umputun / tg-spam
84%

DEFAULT BRANCH: master
Repo Added
Files 42
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

LAST BUILD ON BRANCH master
branch: SELECT
CHANGE BRANCH
x

coverage: 83.723% (+0.02%) from 83.704%
29400565840

push

github

web-flow 
feat: detect spam in external replies (#429)

adds an opt-in meta check that flags a message replying to a message from another chat (Telegram's `external_reply`). Off by default, mirrors the existing `--meta.forward` check.

**why:** a spam wave hit the radio-t chat: premium accounts posting a short fake testimonial like "Пришло быстро!" that *replies* to a post in an external scam channel ("Билайн Информация") and quotes "Написать менеджеру: @Beeline". The bot only saw the tiny comment, so the classifier and similarity stayed at ham and the LLM was inconsistent on the wording. The real signal, a reply to a known scam channel, lived in `external_reply`, which no check ever inspected.

**what it does:** a message whose `external_reply` points at a genuinely different chat is flagged. The flag is gated on the referenced chat being external: `msg.ExternalReply != nil && (Chat == nil || Chat.ID != msg.Chat.ID)`. `external_reply` also fires for a reply across forum topics within the *same* chat, so keying on presence alone would ban legitimate forum replies. The chat-id guard keeps it to real cross-chat replies, and the observed spam still matches (its origin chat differs from the group).

`ExternalReplyInfo` carries no text or caption in the Bot API, so the check keys on presence, not content. The quoted snippet a spammer selects still arrives via `msg.Quote` and hits the existing content checks as before.

wired through the standard meta path: `--meta.external-reply` / `META_EXTERNAL_REPLY`, the `IsMetaEnabled` aggregate, the web settings UI, the Lua plugin meta table (`req.meta.has_external_reply`), and the listener empty-text guard for parity with forward.

34 of 37 new or added lines in 10 files covered. (91.89%)

8909 of 10641 relevant lines covered (83.72%)

236.07 hits per line

Relevant lines Covered
10641 RELEVANT LINES 8909 COVERED LINES
236.07 HITS PER LINE
Source Files on master

Recent builds

Builds Branch Commit Type Ran Committer Via Coverage
29400565840 master feat: detect spam in external replies (#429) adds an opt-in meta check that flags a message replying to a message from another chat (Telegram's `external_reply`). Off by default, mirrors the existing `--meta.forward` check. **why:** a spam wave ... push web-flow github
83.72
29399903250 feat/external-reply-meta test: widen listener_test Do(ctx) deadline from 500ms to 10s the 25 Do(ctx) subtests use a context deadline purely as a safety bound and all assert "telegram update chan closed"; none test the deadline path. under CI load (race detector + postgre... Pull #429 umputun github
83.75
29399171995 feat/external-reply-meta fix(review-loop): final cleanup pass — addressed 1 finding - [docs] [app/webapi/config.go:293,299] correct stale "12" meta-field count to 13 after adding ExternalReply to IsMetaEnabled Pull #429 umputun github
83.73
29379114337 refs/tags/v1.26.0 feat: detect spam in rich messages (telegram articles) (#428) flatten telegram rich_message (bot api 10.2 article compose type) into message.Text in transform() so previously-invisible article spam runs through the existing detection pipeline. bu... push web-flow github
83.69
29372019250 master feat: detect spam in rich messages (telegram articles) (#428) flatten telegram rich_message (bot api 10.2 article compose type) into message.Text in transform() so previously-invisible article spam runs through the existing detection pipeline. bu... push web-flow github
83.7
29371602235 feature/richmsg-detection fix(review-loop): iteration 1 addressed 4 findings - [major] [events.go] concatenate inline rich-message spans and newline-separate only blocks, so a word split across formatting nodes rejoins instead of breaking substring/stopword matching and d... Pull #428 umputun github
83.71
29371604433 feature/richmsg-detection fix(review-loop): iteration 1 addressed 4 findings - [major] [events.go] concatenate inline rich-message spans and newline-separate only blocks, so a word split across formatting nodes rejoins instead of breaking substring/stopword matching and d... Pull #428 umputun github
83.69
29367402134 feature/richmsg-detection feat: detect spam in rich messages (telegram articles) telegram's client "article" compose type arrives as Message.rich_message (bot api 10.2) with an empty msg.Text, so its content bypassed every spam check. flatten rich_message into message.Tex... Pull #428 umputun github
83.68
29366044979 master chore(deps): update telegram-bot-api to bot api 10.2 bump github.com/OvyFlash/telegram-bot-api to the 10.2 release which adds rich_message (article) support. adjust call sites and test fixtures for the int->int64 widening of message time fields (... push umputun github
83.63
29293948129 master feat: prohibited languages spam detector (#427) Add a spam detector that blocks messages written in prohibited Unicode scripts. Configure a blocklist of scripts (chinese=Han, russian=Cyrillic, etc.); a message with enough letters of a banned scri... push web-flow github
83.63
See All Builds (1434)