llm_meta_widget 0.12.0 → 0.12.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b2ba425d2cce6f582baf43df6ca5512bd344ccd78f1e74b56887eb6f3ccdaff6
-  data.tar.gz: a0127db4a9a59f6a88c052e9f6d54d1d0664c168c4de6cd16fe480fa95a73327
+  metadata.gz: 845c38baf1aca6cc874a77c9f324895c6001664dffe4d891d8b31ed2057a3b2b
+  data.tar.gz: 825acda8a84c599e09e8d11cad9fe738092aa807b41911a99889e76dab15c800
 SHA512:
-  metadata.gz: 7b3d3ddd173ef207ccd651ef5a455ffbdc2941fd0d00bc9e1bce687d824cf789b14b8e0d09f898fbfc0db132d4e384cdb7637289467c3259d180f7e71fdd1c16
-  data.tar.gz: eea0ceefa6b8cecc7472d754c16d0de51a04ad92afb69dfe237c31c1448873761d7a1c7fa90c317fc9b29c568c4c2eecf7fe3f7dead0b467e1e9d68b93dafd57
+  metadata.gz: 0244dc664c60f88e1cf29e9cb52ebc9e33bf61e8ab2ab180a8e0873d587bf1cf96d8865eeb9d97b0fe325cec00c43e6e154cef74c511f7f16894a2112d17ad23
+  data.tar.gz: 5584f49e2c7e848ef93fdb5bf6cb1270fa3154f77beee255eb64e221aab20ce8e4828b4b8381c5268e1503b1b0967d4fab81811be718f23f247951ccdf1c142a

data/CHANGELOG.md

@@ -3,6 +3,35 @@
 This file starts at 0.8.0, the first release with a breaking change worth
 calling out. Earlier releases are in the git history (`git log --oneline v0.7.6`).
 
+## 0.12.1
+
+### Fixed: a tool result is counted as JSON, not as prose
+
+The turn was sized at roughly four characters per token, which is right for
+prose and badly wrong for a tool result. A request of 289,101 characters that
+the provider counted as 135,291 tokens was estimated at 72,275 — 53% of its
+real size — so nothing was given up and the provider refused the turn with its
+own error text. That request was 33% letters, 16% digits and 45% punctuation:
+JSON, which tokenizes about twice as densely as prose.
+
+Characters are now counted in three classes — letters and spaces, CJK, and
+everything else — which puts that request at 106% of its real size. Erring
+high is deliberate here: underestimating ends the turn, while overestimating
+gives up an old tool result slightly early. Tool schemas are counted the same
+way, since a schema is JSON too.
+
+### Fixed: following the output no longer fights a reader who scrolls
+
+0.12.0 decided whether to follow the output from the `scroll` event. A browser
+dispatches that event a frame after the position changes, while deltas arrive
+every few milliseconds, so a delta could pin the view before the listener
+heard that the reader had moved — and the pin put them back at the end, so the
+decision never flipped. The question is now asked synchronously, immediately
+before anything is added, which has no such window.
+
+A separate report of the panel jumping to the TOP after a turn ends is not
+explained by this and remains open.
+
 ## 0.12.0
 
 Nothing a page must provide changes in this release: no new attribute, no

data/README.md

@@ -61,7 +61,7 @@ Option 3 if it does not.
 
 ```ruby
 # Gemfile
-gem "llm_meta_widget", "~> 0.12.0"
+gem "llm_meta_widget", "~> 0.12.1"
 ```
 
 ```bash

data/app/assets/javascripts/llm_meta_widget/element.js

@@ -168,15 +168,11 @@ function boot(cfg) {
   	// The reader's own scrolling says which they are doing: leave the bottom
   	// and we stop following; come back to it and we resume.
   	var SCROLL_PIN_SLACK = 40;   // px from the end still counts as "at the end"
-  	var followingOutput  = true;
 
   	function atBottom() {
   		return historyEl.scrollHeight - historyEl.scrollTop - historyEl.clientHeight <= SCROLL_PIN_SLACK;
   	}
   	function pinToBottom() { historyEl.scrollTop = historyEl.scrollHeight; }
-  	// Called wherever content is added mid-turn.
-  	function pinIfFollowing() { if (followingOutput) pinToBottom(); }
-  	historyEl.addEventListener("scroll", function() { followingOutput = atBottom(); });
   	var formEl       = root.querySelector(".lmw-form");
   	var inputEl      = root.querySelector(".lmw-input");
   	var clearBtn     = root.querySelector(".lmw-clear");
@@ -935,6 +931,8 @@ function boot(cfg) {
   	var pendingTurnLabel = null;
 
   	function appendTurn(role, text, turnLabel) {
+  		// Captured BEFORE anything is added: see atBottom.
+  		var following = atBottom();   // a question always wins
   		// Class names mirror llm_meta_chat's chats/_message.html.erb —
   		// `.message.<role>`, `.message-role`, `.message-content` — so the
   		// shared conversation.css styles apply directly (see the <link>
@@ -973,7 +971,9 @@ function boot(cfg) {
   		div.appendChild(document.createTextNode(" "));
   		div.appendChild(body);
   		historyEl.appendChild(div);
-  		pinIfFollowing();
+  		// A question the visitor just asked always goes to the end; anything
+  		// else respects where they were reading.
+  		if (role === "user" || following) pinToBottom();
   		body.roleLabel = label;   // so a turn in flight can show it is working
   		return body;
   	}
@@ -1061,6 +1061,8 @@ function boot(cfg) {
   	}
 
   	function announceToolCall(toolCall) {
+  		// Captured BEFORE anything is added: see atBottom.
+  		var following = atBottom();
   		var row = chipRow();
   		if (!row) return;
   		var chip = document.createElement("span");
@@ -1069,7 +1071,7 @@ function boot(cfg) {
   		chip.dataset.tool = toolCall.name;
   		row.appendChild(chip);
   		pendingChips.push(chip);
-  		pinIfFollowing();
+  		if (following) pinToBottom();
   	}
 
   	function resolveToolCall(outcome) {
@@ -1091,6 +1093,8 @@ function boot(cfg) {
   	var activeAssistantBody  = null;
 
   	function ensureThinkingBlock() {
+  		// Captured BEFORE anything is added: see atBottom.
+  		var following = atBottom();
   		if (currentThinkingBody) return currentThinkingBody;
   		// Class names match llm_meta_chat's .message-thinking pattern so
   		// the shared conversation.css applies directly.
@@ -1117,7 +1121,7 @@ function boot(cfg) {
   		} else {
   			historyEl.appendChild(details);
   		}
-  		pinIfFollowing();
+  		if (following) pinToBottom();
   		currentThinkingBlock = details;
   		currentThinkingBody  = body;
   		return body;
@@ -1266,9 +1270,6 @@ function boot(cfg) {
   		var userText = inputEl.value.trim();
   		if (!userText) return;
   		inputEl.value = "";
-  		// Asking something is a request to see the answer: whatever the reader was
-  		// looking at before, they are following the output again from here.
-  		followingOutput = true;
   		appendTurn("user", userText, pendingTurnLabel);
   		pendingTurnLabel = null;
 
@@ -1356,19 +1357,23 @@ function boot(cfg) {
   					collapseThinkingBlock();
   				},
   				onThinkingDelta: function(delta) {
+  					// Captured BEFORE anything is added: see atBottom.
+  					var following = atBottom();
   					var body = ensureThinkingBlock();
   					body.appendChild(document.createTextNode(delta));
   					body.scrollTop = body.scrollHeight;
-  					pinIfFollowing();
+  					if (following) pinToBottom();
   				},
   				onTextDelta: function(delta) {
+  					// Captured BEFORE anything is added: see atBottom.
+  					var following = atBottom();
   					markDone(assistantBody.roleLabel);
   					collapseThinkingBlock();
   					assistantMarkdown += delta;
   					renderMarkdownInto(assistantBody, assistantMarkdown);
   					liveTurn.content = assistantMarkdown;
   					persistSoon();
-  					pinIfFollowing();
+  					if (following) pinToBottom();
   				}
   			});
   			// The turn was recorded here until 0.11.1, after the loop returned — which

data/app/assets/javascripts/llm_meta_widget/llm-meta-widget.js

@@ -589,17 +589,29 @@ function parseSseFrame(raw) {
 var STATIC_PRIMITIVES_META = "io.modelcontextprotocol/static-primitives";
 var PAGE_CARD_MEDIA_TYPE = "application/mcp-static-primitives+json";
 var RESOURCE_SHARE_OF_WINDOW = 0.0625;
+var RESOURCE_BYTES_PER_TOKEN = 4;
 function resourceBudgetBytes(window2, { floor = 8e3, cap = 64e3 } = {}) {
   if (!window2 || !Number.isFinite(window2)) return floor;
-  const share = Math.round(window2 * RESOURCE_SHARE_OF_WINDOW * CHARS_PER_TOKEN);
+  const share = Math.round(window2 * RESOURCE_SHARE_OF_WINDOW * RESOURCE_BYTES_PER_TOKEN);
   return Math.min(cap, Math.max(floor, share));
 }
-var CHARS_PER_TOKEN = 4;
+var PROSE_CHARS_PER_TOKEN = 3.5;
+var DENSE_CHARS_PER_TOKEN = 1.6;
+var CJK_CHARS_PER_TOKEN = 1;
 var MESSAGE_OVERHEAD_TOKENS = 4;
+var CJK_PATTERN = /[\u3000-\u30ff\u3400-\u4dbf\u4e00-\u9fff\uac00-\ud7af\uf900-\ufaff\uff00-\uffef]/g;
+var PROSE_PATTERN = /[A-Za-z \n\t]/g;
+function estimateTextTokens(text) {
+  const str = typeof text === "string" ? text : String(text ?? "");
+  const cjk = (str.match(CJK_PATTERN) || []).length;
+  const prose = (str.match(PROSE_PATTERN) || []).length;
+  const dense = str.length - cjk - prose;
+  return Math.ceil(cjk / CJK_CHARS_PER_TOKEN + prose / PROSE_CHARS_PER_TOKEN + dense / DENSE_CHARS_PER_TOKEN);
+}
 function estimateTokens(messages) {
   return (messages || []).reduce((sum, m) => {
     const content = typeof m?.content === "string" ? m.content : JSON.stringify(m?.content ?? "");
-    return sum + Math.ceil(content.length / CHARS_PER_TOKEN) + MESSAGE_OVERHEAD_TOKENS;
+    return sum + estimateTextTokens(content) + MESSAGE_OVERHEAD_TOKENS;
   }, 0);
 }
 function estimateToolsTokens(tools) {
@@ -608,7 +620,7 @@ function estimateToolsTokens(tools) {
     description: t?.description,
     schema: t?.input_schema || t?.inputSchema
   })).join("");
-  return Math.ceil(text.length / CHARS_PER_TOKEN);
+  return estimateTextTokens(text);
 }
 function evictOldToolResults({ messages, window: window2, reserve = 0, dropped = [] }) {
   if (!window2 || !Number.isFinite(window2)) return { messages, dropped };
@@ -3940,19 +3952,12 @@ function boot(cfg) {
   var toggleBtn = document.getElementById("llm-meta-widget-toggle");
   var historyEl = root.querySelector(".lmw-messages");
   var SCROLL_PIN_SLACK = 40;
-  var followingOutput = true;
   function atBottom() {
     return historyEl.scrollHeight - historyEl.scrollTop - historyEl.clientHeight <= SCROLL_PIN_SLACK;
   }
   function pinToBottom() {
     historyEl.scrollTop = historyEl.scrollHeight;
   }
-  function pinIfFollowing() {
-    if (followingOutput) pinToBottom();
-  }
-  historyEl.addEventListener("scroll", function() {
-    followingOutput = atBottom();
-  });
   var formEl = root.querySelector(".lmw-form");
   var inputEl = root.querySelector(".lmw-input");
   var clearBtn = root.querySelector(".lmw-clear");
@@ -4505,6 +4510,7 @@ function boot(cfg) {
   }
   var pendingTurnLabel = null;
   function appendTurn(role, text, turnLabel) {
+    var following = atBottom();
     var welcome = historyEl.querySelector(".lmw-welcome");
     if (welcome) welcome.remove();
     var div = document.createElement("div");
@@ -4531,7 +4537,7 @@ function boot(cfg) {
     div.appendChild(document.createTextNode(" "));
     div.appendChild(body);
     historyEl.appendChild(div);
-    pinIfFollowing();
+    if (role === "user" || following) pinToBottom();
     body.roleLabel = label;
     return body;
   }
@@ -4581,6 +4587,7 @@ function boot(cfg) {
     return liveChips;
   }
   function announceToolCall(toolCall) {
+    var following = atBottom();
     var row = chipRow();
     if (!row) return;
     var chip = document.createElement("span");
@@ -4589,7 +4596,7 @@ function boot(cfg) {
     chip.dataset.tool = toolCall.name;
     row.appendChild(chip);
     pendingChips.push(chip);
-    pinIfFollowing();
+    if (following) pinToBottom();
   }
   function resolveToolCall(outcome) {
     var idx = pendingChips.findIndex(function(c) {
@@ -4605,6 +4612,7 @@ function boot(cfg) {
   var currentThinkingBody = null;
   var activeAssistantBody = null;
   function ensureThinkingBlock() {
+    var following = atBottom();
     if (currentThinkingBody) return currentThinkingBody;
     var details = document.createElement("details");
     details.className = "message-thinking";
@@ -4629,7 +4637,7 @@ function boot(cfg) {
     } else {
       historyEl.appendChild(details);
     }
-    pinIfFollowing();
+    if (following) pinToBottom();
     currentThinkingBlock = details;
     currentThinkingBody = body;
     return body;
@@ -4746,7 +4754,6 @@ function boot(cfg) {
     var userText = inputEl.value.trim();
     if (!userText) return;
     inputEl.value = "";
-    followingOutput = true;
     appendTurn("user", userText, pendingTurnLabel);
     pendingTurnLabel = null;
     if (currentAbort) {
@@ -4809,19 +4816,21 @@ function boot(cfg) {
           collapseThinkingBlock();
         },
         onThinkingDelta: function(delta) {
+          var following = atBottom();
           var body = ensureThinkingBlock();
           body.appendChild(document.createTextNode(delta));
           body.scrollTop = body.scrollHeight;
-          pinIfFollowing();
+          if (following) pinToBottom();
         },
         onTextDelta: function(delta) {
+          var following = atBottom();
           markDone(assistantBody.roleLabel);
           collapseThinkingBlock();
           assistantMarkdown += delta;
           renderMarkdownInto(assistantBody, assistantMarkdown);
           liveTurn.content = assistantMarkdown;
           persistSoon();
-          pinIfFollowing();
+          if (following) pinToBottom();
         }
       });
       liveTurn.content = result.content;

data/app/assets/javascripts/llm_meta_widget/orchestrator.js

@@ -937,23 +937,53 @@ export const PAGE_CARD_MEDIA_TYPE = "application/mcp-static-primitives+json"
 // is already generous and a 1M-token model does not need 250 KB of it.
 export const RESOURCE_SHARE_OF_WINDOW = 0.0625
 
+// Bytes per token for SIZING that budget — not the estimator above, which
+// measures real text. Kept at four so the budget stays exactly what it was
+// (8 KB on a 32k model): how many tokens those bytes become depends on the
+// resource, and re-sizing budgets is not this constant's job.
+export const RESOURCE_BYTES_PER_TOKEN = 4
+
 export function resourceBudgetBytes(window, { floor = 8000, cap = 64_000 } = {}) {
   if (!window || !Number.isFinite(window)) return floor
-  const share = Math.round(window * RESOURCE_SHARE_OF_WINDOW * CHARS_PER_TOKEN)
+  const share = Math.round(window * RESOURCE_SHARE_OF_WINDOW * RESOURCE_BYTES_PER_TOKEN)
   return Math.min(cap, Math.max(floor, share))
 }
 
-// A rough token count, the same shape the hub's own estimator uses: characters
-// over four, plus a fixed cost per message. Deliberately approximate — it is
-// used to decide "drop the oldest tool result", never to fill the window
-// exactly, and it errs high so the decision is made early rather than late.
-export const CHARS_PER_TOKEN = 4
+// A rough token count, the same three classes the hub's estimator uses.
+//
+// One ratio cannot describe both prose and a tool result. Letters and spaces
+// tokenize at roughly 3.5 characters each; digits, punctuation and symbols go
+// much denser, and a JSON tool result is mostly those.
+//
+// Measured against a real refusal (2026-10-17): a 289,101-character request
+// the provider counted as 135,291 tokens was 33% letters, 16% digits and 45%
+// punctuation. At a flat four characters per token that is 72,275 — 53% of the
+// truth — so nothing was evicted and the provider refused the turn instead.
+// These classes put it at 106%: high, which is the safe direction here, since
+// underestimating means the turn dies while overestimating only gives up an
+// old tool result slightly early.
+export const PROSE_CHARS_PER_TOKEN = 3.5
+export const DENSE_CHARS_PER_TOKEN = 1.6
+export const CJK_CHARS_PER_TOKEN = 1
 export const MESSAGE_OVERHEAD_TOKENS = 4
 
+const CJK_PATTERN = /[\u3000-\u30ff\u3400-\u4dbf\u4e00-\u9fff\uac00-\ud7af\uf900-\ufaff\uff00-\uffef]/g
+const PROSE_PATTERN = /[A-Za-z \n\t]/g
+
+export function estimateTextTokens(text) {
+  const str = typeof text === "string" ? text : String(text ?? "")
+  const cjk = (str.match(CJK_PATTERN) || []).length
+  const prose = (str.match(PROSE_PATTERN) || []).length
+  const dense = str.length - cjk - prose
+  return Math.ceil(cjk / CJK_CHARS_PER_TOKEN +
+                   prose / PROSE_CHARS_PER_TOKEN +
+                   dense / DENSE_CHARS_PER_TOKEN)
+}
+
 export function estimateTokens(messages) {
   return (messages || []).reduce((sum, m) => {
     const content = typeof m?.content === "string" ? m.content : JSON.stringify(m?.content ?? "")
-    return sum + Math.ceil(content.length / CHARS_PER_TOKEN) + MESSAGE_OVERHEAD_TOKENS
+    return sum + estimateTextTokens(content) + MESSAGE_OVERHEAD_TOKENS
   }, 0)
 }
 
@@ -969,7 +999,8 @@ export function estimateToolsTokens(tools) {
   const text = (tools || []).map((t) => JSON.stringify({
     name: t?.name, description: t?.description, schema: t?.input_schema || t?.inputSchema
   })).join("")
-  return Math.ceil(text.length / CHARS_PER_TOKEN)
+  // Counted with the same classes: a schema is JSON, which is the dense case.
+  return estimateTextTokens(text)
 }
 
 // Makes room for the next round by dropping the OLDEST tool results, which is

data/lib/llm_meta_widget/version.rb

@@ -1,3 +1,3 @@
 module LlmMetaWidget
-  VERSION = "0.12.0"
+  VERSION = "0.12.1"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: llm_meta_widget
 version: !ruby/object:Gem::Version
-  version: 0.12.0
+  version: 0.12.1
 platform: ruby
 authors:
 - jdkim