Merge pull request #5 from alexaubry/develop

Release 2.1.1

Author: alexaubry

.gitignore

@@ -1,4 +1,4 @@
-.DS_Store
+**/.DS_Store
 build/
 DerivedData/
 xcuserdata/

.jazzy.yaml

@@ -3,13 +3,16 @@ author: "Alexis Aubry Radanovic"
 author_url: https://github.com/alexaubry
 github_url: https://github.com/alexaubry/HTMLString
 github_file_prefix: https://github.com/alexaubry/HTMLString/tree/2.1.0
-module_version: 2.1.0
-xcodebuild_arguments: -scheme,HTMLString
+module_version: 2.1.1
+xcodebuild_arguments: 
+- -scheme
+- HTMLString
+- -destination
+- 'platform=OS X,arch=x86_64'
 module: HTMLString
 root_url: https://alexaubry.github.io/HTMLString
 output: docs/
 swift_version: 3.0.2
 copyright: "Copyright © 2016-2017 Alexis Aubry Radanovic. Licensed under the [MIT License](https://github.com/alexaubry/HTMLString/blob/master/LICENSE)"
-podspec: ./HTMLString.podspec
-skip-undocumented: true
+skip_undocumented: true
 theme: apple

Benchmark.md

@@ -4,47 +4,47 @@ A set of measures tracking the performance of the library.
 
 ## ⏱ Compile times
 
-| Configuration | v2.0.1 | v2.1.0 |
-|:---|---|---|
-| Debug | 7.15s | 6.40s |
-| Release | 32.83s | 27.21ss |
+| Configuration | v2.0.1 | v2.1.0 | v2.1.1 |
+|:---|---|---|---|
+| Debug | 7.15s | 6.40s | 5,64s |
+| Release | 32.83s | 27.21s | 23,07s |
 
 ## 🤖 Common Tasks
 
 ### Debug
 
-| Task | v2.0.1 | v2.1.0 |
-|:---|---|---|
-| Unicode-escaping 2 emojis | 0.000083s | 0.000040s |
-| ASCII-escaping 2 emojis | 0.000104s | 0.000035s |
-| Unescaping 2 emojis | 0.000046s | 0.000049s |
-| Unescaping a tweet | 0.000085s | 0.000033s |
-| Unicode-escaping a tweet | 0.000844s | 0.000131s |
-| ASCII-escaping a tweet | 0.001023s | 0.000158s |
-| Unicode-escaping 23,145 characters | 0.159909s | 0.024684s |
-| ASCII-escaping 23,145 characters | 0.198981s | 0.030419s |
-| Unescaping 3,026 words with 366 escapes | 0.165962s | 0.001686s |
+| Task | v2.0.1 | v2.1.0 | v2.1.1 |
+|:---|---|---|---|
+| Unicode-escaping 2 emojis | 0.000083s | 0.000040s | 0.000006s |
+| ASCII-escaping 2 emojis | 0.000104s | 0.000035s | 0.000008s |
+| Unescaping 2 emojis | 0.000046s | 0.000049s | 0.000021s |
+| Unescaping a tweet | 0.000085s | 0.000033s | 0.000029s |
+| Unicode-escaping a tweet | 0.000844s | 0.000131s | 0.000056s |
+| ASCII-escaping a tweet | 0.001023s | 0.000158s | 0.000058s |
+| Unicode-escaping 23,145 characters | 0.159909s | 0.024684s | 0.009895s |
+| ASCII-escaping 23,145 characters | 0.198981s | 0.030419s | 0.010272s |
+| Unescaping 3,026 words with 366 escapes | 0.165962s | 0.001686s | 0.001706s |
 
 ### Release
 
-| Task | v2.0.1 | v2.1.0 |
-|:---|---|---|
-| Unicode-escaping 2 emojis | 0.000022s | 0.000010s |
-| ASCII-escaping 2 emojis | 0.000033s | 0.000015s |
-| Unescaping 2 emojis | 0.000031s | 0.000016s |
-| Unescaping a tweet | 0.000052s | 0.000017s |
-| Unicode-escaping a tweet | 0.000183s | 0.000064s |
-| ASCII-escaping a tweet | 0.000276s | 0.000077s |
-| Unicode-escaping 23,145 characters | 0.034701s | 0.011941s |
-| ASCII-escaping 23,145 characters | 0.052937s | 0.014950s |
-| Unescaping 3,026 words with 366 escapes | 0.014192s | 0.001162s |
-
-## ❓ Complexity
-
-TODO
-
-| Algorithm | v2.0.1 | v2.1.0 |
-|:---|---|---|
-| ASCII escaping | N/A | N/A |
-| Unicode escaping | N/A | N/A |
-| Unescaping | N/A | N/A |
+| Task | v2.0.1 | v2.1.0 | v2.1.1 |
+|:---|---|---|---|
+| Unicode-escaping 2 emojis | 0.000022s | 0.000010s | 0.000004s |
+| ASCII-escaping 2 emojis | 0.000033s | 0.000015s | 0.000008s |
+| Unescaping 2 emojis | 0.000031s | 0.000016s | 0.000012s |
+| Unescaping a tweet | 0.000052s | 0.000017s | 0.000023s |
+| Unicode-escaping a tweet | 0.000183s | 0.000064s | 0.000047s |
+| ASCII-escaping a tweet | 0.000276s | 0.000077s | 0.000047s |
+| Unicode-escaping 23,145 characters | 0.034701s | 0.011941s | 0.006806s |
+| ASCII-escaping 23,145 characters | 0.052937s | 0.014950s | 0.007000s |
+| Unescaping 3,026 words with 366 escapes | 0.014192s | 0.001162s | 0.001204s |
+
+## ❓ Estimated Asymptotic Complexity
+
+Where `N` is the number of characters in the string.
+
+| Algorithm | v2.0.1 | v2.1.0 | v2.1.1 |
+|:---|---|---|---|
+| ASCII escaping | N/A | N/A | O(N) |
+| Unicode escaping | N/A | N/A | O(N) |
+| Unescaping | N/A | N/A | O(N) |

CHANGELOG.md

@@ -1,8 +1,25 @@
 # HTMLString Change Log
 
+## 2017-01-05 — Version 2.1.1
+
+### API
+
+- You can now perform custom escaping by escaping Unicode scalars individually. 
+
+### Improvements
+
+- Further improved the escaping algorithm, which is now up to 6 times faster.
+- Changed the escaping strategy: special characters are now escaped with decimal sequences. This allows for better compatibility with browsers (HTML 4.0 compatible) and better speed
+- Changed the Unicode escaping strategy: only escape characters that could cause an XSS injection
+- Added an asymptotic complexity approximation calculator (every algorithm is now O(N))
+
+### Fixed
+
+- Removed .DS_Store
+
 ## 2017-01-04 — Version 2.1.0
 
-- Change the escaping algorithm (`reduce` instead of `map`)
+- Changed the escaping algorithm (`reduce` instead of `map`)
 - Reduced the size of the escaping mappings
 - Performance improvements: escaping is up to **6.5 times** faster and unescaping is up to **98 times** faster
 - Improved documentation

HTMLString.podspec

@@ -1,7 +1,7 @@
 Pod::Spec.new do |s|
 
   s.name         = "HTMLString"
-  s.version      = "2.1.0"
+  s.version      = "2.1.1"
   s.summary      = "Escape and unescape HTML Strings in Swift"
 
   s.description  = <<-DESC

HTMLString.xcodeproj/xcshareddata/xcschemes/HTMLString.xcscheme

@@ -17,7 +17,7 @@
     selectedDebuggerIdentifier = "Xcode.DebuggerFoundation.Debugger.LLDB"
     selectedLauncherIdentifier = "Xcode.DebuggerFoundation.Launcher.LLDB"
     shouldUseLaunchSchemeArgsEnv = "YES"
-    codeCoverageEnabled = "NO">
+    codeCoverageEnabled = "YES">
     <Testables>
     <TestableReference
       skipped = "NO">

Package.swift

@@ -39,7 +39,7 @@ let targets = [
     // Target(name: "Performance", dependencies: [.Target(name: "HTMLString")])
 ]
 
-let excludes = [
+let excludes: [String] = [
     "Sources/Performance"
 ]
 

Sources/HTMLString/HTMLString.swift

@@ -47,14 +47,16 @@ public extension String {
     ///
     /// | String | Result  | Format                                                  |
     /// |--------|---------|---------------------------------------------------------|
-    /// | `&`    | `&` | Keyword escape (part of the Unicode special characters) |
+    /// | `&`    | `&` | Decimal escape (part of the Unicode special characters) |
     /// | `Σ`    | `Σ`     | Not escaped (Unicode compliant)                         |
     /// | `🇺🇸`   | `🇺🇸`     | Not escaped (Unicode compliant)                         |
     /// | `a`    | `a`     | Not escaped (alphanumerical)                            |
     ///
+    /// **Complexity**: `O(N)` where `N` is the number of characters in the string.
+    ///
 
     public var escapingForUnicodeHTML: String {
-        return escapeHTML(isEncodingUnicode: true)
+        return unicodeScalars.reduce(String()) { $0 + $1.escapingIfNeeded }
     }
 
     ///
@@ -65,7 +67,7 @@ public extension String {
     ///
     /// | String | Result               | Format                                               |
     /// |--------|----------------------|------------------------------------------------------|
-    /// | `&`    | `&`              | Keyword escape                                       |
+    /// | `&`    | `&`              | Keyword escape                                       |
     /// | `Σ`    | `Σ`             | Decimal escape                                       |
     /// | `🇺🇸`   | `🇺🇸` | Combined decimal escapes (extented grapheme cluster) |
     /// | `a`    | `a`                  | Not escaped (alphanumerical)                         |
@@ -76,41 +78,11 @@ public extension String {
     /// as it is faster, and produces less bloated and more readable HTML (as long as you are using 
     /// a unicode compliant HTML reader).
     ///
+    /// **Complexity**: `O(N)` where `N` is the number of characters in the string.
+    ///
 
     public var escapingForASCIIHTML: String {
-        return escapeHTML(isEncodingUnicode: false)
-    }
-
-    private func escapeHTML(isEncodingUnicode: Bool) -> String {
-
-        return self.characters.reduce(String()) {
-
-            let character = String($1)
-
-            // Ignore alphanumerical characters to avoid unnecessary lookups
-            guard character < "\u{30}" || character > "\u{7a}" else {
-                return $0 + character
-            }
-
-            let escaped = isEncodingUnicode ? character.performUnicodeEscaping() : character.performASCIIEscaping()
-            return $0 + escaped
-
-        }
-
-    }
-
-    private func performASCIIEscaping() -> String {
-
-        guard let escapeSequence = HTMLTables.escapingTable[self] else {
-            return unicodeScalars.reduce(String()) { $0 + $1.escapingForASCII }
-        }
-
-        return escapeSequence
-
-    }
-
-    private func performUnicodeEscaping() -> String {
-        return unicodeScalars.reduce(String()) { $0 + $1.escapingIfNeeded }
+        return unicodeScalars.reduce(String()) { $0 + $1.escapingForASCII }
     }
 
 }
@@ -134,6 +106,8 @@ extension String {
     /// | `a`                  | `a`  | Not an escape                                        |
     /// | `&`                  | `&`  | Not an escape                                        |
     ///
+    /// **Complexity**: `O(N)` where `N` is the number of characters in the string.
+    ///
 
     public var unescapingFromHTML: String {
 
@@ -220,19 +194,37 @@ extension String {
 
 extension UnicodeScalar {
 
+    ///
+    /// Returns the decimal HTML escape of the Unicode scalar.
+    ///
+    /// This allows you to perform custom escaping.
+    ///
+
+    public var htmlEscaped: String {
+        return "&#" + String(value) + ";"
+    }
+
+    ///
+    /// The scalar escaped for ASCII encoding.
+    ///
+
     fileprivate var escapingForASCII: String {
-        return isASCII ? escapingIfNeeded : ("&#" + String(value) + ";")
+        return isASCII ? escapingIfNeeded : htmlEscaped
     }
 
+    ///
+    /// Escapes the scalar only if it needs to be escaped for Unicode pages.
+    ///
+    /// [Reference](http://wonko.com/post/html-escaping)
+    /// 
+
     fileprivate var escapingIfNeeded: String {
 
-        // Avoid unnecessary lookups
-        guard value > 0x22 && value < 0x20ac else {
-            return String(self)
+        switch value {
+        case 33, 34, 36, 37, 38, 39, 43, 44, 60, 61, 62, 64, 91, 93, 96, 123, 125: return htmlEscaped
+        default: return String(self)
         }
 
-        return HTMLTables.requiredEscapingsTable[value] ?? String(self)
-
     }
 
 }

Sources/HTMLString/Mappings.swift

@@ -40,16 +40,6 @@ import Foundation
 
 internal struct HTMLTables {
 
-    ///
-    /// The table to use for escaping.
-    ///
-
-    internal static let escapingTable: [String:String] = {
-        return [
-            "\u{22}":""","\u{26}":"&","\u{27}":"'","\u{3c}":"<","\u{3e}":">","\u{a0}":" ","\u{a1}":"¡","\u{a2}":"¢","\u{a3}":"£","\u{a4}":"¤","\u{a5}":"¥","\u{a6}":"¦","\u{a7}":"§","\u{a8}":"¨","\u{a9}":"©","\u{aa}":"ª","\u{ab}":"«","\u{ac}":"¬","\u{ad}":"­","\u{ae}":"®","\u{af}":"¯","\u{b0}":"°","\u{b1}":"±","\u{b2}":"²","\u{b3}":"³","\u{b4}":"´","\u{b5}":"µ","\u{b6}":"¶","\u{b7}":"·","\u{b8}":"¸","\u{b9}":"¹","\u{ba}":"º","\u{bb}":"»","\u{bc}":"¼","\u{bd}":"½","\u{be}":"¾","\u{bf}":"¿","\u{c0}":"À","\u{c1}":"Á","\u{c2}":"Â","\u{c3}":"Ã","\u{c4}":"Ä","\u{c5}":"Å","\u{c6}":"Æ","\u{c7}":"Ç","\u{c8}":"È","\u{c9}":"É","\u{ca}":"Ê","\u{cb}":"Ë","\u{cc}":"Ì","\u{cd}":"Í","\u{ce}":"Î","\u{cf}":"Ï","\u{d0}":"Ð","\u{d1}":"Ñ","\u{d2}":"Ò","\u{d3}":"Ó","\u{d4}":"Ô","\u{d5}":"Õ","\u{d6}":"Ö","\u{d7}":"×","\u{d8}":"Ø","\u{d9}":"Ù","\u{da}":"Ú","\u{db}":"Û","\u{dc}":"Ü","\u{dd}":"Ý","\u{de}":"Þ","\u{df}":"ß","\u{e0}":"à","\u{e1}":"á","\u{e2}":"â","\u{e3}":"ã","\u{e4}":"ä","\u{e5}":"å","\u{e6}":"æ","\u{e7}":"ç","\u{e8}":"è","\u{e9}":"é","\u{ea}":"ê","\u{eb}":"ë","\u{ec}":"ì","\u{ed}":"í","\u{ee}":"î","\u{ef}":"ï","\u{f0}":"ð","\u{f1}":"ñ","\u{f2}":"ò","\u{f3}":"ó","\u{f4}":"ô","\u{f5}":"õ","\u{f6}":"ö","\u{f7}":"÷","\u{f8}":"ø","\u{f9}":"ù","\u{fa}":"ú","\u{fb}":"û","\u{fc}":"ü","\u{fd}":"ý","\u{fe}":"þ","\u{ff}":"ÿ","\u{152}":"Œ","\u{153}":"œ","\u{160}":"Š","\u{161}":"š","\u{178}":"Ÿ","\u{192}":"ƒ","\u{2c6}":"ˆ","\u{2dc}":"˜","\u{391}":"Α","\u{392}":"Β","\u{393}":"Γ","\u{394}":"Δ","\u{395}":"Ε","\u{396}":"Ζ","\u{397}":"Η","\u{398}":"Θ","\u{399}":"Ι","\u{39a}":"Κ","\u{39b}":"Λ","\u{39c}":"Μ","\u{39d}":"Ν","\u{39e}":"Ξ","\u{39f}":"Ο","\u{3a0}":"Π","\u{3a1}":"Ρ","\u{3a3}":"Σ","\u{3a4}":"Τ","\u{3a5}":"Υ","\u{3a6}":"Φ","\u{3a7}":"Χ","\u{3a8}":"Ψ","\u{3a9}":"Ω","\u{3b1}":"α","\u{3b2}":"β","\u{3b3}":"γ","\u{3b4}":"δ","\u{3b5}":"ε","\u{3b6}":"ζ","\u{3b7}":"η","\u{3b8}":"θ","\u{3b9}":"ι","\u{3ba}":"κ","\u{3bb}":"λ","\u{3bc}":"μ","\u{3bd}":"ν","\u{3be}":"ξ","\u{3bf}":"ο","\u{3c0}":"π","\u{3c1}":"ρ","\u{3c2}":"ς","\u{3c3}":"σ","\u{3c4}":"τ","\u{3c5}":"υ","\u{3c6}":"φ","\u{3c7}":"χ","\u{3c8}":"ψ","\u{3c9}":"ω","\u{3d1}":"ϑ","\u{3d2}":"ϒ","\u{3d6}":"ϖ","\u{2002}":" ","\u{2013}":"–","\u{2014}":"—","\u{2018}":"‘","\u{2019}":"’","\u{201a}":"‚","\u{201c}":"“","\u{201d}":"”","\u{201e}":"„","\u{2020}":"†","\u{2021}":"‡","\u{2022}":"•","\u{2026}":"…","\u{2030}":"‰","\u{2032}":"′","\u{2033}":"″","\u{2039}":"‹","\u{203a}":"›","\u{203e}":"‾","\u{2044}":"⁄","\u{20ac}":"€","\u{2111}":"ℑ","\u{2118}":"℘","\u{211c}":"ℜ","\u{2122}":"™","\u{2135}":"ℵ","\u{2190}":"←","\u{2191}":"↑","\u{2192}":"→","\u{2193}":"↓","\u{2194}":"↔","\u{21b5}":"↵","\u{21d0}":"⇐","\u{21d1}":"⇑","\u{21d2}":"⇒","\u{21d3}":"⇓","\u{21d4}":"⇔","\u{2200}":"∀","\u{2202}":"∂","\u{2203}":"∃","\u{2205}":"∅","\u{2207}":"∇","\u{2208}":"∈","\u{2209}":"∉","\u{220b}":"∋","\u{220f}":"∏","\u{2211}":"∑","\u{2212}":"−","\u{2217}":"∗","\u{221a}":"√","\u{221d}":"∝","\u{221e}":"∞","\u{2220}":"∠","\u{2227}":"∧","\u{2228}":"∨","\u{2229}":"∩","\u{222a}":"∪","\u{222b}":"∫","\u{2234}":"∴","\u{223c}":"∼","\u{2245}":"≅","\u{2248}":"≈","\u{2260}":"≠","\u{2261}":"≡","\u{2264}":"≤","\u{2265}":"≥","\u{2282}":"⊂","\u{2283}":"⊃","\u{2284}":"⊄","\u{2286}":"⊆","\u{2287}":"⊇","\u{2295}":"⊕","\u{2297}":"⊗","\u{22a5}":"⊥","\u{22c5}":"⋅","\u{2308}":"⌈","\u{2309}":"⌉","\u{230a}":"⌊","\u{230b}":"⌋","\u{2329}":"⟨","\u{232a}":"⟩","\u{25ca}":"◊","\u{2660}":"♠","\u{2663}":"♣","\u{2665}":"♥","\u{2666}":"♦"
-        ]
-    }()
-
     ///
     /// The table to use for unescaping.
     ///
@@ -60,16 +50,5 @@ internal struct HTMLTables {
         ]
     }()
 
-    ///
-    /// The table for code points of characters that need to be escaped regardless of the encoding.
-    ///
-    /// [Reference](http://wonko.com/post/html-escaping)
-    ///
-
-    internal static let requiredEscapingsTable: [UInt32:String] = {
-        return [
-            0x22:""",0x26:"&",0x27:"'",0x3c:"<",0x3e:">",0x152:"Œ",0x153:"œ",0x160:"Š",0x161:"š",0x178:"Ÿ",0x2c6:"ˆ",0x2dc:"˜",0x2002:" ",0x2003:" ",0x2009:" ",0x200c:"‌",0x200d:"‍",0x200e:"‎",0x200f:"‏",0x2013:"–",0x2014:"—",0x2018:"‘",0x2019:"’",0x201a:"‚",0x201c:"“",0x201d:"”",0x201e:"„",0x2020:"†",0x2021:"‡",0x2030:"‰",0x2039:"‹",0x203a:"›",0x20ac:"euro"
-        ]
-    }()
-
 }
+