Update the fnr-exploration doc

benchmark/benchmark-suite.coffee

@@ -1,5 +1,7 @@
 require './benchmark-helper'
-{$, _, WorkspaceView} = require 'atom'
+{$, _, WorkspaceView, Range, Point} = require 'atom'
+{Emitter} = require 'emissary'
+TextBuffer = require 'text-buffer'
 TokenizedBuffer = require '../src/tokenized-buffer'
 
 describe "editorView.", ->
@@ -29,6 +31,46 @@ describe "editorView.", ->
     benchmark "keydown-event-with-no-binding", 10, ->
       keymap.handleKeyEvent(event)
 
+  describe "markers.", ->
+    [editor, buffer, lineCount, editorPromise] = []
+    class EmitterTest
+      Emitter.includeInto(EmitterTest)
+      emitSomething: ->
+        @emit('something')
+
+    beforeEach ->
+      editorPromise = atom.project.open('medium.coffee').then (ed) => editor = ed
+      waitsForPromise -> editorPromise
+      runs ->
+        lineCount = editor.buffer.getLineCount()
+        buffer = new TextBuffer(text:editor.getText())
+
+    fpbenchmark "marker-creation-from-editor", ->
+      for row in [0...lineCount]
+        if length = editor.buffer.lineLengthForRow(row)
+          editor.markBufferRange(new Range(new Point(row, 0), new Point(row, 1)))
+      return
+
+    fpbenchmark "marker-creation-from-text-buffer", ->
+      for row in [0...lineCount]
+        if length = editor.buffer.lineLengthForRow(row)
+          buffer.markRange(new Range(new Point(row, 0), new Point(row, 1)))
+      return
+
+    fpbenchmark "object-creation-and-emit", ->
+      for row in [0...lineCount]
+        if length = editor.buffer.lineLengthForRow(row)
+          et = new EmitterTest()
+          et.on 'something', ->
+          et.emitSomething()
+      return
+
+    fpbenchmark "object-creation", ->
+      for row in [0...lineCount]
+        if length = editor.buffer.lineLengthForRow(row)
+          new EmitterTest()
+      return
+
   describe "opening-buffers.", ->
     benchmark "300-line-file.", ->
       buffer = project.bufferForPathSync('medium.coffee')

docs/fnr-exploration.md

@@ -0,0 +1,154 @@
+# Find and replace speedup investigation
+
+Find and replace is slow when there are a lot of matches on the page.
+
+## Test junk
+
+I'm testing with `editor-view.coffee` and searching for a single space. There are 9871 spaces in `editor-view.coffee`. I added a command to find and replace that creates a profile.
+
+```coffee
+@subscriber.subscribeToCommand atom.workspaceView, 'find-and-replace:find-space-profile', =>
+  @findView.findEditor.setText ''
+  @findView.updateModel pattern: ''
+
+  console.profile('find spaces')
+  console.time('find spaces')
+  @findView.findEditor.setText ' '
+  @findView.updateModel pattern: ' '
+  console.timeEnd('find spaces')
+  console.profileEnd('find spaces')
+```
+
+## Baseline
+
+As the code is today it takes ~1500ms to do this. Much of the time spent in `SpanSkipList::totalTo`. But there is a whole lot else costing time.
+
+It didnt fluctuate much, ~1490ms - 1590ms.
+
+![baseline](https://cloud.githubusercontent.com/assets/69169/3914724/4337a990-234c-11e4-870c-3ea06c0342dc.png)
+
+## Experiment with removing totalTo
+
+I wondered, was the skip list helping us?
+
+~1280ms; 1260ms - 1360ms
+
+```coffee
+# dumb algorithm
+positionForCharacterIndex: (offset) ->
+  offset = Math.max(0, offset)
+  offset = Math.min(@getText().length, offset)
+
+  row = 0
+  column = 0
+  for line in @lines
+    if line.length < offset
+      offset -= line.length + @lineEndingForRow(row).length
+      row++
+    else
+      column = offset
+      break
+
+  if row > @getLastRow()
+    @getEndPosition()
+  else
+    new Point(row, column)
+```
+
+![no skip list](https://cloud.githubusercontent.com/assets/69169/3914767/36263b8a-234d-11e4-9837-13f302d8805f.png)
+
+It's consistently a litte faster than `totalTo`, but also reduces the GC pressure.
+
+## Remove the places it emits slow
+
+Emitter was recently updated to optimize the single arg case. And all of the decotation events were using 2 params. Reduced them to 1, and git a little bit.
+
+It is consistently in the mid-low 1200ms range, and `emitSlow` is no longer being called.
+
+![fixing decoration-added events](https://cloud.githubusercontent.com/assets/69169/3914801/21e5a740-234e-11e4-9ee3-9fbb692cd53a.png)
+
+## Attempt to optimize the emitter.
+
+Is the emitter doing too much? Making too many temp objects?
+
+## No decorations
+
+How much faster is it with out decorations? A little bit less GC pressure.
+
+820 - 900
+
+![no decorations](https://cloud.githubusercontent.com/assets/69169/3914920/4f1176c4-2351-11e4-80b0-bcd23821cf7a.png)
+
+## No markers
+
+How much faster is it with out markers and no decorations?
+
+210ms - 240ms
+
+Whoa. Now, it's all about finding the position in char range. The GC isnt even in the picture.
+
+![no markers](https://cloud.githubusercontent.com/assets/69169/3914913/db1a84cc-2350-11e4-986f-f2feab722cde.png)
+
+## Optimizing Marker creation
+
+So our largest chunk of time is spent creating markers. How slow is marker creation compared to regular object creation + emit
+
+![benchmarks](https://cloud.githubusercontent.com/assets/69169/3939214/2dc9d96a-24c4-11e4-8d3f-757a14b6a60f.png)
+
+It's a lot slower.
+
+This The profile for marker-creation:
+
+![marker-creation-from-buffer](https://cloud.githubusercontent.com/assets/69169/3939227/74b32da4-24c4-11e4-8468-c3098b35b0a4.png)
+
+The profile looked like we were spending some time in `Range.toObject`. Is there a difference between `new Range(new Point(row, 0), new Point(row, 1))` and `[[row, 0], [row, 1]]`? Yeah, and it's pretty big
+
+![use point and range](https://cloud.githubusercontent.com/assets/69169/3939920/7daf04f8-24d0-11e4-9a0d-9b8cdd87f43b.png)
+
+Could optimize it to not use `args...`? __YES__
+
+```coffee
+@fromObject: (object, copy) ->
+  if Array.isArray(object)
+    new this(...)
+```
+
+to
+
+```coffee
+@fromObject: (object, copy) ->
+  if Array.isArray(object)
+    [pointA, pointB] = object
+    new this(pointA, pointB)
+```
+
+And now they are equal performance!
+
+![Optimize range](https://cloud.githubusercontent.com/assets/69169/3939966/83850264-24d1-11e4-81f0-43fc2ed39dc0.png)
+
+### Marker creation in Atom vs TextBuffer
+
+Creation is fast on the text-buffer side, and ~3x+ slower on the atom side. What are we doing?
+
+![atom vs text-buffer](https://cloud.githubusercontent.com/assets/69169/3940109/6962e758-24d5-11e4-92d7-3384f4f98116.png)
+
+Looks like we're spending a lot more time (~4x!) in the garbage collector and a whole lot more time emitting events.
+
+![profile-from-editor](https://cloud.githubusercontent.com/assets/69169/3940132/15fdc6c2-24d6-11e4-8b31-6d4780bf3fba.png)
+
+![profile-from-text-buffer](https://cloud.githubusercontent.com/assets/69169/3940131/15fab5a4-24d6-11e4-8d57-258c4c390b96.png)
+
+#### Subscribing is slow
+
+Commenting out the changed and destroyed handlers in `DisplayBufferMarker` cut the time in _half_.
+
+```coffee
+# @subscribe @bufferMarker, 'destroyed', => @destroyed()
+# @subscribe @bufferMarker, 'changed', (event) => @notifyObservers(event)
+```
+
+![less subscribing](https://cloud.githubusercontent.com/assets/69169/3940319/a1c6c3de-24db-11e4-8da8-bd320b1b1b34.png)
+
+Even using `on` rather than subscribe is faster:
+
+![using on](https://cloud.githubusercontent.com/assets/69169/3940321/d4f5bd96-24db-11e4-8dd6-397c724adbd6.png)

src/display-buffer.coffee

@@ -786,13 +786,13 @@ class DisplayBuffer extends Model
       # in the change handler. Bookmarks does this.
       if decorations?
         for decoration in decorations
-          @emit 'decoration-changed', marker, decoration, event
+          @emit 'decoration-changed', decoration
 
     decoration = new Decoration(marker, this, decorationParams)
     @decorationsByMarkerId[marker.id] ?= []
     @decorationsByMarkerId[marker.id].push(decoration)
     @decorationsById[decoration.id] = decoration
-    @emit 'decoration-added', marker, decoration
+    @emit 'decoration-added', decoration
     decoration
 
   removeDecoration: (decoration) ->
@@ -803,13 +803,13 @@ class DisplayBuffer extends Model
     if index > -1
       decorations.splice(index, 1)
       delete @decorationsById[decoration.id]
-      @emit 'decoration-removed', marker, decoration
+      @emit 'decoration-removed', decoration
       @removedAllMarkerDecorations(marker) if decorations.length is 0
 
   removeAllDecorationsForMarker: (marker) ->
     decorations = @decorationsByMarkerId[marker.id].slice()
     for decoration in decorations
-      @emit 'decoration-removed', marker, decoration
+      @emit 'decoration-removed', decoration
     @removedAllMarkerDecorations(marker)
 
   removedAllMarkerDecorations: (marker) ->

src/editor.coffee

@@ -227,10 +227,10 @@ class Editor extends Model
     @subscribe @displayBuffer, 'grammar-changed', => @handleGrammarChange()
     @subscribe @displayBuffer, 'tokenized', => @handleTokenization()
     @subscribe @displayBuffer, 'soft-wrap-changed', (args...) => @emit 'soft-wrap-changed', args...
-    @subscribe @displayBuffer, "decoration-added", (args...) => @emit 'decoration-added', args...
-    @subscribe @displayBuffer, "decoration-removed", (args...) => @emit 'decoration-removed', args...
-    @subscribe @displayBuffer, "decoration-changed", (args...) => @emit 'decoration-changed', args...
-    @subscribe @displayBuffer, "decoration-updated", (args...) => @emit 'decoration-updated', args...
+    @subscribe @displayBuffer, "decoration-added", (decoration) => @emit 'decoration-added', decoration
+    @subscribe @displayBuffer, "decoration-removed", (decoration) => @emit 'decoration-removed', decoration
+    @subscribe @displayBuffer, "decoration-changed", (decoration) => @emit 'decoration-changed', decoration
+    @subscribe @displayBuffer, "decoration-updated", (decoration) => @emit 'decoration-updated', decoration
     @subscribe @displayBuffer, "character-widths-changed", (changeCount) => @emit 'character-widths-changed', changeCount
 
   getViewClass: ->