feat(send): quote a long post's text in an expandable blockquote

A post whose text (the split `title` plus `content`, joined by
`site::compose_text`) reaches `CAPTION_QUOTE_TEXT_CHARS` — default 200,
`0` disables — now has that text wrapped in `<blockquote expandable>`
inside its caption, leaving the URL and author line outside the quote.

Applied at the send boundary (`send_media_sequence`, `send_animation` and
the inline answers), where the caption is already truncated and the same
cache snapshot supplies the text, so a fresh send, a link-cache resend
and a queued retry all decide identically. The text is located as what
follows the author link, with the visible prefix accepted as a match
because `truncate_caption` may cut inside it — that keeps the longest
posts, the ones that most need folding, quoted. Captions whose layout
moves the text elsewhere (pixiv's title-inside-a-link, a `/set_format`
that puts `{title}`/`{content}` first) stay unquoted rather than risking
a blockquote nested in a tag, and a caption that already carries one is
never wrapped again.

Telegram measures a caption *after entities parsing*, so the tags cost no
length and the 1024-character limit cannot be breached; retries replay
the unwrapped caption, so a threshold change takes effect immediately.
The edit-before-forward rewrite stays unquoted by design.

Verified against Telegram: a media-group caption built this way comes
back with `caption_entities` `url` @0, `text_link` @50,
`expandable_blockquote` @56 — the quote starts after the author line.

AGENTS.md

@@ -92,7 +92,7 @@ Docker: `docker build -t tgxmb .` then `docker run --rm -d --name tgxmb --env-fi
 - Package manager: **Cargo** (workspace with path dep `x-media` ← `xmedia-bot`). No `[workspace.package]`/shared deps — each crate lists deps independently.
 - **TLS is rustls end-to-end** (no native-tls/openssl in the tree, no libssl in the Docker runtime image): `teloxide` is declared `default-features = false` with `["webhooks-axum", "macros", "rustls", "ctrlc_handler"]` (the removed `default` also carried `native-tls` and `ctrlc_handler` — the latter must stay); x-media's reqwest is `default-features = false` with `["json", "rustls-tls"]` (webpki-roots baked in, so the image ships no CA bundle). One reqwest 0.12.28 in the lock.
 - **Versioning**: bump the version in all three places (`crates/x-media/Cargo.toml`, `crates/xmedia-bot/Cargo.toml`, `Cargo.lock`) and **keep `README.md`, `README.en.md` and `AGENTS.md` in sync with the code on every bump**, then commit (`chore: bump version to X.Y.Z`), create an annotated tag `vX.Y.Z`, and push branch + tag (the tag push triggers the Docker Hub build). The tag must equal both crate versions: `.github/workflows/docker.yml` verifies that before building, and `--locked` verifies the lock file.
-- Config is **environment-variable driven** (dotenv loads `.env`, gitignored; no `.env.example` exists). Key vars: `TELOXIDE_TOKEN` (required), `PIXIV_REFRESH_TOKEN`, `TWITTER_AUTH_TOKEN` (optional; x.com `auth_token` cookie — enables the logged-in GraphQL fallback that fetches NSFW tweets syndication withholds), `BILIBILI_COOKIE` (optional; whole bilibili cookie string — bilibili dynamics fetch anonymously and add their own device cookies, this only rescues an egress IP that bilibili has hard-flagged with `-352`/412), `BOT_ADMIN` (comma-separated ids), `EDIT_MESSAGE_TTL_SECONDS` (default 86400), `LINK_CACHE_TTL_SECONDS` (default 604800), `DATA_DIR` (default `data`, CWD-relative; the SQLite dir, auto-created), `WEBHOOK`/`WEBHOOK_URL`/`WEBHOOK_LISTEN`/`WEBHOOK_PORT`/`WEBHOOK_CERT`/`WEBHOOK_SECRET_TOKEN` (webhook mode requires URL/listen/port, `.expect`ed; `WEBHOOK_CERT` is Telegram-facing self-signed validation only — TLS must be terminated by a reverse proxy), `RUST_LOG`, `TELOXIDE_PROXY`, `LOCAL_USER_ID` (entrypoint only).
+- Config is **environment-variable driven** (dotenv loads `.env`, gitignored; no `.env.example` exists). Key vars: `TELOXIDE_TOKEN` (required), `PIXIV_REFRESH_TOKEN`, `TWITTER_AUTH_TOKEN` (optional; x.com `auth_token` cookie — enables the logged-in GraphQL fallback that fetches NSFW tweets syndication withholds), `BILIBILI_COOKIE` (optional; whole bilibili cookie string — bilibili dynamics fetch anonymously and add their own device cookies, this only rescues an egress IP that bilibili has hard-flagged with `-352`/412), `BOT_ADMIN` (comma-separated ids), `EDIT_MESSAGE_TTL_SECONDS` (default 86400), `LINK_CACHE_TTL_SECONDS` (default 604800), `CAPTION_QUOTE_TEXT_CHARS` (default 200; a post whose text — the `title` plus `content` joined, see `site::compose_text` — reaches this length gets that text wrapped in an expandable blockquote inside its caption, the URL and author line staying outside; `0` disables it. Applied at the send boundary in `send::quote_long_caption`, which locates the text as what follows the author link, so a `/set_format` that moves `{title}`/`{content}` elsewhere and pixiv's title-inside-a-link layout opt out; `copy_messages` forwards and queued retries inherit the wrap, while the edit-before-forward rewrite stays unquoted by design), `DATA_DIR` (default `data`, CWD-relative; the SQLite dir, auto-created), `WEBHOOK`/`WEBHOOK_URL`/`WEBHOOK_LISTEN`/`WEBHOOK_PORT`/`WEBHOOK_CERT`/`WEBHOOK_SECRET_TOKEN` (webhook mode requires URL/listen/port, `.expect`ed; `WEBHOOK_CERT` is Telegram-facing self-signed validation only — TLS must be terminated by a reverse proxy), `RUST_LOG`, `TELOXIDE_PROXY`, `LOCAL_USER_ID` (entrypoint only).
 - SQLite via `rusqlite` with `bundled` feature (no system libsqlite needed). DB file `$DATA_DIR/task_queue.db` (default `data/task_queue.db`, CWD-relative — run from the workspace root, or `/app` in Docker; set `DATA_DIR` to pin state anywhere). Mount `./data` and `./cert` volumes.
 - `.gitattributes` enforces LF for `*.sh` (CRLF breaks shebangs in containers). `.gitignore`: `.env`, `data/`, `cert/`, `docker-compose.yml`, `/target`, `.idea/`.
 - Docs are in Chinese (README, AGENTS.md); user-facing bot strings are in English. Keep that split when editing user-facing strings and docs.

README.en.md

@@ -6,6 +6,7 @@ A Telegram bot that turns post links from X / Twitter, Pixiv, Bluesky, Misskey (
 
 - Sending a link in a private chat fetches and sends the images, videos and GIFs automatically; oversized media is split into batches
 - Text-only posts report "no media"; unsupported links are silently ignored
+- Long posts (text ≥ `CAPTION_QUOTE_TEXT_CHARS`, default 200) show **the text part** of their caption inside a collapsible blockquote, with the link and author line left outside it
 - Inline queries (`@bot <link>`)
 - Bind a forward channel for automatic forwarding; edit the caption before forwarding and apply custom templates
 - Failed sends are retried automatically with persistence; the user is notified after retries are exhausted
@@ -87,6 +88,7 @@ Telegram only accepts ports 443/80/88/8443.
 | `BOT_ADMIN` | Admin chat IDs, comma-separated; receives start/stop notifications |
 | `EDIT_MESSAGE_TTL_SECONDS` | Edit-before-forward record expiry in seconds, default 86400 |
 | `LINK_CACHE_TTL_SECONDS` | Link-result cache expiry in seconds, default 604800 (7 days) |
+| `CAPTION_QUOTE_TEXT_CHARS` | **The text part** of the caption (the joined `{title}` + `{content}`) is wrapped in a collapsible blockquote once it reaches this many characters, default 200; `0` disables |
 | `DATA_DIR` | Data directory (where the SQLite `task_queue.db` lives), default `data` (relative to the working directory, created automatically) |
 | `RUST_LOG` | Log level |
 | `TELOXIDE_PROXY` | HTTP proxy (e.g. `http://127.0.0.1:10808`); applies to both the Telegram Bot API and site fetches — required on restricted networks (e.g. behind the GFW) |

README.md

@@ -6,6 +6,7 @@ Telegram 机器人，将 X / Twitter、Pixiv、Bluesky、Misskey (misskey.io)、
 
 - 私聊发送链接后自动抓取并发送图片、视频与 GIF，超量图片自动分批
 - 纯文字帖提示无媒体；不支持的链接静默忽略
+- 长帖（正文 ≥ `CAPTION_QUOTE_TEXT_CHARS`，默认 200）的**正文部分**用可折叠引用块展示，链接与作者行留在引用块外
 - 支持内联查询（`@机器人 <链接>`）
 - 可绑定转发频道自动转发；支持转发前编辑 caption 与自定义模板
 - 发送失败自动重试并持久化，重试耗尽后通知用户
@@ -87,6 +88,7 @@ Telegram 只接受 443/80/88/8443 端口。
 | `BOT_ADMIN` | 管理员聊天 ID，逗号分隔；接收启动/停止通知 |
 | `EDIT_MESSAGE_TTL_SECONDS` | 转发前编辑记录过期秒数，默认 86400 |
 | `LINK_CACHE_TTL_SECONDS` | 链接结果缓存过期秒数，默认 604800（7 天） |
+| `CAPTION_QUOTE_TEXT_CHARS` | 正文（`{title}` + `{content}` 合计）达到该长度（字符）时，caption 的**正文部分**用可折叠引用块包裹，默认 200；`0` 关闭 |
 | `DATA_DIR` | 数据目录（SQLite 数据库 `task_queue.db` 所在目录），默认 `data`（相对工作目录，会自动创建） |
 | `RUST_LOG` | 日志级别 |
 | `TELOXIDE_PROXY` | HTTP 代理（如 `http://127.0.0.1:10808`）；同时作用于 Telegram Bot API 与站点抓取请求，网络受限环境（如 GFW）必需 |

crates/xmedia-bot/src/config.rs

@@ -13,6 +13,10 @@ pub struct Config {
     pub edit_message_ttl: Duration,
     /// LINK_CACHE_TTL_SECONDS, default 604800 (7 days).
     pub link_cache_ttl: Duration,
+    /// CAPTION_QUOTE_TEXT_CHARS, default 200: a post whose text (title plus
+    /// content) is at least this many characters gets that text wrapped in an
+    /// expandable blockquote inside its caption. `0` disables the wrap.
+    pub caption_quote_text_chars: usize,
     // Webhook settings (moved out of main; names/defaults unchanged).
     pub webhook_enabled: bool,
     pub webhook_url: Option<url::Url>,
@@ -58,6 +62,7 @@ impl Config {
             Duration::from_secs(parse_u64("EDIT_MESSAGE_TTL_SECONDS", 24 * 3600));
         let link_cache_ttl =
             Duration::from_secs(parse_u64("LINK_CACHE_TTL_SECONDS", 7 * 24 * 3600));
+        let caption_quote_text_chars = parse_u64("CAPTION_QUOTE_TEXT_CHARS", 200) as usize;
 
         let webhook_enabled = env::var("WEBHOOK")
             .is_ok_and(|v| matches!(v.to_lowercase().as_str(), "true" | "yes" | "1"));
@@ -93,6 +98,7 @@ impl Config {
             admin_ids,
             edit_message_ttl,
             link_cache_ttl,
+            caption_quote_text_chars,
             webhook_enabled,
             webhook_url,
             webhook_listen,

crates/xmedia-bot/src/ctx.rs

@@ -88,6 +88,12 @@ pub(crate) mod test_support {
             &self.chat_store
         }
 
+        /// The parsed config, mutable so a test can pin a knob (e.g. the
+        /// caption-quote threshold) instead of depending on the environment.
+        pub(crate) fn config_mut(&mut self) -> &mut Config {
+            &mut self.config
+        }
+
         pub(crate) fn link_cache(&self) -> &LinkCache {
             &self.link_cache
         }

crates/xmedia-bot/src/handlers/inline.rs

@@ -127,8 +127,23 @@ async fn answer_inline_query(bot: Bot, query: InlineQuery) -> Result<bool, Reque
         Ok(Some(fetched)) => {
             let mut results: Vec<InlineQueryResult> = Vec::new();
             // Inline results have the same 1024-char caption limit as regular
-            // messages; truncate once here for all items.
+            // messages; truncate once here for all items, then apply the same
+            // long-post quoting as the send paths. `answer_inline_query` has no
+            // `AppContext` (the debounce spawns it), so the parsed config comes
+            // from the process-wide static, and the text is the *escaped*
+            // title/content the built-in caption embeds (the raw
+            // `Fetched.title`/`content` differ whenever the post contains
+            // `<`/`&`).
             let caption = x_media::site::truncate_caption(&fetched.caption);
+            let text = fetched
+                .render_fields()
+                .map(|(_, _, title, content, _)| x_media::site::compose_text(title, content))
+                .unwrap_or_default();
+            let caption = crate::send::quote_long_caption(
+                &caption,
+                &text,
+                super::CONFIG.caption_quote_text_chars,
+            );
             for (i, media) in fetched.media.iter().enumerate() {
                 let id = format!("{i}");
                 let Some(url) = url::Url::parse(media.url()).ok() else {
@@ -138,7 +153,7 @@ async fn answer_inline_query(bot: Bot, query: InlineQuery) -> Result<bool, Reque
                     .thumbnail_url()
                     .and_then(|t| url::Url::parse(t).ok())
                     .unwrap_or_else(|| url.clone());
-                let caption = caption.clone();
+                let caption = caption.clone().into_owned();
                 let result = match media {
                     Media::Illustration { .. } => {
                         // Inline photo results have their own (smaller) size

crates/xmedia-bot/src/handlers/urls.rs

@@ -534,6 +534,35 @@ mod tests {
         );
     }
 
+    /// The caption-quote threshold matches the post's text inside the caption,
+    /// so a long-text cache hit is quoted and a short-text one is not.
+    #[tokio::test]
+    async fn cache_hit_quotes_a_long_text_caption() {
+        let mut stores = TestStores::new();
+        stores.config_mut().caption_quote_text_chars = 3;
+        let prefix = "https://x.com/u/status/1\n<a href=\"au\">a</a>: ";
+
+        for (text, expected) in [
+            (
+                "abc",
+                format!("{prefix}<blockquote expandable>abc</blockquote>"),
+            ),
+            ("ab", format!("{prefix}ab")),
+        ] {
+            let sender = MockSender::scripted(vec![Outcome::GroupOk], permanent_error);
+            let ctx = stores.ctx(&sender);
+            let mut entry = cached_photo_entry();
+            entry.caption = format!("{prefix}{text}");
+            entry.title = String::new();
+            entry.content = text.into();
+            stores.link_cache().put("twitter:1", &entry).await;
+
+            url_media(&ctx, 1, 2, "https://x.com/u/status/1", PostSend::FromChat).await;
+
+            assert_eq!(sender.captions(), vec![expected], "text {text:?}");
+        }
+    }
+
     #[tokio::test]
     async fn unsupported_url_is_ignored_silently() {
         let stores = TestStores::new();

crates/xmedia-bot/src/media_sender.rs

@@ -327,9 +327,20 @@ pub(crate) mod test_support {
             &self,
             _chat_id: ChatId,
             _reply_to: MessageId,
-            _items: Vec<InputMedia>,
+            items: Vec<InputMedia>,
         ) -> BoxFuture<'_, Result<Vec<Message>, RequestError>> {
             Box::pin(async move {
+                // Record the captions exactly as Telegram receives them (only
+                // the first item of a group carries one), so tests can assert
+                // what a recipient sees.
+                self.captions
+                    .lock()
+                    .extend(items.iter().filter_map(|item| match item {
+                        InputMedia::Photo(photo) => photo.caption.clone(),
+                        InputMedia::Video(video) => video.caption.clone(),
+                        InputMedia::Animation(animation) => animation.caption.clone(),
+                        _ => None,
+                    }));
                 match self.next("send_media_group") {
                     Outcome::GroupOk => Ok(Vec::new()),
                     Outcome::GroupErr => Err(self.error()),

crates/xmedia-bot/src/send/mod.rs

@@ -18,6 +18,7 @@ use crate::media_sender::MediaSender;
 use input_media::{build_media_group, input_file_for, item_url};
 use post_send::{cache_animation_send, cache_sent_task};
 use serde::{Deserialize, Serialize};
+use std::borrow::Cow;
 use std::sync::LazyLock;
 use teloxide::prelude::*;
 use teloxide::types::{ChatId, InputFile, InputMedia, MessageId};
@@ -388,6 +389,61 @@ fn updated_sequence_task(task: &Task, batch_index: usize, sent_message_ids: Vec<
     updated
 }
 
+/// The caption's text tail: everything after the author link, provided it
+/// really is the post's text.
+///
+/// `text` is the *escaped* title + content the caption embeds; the caption may
+/// have been truncated inside it, in which case only its prefix is present, so
+/// the tail only has to match the text's start. `None` for a caption with
+/// another layout — pixiv's title-inside-a-link, a `/set_format` that moves
+/// `{title}`/`{content}` off the author line — which is left unquoted instead
+/// of guessing where the text begins.
+fn text_tail<'c>(caption: &'c str, text: &str) -> Option<&'c str> {
+    let (_, tail) = caption.rsplit_once("</a>: ")?;
+    let visible = tail.strip_suffix('\u{2026}').unwrap_or(tail);
+    (!visible.is_empty() && text.starts_with(visible)).then_some(tail)
+}
+
+/// The text a task's caption embeds, read from the same cache snapshot the
+/// caption came from: `title` and `content` joined the way the sites' built-in
+/// captions join them.
+fn task_text(task: &Task) -> String {
+    task.cache_data()
+        .map(|data| x_media::site::compose_text(&data.title, &data.content))
+        .unwrap_or_default()
+}
+
+/// Wraps the post's text inside the caption in an expandable blockquote once
+/// that text is long enough that the message would otherwise be a wall of text
+/// (`threshold` is `CAPTION_QUOTE_TEXT_CHARS`; `0` disables the wrap). The URL
+/// and the author line stay outside the quote.
+///
+/// Applied at the send boundary, after the caller's `truncate_caption`:
+/// Telegram measures a caption *after entities parsing*, so the tags cost no
+/// length and a wrapped caption cannot exceed the 1024-character limit.
+/// Retries replay the task's (unwrapped) caption, so the decision is remade on
+/// every attempt — changing the threshold takes effect immediately.
+///
+/// A caption that already carries a blockquote is left as it is: the API
+/// rejects nested ones ("all other entities can't contain each other"), and a
+/// user-written `/set_format` template may contain one.
+pub(crate) fn quote_long_caption<'a>(
+    caption: &'a str,
+    text: &str,
+    threshold: usize,
+) -> Cow<'a, str> {
+    if threshold == 0 || caption.contains("<blockquote") || text.chars().count() < threshold {
+        return Cow::Borrowed(caption);
+    }
+    let Some(tail) = text_tail(caption, text) else {
+        return Cow::Borrowed(caption);
+    };
+    let prefix = &caption[..caption.len() - tail.len()];
+    Cow::Owned(format!(
+        "{prefix}<blockquote expandable>{tail}</blockquote>"
+    ))
+}
+
 /// Sends the media batches starting at `task.batch_index`, extending
 /// `sent_message_ids`. Returns all sent message ids on full success; on
 /// failure returns a [`SendError`] whose task carries the resumed state.
@@ -406,6 +462,10 @@ pub async fn send_media_sequence(ctx: &AppContext<'_>, task: &Task) -> Result<Ve
     };
     let chat_id = *chat_id;
     let reply_to = *reply_to_message_id;
+    // A long post is quoted so the message reads as a card rather than a wall
+    // of text; the text comes from the same cache snapshot as the caption.
+    let text = task_text(task);
+    let caption = quote_long_caption(caption, &text, ctx.config.caption_quote_text_chars);
     let mut sent = sent_message_ids.clone();
     // File ids accumulated across batches for the link cache. Only a fresh
     // (non-resumed) full send populates the cache.
@@ -414,7 +474,7 @@ pub async fn send_media_sequence(ctx: &AppContext<'_>, task: &Task) -> Result<Ve
     for idx in *batch_index..media_batches.len() {
         let batch = &media_batches[idx];
         let caption = if idx == 0 {
-            Some(caption.as_str())
+            Some(caption.as_ref())
         } else {
             None
         };
@@ -515,6 +575,10 @@ pub async fn send_animation(ctx: &AppContext<'_>, task: &Task) -> Result<Vec<i64
     };
     let chat_id = *chat_id;
     let reply_to = *reply_to_message_id;
+    // Same long-post quoting as the media-group path (see
+    // `quote_long_caption`); both sends below share this string.
+    let text = task_text(task);
+    let caption = quote_long_caption(caption, &text, ctx.config.caption_quote_text_chars);
     let (media_url, has_spoiler) = match animation {
         MediaItemPayload::Animation {
             media, has_spoiler, ..
@@ -536,7 +600,7 @@ pub async fn send_animation(ctx: &AppContext<'_>, task: &Task) -> Result<Vec<i64
         ctx.sender,
         chat_id,
         reply_to,
-        caption,
+        &caption,
         has_spoiler,
         url_file,
     )
@@ -571,7 +635,7 @@ pub async fn send_animation(ctx: &AppContext<'_>, task: &Task) -> Result<Vec<i64
                         ctx.sender,
                         chat_id,
                         reply_to,
-                        caption,
+                        &caption,
                         has_spoiler,
                         animation.media,
                     )
@@ -930,10 +994,16 @@ mod tests {
     }
 
     fn sequence_task(media: &str) -> Task {
+        sequence_task_with(media, "cap", None)
+    }
+
+    /// A media-group task; `text` (when given) rides in the link-cache
+    /// snapshot as `content`, which is where the quote threshold reads it.
+    fn sequence_task_with(media: &str, caption: &str, text: Option<&str>) -> Task {
         Task::SendMediaSequence {
             chat_id: 1,
             reply_to_message_id: 2,
-            caption: "cap".into(),
+            caption: caption.into(),
             media_batches: vec![vec![MediaItemPayload::Photo {
                 media: media.to_string(),
                 has_spoiler: false,
@@ -947,7 +1017,99 @@ mod tests {
             forward_channel_id: None,
             notify_chat_id: Some(1),
             notify_message_id: Some(2),
-            cache_data: None,
+            // The snapshot splits the post's text into title/content the way a
+            // real fetch does; the quote threshold joins them again.
+            cache_data: text.map(|text| CachedPost {
+                url: "https://x.com/u/status/1".into(),
+                caption: caption.into(),
+                title: String::new(),
+                content: text.into(),
+                author: "me".into(),
+                author_url: "https://x.com/u".into(),
+                tags: String::new(),
+                sensitive: false,
+                media: vec![],
+            }),
+        }
+    }
+
+    /// A media-group task whose built-in caption carries `text` behind the
+    /// author link — the shape the quote threshold locates the text in.
+    fn sequence_task_with_text(media: &str, text: &str) -> Task {
+        let caption =
+            format!("https://x.com/u/status/1\n<a href=\"https://x.com/u\">me</a>: {text}");
+        sequence_task_with(media, &caption, Some(text))
+    }
+
+    #[test]
+    fn quote_long_caption_wraps_only_the_text_tail() {
+        let text = "一二三四五";
+        let prefix = "https://x.com/u/status/1\n<a href=\"https://x.com/u\">me</a>: ";
+        let caption = format!("{prefix}{text}");
+
+        // Only the text goes inside the quote; the URL and author line stay
+        // outside.
+        assert_eq!(
+            quote_long_caption(&caption, text, 5),
+            format!("{prefix}<blockquote expandable>{text}</blockquote>")
+        );
+        // One char below the threshold, disabled, and a short text: untouched.
+        assert_eq!(
+            quote_long_caption(&caption, text, 6),
+            format!("{prefix}{text}")
+        );
+        assert_eq!(quote_long_caption(&caption, text, 0), caption);
+        // No author-line anchor means no text to locate — a pixiv caption
+        // (title inside the link) and a `{content}`-first format stay as they
+        // are rather than risking a blockquote nested in a tag.
+        let pixiv =
+            format!("<a href=\"https://pixiv.net/1\">{text}</a> / <a href=\"u\">me</a>\ntag");
+        assert_eq!(quote_long_caption(&pixiv, text, 5), pixiv);
+        let content_first = format!("{text}\nhttps://x.com/u/status/1");
+        assert_eq!(quote_long_caption(&content_first, text, 5), content_first);
+        // An empty body has nothing to quote.
+        assert_eq!(quote_long_caption(prefix, text, 5), prefix);
+        // A caption that already carries a blockquote is never nested.
+        let quoted = format!("<blockquote>{caption}</blockquote>");
+        assert_eq!(quote_long_caption(&quoted, text, 5), quoted);
+    }
+
+    /// `truncate_caption` cuts inside the text and appends an ellipsis; the
+    /// visible prefix still marks it, so the long-text case that most needs
+    /// quoting is still quoted.
+    #[test]
+    fn quote_long_caption_wraps_a_truncated_text() {
+        let text = "一二三四五六七八九十";
+        let prefix = "https://x.com/u/status/1\n<a href=\"https://x.com/u\">me</a>: ";
+        let caption = format!("{prefix}一二三四五…");
+        assert_eq!(
+            quote_long_caption(&caption, text, 5),
+            format!("{prefix}<blockquote expandable>一二三四五…</blockquote>")
+        );
+    }
+
+    #[tokio::test]
+    async fn long_text_caption_reaches_telegram_quoted() {
+        // The threshold is pinned here instead of read from the environment.
+        let dir = tempfile::tempdir().unwrap();
+        let file = dir.path().join("media.jpg");
+        std::fs::write(&file, b"not-a-real-jpeg").unwrap();
+        let mut stores = TestStores::new();
+        stores.config_mut().caption_quote_text_chars = 5;
+        let prefix = "https://x.com/u/status/1\n<a href=\"https://x.com/u\">me</a>: ";
+
+        for (text, expected) in [
+            (
+                "一二三四五",
+                format!("{prefix}<blockquote expandable>一二三四五</blockquote>"),
+            ),
+            ("一二三四", format!("{prefix}一二三四")),
+        ] {
+            let sender = MockSender::scripted(vec![Outcome::GroupOk], media_fetch_error);
+            let ctx = stores.ctx(&sender);
+            let task = sequence_task_with_text(file.to_str().unwrap(), text);
+            assert!(send_media_sequence(&ctx, &task).await.is_ok());
+            assert_eq!(sender.captions(), vec![expected], "text {text:?}");
         }
     }