Merge pull request 'Post translation via Interpreter (milestone 143)' (#201) from dev into main
Build images / sign-extension (push) Successful in 3s
CI / lint (push) Successful in 3s
Build images / build-agent (push) Successful in 7s
Build images / build-ml (push) Successful in 8s
Build images / build-web (push) Successful in 11s
CI / frontend-build (push) Successful in 19s
CI / backend-lint-and-test (push) Successful in 32s
CI / integration (push) Successful in 3m44s

This commit was merged in pull request #201.
This commit is contained in:
2026-07-07 13:05:18 -04:00
17 changed files with 975 additions and 7 deletions
+73
View File
@@ -0,0 +1,73 @@
"""post-text translation via Interpreter (milestone 143) — Post columns + settings
Post gains the translated title/description + the detected source language,
Interpreter engine_version (cache key), and translated_at — filled by the
translate sweep. ImportSettings gains translation_enabled (OFF by default),
interpreter_base_url (EMPTY — the operator sets their own, behind a reverse
proxy), and translation_target_lang (en).
Revision ID: 0083
Revises: 0082
Create Date: 2026-07-07
"""
from typing import Sequence, Union
import sqlalchemy as sa
from alembic import op
revision: str = "0083"
down_revision: Union[str, None] = "0082"
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
op.add_column(
"post", sa.Column("post_title_translated", sa.Text(), nullable=True)
)
op.add_column(
"post", sa.Column("description_translated", sa.Text(), nullable=True)
)
op.add_column(
"post",
sa.Column("translated_source_lang", sa.String(8), nullable=True),
)
op.add_column(
"post",
sa.Column("translation_engine_version", sa.String(128), nullable=True),
)
op.add_column(
"post",
sa.Column("translated_at", sa.DateTime(timezone=True), nullable=True),
)
op.add_column(
"import_settings",
sa.Column(
"translation_enabled", sa.Boolean(), nullable=False,
server_default=sa.text("false"),
),
)
op.add_column(
"import_settings",
sa.Column(
"interpreter_base_url", sa.Text(), nullable=False, server_default="",
),
)
op.add_column(
"import_settings",
sa.Column(
"translation_target_lang", sa.Text(), nullable=False,
server_default="en",
),
)
def downgrade() -> None:
op.drop_column("import_settings", "translation_target_lang")
op.drop_column("import_settings", "interpreter_base_url")
op.drop_column("import_settings", "translation_enabled")
op.drop_column("post", "translated_at")
op.drop_column("post", "translation_engine_version")
op.drop_column("post", "translated_source_lang")
op.drop_column("post", "description_translated")
op.drop_column("post", "post_title_translated")
+83 -2
View File
@@ -1,12 +1,23 @@
"""Settings API: import filters, system stats.""" """Settings API: import filters, system stats."""
import asyncio
import secrets import secrets
from quart import Blueprint, jsonify, request from quart import Blueprint, jsonify, request
from sqlalchemy import func, select from sqlalchemy import func, or_, select
from ..extensions import get_session from ..extensions import get_session
from ..models import AppSetting, Artist, ImageRecord, ImportBatch, ImportSettings, ImportTask, Tag from ..models import (
AppSetting,
Artist,
ImageRecord,
ImportBatch,
ImportSettings,
ImportTask,
Post,
Tag,
)
from ..services import interpreter_client as ic
settings_bp = Blueprint("settings", __name__, url_prefix="/api") settings_bp = Blueprint("settings", __name__, url_prefix="/api")
@@ -32,6 +43,9 @@ _EDITABLE_FIELDS = (
"extdl_mediafire_enabled", "extdl_mediafire_enabled",
"extdl_dropbox_enabled", "extdl_dropbox_enabled",
"extdl_pixeldrain_enabled", "extdl_pixeldrain_enabled",
"translation_enabled",
"interpreter_base_url",
"translation_target_lang",
) )
# Per-host external-download toggles — all plain booleans, validated uniformly. # Per-host external-download toggles — all plain booleans, validated uniformly.
@@ -69,6 +83,9 @@ async def get_import_settings():
"extdl_mediafire_enabled": row.extdl_mediafire_enabled, "extdl_mediafire_enabled": row.extdl_mediafire_enabled,
"extdl_dropbox_enabled": row.extdl_dropbox_enabled, "extdl_dropbox_enabled": row.extdl_dropbox_enabled,
"extdl_pixeldrain_enabled": row.extdl_pixeldrain_enabled, "extdl_pixeldrain_enabled": row.extdl_pixeldrain_enabled,
"translation_enabled": row.translation_enabled,
"interpreter_base_url": row.interpreter_base_url,
"translation_target_lang": row.translation_target_lang,
}) })
@@ -128,6 +145,15 @@ async def update_import_settings():
for tog in _EXTDL_TOGGLE_FIELDS: for tog in _EXTDL_TOGGLE_FIELDS:
if tog in body and not isinstance(body[tog], bool): if tog in body and not isinstance(body[tog], bool):
return jsonify({"error": f"{tog} must be a boolean"}), 400 return jsonify({"error": f"{tog} must be a boolean"}), 400
# Translation (#143): base URL may be empty (feature off until set — no
# default host; the operator points it at their own Interpreter proxy).
if "translation_enabled" in body and not isinstance(
body["translation_enabled"], bool
):
return jsonify({"error": "translation_enabled must be a boolean"}), 400
for key in ("interpreter_base_url", "translation_target_lang"):
if key in body and not isinstance(body[key], str):
return jsonify({"error": f"{key} must be a string"}), 400
if "series_suggest_threshold" in body: if "series_suggest_threshold" in body:
v = body["series_suggest_threshold"] v = body["series_suggest_threshold"]
if not isinstance(v, (int, float)) or isinstance(v, bool) or v < 0 or v > 1: if not isinstance(v, (int, float)) or isinstance(v, bool) or v < 0 or v > 1:
@@ -270,3 +296,58 @@ async def rotate_extension_api_key():
row.value = new_value row.value = new_value
await session.commit() await session.commit()
return jsonify({"key": new_value}) return jsonify({"key": new_value})
# --- Translation (#143): live status + manual "Translate now" --------------
@settings_bp.route("/settings/translation/status", methods=["GET"])
async def translation_status():
"""For the Settings card: is it on, is a URL set, is the service reachable,
and how many posts still await translation. Health runs the sync client in a
thread so the event loop isn't blocked."""
async with get_session() as session:
cfg = await ImportSettings.load(session)
untranslated = (await session.execute(
select(func.count(Post.id))
.where(Post.translated_source_lang.is_(None))
.where(or_(
Post.post_title.is_not(None), Post.description.is_not(None),
))
)).scalar_one()
base_url = (cfg.interpreter_base_url or "").strip()
healthy = await asyncio.to_thread(ic.health, base_url) if base_url else False
return jsonify({
"enabled": cfg.translation_enabled,
"base_url_set": bool(base_url),
"healthy": healthy,
"untranslated_count": int(untranslated),
})
@settings_bp.route("/settings/translation/test", methods=["POST"])
async def translation_test():
"""On-demand reachability check for a GIVEN Interpreter base URL (the Settings
'Test connection' button) — pings /v1/health without saving, so the operator
can verify a URL before enabling. Health runs in a thread (sync client)."""
body = await request.get_json()
base_url = ""
if isinstance(body, dict):
base_url = (body.get("base_url") or "").strip()
healthy = await asyncio.to_thread(ic.health, base_url) if base_url else False
return jsonify({"healthy": healthy})
@settings_bp.route("/settings/translation/run", methods=["POST"])
async def translation_run():
"""Enqueue the translate sweep now (the Settings 'Translate now' button)."""
async with get_session() as session:
cfg = await ImportSettings.load(session)
if not cfg.translation_enabled or not (cfg.interpreter_base_url or "").strip():
return jsonify(
{"error": "translation is disabled or no base URL is set"}
), 400
from ..tasks.translation import translate_posts
r = translate_posts.delay()
return jsonify({"celery_task_id": r.id}), 202
+8
View File
@@ -35,6 +35,7 @@ def make_celery() -> Celery:
"backend.app.tasks.backup", "backend.app.tasks.backup",
"backend.app.tasks.admin", "backend.app.tasks.admin",
"backend.app.tasks.library_audit", "backend.app.tasks.library_audit",
"backend.app.tasks.translation",
], ],
) )
app.conf.update( app.conf.update(
@@ -63,6 +64,9 @@ def make_celery() -> Celery:
"backend.app.tasks.backup.*": {"queue": "maintenance_long"}, "backend.app.tasks.backup.*": {"queue": "maintenance_long"},
"backend.app.tasks.admin.*": {"queue": "maintenance_long"}, "backend.app.tasks.admin.*": {"queue": "maintenance_long"},
"backend.app.tasks.library_audit.*": {"queue": "maintenance_long"}, "backend.app.tasks.library_audit.*": {"queue": "maintenance_long"},
# Translation backfill hits the LLM (~16s/item) → the long lane so it
# never starves the quick self-healing sweeps (#143).
"backend.app.tasks.translation.*": {"queue": "maintenance_long"},
}, },
# Heavy ML tasks need fair dispatch — see ImageRepo's precedent. # Heavy ML tasks need fair dispatch — see ImageRepo's precedent.
task_acks_late=True, task_acks_late=True,
@@ -161,6 +165,10 @@ def make_celery() -> Celery:
"task": "backend.app.tasks.ml.prune_presentation_reviews", "task": "backend.app.tasks.ml.prune_presentation_reviews",
"schedule": 86400.0, # retention: drop resolved review flags >30d "schedule": 86400.0, # retention: drop resolved review flags >30d
}, },
"translate-posts-daily": {
"task": "backend.app.tasks.translation.translate_posts",
"schedule": 86400.0, # no-op unless translation configured + healthy
},
"snapshot-head-metrics-daily": { "snapshot-head-metrics-daily": {
"task": "backend.app.tasks.maintenance.snapshot_head_metrics", "task": "backend.app.tasks.maintenance.snapshot_head_metrics",
"schedule": 86400.0, "schedule": 86400.0,
+15
View File
@@ -92,6 +92,21 @@ class ImportSettings(Base):
Boolean, nullable=False, default=True, server_default="true", Boolean, nullable=False, default=True, server_default="true",
) )
# -- Post-text translation via the Interpreter LAN service (milestone 143).
# Off by default with NO default host — it needs a reachable Interpreter
# service (the operator's, behind a reverse proxy), which not every install
# has; the operator sets the URL and flips it on. Empty base_url OR disabled
# → the translate sweep no-ops.
translation_enabled: Mapped[bool] = mapped_column(
Boolean, nullable=False, default=False, server_default="false",
)
interpreter_base_url: Mapped[str] = mapped_column(
Text, nullable=False, default="", server_default="",
)
translation_target_lang: Mapped[str] = mapped_column(
Text, nullable=False, default="en", server_default="en",
)
@classmethod @classmethod
async def load(cls, session) -> ImportSettings: async def load(cls, session) -> ImportSettings:
"""The singleton settings row (id=1), via an async session.""" """The singleton settings row (id=1), via an async session."""
+18
View File
@@ -47,6 +47,24 @@ class Post(Base):
description: Mapped[str | None] = mapped_column(Text, nullable=True) description: Mapped[str | None] = mapped_column(Text, nullable=True)
attachment_count: Mapped[int | None] = mapped_column(Integer, nullable=True) attachment_count: Mapped[int | None] = mapped_column(Integer, nullable=True)
# -- Post-text translation (milestone 143). Filled by the translate_posts
# sweep via the Interpreter LAN service so viewing is instant.
# translated_source_lang is the DETECTED original language; "en" (or a
# passthrough) means nothing to translate and the *_translated columns stay
# NULL. engine_version keys the Interpreter cache — re-runs are ~1ms and a
# model upgrade re-translates instead of serving stale.
post_title_translated: Mapped[str | None] = mapped_column(Text, nullable=True)
description_translated: Mapped[str | None] = mapped_column(Text, nullable=True)
translated_source_lang: Mapped[str | None] = mapped_column(
String(8), nullable=True
)
translation_engine_version: Mapped[str | None] = mapped_column(
String(128), nullable=True
)
translated_at: Mapped[datetime | None] = mapped_column(
DateTime(timezone=True), nullable=True
)
downloaded_at: Mapped[datetime] = mapped_column( downloaded_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False, server_default=func.now() DateTime(timezone=True), nullable=False, server_default=func.now()
) )
@@ -0,0 +1,88 @@
"""Interpreter translation client (milestone 143) — a thin SYNC wrapper over the
self-hosted Interpreter LAN service (LibreTranslate-compatible `/v1/translate`).
Sync (requests) because the only caller is the sync celery translate sweep, and
it mirrors FC's other platform clients. The service knows nothing about Curator —
all Curator logic stays here. base_url is operator-configured (empty until set;
behind a reverse proxy). Full API contract: Scribe note #1347.
"""
from __future__ import annotations
import requests
class InterpreterUnavailable(Exception):
"""The translation engine is down / unreachable (HTTP 503 or a connection
error) — retry later, don't drop the item."""
class InterpreterBadRequest(Exception):
"""Bad request params (HTTP 400)."""
def _url(base_url: str, path: str) -> str:
return f"{base_url.rstrip('/')}{path}"
def health(base_url: str, *, timeout: float = 5.0) -> bool:
"""True iff the Interpreter LLM engine is up. Any error (unset URL, network,
non-200, engine down) → False, so the sweep just no-ops rather than raising."""
if not base_url:
return False
try:
r = requests.get(_url(base_url, "/v1/health"), timeout=timeout)
except requests.RequestException:
return False
if r.status_code != 200:
return False
engines = (r.json() or {}).get("engines") or {}
return bool(engines.get("llm"))
def translate(
texts: list[str], *, base_url: str, target: str = "en",
source: str = "auto", timeout: float = 120.0,
) -> dict:
"""Translate a batch. Returns::
{"translations": [str, ...], # SAME order & length as `texts`
"detected_lang": str | None, # aggregate: describes the FIRST item
"engine": str | None,
"engine_version": str | None}
Interpreter batch metadata is aggregate (first item only) — fine for one
post's ``[title, description]`` batch since they share a language. Passthrough
items (already target-language / emoji-only) come back UNCHANGED in their
slot. Raises InterpreterUnavailable on 503 / connection error (retry later),
InterpreterBadRequest on 400. (Scribe note #1347.)
"""
if not texts:
return {"translations": [], "detected_lang": None,
"engine": None, "engine_version": None}
try:
r = requests.post(
_url(base_url, "/v1/translate"),
json={"q": list(texts), "source": source,
"target": target, "engine": "auto"},
timeout=timeout,
)
except requests.RequestException as e:
raise InterpreterUnavailable(str(e)) from e
if r.status_code == 503:
raise InterpreterUnavailable("interpreter engine unavailable")
if r.status_code == 400:
raise InterpreterBadRequest((r.json() or {}).get("error", "bad request"))
r.raise_for_status()
body = r.json() or {}
out = body.get("translatedText")
# q was an array → translatedText is an array (same order & length). Guard a
# scalar reply just in case (shouldn't happen for an array q).
if not isinstance(out, list):
out = [out]
interp = body.get("interpreter") or {}
return {
"translations": out,
"detected_lang": (body.get("detectedLanguage") or {}).get("language"),
"engine": interp.get("engine"),
"engine_version": interp.get("engine_version"),
}
+13
View File
@@ -200,6 +200,8 @@ class PostFeedService:
atts_map = await self._attachments_for([post.id]) atts_map = await self._attachments_for([post.id])
item = self._to_dict(post, artist, source, thumbs_map, atts_map) item = self._to_dict(post, artist, source, thumbs_map, atts_map)
item["description_full"] = html_to_plain(post.description) item["description_full"] = html_to_plain(post.description)
# Full (uncapped) translated description for the detail view (#143).
item["description_translated_full"] = post.description_translated
# Sanitized HTML body for faithful (semantic) rendering in the post view; # Sanitized HTML body for faithful (semantic) rendering in the post view;
# detail-only (the feed list stays lightweight plain text). None when the # detail-only (the feed list stays lightweight plain text). None when the
# post has no body. Inline `<img>` sources are remapped to locally-served # post has no body. Inline `<img>` sources are remapped to locally-served
@@ -371,6 +373,12 @@ class PostFeedService:
description_plain, truncated = None, False description_plain, truncated = None, False
else: else:
description_plain, truncated = truncate_at_word(plain_full, DESCRIPTION_LIMIT) description_plain, truncated = truncate_at_word(plain_full, DESCRIPTION_LIMIT)
# Translation (#143): the stored translated description is already plain
# text; truncate it the same way for the card.
desc_trans = post.description_translated
desc_trans_short = (
truncate_at_word(desc_trans, DESCRIPTION_LIMIT)[0] if desc_trans else None
)
thumbs_entry = thumbs_map.get(post.id, {"thumbs": [], "more": 0}) thumbs_entry = thumbs_map.get(post.id, {"thumbs": [], "more": 0})
# `source` is null for filesystem-imported posts with no live # `source` is null for filesystem-imported posts with no live
# subscription (alembic 0030). Frontend renders that as a # subscription (alembic 0030). Frontend renders that as a
@@ -384,6 +392,11 @@ class PostFeedService:
"downloaded_at": post.downloaded_at.isoformat(), "downloaded_at": post.downloaded_at.isoformat(),
"description_plain": description_plain, "description_plain": description_plain,
"description_truncated": truncated, "description_truncated": truncated,
# Translation-forward fields (#143): shown by default when present;
# UI toggles back to the originals above. Source lang labels the toggle.
"post_title_translated": post.post_title_translated,
"description_translated": desc_trans_short,
"translated_source_lang": post.translated_source_lang,
"artist": {"id": artist.id, "name": artist.name, "slug": artist.slug}, "artist": {"id": artist.id, "name": artist.name, "slug": artist.slug},
"source": ( "source": (
{"id": source.id, "platform": source.platform} {"id": source.id, "platform": source.platform}
@@ -29,6 +29,12 @@ def _post_dict(p: Post) -> dict:
"date": p.post_date.isoformat() if p.post_date else None, "date": p.post_date.isoformat() if p.post_date else None,
"description_html": sanitize_post_html(p.description), "description_html": sanitize_post_html(p.description),
"attachment_count": p.attachment_count, "attachment_count": p.attachment_count,
# Translation (#143): the English title/description shown by default when
# a translation exists; the UI toggles to the original. Source lang labels
# the original.
"title_translated": p.post_title_translated,
"description_translated": p.description_translated,
"translated_source_lang": p.translated_source_lang,
} }
+101
View File
@@ -0,0 +1,101 @@
"""Post-text translation Celery task (milestone 143).
Backfills non-English post title/description to the target language via the
self-hosted Interpreter LAN service, storing the result + engine_version so
viewing is instant and re-runs are cache-fast. Sync (Celery workers are sync),
same pattern as the other backfill sweeps. No-op unless translation is enabled,
a base URL is set, and the service is healthy.
"""
import logging
from datetime import UTC, datetime
from celery.exceptions import SoftTimeLimitExceeded
from sqlalchemy import or_, select
from ..celery_app import celery
from ..models import ImportSettings, Post
from ..services import interpreter_client as ic
from ..utils.text import html_to_plain
from ._sync_engine import sync_session_factory as _sync_session_factory
log = logging.getLogger(__name__)
# Bound one run's work so it commits progress rather than dying at the time
# limit; the rest resumes next cycle (idempotent — only untranslated posts are
# picked, so an interrupted run just re-runs = rule-89 recovery).
_MAX_POSTS_PER_RUN = 300
@celery.task(
name="backend.app.tasks.translation.translate_posts",
soft_time_limit=1800, time_limit=2100,
)
def translate_posts() -> str:
"""Translate untranslated non-English post title/description (default target
en) via Interpreter. Per-post [title, description] batch → per-post detected
language. Passthrough / already-target posts are marked handled (source ==
target) with no stored translation. 503 or a connection error interrupts the
run (retry next cycle); 400 stops it (fix config); posts done so far stay
committed. No-op unless configured + healthy. Returns a summary string."""
SessionLocal = _sync_session_factory()
with SessionLocal() as session:
cfg = ImportSettings.load_sync(session)
if not cfg.translation_enabled or not cfg.interpreter_base_url.strip():
return "disabled"
base_url = cfg.interpreter_base_url.strip()
target = (cfg.translation_target_lang or "en").strip() or "en"
if not ic.health(base_url):
return "interpreter unavailable"
posts = session.execute(
select(Post)
.where(Post.translated_source_lang.is_(None))
.where(or_(
Post.post_title.is_not(None), Post.description.is_not(None),
))
.limit(_MAX_POSTS_PER_RUN)
).scalars().all()
translated = 0
try:
for post in posts:
translated += _translate_one(session, post, base_url, target)
session.commit() # per-post → an interrupted run keeps progress
except ic.InterpreterUnavailable:
session.rollback()
return f"interrupted (service down) — translated={translated}"
except ic.InterpreterBadRequest as e:
session.rollback()
log.warning("translate_posts bad request: %s", e)
return f"stopped (bad request) — translated={translated}"
except SoftTimeLimitExceeded:
return f"time limit — translated={translated}"
return f"translated={translated} scanned={len(posts)}"
def _translate_one(session, post, base_url: str, target: str) -> int:
"""Translate one post's title/description in place. Returns 1 if it stored a
translation, 0 for passthrough/empty (which still marks the post handled)."""
title = (post.post_title or "").strip()
desc = (html_to_plain(post.description) if post.description else "") or ""
desc = desc.strip()
fields: list[tuple[str, str]] = []
if title:
fields.append(("title", title))
if desc:
fields.append(("desc", desc))
if not fields:
post.translated_source_lang = target # nothing to translate → handled
return 0
res = ic.translate([t for _, t in fields], base_url=base_url, target=target)
detected = res["detected_lang"] or target
if detected == target or res["engine"] == "none":
post.translated_source_lang = target # already target language → handled
return 0
by_field = {f: res["translations"][i] for i, (f, _) in enumerate(fields)}
post.post_title_translated = by_field.get("title")
post.description_translated = by_field.get("desc")
post.translated_source_lang = detected
post.translation_engine_version = res["engine_version"]
post.translated_at = datetime.now(UTC)
return 1
@@ -42,13 +42,23 @@
· {{ e.post.attachment_count }} files · {{ e.post.attachment_count }} files
</span> </span>
</div> </div>
<div v-if="hasTranslation(e)" class="fc-prov__actions">
<a
href="#" @click.prevent="toggleOrig(e.provenance_id)"
>{{ showOrig[e.provenance_id] ? 'Show translation' : `Show original${langLabel(e)}` }}</a>
</div>
<div v-if="e.post.description_html" class="fc-prov__actions"> <div v-if="e.post.description_html" class="fc-prov__actions">
<a <a
href="#" @click.prevent="toggleDesc(e.provenance_id)" href="#" @click.prevent="toggleDesc(e.provenance_id)"
>{{ expanded[e.provenance_id] ? 'Hide description ' : 'Show description ' }}</a> >{{ expanded[e.provenance_id] ? 'Hide description ' : 'Show description ' }}</a>
</div> </div>
<!-- Translated body = plain text; original = sanitized HTML (#143). -->
<p
v-if="showTranslated(e) && expanded[e.provenance_id] && e.post.description_translated"
class="fc-prov__desc"
>{{ e.post.description_translated }}</p>
<div <div
v-if="e.post.description_html && expanded[e.provenance_id]" v-else-if="e.post.description_html && expanded[e.provenance_id]"
class="fc-prov__desc" v-html="e.post.description_html" class="fc-prov__desc" v-html="e.post.description_html"
/> />
</article> </article>
@@ -117,8 +127,12 @@ const effectiveImage = computed(() => props.image ?? modal.current)
// 2026-05-28. Reset when the viewed image changes. // 2026-05-28. Reset when the viewed image changes.
const expanded = reactive({}) const expanded = reactive({})
function toggleDesc(id) { expanded[id] = !expanded[id] } function toggleDesc(id) { expanded[id] = !expanded[id] }
// Per-entry "show the original (untranslated) text" toggle (#143).
const showOrig = reactive({})
function toggleOrig(id) { showOrig[id] = !showOrig[id] }
watch(() => effectiveId.value, () => { watch(() => effectiveId.value, () => {
for (const k of Object.keys(expanded)) delete expanded[k] for (const k of Object.keys(expanded)) delete expanded[k]
for (const k of Object.keys(showOrig)) delete showOrig[k]
}) })
watch( watch(
@@ -157,10 +171,23 @@ const show = computed(() => {
const attachments = computed(() => state.value?.attachments || []) const attachments = computed(() => state.value?.attachments || [])
function postDate(e) { return formatPostDate(e.post.date) } function postDate(e) { return formatPostDate(e.post.date) }
// Translation-forward (#143): show the English by default when present.
function hasTranslation(e) {
return !!(e.post.title_translated || e.post.description_translated)
}
function showTranslated(e) {
return hasTranslation(e) && !showOrig[e.provenance_id]
}
function langLabel(e) {
return e.post.translated_source_lang ? ` (${e.post.translated_source_lang})` : ''
}
function postTitle(e) { function postTitle(e) {
// Titles can arrive as stored HTML (e.g. "<strong>…</strong>"); render // Titles can arrive as stored HTML (e.g. "<strong>…</strong>"); render
// as plain text (the CSS makes it bold). // as plain text (the CSS makes it bold).
return toPlainText(e.post.title) || `Post ${e.post.external_post_id}` const t = showTranslated(e) && e.post.title_translated
? e.post.title_translated
: e.post.title
return toPlainText(t) || `Post ${e.post.external_post_id}`
} }
function openPost(postId, artistId) { function openPost(postId, artistId) {
+49 -3
View File
@@ -58,16 +58,23 @@
</div> </div>
<div class="fc-post-card__text"> <div class="fc-post-card__text">
<h3 v-if="plainTitle" class="fc-post-card__title">{{ plainTitle }}</h3> <h3 v-if="displayTitle" class="fc-post-card__title">{{ displayTitle }}</h3>
<h3 v-else class="fc-post-card__title fc-post-card__title--missing"> <h3 v-else class="fc-post-card__title fc-post-card__title--missing">
Post {{ post.external_post_id }} Post {{ post.external_post_id }}
</h3> </h3>
<!-- Translation toggle (#143): English shown by default; reveal the
original, labelled with its detected language. -->
<button
v-if="hasTranslation" type="button" class="fc-post-card__lang"
:title="showOriginal ? 'Show the English translation' : `Show the original${sourceLangLabel} text`"
@click.stop="showOriginal = !showOriginal"
>{{ showOriginal ? 'Show translation' : `Show original${sourceLangLabel}` }}</button>
<!-- Faithful (semantic) body render once expanded: backend-sanitized <!-- Faithful (semantic) body render once expanded: backend-sanitized
HTML (headings, lists, links, inline images). Collapsed and the HTML (headings, lists, links, inline images). Collapsed and the
no-detail fallback stay plain text. --> no-detail fallback stay plain text. -->
<div <div
v-if="hasDescription && descExpanded && descHtml" v-if="hasDescription && showHtmlBody"
class="fc-post-card__desc fc-post-card__desc--html" class="fc-post-card__desc fc-post-card__desc--html"
v-html="descHtml" v-html="descHtml"
/> />
@@ -75,7 +82,7 @@
v-else-if="hasDescription" ref="descEl" v-else-if="hasDescription" ref="descEl"
class="fc-post-card__desc" class="fc-post-card__desc"
:class="{ 'fc-post-card__desc--clamped': !descExpanded }" :class="{ 'fc-post-card__desc--clamped': !descExpanded }"
>{{ descText }}</p> >{{ descTextDisplay }}</p>
<p v-else class="fc-post-card__desc fc-post-card__desc--missing"> <p v-else class="fc-post-card__desc fc-post-card__desc--missing">
(no description) (no description)
</p> </p>
@@ -231,6 +238,32 @@ const canExpand = computed(
() => props.post.description_truncated === true || cssOverflow.value, () => props.post.description_truncated === true || cssOverflow.value,
) )
// --- translation-forward (#143): the English is shown by default when a
// translation exists; a per-card toggle reveals the original. Translations are
// plain text, so showTranslated also decides HTML-vs-plain body rendering.
const showOriginal = ref(false)
const hasTranslation = computed(
() => !!(props.post.post_title_translated || props.post.description_translated),
)
const showTranslated = computed(() => hasTranslation.value && !showOriginal.value)
const sourceLangLabel = computed(() =>
props.post.translated_source_lang ? ` (${props.post.translated_source_lang})` : '',
)
const displayTitle = computed(() =>
showTranslated.value && props.post.post_title_translated
? toPlainText(props.post.post_title_translated)
: plainTitle.value,
)
const descTextDisplay = computed(() => {
if (!showTranslated.value) return descText.value
return descExpanded.value
? (detail.value?.description_translated_full || props.post.description_translated)
: props.post.description_translated
})
const showHtmlBody = computed(
() => !showTranslated.value && descExpanded.value && !!descHtml.value,
)
function measureOverflow () { function measureOverflow () {
const el = descEl.value const el = descEl.value
cssOverflow.value = !!el && el.scrollHeight > el.clientHeight + 1 cssOverflow.value = !!el && el.scrollHeight > el.clientHeight + 1
@@ -467,6 +500,19 @@ function formatBytes (n) {
font-size: 0.85rem; font-weight: 600; font-size: 0.85rem; font-weight: 600;
} }
.fc-post-card__more:hover { text-decoration: underline; } .fc-post-card__more:hover { text-decoration: underline; }
/* Translation toggle (#143) — quiet secondary affordance under the title. */
.fc-post-card__lang {
display: block; margin: 2px 0 6px; padding: 0;
background: none; border: 0; cursor: pointer;
color: rgb(var(--v-theme-on-surface-variant));
font-size: 0.75rem; font-weight: 600;
}
.fc-post-card__lang:hover {
text-decoration: underline; color: rgb(var(--v-theme-accent));
}
.fc-post-card__lang:focus-visible {
outline: 2px solid rgb(var(--v-theme-accent)); outline-offset: 1px;
}
.fc-post-card__atts { .fc-post-card__atts {
display: flex; flex-wrap: wrap; gap: 8px; display: flex; flex-wrap: wrap; gap: 8px;
@@ -13,6 +13,7 @@
</p> </p>
<div class="fc-tile-stack"> <div class="fc-tile-stack">
<ImportFiltersForm /> <ImportFiltersForm />
<TranslationCard />
</div> </div>
</section> </section>
@@ -69,6 +70,7 @@
import { onMounted, onUnmounted } from 'vue' import { onMounted, onUnmounted } from 'vue'
import ImportFiltersForm from './ImportFiltersForm.vue' import ImportFiltersForm from './ImportFiltersForm.vue'
import TranslationCard from './TranslationCard.vue'
import MLBackfillCard from './MLBackfillCard.vue' import MLBackfillCard from './MLBackfillCard.vue'
import ThumbnailBackfillCard from './ThumbnailBackfillCard.vue' import ThumbnailBackfillCard from './ThumbnailBackfillCard.vue'
import ArchiveReextractCard from './ArchiveReextractCard.vue' import ArchiveReextractCard from './ArchiveReextractCard.vue'
@@ -0,0 +1,162 @@
<template>
<MaintenanceTile
icon="mdi-translate"
title="Post translation (Interpreter)"
blurb="Translate non-English post titles + descriptions to English via your self-hosted Interpreter service, so archived posts read naturally."
>
<div class="d-flex align-center mb-3" style="gap: 10px;">
<span class="fc-section-h">Enabled</span>
<v-switch
v-model="enabled" :loading="busy" hide-details density="compact"
color="success" class="ml-auto" @update:model-value="onToggle"
/>
</div>
<p class="fc-muted text-body-2 mb-3">
Point this at your Interpreter service. It stays off until a URL is set, and
only translates posts whose text isn't already in the target language — the
English is shown by default on post cards, with a toggle back to the original.
</p>
<v-text-field
v-model="baseUrl" label="Interpreter base URL"
placeholder="http://interpreter.your-domain/" density="compact"
hide-details class="mb-3" :disabled="busy" @change="onSave"
/>
<v-text-field
v-model="targetLang" label="Target language" density="compact"
hide-details style="max-width: 160px;" class="mb-3" :disabled="busy"
@change="onSave"
/>
<div class="d-flex align-center flex-wrap mb-3" style="gap: 8px;">
<v-icon size="12" :color="statusColor">mdi-circle</v-icon>
<span class="fc-muted text-body-2">{{ statusText }}</span>
<v-btn
size="x-small" variant="tonal" rounded="pill" class="ml-2"
:loading="testing" :disabled="!baseUrl" @click="onTest"
>Test connection</v-btn>
<span
v-if="testResult !== null" class="text-caption"
:class="testResult ? 'text-success' : 'text-error'"
>{{ testResult ? ' reachable' : ' unreachable' }}</span>
</div>
<div class="d-flex align-center flex-wrap" style="gap: 10px;">
<v-btn
size="small" variant="flat" color="accent" rounded="pill"
prepend-icon="mdi-translate" :loading="running"
:disabled="!enabled || !baseUrl" @click="onRun"
>Translate now</v-btn>
<span v-if="status" class="fc-muted text-caption">
{{ status.untranslated_count }}
post{{ status.untranslated_count === 1 ? '' : 's' }} awaiting translation
</span>
</div>
<v-alert
v-if="err" type="error" variant="tonal" density="compact"
class="mt-3" closable @click:close="err = null"
>{{ err }}</v-alert>
</MaintenanceTile>
</template>
<script setup>
import { computed, onMounted, ref } from 'vue'
import { toast } from '../../utils/toast.js'
import { useApi } from '../../composables/useApi.js'
import { useImportStore } from '../../stores/import.js'
import MaintenanceTile from '../common/MaintenanceTile.vue'
const api = useApi()
const store = useImportStore()
const enabled = ref(false)
const baseUrl = ref('')
const targetLang = ref('en')
const busy = ref(false)
const running = ref(false)
const testing = ref(false)
const testResult = ref(null) // null = not tested this session; true/false = last result
const status = ref(null)
const err = ref(null)
const statusColor = computed(() => {
if (!enabled.value || !baseUrl.value) return 'grey'
return status.value?.healthy ? 'success' : 'error'
})
const statusText = computed(() => {
if (!enabled.value) return 'Off'
if (!baseUrl.value) return 'No base URL set'
if (status.value == null) return 'Checking'
return status.value.healthy ? 'Interpreter reachable' : 'Interpreter unreachable'
})
async function loadStatus() {
try { status.value = await api.get('/api/settings/translation/status') }
catch { status.value = null }
}
async function onTest() {
// Ping /v1/health for the CURRENTLY-typed URL (not the saved one), so a new
// URL can be verified before enabling. Also reflects in the status dot.
testing.value = true
err.value = null
testResult.value = null
try {
const r = await api.post('/api/settings/translation/test', {
body: { base_url: baseUrl.value.trim() },
})
testResult.value = !!r.healthy
if (status.value) status.value.healthy = !!r.healthy
} catch (e) {
err.value = e.message
} finally {
testing.value = false
}
}
onMounted(async () => {
try {
await store.loadSettings()
const s = store.settings || {}
enabled.value = !!s.translation_enabled
baseUrl.value = s.interpreter_base_url || ''
targetLang.value = s.translation_target_lang || 'en'
} catch { /* non-fatal */ }
await loadStatus()
})
async function save(patch) {
busy.value = true
err.value = null
try { await store.patchSettings(patch) }
catch (e) { err.value = e.message }
finally { busy.value = false }
}
async function onToggle(val) {
await save({ translation_enabled: !!val })
toast({ text: val ? 'Translation on' : 'Translation off', type: 'success' })
await loadStatus()
}
async function onSave() {
await save({
interpreter_base_url: baseUrl.value.trim(),
translation_target_lang: (targetLang.value || 'en').trim() || 'en',
})
await loadStatus()
}
async function onRun() {
running.value = true
err.value = null
try {
await api.post('/api/settings/translation/run')
toast({ text: 'Translation sweep started', type: 'success' })
} catch (e) {
err.value = e.message
} finally {
running.value = false
setTimeout(loadStatus, 1500) // let the sweep make a dent, then refresh
}
}
</script>
+69
View File
@@ -28,6 +28,75 @@ async def test_patch_import_settings(client):
assert body["transparency_threshold"] == 0.7 assert body["transparency_threshold"] == 0.7
@pytest.mark.asyncio
async def test_translation_settings_defaults_and_patch(client):
# #143: translation is OFF by default with NO default host — the operator
# points it at their own Interpreter proxy and enables it.
body = await (await client.get("/api/settings/import")).get_json()
assert body["translation_enabled"] is False
assert body["interpreter_base_url"] == ""
assert body["translation_target_lang"] == "en"
ok = await client.patch("/api/settings/import", json={
"translation_enabled": True,
"interpreter_base_url": "http://interpreter.lan",
})
assert ok.status_code == 200
out = await ok.get_json()
assert out["translation_enabled"] is True
assert out["interpreter_base_url"] == "http://interpreter.lan"
# Type validation.
assert (await client.patch(
"/api/settings/import", json={"translation_enabled": "yes"})).status_code == 400
assert (await client.patch(
"/api/settings/import", json={"interpreter_base_url": 123})).status_code == 400
@pytest.mark.asyncio
async def test_translation_status_defaults(client):
# Off + no URL → no network health call, healthy False (#143).
body = await (await client.get("/api/settings/translation/status")).get_json()
assert body["enabled"] is False
assert body["base_url_set"] is False
assert body["healthy"] is False
assert "untranslated_count" in body
@pytest.mark.asyncio
async def test_translation_run_requires_config(client):
resp = await client.post("/api/settings/translation/run")
assert resp.status_code == 400 # disabled / no base URL
@pytest.mark.asyncio
async def test_translation_test_connection(client, monkeypatch):
# Tests a GIVEN url (not the saved one) without persisting anything.
monkeypatch.setattr("backend.app.api.settings.ic.health", lambda *a, **k: True)
resp = await client.post(
"/api/settings/translation/test", json={"base_url": "http://i.lan"})
assert resp.status_code == 200
assert (await resp.get_json())["healthy"] is True
# Empty URL → False, no health call.
empty = await client.post(
"/api/settings/translation/test", json={"base_url": ""})
assert (await empty.get_json())["healthy"] is False
@pytest.mark.asyncio
async def test_translation_run_enqueues_when_configured(client, monkeypatch):
monkeypatch.setattr(
"backend.app.tasks.translation.translate_posts.delay",
lambda *a, **k: type("R", (), {"id": "t1"})(),
)
await client.patch("/api/settings/import", json={
"translation_enabled": True, "interpreter_base_url": "http://i.lan",
})
resp = await client.post("/api/settings/translation/run")
assert resp.status_code == 202
assert (await resp.get_json())["celery_task_id"] == "t1"
@pytest.mark.asyncio @pytest.mark.asyncio
async def test_patch_rejects_non_object(client): async def test_patch_rejects_non_object(client):
resp = await client.patch("/api/settings/import", json=[1, 2, 3]) resp = await client.patch("/api/settings/import", json=[1, 2, 3])
+108
View File
@@ -0,0 +1,108 @@
"""Interpreter translation client (#143) — unit tests with mocked HTTP (no DB,
no real network)."""
import pytest
from backend.app.services import interpreter_client as ic
class _Resp:
def __init__(self, status_code=200, payload=None):
self.status_code = status_code
self._payload = payload or {}
def json(self):
return self._payload
def raise_for_status(self):
if self.status_code >= 400:
raise ic.requests.HTTPError(f"status {self.status_code}")
def test_translate_maps_batch_in_order(monkeypatch):
captured = {}
def fake_post(url, json, timeout):
captured["json"] = json
return _Resp(200, {
"translatedText": ["The cat is cute", "Blonde gal!"],
"detectedLanguage": {"language": "ja", "confidence": 0.98},
"interpreter": {"engine": "llm", "engine_version": "ollama:x:12b"},
})
monkeypatch.setattr(ic.requests, "post", fake_post)
out = ic.translate(["ねこが可愛い", "金髪ギャル!"], base_url="http://i.lan")
assert out["translations"] == ["The cat is cute", "Blonde gal!"]
assert out["detected_lang"] == "ja"
assert out["engine_version"] == "ollama:x:12b"
assert captured["json"]["q"] == ["ねこが可愛い", "金髪ギャル!"]
assert captured["json"]["target"] == "en"
def test_translate_passthrough_unchanged(monkeypatch):
# Already-English items come back unchanged in their slot (engine "none").
monkeypatch.setattr(ic.requests, "post", lambda *a, **k: _Resp(200, {
"translatedText": ["already english"],
"detectedLanguage": {"language": "en"},
"interpreter": {"engine": "none", "engine_version": None},
}))
out = ic.translate(["already english"], base_url="http://i.lan")
assert out["translations"] == ["already english"]
assert out["detected_lang"] == "en"
assert out["engine"] == "none"
def test_translate_503_raises_unavailable(monkeypatch):
monkeypatch.setattr(ic.requests, "post", lambda *a, **k: _Resp(503, {}))
with pytest.raises(ic.InterpreterUnavailable):
ic.translate(["x"], base_url="http://i.lan")
def test_translate_connection_error_raises_unavailable(monkeypatch):
def boom(*a, **k):
raise ic.requests.ConnectionError("refused")
monkeypatch.setattr(ic.requests, "post", boom)
with pytest.raises(ic.InterpreterUnavailable):
ic.translate(["x"], base_url="http://i.lan")
def test_translate_400_raises_bad_request(monkeypatch):
monkeypatch.setattr(
ic.requests, "post", lambda *a, **k: _Resp(400, {"error": "bad target"})
)
with pytest.raises(ic.InterpreterBadRequest):
ic.translate(["x"], base_url="http://i.lan")
def test_translate_empty_is_noop(monkeypatch):
def boom(*a, **k):
raise AssertionError("should not call the service for an empty batch")
monkeypatch.setattr(ic.requests, "post", boom)
assert ic.translate([], base_url="http://i.lan")["translations"] == []
def test_health_true_when_llm_up(monkeypatch):
monkeypatch.setattr(
ic.requests, "get", lambda *a, **k: _Resp(200, {"engines": {"llm": True}})
)
assert ic.health("http://i.lan") is True
def test_health_false_when_engine_down(monkeypatch):
monkeypatch.setattr(
ic.requests, "get", lambda *a, **k: _Resp(200, {"engines": {"llm": False}})
)
assert ic.health("http://i.lan") is False
def test_health_false_on_error(monkeypatch):
def boom(*a, **k):
raise ic.requests.ConnectionError("refused")
monkeypatch.setattr(ic.requests, "get", boom)
assert ic.health("http://i.lan") is False
def test_health_false_when_url_empty():
assert ic.health("") is False
+24
View File
@@ -283,6 +283,30 @@ async def test_scroll_item_shape_minimal(db):
assert "description_full" not in item assert "description_full" not in item
@pytest.mark.asyncio
async def test_scroll_surfaces_translation_fields(db):
# #143: a translated post exposes the translated title/description + source
# lang, with the originals still present for the toggle.
artist = await _seed_artist(db, "alice-tr")
src = await _seed_source(db, artist.id, "pixiv", "https://p/alice-tr")
p = await _seed_post(
db, src.id, external_id="TR", post_date=datetime.now(UTC),
title="ねこ", description="<p>かわいい</p>",
)
p.post_title_translated = "cat"
p.description_translated = "cute"
p.translated_source_lang = "ja"
await db.commit()
page = await PostFeedService(db).scroll(cursor=None, limit=10)
item = page["items"][0]
assert item["post_title_translated"] == "cat"
assert item["description_translated"] == "cute"
assert item["translated_source_lang"] == "ja"
assert item["post_title"] == "ねこ" # original kept for the toggle
assert item["description_plain"] == "かわいい"
@pytest.mark.asyncio @pytest.mark.asyncio
async def test_scroll_description_truncates_long_text(db): async def test_scroll_description_truncates_long_text(db):
artist = await _seed_artist(db, "alice-long") artist = await _seed_artist(db, "alice-long")
+127
View File
@@ -0,0 +1,127 @@
"""translate_posts sweep (#143) — integration, with a stubbed Interpreter client
and the task's session factory pointed at the test's db_sync."""
import pytest
from sqlalchemy import select
from backend.app.models import Artist, ImportSettings, Post
from backend.app.tasks.translation import translate_posts
pytestmark = pytest.mark.integration
class _Ctx:
def __init__(self, s):
self.s = s
def __enter__(self):
return self.s
def __exit__(self, *a):
return False
def _sf(db_sync):
class _SM:
def __call__(self):
return _Ctx(db_sync)
return _SM()
def _artist(db):
a = Artist(name="A", slug="a")
db.add(a)
db.flush()
return a
def _post(db, artist_id, *, title=None, desc=None, ext="p1"):
p = Post(
artist_id=artist_id, external_post_id=ext,
post_title=title, description=desc,
)
db.add(p)
db.flush()
return p
def _enable(db, url="http://i.lan"):
cfg = db.execute(
select(ImportSettings).where(ImportSettings.id == 1)
).scalar_one()
cfg.translation_enabled = True
cfg.interpreter_base_url = url
db.flush()
def _patch(monkeypatch, db_sync):
monkeypatch.setattr(
"backend.app.tasks.translation._sync_session_factory",
lambda: _sf(db_sync),
)
def test_translate_posts_stores_translation(db_sync, monkeypatch):
_patch(monkeypatch, db_sync)
monkeypatch.setattr("backend.app.tasks.translation.ic.health", lambda *a, **k: True)
monkeypatch.setattr(
"backend.app.tasks.translation.ic.translate",
lambda texts, **k: {
"translations": [f"EN:{t}" for t in texts],
"detected_lang": "ja", "engine": "llm", "engine_version": "v1",
},
)
a = _artist(db_sync)
p = _post(db_sync, a.id, title="ねこ", desc="かわいい")
_enable(db_sync)
db_sync.commit()
assert "translated=1" in translate_posts()
db_sync.refresh(p)
assert p.post_title_translated == "EN:ねこ"
assert p.description_translated == "EN:かわいい"
assert p.translated_source_lang == "ja"
assert p.translation_engine_version == "v1"
def test_translate_posts_passthrough_english_marks_handled(db_sync, monkeypatch):
_patch(monkeypatch, db_sync)
monkeypatch.setattr("backend.app.tasks.translation.ic.health", lambda *a, **k: True)
monkeypatch.setattr(
"backend.app.tasks.translation.ic.translate",
lambda texts, **k: {
"translations": list(texts), "detected_lang": "en",
"engine": "none", "engine_version": None,
},
)
a = _artist(db_sync)
p = _post(db_sync, a.id, title="hello world", ext="p2")
_enable(db_sync)
db_sync.commit()
translate_posts()
db_sync.refresh(p)
assert p.translated_source_lang == "en" # marked handled...
assert p.post_title_translated is None # ...but nothing to show
def test_translate_posts_disabled_is_noop(db_sync, monkeypatch):
_patch(monkeypatch, db_sync)
a = _artist(db_sync)
p = _post(db_sync, a.id, title="ねこ", ext="p3")
db_sync.commit() # translation_enabled defaults False
assert translate_posts() == "disabled"
db_sync.refresh(p)
assert p.translated_source_lang is None
def test_translate_posts_service_down_leaves_untranslated(db_sync, monkeypatch):
_patch(monkeypatch, db_sync)
monkeypatch.setattr("backend.app.tasks.translation.ic.health", lambda *a, **k: False)
a = _artist(db_sync)
p = _post(db_sync, a.id, title="ねこ", ext="p4")
_enable(db_sync)
db_sync.commit()
assert translate_posts() == "interpreter unavailable"
db_sync.refresh(p)
assert p.translated_source_lang is None # will retry next run