feat: normalize markdown message send/reply output (#28)

* feat(IM):  im markdown send/reply

Change-Id: I6b53eb2207d7c2393d3c7d108df3ba197b9eae46

* add Resolve content type

Change-Id: I71a0cdb8b500ca1496b23fede1f2b2617f16ec63

Author: 91-enjoy

shortcuts/im/coverage_additional_test.go

@@ -101,6 +101,9 @@ func TestResolveMarkdownAsPost(t *testing.T) {
 	if !strings.Contains(got, `#### Title`) || !strings.Contains(got, `##### Subtitle`) {
 		t.Fatalf("resolveMarkdownAsPost() = %q, want optimized heading levels", got)
 	}
+	if strings.Contains(got, `<br>`) {
+		t.Fatalf("resolveMarkdownAsPost() = %q, want no literal <br>", got)
+	}
 }
 
 func TestValidateContentFlags(t *testing.T) {

shortcuts/im/helpers.go

@@ -619,31 +619,22 @@ func readMp4Duration(f *os.File, fileSize int64) int64 {
 // Steps:
 //  1. Extract code blocks with placeholders to protect them
 //  2. Downgrade headings: H1 → H4, H2~H6 → H5 (only when H1~H3 present)
-//  3. Add <br> between consecutive headings
-//  4. Add spacing around tables with <br>
-//  5. Restore code blocks with <br> wrappers
-//  6. Compress excess blank lines
-//  7. Strip invalid image references (keep only img_xxx keys)
+//  3. Normalize spacing between consecutive headings and tables with blank lines
+//  4. Restore code blocks
+//  5. Compress excess blank lines
+//  6. Strip invalid image references (keep only img_xxx keys)
 var (
-	reH2toH6       = regexp.MustCompile(`(?m)^#{2,6} (.+)$`)
-	reH1           = regexp.MustCompile(`(?m)^# (.+)$`)
-	reHasH1toH3    = regexp.MustCompile(`(?m)^#{1,3} `)
-	reConsecH      = regexp.MustCompile(`(?m)^(#{4,5} .+)\n{1,2}(#{4,5} )`)
-	reTableNoGap   = regexp.MustCompile(`(?m)^([^|\n].*)\n(\|.+\|)`)
-	reTableBefore  = regexp.MustCompile(`\n\n((?:\|.+\|[^\S\n]*\n?)+)`)
-	reTableAfter   = regexp.MustCompile(`(?m)((?:^\|.+\|[^\S\n]*\n?)+)`)
-	reTableTxtPre  = regexp.MustCompile(`(?m)^([^\n]+)\n\n(<br>)\n\n(\|)`)
-	reTableBoldPre = regexp.MustCompile(`(?m)^(\*\*.+)\n\n(<br>)\n\n(\|)`)
-	reTableTxtPost = regexp.MustCompile(`(?m)(\|[^\n]*\n)\n(<br>\n)([^\n]+)`)
-	reExcessNL     = regexp.MustCompile(`\n{3,}`)
-	reInvalidImg   = regexp.MustCompile(`!\[[^\]]*\]\(([^)\s]+)\)`)
-	reCodeBlock    = regexp.MustCompile("```[\\s\\S]*?```")
+	reH2toH6     = regexp.MustCompile(`(?m)^#{2,6} (.+)$`)
+	reH1         = regexp.MustCompile(`(?m)^# (.+)$`)
+	reHasH1toH3  = regexp.MustCompile(`(?m)^#{1,3} `)
+	reConsecH    = regexp.MustCompile(`(?m)^(#{4,5} .+)\n{1,2}(#{4,5} )`)
+	reTableNoGap = regexp.MustCompile(`(?m)^([^|\n].*)\n(\|.+\|)`)
+	reTableAfter = regexp.MustCompile(`(?m)((?:^\|.+\|[^\S\n]*\n?)+)`)
+	reExcessNL   = regexp.MustCompile(`\n{3,}`)
+	reInvalidImg = regexp.MustCompile(`!\[[^\]]*\]\(([^)\s]+)\)`)
+	reCodeBlock  = regexp.MustCompile("```[\\s\\S]*?```")
 )
 
-func isTableSpacingProtectedLine(line string) bool {
-	return strings.HasPrefix(line, "#### ") || strings.HasPrefix(line, "##### ") || strings.HasPrefix(line, "**")
-}
-
 func optimizeMarkdownStyle(text string) string {
 	const mark = "___CB_"
 	var codeBlocks []string
@@ -659,29 +650,13 @@ func optimizeMarkdownStyle(text string) string {
 		r = reH1.ReplaceAllString(r, "#### $1")
 	}
 
-	r = reConsecH.ReplaceAllString(r, "$1\n<br>\n$2")
+	r = reConsecH.ReplaceAllString(r, "$1\n\n$2")
 
 	r = reTableNoGap.ReplaceAllString(r, "$1\n\n$2")
-	r = reTableBefore.ReplaceAllString(r, "\n\n<br>\n\n$1")
-	r = reTableAfter.ReplaceAllString(r, "$1\n<br>\n")
-	r = reTableTxtPre.ReplaceAllStringFunc(r, func(m string) string {
-		sub := reTableTxtPre.FindStringSubmatch(m)
-		if len(sub) != 4 || isTableSpacingProtectedLine(sub[1]) {
-			return m
-		}
-		return sub[1] + "\n" + sub[2] + "\n" + sub[3]
-	})
-	r = reTableBoldPre.ReplaceAllString(r, "$1\n$2\n\n$3")
-	r = reTableTxtPost.ReplaceAllStringFunc(r, func(m string) string {
-		sub := reTableTxtPost.FindStringSubmatch(m)
-		if len(sub) != 4 || isTableSpacingProtectedLine(sub[3]) {
-			return m
-		}
-		return sub[1] + sub[2] + sub[3]
-	})
+	r = reTableAfter.ReplaceAllString(r, "$1\n")
 
 	for i, block := range codeBlocks {
-		r = strings.Replace(r, fmt.Sprintf("%s%d___", mark, i), "\n<br>\n"+block+"\n<br>\n", 1)
+		r = strings.Replace(r, fmt.Sprintf("%s%d___", mark, i), block, 1)
 	}
 
 	r = reExcessNL.ReplaceAllString(r, "\n\n")

shortcuts/im/helpers_test.go

@@ -282,7 +282,7 @@ func TestOptimizeMarkdownStyle(t *testing.T) {
 		{
 			name:  "heading downgrade H1 and H2",
 			input: "# Title\n## Section\ntext",
-			want:  "#### Title\n<br>\n##### Section\ntext",
+			want:  "#### Title\n\n##### Section\ntext",
 		},
 		{
 			name:  "no downgrade when no H1-H3",
@@ -292,17 +292,17 @@ func TestOptimizeMarkdownStyle(t *testing.T) {
 		{
 			name:  "code block protected",
 			input: "# Title\n```\n# not a heading\n```\ntext",
-			want:  "#### Title\n\n<br>\n```\n# not a heading\n```\n<br>\n\ntext",
+			want:  "#### Title\n```\n# not a heading\n```\ntext",
 		},
 		{
 			name:  "table spacing",
 			input: "text\n| A | B |\n| - | - |\n| 1 | 2 |\nafter",
-			want:  "text\n<br>\n| A | B |\n| - | - |\n| 1 | 2 |\n<br>\nafter",
+			want:  "text\n\n| A | B |\n| - | - |\n| 1 | 2 |\n\nafter",
 		},
 		{
 			name:  "table spacing keeps heading separation",
 			input: "# Title\n| A | B |\n| - | - |\n| 1 | 2 |\n## Next",
-			want:  "#### Title\n\n<br>\n\n| A | B |\n| - | - |\n| 1 | 2 |\n\n<br>\n##### Next",
+			want:  "#### Title\n\n| A | B |\n| - | - |\n| 1 | 2 |\n\n##### Next",
 		},
 		{
 			name:  "excess blank lines compressed",

shortcuts/im/im_messages_send.go

@@ -188,7 +188,7 @@ var ImMessagesSend = common.Shortcut{
 				return output.ErrValidation("%v", err)
 			}
 		}
-
+		// Resolve content type
 		if markdown != "" {
 			msgType, content = "post", resolveMarkdownAsPost(ctx, runtime, markdown)
 		} else if mt, c, err := resolveMediaContent(ctx, runtime, text, imageVal, fileVal, videoVal, videoCoverVal, audioVal); err != nil {

skills/lark-im/references/lark-im-messages-reply.md

@@ -18,10 +18,88 @@ Replies sent by this tool are visible to other people. Before calling it, you **
 
 When using `--as bot`, the reply is sent in the app's name, so make sure the app has already been added to the target chat.
 
+## Choose The Right Content Flag
+
+| Need | Recommended flag | Why |
+|------|------|------|
+| Reply with plain text exactly as written | `--text` | Wrapped directly to `{"text":"..."}` |
+| Reply with simple Markdown and accept conversion | `--markdown` | Automatically converted to `post` JSON |
+| Precisely control the reply payload | `--content` | You provide the exact JSON |
+| Reply with media | `--image` / `--file` / `--video` / `--audio` | Shortcut uploads local files automatically |
+
+### `--text` vs `--markdown`
+
+- Use `--text` when the reply should remain plain text and you want exact control over line breaks, spacing, indentation, code samples, or literal Markdown characters.
+- Use `--markdown` when you want a lightweight formatted reply and you accept that the shortcut will normalize and rewrite parts of the content before sending.
+- Use `--content` when you need exact `post` JSON, a card, a title, multiple locales, or any structure that `--markdown` cannot express reliably.
+
+## What `--markdown` Really Does
+
+`--markdown` does **not** send arbitrary raw Markdown to the API.
+
+The shortcut:
+
+1. Forces `msg_type=post`
+2. Resolves remote Markdown images like `![x](https://...)`
+3. Normalizes the Markdown for Feishu post rendering
+4. Wraps the final content as:
+
+```json
+{"zh_cn":{"content":[[{"tag":"md","text":"..."}]]}}
+```
+
+So `--markdown` is a convenience mode, not a full Markdown compatibility layer.
+
+### Current Markdown Caveats
+
+- It does **not** promise full CommonMark / GitHub Flavored Markdown support.
+- It always becomes a `post` payload with a single `zh_cn` locale.
+- It does **not** let you set a `post` title.
+- Headings are rewritten:
+    - `# Title` becomes `#### Title`
+    - `##` to `######` are normalized to `#####` when the content contains H1-H3
+- Consecutive headings are separated with blank lines after heading normalization.
+- Block spacing and line breaks may be normalized during conversion.
+- Code blocks are preserved as code blocks.
+- Excess blank lines are compressed.
+- Only remote `http://...`, `https://...`, or already-uploaded `img_xxx` Markdown images are kept reliably.
+- Local paths in Markdown image syntax like `![x](./a.png)` are **not** auto-uploaded by `--markdown`.
+- If remote Markdown image handling fails, that image is removed with a warning.
+
+If you need exact output, use `--msg-type post --content ...` instead of `--markdown`.
+
+## Preserving Formatting
+
+If the reply contains multiple lines, code blocks, indentation, tabs, or a lot of escaping, prefer `$'...'`.
+
+### When formatting must be preserved
+
+Use `--text` plus `$'...'`:
+
+```bash
+lark-cli im +messages-reply --message-id om_xxx --text $'Received\nI will check this today.\nOwner: alice'
+```
+
+```bash
+lark-cli im +messages-reply --message-id om_xxx --text $'```sql\nselect * from jobs;\n```'
+```
+
+This keeps the reply as plain text instead of converting it to a `post`.
+
+### When formatting does not need exact preservation
+
+Use `--markdown`:
+
+```bash
+lark-cli im +messages-reply --message-id om_xxx --markdown $'## Follow-up\n\n- I reproduced it\n- I am fixing it'
+```
+
+This is better for quick readable formatting, but the final payload may still differ from the source text because headings and spacing are normalized before sending.
+
 ## Commands
 
 ```bash
-# Reply to a message (plain text, bot identity, --text is recommended)
+# Reply to a message (plain text, --text is recommended for normal replies)
 lark-cli im +messages-reply --message-id om_xxx --text "Received"
 
 # Equivalent manual JSON
@@ -30,13 +108,16 @@ lark-cli im +messages-reply --message-id om_xxx --content '{"text":"Received"}'
 # Reply as a bot
 lark-cli im +messages-reply --message-id om_xxx --text "bot reply" --as bot
 
+# Reply with preserved multi-line text
+lark-cli im +messages-reply --message-id om_xxx --text $'Line 1\nLine 2\n  indented line'
+
 # Reply inside the thread (message appears in the target thread)
 lark-cli im +messages-reply --message-id om_xxx --text "Let's discuss this" --reply-in-thread
 
-# Bot identity + thread reply
-lark-cli im +messages-reply --message-id om_xxx --text "bot reply" --as bot --reply-in-thread
+# Reply with basic Markdown (will be converted to post JSON)
+lark-cli im +messages-reply --message-id om_xxx --markdown $'## Reply\n\n- item 1\n- item 2'
 
-# Reply with a rich-text message
+# If you need exact post structure, send JSON directly
 lark-cli im +messages-reply --message-id om_xxx --msg-type post --content '{"zh_cn":{"title":"Reply","content":[[{"tag":"text","text":"Detailed content"}]]}}'
 
 # Reply with a local image (uploaded automatically before sending)
@@ -52,23 +133,23 @@ lark-cli im +messages-reply --message-id om_xxx --video ./demo.mp4 --video-cover
 lark-cli im +messages-reply --message-id om_xxx --text "Received" --idempotency-key my-unique-id
 
 # Preview the request without executing it
-lark-cli im +messages-reply --message-id om_xxx --text "Test" --dry-run
+lark-cli im +messages-reply --message-id om_xxx --markdown $'## Test\n\nhello' --dry-run
 ```
 
 ## Parameters
 
 | Parameter | Required | Description |
 |------|------|------|
 | `--message-id <id>` | Yes | ID of the message being replied to (`om_xxx`) |
-| `--msg-type <type>` | No | Message type (default `text`): `text`, `post`, `image`, `file`, `audio`, `media`, `interactive`, `share_chat`, `share_user` |
-| `--content <json>` | One of content options | Reply content as a JSON string; format depends on `msg_type` |
-| `--text <string>` | One of content options | Plain text message (automatically wrapped as `{"text":"..."}` JSON) |
-| `--markdown <string>` | One of content options | Markdown text (auto-wrapped as post format with style optimization; image URLs auto-resolved) |
-| `--image <path\|key>` | One of content options | Local image path, `image_key` (`img_xxx`)|
-| `--file <path\|key>` | One of content options | Local file path, `file_key` (`file_xxx`)|
-| `--video <path\|key>` | One of content options | Local video path, `file_key`; **must be used together with `--video-cover`** |
-| `--video-cover <path\|key>` | **Required with `--video`** | Video cover image path, `image_key` (`img_xxx`) |
-| `--audio <path\|key>` | One of content options | Local audio path, `file_key` |
+| `--msg-type <type>` | No | Message type (default `text`). If you use `--text` / `--markdown` / media flags, the effective type is inferred automatically. Explicitly setting a conflicting `--msg-type` fails validation |
+| `--content <json>` | One content option | Exact reply content as JSON. The JSON must match the effective `--msg-type` |
+| `--text <string>` | One content option | Plain text reply. Best default when you need exact text and formatting preservation |
+| `--markdown <string>` | One content option | Convenience Markdown input. Internally converted to `post` JSON with Feishu-specific normalization |
+| `--image <path\|key>` | One content option | Local image path or `image_key` (`img_xxx`) |
+| `--file <path\|key>` | One content option | Local file path or `file_key` (`file_xxx`) |
+| `--video <path\|key>` | One content option | Local video path or `file_key`; **must be used together with `--video-cover`** |
+| `--video-cover <path\|key>` | **Required with `--video`** | Video cover image path or `image_key` (`img_xxx`) |
+| `--audio <path\|key>` | One content option | Local audio path or `file_key` |
 | `--reply-in-thread` | No | Reply inside the thread. The reply appears in the target message's thread instead of the main chat stream |
 | `--idempotency-key <key>` | No | Idempotency key; the same key sends only one reply within 1 hour |
 | `--as <identity>` | No | Identity type: `bot` only |
@@ -78,6 +159,15 @@ lark-cli im +messages-reply --message-id om_xxx --text "Test" --dry-run
 >
 > **Video cover rule:** `--video` **must** be accompanied by `--video-cover`. Omitting `--video-cover` when using `--video` will fail validation. `--video-cover` cannot be used without `--video`.
 
+## Common Mistakes
+
+- Choosing `--markdown` when you actually need exact plain text. If exact line breaks and spacing matter, use `--text`, usually with `$'...'`.
+- Assuming `--markdown` supports all Markdown features. It does not; it is converted into a Feishu `post` payload and rewritten first.
+- Putting local image paths inside Markdown like `![x](./a.png)`. `--markdown` does not auto-upload those paths.
+- Using `--content` without making the JSON match the effective `--msg-type`.
+- Explicitly setting `--msg-type` to something that conflicts with `--text`, `--markdown`, or media flags.
+- Mixing `--text`, `--markdown`, or `--content` with media flags in one command.
+
 ## Return Value
 
 ```json
@@ -108,16 +198,22 @@ The reply appears in the target message's thread and does not show up in the mai
 
 ## @Mention Format (text / post)
 
-- @specific user: `<at user_id="ou_xxx">name</at>`
+- Recommended format: `<at user_id="ou_xxx">name</at>`
 - @all: `<at user_id="all"></at>`
+- The shortcut normalizes common variants like `<at id=...>` and `<at open_id=...>` into `user_id`, but `user_id` remains the recommended documented form
 
 ## Notes
 
 - `--message-id` must be a valid message ID in `om_xxx` format
-- `--content` must be a valid JSON string
-- `--reply-in-thread` is only meaningful in group chats
-- `--image`/`--file`/`--video`/`--audio`/`--video-cover` support local file paths; use relative paths within the current working directory. The shortcut automatically uploads the file first and then sends the reply
-- If the provided value starts with `img_` or `file_`, it is treated as an existing key and used directly
-- When using `--video`, `--video-cover` is **required** as the video cover. Omitting `--video-cover` with `--video` will produce a validation error. `--video-cover` cannot be used without `--video`
+- `--content` must be valid JSON
+- When using `--content`, you are responsible for making the JSON structure match the effective `msg_type`
+- `--reply-in-thread` adds `reply_in_thread=true` to the API request
+- `--reply-in-thread` is mainly meaningful in chats that support thread replies
+- `--image`/`--file`/`--video`/`--audio`/`--video-cover` support local file paths; the shortcut uploads first and then sends the reply
+- If the provided media value starts with `img_` or `file_`, it is treated as an existing key and used directly
+- `--markdown` always sends `msg_type=post`
+- If you explicitly set `--msg-type` and it conflicts with the chosen content flag, validation fails
+- When using `--video`, `--video-cover` is required as the video cover
+- `--dry-run` uses placeholder image keys for remote Markdown images and placeholder media keys for local uploads
 - Failures return error codes and messages
-- `--as bot` uses a tenant access token (TAT), and requires the `im:message:send_as_bot` scope
+- `--as bot` uses a tenant access token (TAT), and requires the `im:message:send_as_bot` scope