroma.Application do
config = Config.get(ConcurrentLimiter, [])
[
Pleroma.Web.RichMedia.Helpers,
Pleroma.Web.ActivityPub.MRF.MediaProxyWarmingPolicy,
Pleroma.Search
]
|> Enum.each(fn module ->

View file

@ -16,4 +16,15 @@ defmodule Pleroma.Helpers.InetHelper do
def parse_address(ip) do
:inet.parse_address(ip)
end
def parse_cidr(proxy) when is_binary(proxy) do
proxy =
cond do
"/" in String.codepoints(proxy) -> proxy
InetCidr.v4?(InetCidr.parse_address!(proxy)) -> proxy <> "/32"
InetCidr.v6?(InetCidr.parse_address!(proxy)) -> proxy <> "/128"
end
InetCidr.parse_cidr!(proxy, true)
end
end

View file

@ -25,7 +25,7 @@ defmodule Pleroma.Helpers.MediaHelper do
end
def image_resize(url, options) do
with {:ok, env} <- HTTP.get(url, [], pool: :media),
with {:ok, env} <- HTTP.get(url, [], http_client_opts()),
{:ok, resized} <-
Operation.thumbnail_buffer(env.body, options.max_width,
height: options.max_height,
@ -45,8 +45,8 @@ defmodule Pleroma.Helpers.MediaHelper do
@spec video_framegrab(String.t()) :: {:ok, binary()} | {:error, any()}
def video_framegrab(url) do
with executable when is_binary(executable) <- System.find_executable("ffmpeg"),
false <- @cachex.exists?(:failed_media_helper_cache, url),
{:ok, env} <- HTTP.get(url, [], pool: :media),
{:ok, false} <- @cachex.exists?(:failed_media_helper_cache, url),
{:ok, env} <- HTTP.get(url, [], http_client_opts()),
{:ok, pid} <- StringIO.open(env.body) do
body_stream = IO.binstream(pid, 1)
@ -71,17 +71,19 @@ defmodule Pleroma.Helpers.MediaHelper do
end)
case Task.yield(task, 5_000) do
nil ->
{:ok, result} ->
{:ok, result}
_ ->
Task.shutdown(task)
@cachex.put(:failed_media_helper_cache, url, nil)
{:error, {:ffmpeg, :timeout}}
result ->
{:ok, result}
end
else
nil -> {:error, {:ffmpeg, :command_not_found}}
{:error, _} = error -> error
end
end
defp http_client_opts, do: Pleroma.Config.get([:media_proxy, :proxy_opts, :http], pool: :media)
end

View file

@ -37,7 +37,7 @@ defmodule Pleroma.HTTP do
See `Pleroma.HTTP.request/5`
"""
@spec post(Request.url(), String.t(), Request.headers(), keyword()) ::
@spec post(Request.url(), Tesla.Env.body(), Request.headers(), keyword()) ::
{:ok, Env.t()} | {:error, any()}
def post(url, body, headers \\ [], options \\ []),
do: request(:post, url, body, headers, options)
@ -56,7 +56,7 @@ defmodule Pleroma.HTTP do
`{:ok, %Tesla.Env{}}` or `{:error, error}`
"""
@spec request(method(), Request.url(), String.t(), Request.headers(), keyword()) ::
@spec request(method(), Request.url(), Tesla.Env.body(), Request.headers(), keyword()) ::
{:ok, Env.t()} | {:error, any()}
def request(method, url, body, headers, options) when is_binary(url) do
uri = URI.parse(url)

View file

@ -15,7 +15,7 @@ defmodule Pleroma.HTTP.AdapterHelper.Gun do
retry_timeout: 1_000
]
@type pool() :: :federation | :upload | :media | :default
@type pool() :: :federation | :upload | :media | :rich_media | :default
@spec options(keyword(), URI.t()) :: keyword()
def options(incoming_opts \\ [], %URI{} = uri) do

View file

@ -0,0 +1,4 @@
defmodule Pleroma.HTTPSignaturesAPI do
@callback validate_conn(conn :: Plug.Conn.t()) :: boolean
@callback signature_for_conn(conn :: Plug.Conn.t()) :: map
end

View file

@ -73,6 +73,7 @@ defmodule Pleroma.Notification do
pleroma:report
reblog
poll
status
}
def changeset(%Notification{} = notification, attrs) do
@ -280,15 +281,10 @@ defmodule Pleroma.Notification do
select: n.id
)
{:ok, %{ids: {_, notification_ids}}} =
Multi.new()
|> Multi.update_all(:ids, query, set: [seen: true, updated_at: NaiveDateTime.utc_now()])
|> Marker.multi_set_last_read_id(user, "notifications")
|> Repo.transaction()
for_user_query(user)
|> where([n], n.id in ^notification_ids)
|> Repo.all()
Multi.new()
|> Multi.update_all(:ids, query, set: [seen: true, updated_at: NaiveDateTime.utc_now()])
|> Marker.multi_set_last_read_id(user, "notifications")
|> Repo.transaction()
end
@spec read_one(User.t(), String.t()) ::
@ -299,10 +295,6 @@ defmodule Pleroma.Notification do
|> Multi.update(:update, changeset(notification, %{seen: true}))
|> Marker.multi_set_last_read_id(user, "notifications")
|> Repo.transaction()
|> case do
{:ok, %{update: notification}} -> {:ok, notification}
{:error, :update, changeset, _} -> {:error, changeset}
end
end
end
@ -384,10 +376,15 @@ defmodule Pleroma.Notification do
defp do_create_notifications(%Activity{} = activity) do
enabled_receivers = get_notified_from_activity(activity)
enabled_subscribers = get_notified_subscribers_from_activity(activity)
notifications =
Enum.map(enabled_receivers, fn user ->
create_notification(activity, user)
end)
(Enum.map(enabled_receivers, fn user ->
create_notification(activity, user)
end) ++
Enum.map(enabled_subscribers -- enabled_receivers, fn user ->
create_notification(activity, user, type: "status")
end))
|> Enum.reject(&is_nil/1)
{:ok, notifications}
@ -492,7 +489,7 @@ defmodule Pleroma.Notification do
NOTE: might be called for FAKE Activities, see ActivityPub.Utils.get_notified_from_object/1
"""
@spec get_notified_from_activity(Activity.t(), boolean()) :: {list(User.t()), list(User.t())}
@spec get_notified_from_activity(Activity.t(), boolean()) :: list(User.t())
def get_notified_from_activity(activity, local_only \\ true)
def get_notified_from_activity(%Activity{data: %{"type" => type}} = activity, local_only)
@ -520,7 +517,25 @@ defmodule Pleroma.Notification do
Enum.filter(potential_receivers, fn u -> u.ap_id in notification_enabled_ap_ids end)
end
def get_notified_from_activity(_, _local_only), do: {[], []}
def get_notified_from_activity(_, _local_only), do: []
def get_notified_subscribers_from_activity(activity, local_only \\ true)
def get_notified_subscribers_from_activity(
%Activity{data: %{"type" => "Create"}} = activity,
local_only
) do
notification_enabled_ap_ids =
[]
|> Utils.maybe_notify_subscribers(activity)
potential_receivers =
User.get_users_from_set(notification_enabled_ap_ids, local_only: local_only)
Enum.filter(potential_receivers, fn u -> u.ap_id in notification_enabled_ap_ids end)
end
def get_notified_subscribers_from_activity(_, _), do: []
# For some activities, only notify the author of the object
def get_potential_receiver_ap_ids(%{data: %{"type" => type, "object" => object_id}})
@ -563,7 +578,6 @@ defmodule Pleroma.Notification do
[]
|> Utils.maybe_notify_to_recipients(activity)
|> Utils.maybe_notify_mentioned_recipients(activity)
|> Utils.maybe_notify_subscribers(activity)
|> Utils.maybe_notify_followers(activity)
|> Enum.uniq()
end
@ -743,8 +757,9 @@ defmodule Pleroma.Notification do
|> Repo.update_all(set: [seen: true])
end
@spec send(list(Notification.t())) :: :ok
def send(notifications) do
@doc "Streams a list of notifications over websockets and web push"
@spec stream(list(Notification.t())) :: :ok
def stream(notifications) do
Enum.each(notifications, fn notification ->
Streamer.stream(["user", "user:notification"], notification)
Push.send(notification)

View file

@ -8,7 +8,7 @@ defmodule Pleroma.ReverseProxy do
~w(if-unmodified-since if-none-match) ++ @range_headers
@resp_cache_headers ~w(etag date last-modified)
@keep_resp_headers @resp_cache_headers ++
~w(content-type content-disposition content-encoding) ++
~w(content-length content-type content-disposition content-encoding) ++
~w(content-range accept-ranges vary)
@default_cache_control_header "public, max-age=1209600"
@valid_resp_codes [200, 206, 304]
@ -180,6 +180,7 @@ defmodule Pleroma.ReverseProxy do
result =
conn
|> put_resp_headers(build_resp_headers(headers, opts))
|> streaming_compat
|> send_chunked(status)
|> chunk_reply(client, opts)
@ -417,4 +418,17 @@ defmodule Pleroma.ReverseProxy do
@cachex.put(:failed_proxy_url_cache, url, true, ttl: ttl)
end
# When Cowboy handles a chunked response with a content-length header it streams
# over HTTP 1.1 instead of chunking. Bandit cannot stream over HTTP 1.1 so the header
# must be stripped or it breaks RFC compliance for Transfer Encoding: Chunked. RFC9112§6.2
#
# HTTP2 is always streamed for all adapters.
defp streaming_compat(conn) do
with Phoenix.Endpoint.Cowboy2Adapter <- Pleroma.Web.Endpoint.config(:adapter) do
conn
else
_ -> delete_resp_header(conn, "content-length")
end
end
end

View file

@ -204,7 +204,7 @@ defmodule Pleroma.ScheduledActivity do
def job_query(scheduled_activity_id) do
from(j in Oban.Job,
where: j.queue == "scheduled_activities",
where: j.queue == "federator_outgoing",
where: fragment("args ->> 'activity_id' = ?::text", ^to_string(scheduled_activity_id))
)
end

View file

@ -10,8 +10,12 @@ defmodule Pleroma.Search do
end
def search(query, options) do
search_module = Pleroma.Config.get([Pleroma.Search, :module], Pleroma.Activity)
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
end

View file

@ -28,7 +28,7 @@ defmodule Pleroma.Search.DatabaseSearch do
|> Activity.with_preloaded_object()
|> Activity.restrict_deactivated_users()
|> restrict_public(user)
|> query_with(index_type, search_query, :websearch)
|> query_with(index_type, search_query)
|> maybe_restrict_local(user)
|> maybe_restrict_author(author)
|> maybe_restrict_blocked(user)
@ -48,6 +48,15 @@ defmodule Pleroma.Search.DatabaseSearch do
@impl true
def remove_from_index(_object), do: :ok
@impl true
def create_index, do: :ok
@impl true
def drop_index, do: :ok
@impl true
def healthcheck_endpoints, do: nil
def maybe_restrict_author(query, %User{} = author) do
Activity.Queries.by_author(query, author)
end
@ -79,25 +88,7 @@ defmodule Pleroma.Search.DatabaseSearch do
)
end
defp query_with(q, :gin, search_query, :plain) do
%{rows: [[tsc]]} =
Ecto.Adapters.SQL.query!(
Pleroma.Repo,
"select current_setting('default_text_search_config')::regconfig::oid;"
)
from([a, o] in q,
where:
fragment(
"to_tsvector(?::oid::regconfig, ?->>'content') @@ plainto_tsquery(?)",
^tsc,
o.data,
^search_query
)
)
end
defp query_with(q, :gin, search_query, :websearch) do
defp query_with(q, :gin, search_query) do
%{rows: [[tsc]]} =
Ecto.Adapters.SQL.query!(
Pleroma.Repo,
@ -115,19 +106,7 @@ defmodule Pleroma.Search.DatabaseSearch do
)
end
defp query_with(q, :rum, search_query, :plain) do
from([a, o] in q,
where:
fragment(
"? @@ plainto_tsquery(?)",
o.fts_content,
^search_query
),
order_by: [fragment("? <=> now()::date", o.inserted_at)]
)
end
defp query_with(q, :rum, search_query, :websearch) do
defp query_with(q, :rum, search_query) do
from([a, o] in q,
where:
fragment(

View file

@ -0,0 +1,86 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2024 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Search.Healthcheck do
@doc """
Monitors health of search backend to control processing of events based on health and availability.
"""
use GenServer
require Logger
@queue :search_indexing
@tick :timer.seconds(5)
@timeout :timer.seconds(2)
def start_link(_) do
GenServer.start_link(__MODULE__, [], name: __MODULE__)
end
@impl true
def init(_) do
state = %{healthy: false}
{:ok, state, {:continue, :start}}
end
@impl true
def handle_continue(:start, state) do
tick()
{:noreply, state}
end
@impl true
def handle_info(:check, state) do
urls = Pleroma.Search.healthcheck_endpoints()
new_state =
if check(urls) do
Oban.resume_queue(queue: @queue)
Map.put(state, :healthy, true)
else
Oban.pause_queue(queue: @queue)
Map.put(state, :healthy, false)
end
maybe_log_state_change(state, new_state)
tick()
{:noreply, new_state}
end
@impl true
def handle_call(:state, _from, state) do
{:reply, state, state, :hibernate}
end
def state, do: GenServer.call(__MODULE__, :state)
def check([]), do: true
def check(urls) when is_list(urls) do
Enum.all?(
urls,
fn url ->
case Pleroma.HTTP.get(url, [], recv_timeout: @timeout) do
{:ok, %{status: 200}} -> true
_ -> false
end
end
)
end
def check(_), do: true
defp tick do
Process.send_after(self(), :check, @tick)
end
defp maybe_log_state_change(%{healthy: true}, %{healthy: false}) do
Logger.error("Pausing Oban queue #{@queue} due to search backend healthcheck failure")
end
defp maybe_log_state_change(%{healthy: false}, %{healthy: true}) do
Logger.info("Resuming Oban queue #{@queue} due to search backend healthcheck pass")
end
defp maybe_log_state_change(_, _), do: :ok
end

View file

@ -10,6 +10,12 @@ defmodule Pleroma.Search.Meilisearch do
@behaviour Pleroma.Search.SearchBackend
@impl true
def create_index, do: :ok
@impl true
def drop_index, do: :ok
defp meili_headers do
private_key = Config.get([Pleroma.Search.Meilisearch, :private_key])
@ -178,4 +184,15 @@ defmodule Pleroma.Search.Meilisearch do
def remove_from_index(object) do
meili_delete("/indexes/objects/documents/#{object.id}")
end
@impl true
def healthcheck_endpoints do
endpoint =
Config.get([Pleroma.Search.Meilisearch, :url])
|> URI.parse()
|> Map.put(:path, "/health")
|> URI.to_string()
[endpoint]
end
end

View file

@ -0,0 +1,182 @@
defmodule Pleroma.Search.QdrantSearch do
@behaviour Pleroma.Search.SearchBackend
import Ecto.Query
alias Pleroma.Activity
alias Pleroma.Config.Getting, as: Config
alias __MODULE__.OpenAIClient
alias __MODULE__.QdrantClient
import Pleroma.Search.Meilisearch, only: [object_to_search_data: 1]
import Pleroma.Search.DatabaseSearch, only: [maybe_fetch: 3]
@impl true
def create_index do
payload = Config.get([Pleroma.Search.QdrantSearch, :qdrant_index_configuration])
with {:ok, %{status: 200}} <- QdrantClient.put("/collections/posts", payload) do
:ok
else
e -> {:error, e}
end
end
@impl true
def drop_index do
with {:ok, %{status: 200}} <- QdrantClient.delete("/collections/posts") do
:ok
else
e -> {:error, e}
end
end
def get_embedding(text) do
with {:ok, %{body: %{"data" => [%{"embedding" => embedding}]}}} <-
OpenAIClient.post("/v1/embeddings", %{
input: text,
model: Config.get([Pleroma.Search.QdrantSearch, :openai_model])
}) do
{:ok, embedding}
else
_ ->
{:error, "Failed to get embedding"}
end
end
defp actor_from_activity(%{data: %{"actor" => actor}}) do
actor
end
defp actor_from_activity(_), do: nil
defp build_index_payload(activity, embedding) do
actor = actor_from_activity(activity)
published_at = activity.data["published"]
%{
points: [
%{
id: activity.id |> FlakeId.from_string() |> Ecto.UUID.cast!(),
vector: embedding,
payload: %{actor: actor, published_at: published_at}
}
]
}
end
defp build_search_payload(embedding, options) do
base = %{
vector: embedding,
limit: options[:limit] || 20,
offset: options[:offset] || 0
}
if author = options[:author] do
Map.put(base, :filter, %{
must: [%{key: "actor", match: %{value: author.ap_id}}]
})
else
base
end
end
@impl true
def add_to_index(activity) do
# This will only index public or unlisted notes
maybe_search_data = object_to_search_data(activity.object)
if activity.data["type"] == "Create" and maybe_search_data do
with {:ok, embedding} <- get_embedding(maybe_search_data.content),
{:ok, %{status: 200}} <-
QdrantClient.put(
"/collections/posts/points",
build_index_payload(activity, embedding)
) do
:ok
else
e -> {:error, e}
end
else
:ok
end
end
@impl true
def remove_from_index(object) do
activity = Activity.get_by_object_ap_id_with_object(object.data["id"])
id = activity.id |> FlakeId.from_string() |> Ecto.UUID.cast!()
with {:ok, %{status: 200}} <-
QdrantClient.post("/collections/posts/points/delete", %{"points" => [id]}) do
:ok
else
e -> {:error, e}
end
end
@impl true
def search(user, original_query, options) do
query = "Represent this sentence for searching relevant passages: #{original_query}"
with {:ok, embedding} <- get_embedding(query),
{:ok, %{body: %{"result" => result}}} <-
QdrantClient.post(
"/collections/posts/points/search",
build_search_payload(embedding, options)
) do
ids =
Enum.map(result, fn %{"id" => id} ->
Ecto.UUID.dump!(id)
end)
from(a in Activity, where: a.id in ^ids)
|> Activity.with_preloaded_object()
|> Activity.restrict_deactivated_users()
|> Ecto.Query.order_by([a], fragment("array_position(?, ?)", ^ids, a.id))
|> Pleroma.Repo.all()
|> maybe_fetch(user, original_query)
else
_ ->
[]
end
end
@impl true
def healthcheck_endpoints do
qdrant_health =
Config.get([Pleroma.Search.QdrantSearch, :qdrant_url])
|> URI.parse()
|> Map.put(:path, "/healthz")
|> URI.to_string()
openai_health = Config.get([Pleroma.Search.QdrantSearch, :openai_healthcheck_url])
[qdrant_health, openai_health] |> Enum.filter(& &1)
end
end
defmodule Pleroma.Search.QdrantSearch.OpenAIClient do
use Tesla
alias Pleroma.Config.Getting, as: Config
plug(Tesla.Middleware.BaseUrl, Config.get([Pleroma.Search.QdrantSearch, :openai_url]))
plug(Tesla.Middleware.JSON)
plug(Tesla.Middleware.Headers, [
{"Authorization",
"Bearer #{Pleroma.Config.get([Pleroma.Search.QdrantSearch, :openai_api_key])}"}
])
end
defmodule Pleroma.Search.QdrantSearch.QdrantClient do
use Tesla
alias Pleroma.Config.Getting, as: Config
plug(Tesla.Middleware.BaseUrl, Config.get([Pleroma.Search.QdrantSearch, :qdrant_url]))
plug(Tesla.Middleware.JSON)
plug(Tesla.Middleware.Headers, [
{"api-key", Pleroma.Config.get([Pleroma.Search.QdrantSearch, :qdrant_api_key])}
])
end

View file

@ -21,4 +21,22 @@ defmodule Pleroma.Search.SearchBackend do
from index.
"""
@callback remove_from_index(object :: Pleroma.Object.t()) :: :ok | {:error, any()}
@doc """
Create the index
"""
@callback create_index() :: :ok | {:error, any()}
@doc """
Drop the index
"""
@callback drop_index() :: :ok | {:error, any()}
@doc """
Healthcheck endpoints of search backend infrastructure to monitor for controlling
processing of jobs in the Oban queue.
It is expected a 200 response is healthy and other responses are unhealthy.
"""
@callback healthcheck_endpoints :: list() | nil
end

View file

@ -44,8 +44,7 @@ defmodule Pleroma.Signature do
defp remove_suffix(uri, []), do: uri
def fetch_public_key(conn) do
with %{"keyId" => kid} <- HTTPSignatures.signature_for_conn(conn),
{:ok, actor_id} <- key_id_to_actor_id(kid),
with {:ok, actor_id} <- get_actor_id(conn),
{:ok, public_key} <- User.get_public_key_for_ap_id(actor_id) do
{:ok, public_key}
else
@ -55,8 +54,7 @@ defmodule Pleroma.Signature do
end
def refetch_public_key(conn) do
with %{"keyId" => kid} <- HTTPSignatures.signature_for_conn(conn),
{:ok, actor_id} <- key_id_to_actor_id(kid),
with {:ok, actor_id} <- get_actor_id(conn),
{:ok, _user} <- ActivityPub.make_user_from_ap_id(actor_id),
{:ok, public_key} <- User.get_public_key_for_ap_id(actor_id) do
{:ok, public_key}
@ -66,6 +64,16 @@ defmodule Pleroma.Signature do
end
end
def get_actor_id(conn) do
with %{"keyId" => kid} <- HTTPSignatures.signature_for_conn(conn),
{:ok, actor_id} <- key_id_to_actor_id(kid) do
{:ok, actor_id}
else
e ->
{:error, e}
end
end
def sign(%User{keys: keys} = user, headers) do
with {:ok, private_key, _} <- Keys.keys_from_pem(keys) do
HTTPSignatures.sign(private_key, user.ap_id <> "#main-key", headers)

View file

@ -239,8 +239,12 @@ defmodule Pleroma.Upload do
""
end
[base_url, path]
|> Path.join()
if String.contains?(base_url, Pleroma.Uploaders.IPFS.placeholder()) do
String.replace(base_url, Pleroma.Uploaders.IPFS.placeholder(), path)
else
[base_url, path]
|> Path.join()
end
end
defp url_from_spec(_upload, _base_url, {:url, url}), do: url
@ -277,6 +281,9 @@ defmodule Pleroma.Upload do
Path.join([upload_base_url, bucket_with_namespace])
end
Pleroma.Uploaders.IPFS ->
@config_impl.get([Pleroma.Uploaders.IPFS, :get_gateway_url])
_ ->
public_endpoint || upload_base_url || Pleroma.Web.Endpoint.url() <> "/media/"
end

View file

@ -9,8 +9,6 @@ defmodule Pleroma.Upload.Filter.Exiftool.StripLocation do
"""
@behaviour Pleroma.Upload.Filter
@spec filter(Pleroma.Upload.t()) :: {:ok, any()} | {:error, String.t()}
# Formats not compatible with exiftool at this time
def filter(%Pleroma.Upload{content_type: "image/heic"}), do: {:ok, :noop}
def filter(%Pleroma.Upload{content_type: "image/webp"}), do: {:ok, :noop}

View file

@ -38,7 +38,6 @@ defmodule Pleroma.Upload.Filter.Mogrifun do
[{"fill", "yellow"}, {"tint", "40"}]
]
@spec filter(Pleroma.Upload.t()) :: {:ok, atom()} | {:error, String.t()}
def filter(%Pleroma.Upload{tempfile: file, content_type: "image" <> _}) do
try do
Filter.Mogrify.do_filter(file, [Enum.random(@filters)])

View file

@ -8,7 +8,6 @@ defmodule Pleroma.Upload.Filter.Mogrify do
@type conversion :: action :: String.t() | {action :: String.t(), opts :: String.t()}
@type conversions :: conversion() | [conversion()]
@spec filter(Pleroma.Upload.t()) :: {:ok, :atom} | {:error, String.t()}
def filter(%Pleroma.Upload{tempfile: file, content_type: "image" <> _}) do
try do
do_filter(file, Pleroma.Config.get!([__MODULE__, :args]))

View file

@ -0,0 +1,72 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2022 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Uploaders.IPFS do
@behaviour Pleroma.Uploaders.Uploader
require Logger
alias Tesla.Multipart
@api_add "/api/v0/add"
@api_delete "/api/v0/files/rm"
@config_impl Application.compile_env(:pleroma, [__MODULE__, :config_impl], Pleroma.Config)
@placeholder "{CID}"
def placeholder, do: @placeholder
@impl true
def get_file(file) do
b_url = Pleroma.Upload.base_url()
if String.contains?(b_url, @placeholder) do
{:ok, {:url, String.replace(b_url, @placeholder, URI.decode(file))}}
else
{:error, "IPFS Get URL doesn't contain 'cid' placeholder"}
end
end
@impl true
def put_file(%Pleroma.Upload{tempfile: tempfile}) do
mp =
Multipart.new()
|> Multipart.add_content_type_param("charset=utf-8")
|> Multipart.add_file(tempfile)
endpoint = ipfs_endpoint(@api_add)
with {:ok, %{body: body}} when is_binary(body) <-
Pleroma.HTTP.post(endpoint, mp, [], params: ["cid-version": "1"], pool: :upload),
{_, {:ok, decoded}} <- {:json, Jason.decode(body)},
{_, true} <- {:hash, Map.has_key?(decoded, "Hash")} do
{:ok, {:file, decoded["Hash"]}}
else
{:hash, false} ->
{:error, "JSON doesn't contain Hash key"}
{:json, error} ->
Logger.error("#{__MODULE__}: #{inspect(error)}")
{:error, "JSON decode failed"}
error ->
Logger.error("#{__MODULE__}: #{inspect(error)}")
{:error, "IPFS Gateway upload failed"}
end
end
@impl true
def delete_file(file) do
endpoint = ipfs_endpoint(@api_delete)
case Pleroma.HTTP.post(endpoint, "", [], params: [arg: file]) do
{:ok, %{status: 204}} -> :ok
error -> {:error, inspect(error)}
end
end
defp ipfs_endpoint(path) do
URI.parse(@config_impl.get([__MODULE__, :post_gateway_url]))
|> Map.put(:path, path)
|> URI.to_string()
end
end

View file

@ -1404,6 +1404,40 @@ defmodule Pleroma.User do
|> Repo.all()
end
@spec get_familiar_followers_query(User.t(), User.t(), pos_integer() | nil) :: Ecto.Query.t()
def get_familiar_followers_query(%User{} = user, %User{} = current_user, nil) do
friends =
get_friends_query(current_user)
|> where([u], not u.hide_follows)
|> select([u], u.id)
User.Query.build(%{is_active: true})
|> where([u], u.id not in ^[user.id, current_user.id])
|> join(:inner, [u], r in FollowingRelationship,
as: :followers_relationships,
on: r.following_id == ^user.id and r.follower_id == u.id
)
|> where([followers_relationships: r], r.state == ^:follow_accept)
|> where([followers_relationships: r], r.follower_id in subquery(friends))
end
def get_familiar_followers_query(%User{} = user, %User{} = current_user, page) do
user
|> get_familiar_followers_query(current_user, nil)
|> User.Query.paginate(page, 20)
end
@spec get_familiar_followers_query(User.t(), User.t()) :: Ecto.Query.t()
def get_familiar_followers_query(%User{} = user, %User{} = current_user),
do: get_familiar_followers_query(user, current_user, nil)
@spec get_familiar_followers(User.t(), User.t(), pos_integer() | nil) :: {:ok, list(User.t())}
def get_familiar_followers(%User{} = user, %User{} = current_user, page \\ nil) do
user
|> get_familiar_followers_query(current_user, page)
|> Repo.all()
end
def increase_note_count(%User{} = user) do
User
|> where(id: ^user.id)
@ -2019,7 +2053,8 @@ defmodule Pleroma.User do
%{scheme: scheme, userinfo: nil, host: host}
when not_empty_string(host) and scheme in ["http", "https"] <-
URI.parse(value),
{:not_idn, true} <- {:not_idn, to_string(:idna.encode(host)) == host},
{:not_idn, true} <-
{:not_idn, match?(^host, to_string(:idna.encode(to_charlist(host))))},
"me" <- Pleroma.Web.RelMe.maybe_put_rel_me(value, profile_urls) do
CommonUtils.to_masto_date(NaiveDateTime.utc_now())
else
@ -2693,7 +2728,7 @@ defmodule Pleroma.User do
end
end
@spec add_to_block(User.t(), User.t()) ::
@spec remove_from_block(User.t(), User.t()) ::
{:ok, UserRelationship.t()} | {:ok, nil} | {:error, Ecto.Changeset.t()}
defp remove_from_block(%User{} = user, %User{} = blocked) do
with {:ok, relationship} <- UserRelationship.delete_block(user, blocked) do

View file

@ -201,7 +201,7 @@ defmodule Pleroma.Web.ActivityPub.ActivityPub do
def notify_and_stream(activity) do
{:ok, notifications} = Notification.create_notifications(activity)
Notification.send(notifications)
Notification.stream(notifications)
original_activity =
case activity do
@ -979,8 +979,9 @@ defmodule Pleroma.Web.ActivityPub.ActivityPub do
defp restrict_replies(query, %{exclude_replies: true}) do
from(
[_activity, object] in query,
where: fragment("?->>'inReplyTo' is null", object.data)
[activity, object] in query,
where:
fragment("?->>'inReplyTo' is null or ?->>'type' = 'Announce'", object.data, activity.data)
)
end
@ -1793,24 +1794,25 @@ defmodule Pleroma.Web.ActivityPub.ActivityPub do
end
end
def pinned_fetch_task(nil), do: nil
def pinned_fetch_task(%{pinned_objects: pins}) do
if Enum.all?(pins, fn {ap_id, _} ->
Object.get_cached_by_ap_id(ap_id) ||
match?({:ok, _object}, Fetcher.fetch_object_from_id(ap_id))
end) do
:ok
else
:error
end
def enqueue_pin_fetches(%{pinned_objects: pins}) do
# enqueue a task to fetch all pinned objects
Enum.each(pins, fn {ap_id, _} ->
if is_nil(Object.get_cached_by_ap_id(ap_id)) do
Pleroma.Workers.RemoteFetcherWorker.enqueue("fetch_remote", %{
"id" => ap_id,
"depth" => 1
})
end
end)
end
def enqueue_pin_fetches(_), do: nil
def make_user_from_ap_id(ap_id, additional \\ []) do
user = User.get_cached_by_ap_id(ap_id)
with {:ok, data} <- fetch_and_prepare_user_from_ap_id(ap_id, additional) do
{:ok, _pid} = Task.start(fn -> pinned_fetch_task(data) end)
enqueue_pin_fetches(data)
if user do
user

View file

@ -52,6 +52,7 @@ defmodule Pleroma.Web.ActivityPub.ActivityPubController do
when action in [:activity, :object]
)
plug(:log_inbox_metadata when action in [:inbox])
plug(:set_requester_reachable when action in [:inbox])
plug(:relay_active? when action in [:relay])
@ -521,6 +522,13 @@ defmodule Pleroma.Web.ActivityPub.ActivityPubController do
conn
end
defp log_inbox_metadata(%{params: %{"actor" => actor, "type" => type}} = conn, _) do
Logger.metadata(actor: actor, type: type)
conn
end
defp log_inbox_metadata(conn, _), do: conn
def upload_media(%{assigns: %{user: %User{} = user}} = conn, %{"file" => file} = data) do
with {:ok, object} <-
ActivityPub.upload(

View file

@ -0,0 +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.AntiMentionSpamPolicy do
alias Pleroma.Config
alias Pleroma.User
require Pleroma.Constants
@behaviour Pleroma.Web.ActivityPub.MRF.Policy
defp user_has_posted?(%User{} = u), do: u.note_count > 0
defp user_has_age?(%User{} = u) do
user_age_limit = Config.get([:mrf_antimentionspam, :user_age_limit], 30_000)
diff = NaiveDateTime.utc_now() |> NaiveDateTime.diff(u.inserted_at, :millisecond)
diff >= user_age_limit
end
defp good_reputation?(%User{} = u) do
user_has_age?(u) and user_has_posted?(u)
end
# copied from HellthreadPolicy
defp get_recipient_count(message) do
recipients = (message["to"] || []) ++ (message["cc"] || [])
follower_collection =
User.get_cached_by_ap_id(message["actor"] || message["attributedTo"]).follower_address
if Enum.member?(recipients, Pleroma.Constants.as_public()) do
recipients =
recipients
|> List.delete(Pleroma.Constants.as_public())
|> List.delete(follower_collection)
{:public, length(recipients)}
else
recipients =
recipients
|> List.delete(follower_collection)
{:not_public, length(recipients)}
end
end
defp object_has_recipients?(%{"object" => object} = activity) do
{_, object_count} = get_recipient_count(object)
{_, activity_count} = get_recipient_count(activity)
object_count + activity_count > 0
end
defp object_has_recipients?(object) do
{_, count} = get_recipient_count(object)
count > 0
end
@impl true
def filter(%{"type" => "Create", "actor" => actor} = activity) do
with {:ok, %User{local: false} = u} <- User.get_or_fetch_by_ap_id(actor),
{:has_mentions, true} <- {:has_mentions, object_has_recipients?(activity)},
{:good_reputation, true} <- {:good_reputation, good_reputation?(u)} do
{:ok, activity}
else
{:ok, %User{local: true}} ->
{:ok, activity}
{:has_mentions, false} ->
{:ok, activity}
{:good_reputation, false} ->
{:reject, "[AntiMentionSpamPolicy] User rejected"}
{:error, _} ->
{:reject, "[AntiMentionSpamPolicy] Failed to get or fetch user by ap_id"}
e ->
{:reject, "[AntiMentionSpamPolicy] Unhandled error #{inspect(e)}"}
end
end
# in all other cases, pass through
def filter(message), do: {:ok, message}
@impl true
def describe, do: {:ok, %{}}
end

View file

@ -0,0 +1,146 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2024 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Web.ActivityPub.MRF.DNSRBLPolicy do
@moduledoc """
Dynamic activity filtering based on an RBL database
This MRF makes queries to a custom DNS server which will
respond with values indicating the classification of the domain
the activity originated from. This method has been widely used
in the email anti-spam industry for very fast reputation checks.
e.g., if the DNS response is 127.0.0.1 or empty, the domain is OK
Other values such as 127.0.0.2 may be used for specific classifications.
Information for why the host is blocked can be stored in a corresponding TXT record.
This method is fail-open so if the queries fail the activites are accepted.
An example of software meant for this purpsoe is rbldnsd which can be found
at http://www.corpit.ru/mjt/rbldnsd.html or mirrored at
https://git.pleroma.social/feld/rbldnsd
It is highly recommended that you run your own copy of rbldnsd and use an
external mechanism to sync/share the contents of the zone file. This is
important to keep the latency on the queries as low as possible and prevent
your DNS server from being attacked so it fails and content is permitted.
"""
@behaviour Pleroma.Web.ActivityPub.MRF.Policy
alias Pleroma.Config
require Logger
@query_retries 1
@query_timeout 500
@impl true
def filter(%{"actor" => actor} = object) do
actor_info = URI.parse(actor)
with {:ok, object} <- check_rbl(actor_info, object) do
{:ok, object}
else
_ -> {:reject, "[DNSRBLPolicy]"}
end
end
@impl true
def filter(object), do: {:ok, object}
@impl true
def describe do
mrf_dnsrbl =
Config.get(:mrf_dnsrbl)
|> Enum.into(%{})
{:ok, %{mrf_dnsrbl: mrf_dnsrbl}}
end
@impl true
def config_description do
%{
key: :mrf_dnsrbl,
related_policy: "Pleroma.Web.ActivityPub.MRF.DNSRBLPolicy",
label: "MRF DNSRBL",
description: "DNS RealTime Blackhole Policy",
children: [
%{
key: :nameserver,
type: {:string},
description: "DNSRBL Nameserver to Query (IP or hostame)",
suggestions: ["127.0.0.1"]
},
%{
key: :port,
type: {:string},
description: "Nameserver port",
suggestions: ["53"]
},
%{
key: :zone,
type: {:string},
description: "Root zone for querying",
suggestions: ["bl.pleroma.com"]
}
]
}
end
defp check_rbl(%{host: actor_host}, object) do
with false <- match?(^actor_host, Pleroma.Web.Endpoint.host()),
zone when not is_nil(zone) <- Keyword.get(Config.get([:mrf_dnsrbl]), :zone) do
query =
Enum.join([actor_host, zone], ".")
|> String.to_charlist()
rbl_response = rblquery(query)
if Enum.empty?(rbl_response) do
{:ok, object}
else
Task.start(fn ->
reason =
case rblquery(query, :txt) do
[[result]] -> result
_ -> "undefined"
end
Logger.warning(
"DNSRBL Rejected activity from #{actor_host} for reason: #{inspect(reason)}"
)
end)
:error
end
else
_ -> {:ok, object}
end
end
defp get_rblhost_ip(rblhost) do
case rblhost |> String.to_charlist() |> :inet_parse.address() do
{:ok, _} -> rblhost |> String.to_charlist() |> :inet_parse.address()
_ -> {:ok, rblhost |> String.to_charlist() |> :inet_res.lookup(:in, :a) |> Enum.random()}
end
end
defp rblquery(query, type \\ :a) do
config = Config.get([:mrf_dnsrbl])
case get_rblhost_ip(config[:nameserver]) do
{:ok, rblnsip} ->
:inet_res.lookup(query, :in, type,
nameservers: [{rblnsip, config[:port]}],
timeout: @query_timeout,
retry: @query_retries
)
_ ->
[]
end
end
end

View file

@ -11,11 +11,6 @@ defmodule Pleroma.Web.ActivityPub.MRF.MediaProxyWarmingPolicy do
require Logger
@adapter_options [
pool: :media,
recv_timeout: 10_000
]
@impl true
def history_awareness, do: :auto
@ -27,17 +22,14 @@ defmodule Pleroma.Web.ActivityPub.MRF.MediaProxyWarmingPolicy do
Logger.debug("Prefetching #{inspect(url)} as #{inspect(prefetch_url)}")
if Pleroma.Config.get(:env) == :test do
fetch(prefetch_url)
else
ConcurrentLimiter.limit(__MODULE__, fn ->
Task.start(fn -> fetch(prefetch_url) end)
end)
end
fetch(prefetch_url)
end
end
defp fetch(url), do: HTTP.get(url, [], @adapter_options)
defp fetch(url) do
http_client_opts = Pleroma.Config.get([:media_proxy, :proxy_opts, :http], pool: :media)
HTTP.get(url, [], http_client_opts)
end
defp preload(%{"object" => %{"attachment" => attachments}} = _message) do
Enum.each(attachments, fn

View file

@ -0,0 +1,265 @@
# Pleroma: A lightweight social networking server
# Copyright © 2017-2021 Pleroma Authors <https://pleroma.social/>
# SPDX-License-Identifier: AGPL-3.0-only
defmodule Pleroma.Web.ActivityPub.MRF.NsfwApiPolicy do
@moduledoc """
Hide, delete, or mark sensitive NSFW content with artificial intelligence.
Requires a NSFW API server, configured like so:
config :pleroma, Pleroma.Web.ActivityPub.MRF.NsfwMRF,
url: "http://127.0.0.1:5000/",
threshold: 0.7,
mark_sensitive: true,
unlist: false,
reject: false
The NSFW API server must implement an HTTP endpoint like this:
curl http://localhost:5000/?url=https://fedi.com/images/001.jpg
Returning a response like this:
{"score", 0.314}
Where a score is 0-1, with `1` being definitely NSFW.
A good API server is here: https://github.com/EugenCepoi/nsfw_api
You can run it with Docker with a one-liner:
docker run -it -p 127.0.0.1:5000:5000/tcp --env PORT=5000 eugencepoi/nsfw_api:latest
Options:
- `url`: Base URL of the API server. Default: "http://127.0.0.1:5000/"
- `threshold`: Lowest score to take action on. Default: `0.7`
- `mark_sensitive`: Mark sensitive all detected NSFW content? Default: `true`
- `unlist`: Unlist all detected NSFW content? Default: `false`
- `reject`: Reject all detected NSFW content (takes precedence)? Default: `false`
"""
alias Pleroma.Config
alias Pleroma.Constants
alias Pleroma.HTTP
alias Pleroma.User
require Logger
require Pleroma.Constants
@behaviour Pleroma.Web.ActivityPub.MRF.Policy
@policy :mrf_nsfw_api
def build_request_url(url) do
Config.get([@policy, :url])
|> URI.parse()
|> fix_path()
|> Map.put(:query, "url=#{url}")
|> URI.to_string()
end
def parse_url(url) do
request = build_request_url(url)
with {:ok, %Tesla.Env{body: body}} <- HTTP.get(request) do
Jason.decode(body)
else
error ->
Logger.warning("""
[NsfwApiPolicy]: The API server failed. Skipping.
#{inspect(error)}
""")
error
end
end
def check_url_nsfw(url) when is_binary(url) do
threshold = Config.get([@policy, :threshold])
case parse_url(url) do
{:ok, %{"score" => score}} when score >= threshold ->
{:nsfw, %{url: url, score: score, threshold: threshold}}
{:ok, %{"score" => score}} ->
{:sfw, %{url: url, score: score, threshold: threshold}}
_ ->
{:sfw, %{url: url, score: nil, threshold: threshold}}
end
end
def check_url_nsfw(%{"href" => url}) when is_binary(url) do
check_url_nsfw(url)
end
def check_url_nsfw(url) do
threshold = Config.get([@policy, :threshold])
{:sfw, %{url: url, score: nil, threshold: threshold}}
end
def check_attachment_nsfw(%{"url" => urls} = attachment) when is_list(urls) do
if Enum.all?(urls, &match?({:sfw, _}, check_url_nsfw(&1))) do
{:sfw, attachment}
else
{:nsfw, attachment}
end
end
def check_attachment_nsfw(%{"url" => url} = attachment) when is_binary(url) do
case check_url_nsfw(url) do
{:sfw, _} -> {:sfw, attachment}
{:nsfw, _} -> {:nsfw, attachment}
end
end
def check_attachment_nsfw(attachment), do: {:sfw, attachment}
def check_object_nsfw(%{"attachment" => attachments} = object) when is_list(attachments) do
if Enum.all?(attachments, &match?({:sfw, _}, check_attachment_nsfw(&1))) do
{:sfw, object}
else
{:nsfw, object}
end
end
def check_object_nsfw(%{"object" => %{} = child_object} = object) do
case check_object_nsfw(child_object) do
{:sfw, _} -> {:sfw, object}
{:nsfw, _} -> {:nsfw, object}
end
end
def check_object_nsfw(object), do: {:sfw, object}
@impl true
def filter(object) do
with {:sfw, object} <- check_object_nsfw(object) do
{:ok, object}
else
{:nsfw, _data} -> handle_nsfw(object)
_ -> {:reject, "NSFW: Attachment rejected"}
end
end
defp handle_nsfw(object) do
if Config.get([@policy, :reject]) do
{:reject, object}
else
{:ok,
object
|> maybe_unlist()
|> maybe_mark_sensitive()}
end
end
defp maybe_unlist(object) do
if Config.get([@policy, :unlist]) do
unlist(object)
else
object
end
end
defp maybe_mark_sensitive(object) do
if Config.get([@policy, :mark_sensitive]) do
mark_sensitive(object)
else
object
end
end
def unlist(%{"to" => to, "cc" => cc, "actor" => actor} = object) do
with %User{} = user <- User.get_cached_by_ap_id(actor) do
to =
[user.follower_address | to]
|> List.delete(Constants.as_public())
|> Enum.uniq()
cc =
[Constants.as_public() | cc]
|> List.delete(user.follower_address)
|> Enum.uniq()
object
|> Map.put("to", to)
|> Map.put("cc", cc)
else
_ -> raise "[NsfwApiPolicy]: Could not find user #{actor}"
end
end
def mark_sensitive(%{"object" => child_object} = object) when is_map(child_object) do
Map.put(object, "object", mark_sensitive(child_object))
end
def mark_sensitive(object) when is_map(object) do
tags = (object["tag"] || []) ++ ["nsfw"]
object
|> Map.put("tag", tags)
|> Map.put("sensitive", true)
end
# Hackney needs a trailing slash
defp fix_path(%URI{path: path} = uri) when is_binary(path) do
path = String.trim_trailing(path, "/") <> "/"
Map.put(uri, :path, path)
end
defp fix_path(%URI{path: nil} = uri), do: Map.put(uri, :path, "/")
@impl true
def describe do
options = %{
threshold: Config.get([@policy, :threshold]),
mark_sensitive: Config.get([@policy, :mark_sensitive]),
unlist: Config.get([@policy, :unlist]),
reject: Config.get([@policy, :reject])
}
{:ok, %{@policy => options}}
end
@impl true
def config_description do
%{
key: @policy,
related_policy: to_string(__MODULE__),
label: "NSFW API Policy",
description:
"Hide, delete, or mark sensitive NSFW content with artificial intelligence. Requires running an external API server.",
children: [
%{
key: :url,
type: :string,
description: "Base URL of the API server.",
suggestions: ["http://127.0.0.1:5000/"]
},
%{
key: :threshold,
type: :float,
description: "Lowest score to take action on. Between 0 and 1.",
suggestions: [0.7]
},
%{
key: :mark_sensitive,
type: :boolean,
description: "Mark sensitive all detected NSFW content?",
suggestions: [true]
},
%{
key: :unlist,
type: :boolean,
description: "Unlist sensitive all detected NSFW content?",
suggestions: [false]
},
%{
key: :reject,
type: :boolean,
description: "Reject sensitive all detected NSFW content (takes precedence)?",
suggestions: [false]
}
]
}
end
end

View file

@ -15,6 +15,7 @@ defmodule Pleroma.Web.ActivityPub.ObjectValidators.AttachmentValidator do
field(:type, :string, default: "Link")
field(:mediaType, ObjectValidators.MIME, default: "application/octet-stream")
field(:name, :string)
field(:summary, :string)
field(:blurhash, :string)
embeds_many :url, UrlObjectValidator, primary_key: false do
@ -44,7 +45,7 @@ defmodule Pleroma.Web.ActivityPub.ObjectValidators.AttachmentValidator do
|> fix_url()
struct
|> cast(data, [:id, :type, :mediaType, :name, :blurhash])
|> cast(data, [:id, :type, :mediaType, :name, :summary, :blurhash])
|> cast_embed(:url, with: &url_changeset/2, required: true)
|> validate_inclusion(:type, ~w[Link Document Audio Image Video])
|> validate_required([:type, :mediaType])

View file

@ -592,9 +592,9 @@ defmodule Pleroma.Web.ActivityPub.SideEffects do
with {:ok, _} <- Repo.delete(object), do: :ok
end
defp send_notifications(meta) do
defp stream_notifications(meta) do
Keyword.get(meta, :notifications, [])
|> Notification.send()
|> Notification.stream()
meta
end
@ -625,7 +625,7 @@ defmodule Pleroma.Web.ActivityPub.SideEffects do
@impl true
def handle_after_transaction(meta) do
meta
|> send_notifications()
|> stream_notifications()
|> send_streamables()
end
end

View file

@ -912,9 +912,11 @@ defmodule Pleroma.Web.ActivityPub.Transmogrifier do
def add_emoji_tags(object), do: object
defp build_emoji_tag({name, url}) do
def build_emoji_tag({name, url}) do
url = URI.encode(url)
%{
"icon" => %{"url" => "#{URI.encode(url)}", "type" => "Image"},
"icon" => %{"url" => "#{url}", "type" => "Image"},
"name" => ":" <> name <> ":",
"type" => "Emoji",
"updated" => "1970-01-01T00:00:00Z",

View file

@ -18,6 +18,8 @@ defmodule Pleroma.Web.ApiSpec.CastAndValidate do
alias OpenApiSpex.Plug.PutApiSpec
alias Plug.Conn
require Logger
@impl Plug
def init(opts) do
opts
@ -51,6 +53,10 @@ defmodule Pleroma.Web.ApiSpec.CastAndValidate do
conn
{:error, reason} ->
Logger.error(
"Strict ApiSpec: request denied to #{conn.request_path} with params #{inspect(conn.params)}"
)
opts = render_error.init(reason)
conn

Some files were not shown because too many files have changed in this diff Show more