Merge pull request #169 from StartAutomating/Irregular-Improvement

Irregular improvement
StartAutomating · Oct 29, 2022 · 9861ccd · 9861ccd
2 parents 0bfcfd2 + e037377
commit 9861ccd
Show file tree

Hide file tree

Showing 59 changed files with 232 additions and 117 deletions.
diff --git a/.github/workflows/RunIrregularAction.yml b/.github/workflows/RunIrregularAction.yml
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,4 +1,18 @@
-## 0.7.5:
+## 0.7.6:
+
+* New-RegEx improvements:
+  * Now supporting new character classes:  MarkSpacing, MarkEnclosing, MarkNonEnclosing (Fixes #168)
+
+* More CSharp Patterns:
+  * ?<CSharp_Class> (Fixes #164)
+  * ?<CSharp_Identifier> (Fixes #165)
+
+* Improved Patterns:
+  * ?<ANSI_4BitColor>/?<ANSI_Style>: Handling brightness bit before or after color (Fixes #166)
+
+---
+
+## 0.7.5:
 
 * Added Patterns for Liquid:
   * ?<Liquid_Expression> (Fixes #153)

diff --git a/Irregular.GitHubWorkflow.PSDevops.ps1 b/Irregular.GitHubWorkflow.PSDevops.ps1
@@ -2,7 +2,7 @@
 Push-Location $PSScriptRoot
 
 Import-BuildStep -ModuleName Irregular
-New-GitHubWorkflow -Name "Test, Tag, Release, and Publish" -Job PowerShellStaticAnalysis, TestPowerShellOnLinux, TagReleaseAndPublish, BuildModule -On Push, Demand |
+New-GitHubWorkflow -Name "Test, Tag, Release, and Publish" -Job PowerShellStaticAnalysis, TestPowerShellOnLinux, TagReleaseAndPublish, BuildIrregular -On Push, Demand |
     Set-Content .\.github\workflows\IrregularTests.yml -Encoding UTF8 -PassThru
 
 New-GitHubWorkflow -Name "Run GitHub Action" -On Push, Demand -Job UseIrregularAction |

diff --git a/Irregular.format.ps1xml b/Irregular.format.ps1xml
@@ -1,5 +1,5 @@
 <?xml version="1.0" encoding="utf-16"?>
-<!-- Generated with EZOut 1.9.6: Install-Module EZOut or https://github.com/StartAutomating/EZOut -->
+<!-- Generated with EZOut 1.9.7: Install-Module EZOut or https://github.com/StartAutomating/EZOut -->
 <Configuration>
   <ViewDefinitions>
     <View>

diff --git a/Irregular.psd1 b/Irregular.psd1
@@ -1,5 +1,5 @@
 @{
-    ModuleVersion = '0.7.5'
+    ModuleVersion = '0.7.6'
     RootModule = 'Irregular.psm1'
     Description = 'Regular Expressions made Strangely Simple'
     FormatsToProcess = 'Irregular.format.ps1xml'
@@ -14,6 +14,20 @@
             LicenseURI = 'https://github.com/StartAutomating/Irregular/blob/master/LICENSE'
             IconURI    = 'https://github.com/StartAutomating/Irregular/blob/master/Assets/Irregular_600_Square.png'        
             ReleaseNotes = @'
+## 0.7.6:
+
+* New-RegEx improvements:
+  * Now supporting new character classes:  MarkSpacing, MarkEnclosing, MarkNonEnclosing (Fixes #168)
+
+* More CSharp Patterns:
+  * ?<CSharp_Class> (Fixes #164)
+  * ?<CSharp_Identifier> (Fixes #165)
+
+* Improved Patterns:
+  * ?<ANSI_4BitColor>/?<ANSI_Style>: Handling brightness bit before or after color (Fixes #166)
+
+---
+
 ## 0.7.5:
 
 * Added Patterns for Liquid:

diff --git a/Irregular.types.ps1xml b/Irregular.types.ps1xml
@@ -1,5 +1,5 @@
 <?xml version="1.0" encoding="utf-16"?>
-<!-- Generated with EZOut 1.9.6: Install-Module EZOut or https://github.com/StartAutomating/EZOut -->
+<!-- Generated with EZOut 1.9.7: Install-Module EZOut or https://github.com/StartAutomating/EZOut -->
 <Types>
   <Type>
     <Name>Irregular.Match.Extract</Name>

diff --git a/New-RegEx.ps1 b/New-RegEx.ps1
@@ -82,6 +82,9 @@
         'CombiningMark' ,'\p{Mc}',
         'EnclosingMark' , '\p{Me}',
         'Mark' , '\p{M}',
+        'MarkSpacing' , '\p{Mc}',
+        'MarkNonSpacing' , '\p{Mn}',
+        'MarkEnclosing' , '\p{Me}',
         'Number' , '\p{N}',
         'NumberDecimalDigit' , '\p{Nd}',
         'NumberLetter' , '\p{Nl}',

diff --git a/README.md b/README.md
@@ -3,7 +3,7 @@
 <h2>Regular Expressions made Strangely Simple</h2>
 <h3>A PowerShell module that helps you understand, use, and build Regular Expressions.</h3>
 <h4>
-<a href='https://github.com/StartAutomating/Irregular/releases/tag/v0.7.5'>v 0.7.5 </a>
+<a href='https://github.com/StartAutomating/Irregular/releases/tag/v0.7.6'>v 0.7.6 </a>
 </h4>
 <a href='https://www.powershellgallery.com/packages/Irregular/'>
 <img src='https://img.shields.io/powershellgallery/dt/Irregular' />
@@ -32,7 +32,7 @@ Once you understand some basics of that syntax, regular expressions become a lot
 3. A Regex can have comments! ( # Like this in .NET  ( or like (?#this comment) in ECMAScript ) ).
 4. You don't have to do it all in one expression! 
 
-Irregular comes with 133 useful [named expressions](SavedPatterns.md), and lets you create more.
+Irregular comes with 134 useful [named expressions](SavedPatterns.md), and lets you create more.
 
 To see the expressions that ship with Irregular, run:
 
@@ -72,7 +72,7 @@ New-RegEx -StartAnchor StringStart -Pattern @(
 
 Irregular also contains a colorized PowerShell formatter for all Regular Expressions.
 This provides syntax highlighting that can make complicated expressions easier to read.
-![RegexSyntaxHighlighting](assets/images/RegexSyntaxHighlighting.gif)
+![RegexSyntaxHighlighting](assets/RegexSyntaxHighlighting.gif)
 
 
 #### Building Regular Expressions

diff --git a/README.ps1.md b/README.ps1.md
@@ -82,7 +82,7 @@ New-RegEx -StartAnchor StringStart -Pattern @(
 
 Irregular also contains a colorized PowerShell formatter for all Regular Expressions.
 This provides syntax highlighting that can make complicated expressions easier to read.
-![RegexSyntaxHighlighting](assets/images/RegexSyntaxHighlighting.gif)
+![RegexSyntaxHighlighting](assets/RegexSyntaxHighlighting.gif)
 
 
 #### Building Regular Expressions

diff --git a/RegEx/ANSI/4BitColor.regex.source.ps1 b/RegEx/ANSI/4BitColor.regex.source.ps1
@@ -7,23 +7,28 @@ New-RegEx -Description "Matches an ANSI 3 or 4-bit color" |
     New-RegEx -Pattern (
 
         New-RegEx -Atomic -Or @(
+
             New-RegEx -Pattern 1 -Optional -Name IsBright |
-                New-RegEx -LiteralCharacter ';' -Min 0 -Max 1 |    
-                New-RegEx -Pattern '3' -Name IsForegroundColor
-
+                New-RegEx -LiteralCharacter ';' -Min 0 -Max 1 -Comment "A 1 and a semicolon indicate a bright color" |    
+                New-RegEx -Pattern '3' -Name IsForegroundColor -Comment "A number that starts with 3 indicates foreground color"
+                        
             New-RegEx -Name IsBright (
                 New-RegEx -Pattern '9' -Name IsForegroundColor
-            )
+            ) -Comment "OR it could be a less common bright foreground color, which starts with 9"
 
             New-RegEx -Pattern 1 -Optional -Name IsBright |
-                New-RegEx -LiteralCharacter ';' -Min 0 -Max 1 |
-                New-RegEx -Pattern '4' -Name IsBackgroundColor
+                New-RegEx -LiteralCharacter ';' -Min 0 -Max 1 -Comment "A 1 and a semicolon indicate a bright color"|
+                New-RegEx -Pattern '4' -Name IsBackgroundColor -Comment "A number that starts with 3 indicates foreground color"
 
             New-RegEx -Name IsBright (
                 New-RegEx -Pattern '10' -Name IsBackgroundColor
-            )
-        ) |
-        New-RegEx -Pattern '[0-7]' -Name ColorNumber |
+            ) -Comment "OR it could be a less common bright foreground color, which starts with 9"
+        ) |        
+        New-RegEx -Pattern '[0-7]' -Name ColorNumber -Comment "The color number will be between 0 and 7" |
+        New-RegEx -NoCapture -Optional @(
+            New-RegEx -LiteralCharacter ';' -Min 0 -Max 1 |
+                New-RegEx -Pattern 1 -Optional -Name IsBright
+        ) -Comment "Brightness can also come after a color" |
         New-RegEx -LiteralCharacter 'm'
-    ) -Name Color | #>   
-    Set-Content -Path (Join-Path $myRoot $myName)
+    ) -Name Color |
+    Set-Content -Path (Join-Path $myRoot $myName) -PassThru
diff --git a/RegEx/ANSI/4BitColor.regex.txt b/RegEx/ANSI/4BitColor.regex.txt
@@ -1,8 +1,16 @@
 # Matches an ANSI 3 or 4-bit color
-\e # An Escape
-\[ # Followed by a bracket
+\e                                      # An Escape
+\[                                      # Followed by a bracket
 (?<Color>(?>
-  (?<IsBright>1)?\;{0,1}(?<IsForegroundColor>3)  |
-  (?<IsBright>(?<IsForegroundColor>9))  |
-  (?<IsBright>1)?\;{0,1}(?<IsBackgroundColor>4)  |
-  (?<IsBright>(?<IsBackgroundColor>10)))(?<ColorNumber>[0-7])m)
+  (?<IsBright>1)?\;{0,1}                # A 1 and a semicolon indicate a bright color
+(?<IsForegroundColor>3)                 # A number that starts with 3 indicates foreground color
+  |
+  (?<IsBright>(?<IsForegroundColor>9))  # OR it could be a less common bright foreground color, which starts with 9
+  |
+  (?<IsBright>1)?\;{0,1}                # A 1 and a semicolon indicate a bright color
+(?<IsBackgroundColor>4)                 # A number that starts with 3 indicates foreground color
+  |
+  (?<IsBright>(?<IsBackgroundColor>10)) # OR it could be a less common bright foreground color, which starts with 9
+)(?<ColorNumber>[0-7])                  # The color number will be between 0 and 7
+(?:\;{0,1}(?<IsBright>1)?)?             # Brightness can also come after a color
+m)
diff --git a/RegEx/ANSI/Style.regex.txt b/RegEx/ANSI/Style.regex.txt
@@ -123,10 +123,18 @@ m))
 \e                                                                                     # An Escape
 \[                                                                                     # Followed by a bracket
 (?<Color>(?>
-  (?<IsBright>1)?\;{0,1}(?<IsForegroundColor>3)  |
-  (?<IsBright>(?<IsForegroundColor>9))  |
-  (?<IsBright>1)?\;{0,1}(?<IsBackgroundColor>4)  |
-  (?<IsBright>(?<IsBackgroundColor>10)))(?<ColorNumber>[0-7])m)
+  (?<IsBright>1)?\;{0,1}                                                               # A 1 and a semicolon indicate a bright color
+(?<IsForegroundColor>3)                                                                # A number that starts with 3 indicates foreground color
+  |
+  (?<IsBright>(?<IsForegroundColor>9))                                                 # OR it could be a less common bright foreground color, which starts with 9
+  |
+  (?<IsBright>1)?\;{0,1}                                                               # A 1 and a semicolon indicate a bright color
+(?<IsBackgroundColor>4)                                                                # A number that starts with 3 indicates foreground color
+  |
+  (?<IsBright>(?<IsBackgroundColor>10))                                                # OR it could be a less common bright foreground color, which starts with 9
+)(?<ColorNumber>[0-7])                                                                 # The color number will be between 0 and 7
+(?:\;{0,1}(?<IsBright>1)?)?                                                            # Brightness can also come after a color
+m)
 )
   |
   (?<ANSI_DefaultColor>

diff --git a/RegEx/CSharp/Class.regex.source.ps1 b/RegEx/CSharp/Class.regex.source.ps1
@@ -0,0 +1,43 @@
+$myName = ($MyInvocation.MyCommand.ScriptBlock.File | Split-Path -Leaf) -replace '\.source', '' -replace '\.ps1', '.txt'
+$myRoot = $MyInvocation.MyCommand.ScriptBlock.File | Split-Path
+New-RegEx -Description 'Matches a CSharp class' | 
+    New-RegEx -Name AccessModifier @(
+        New-Regex -Modifier IgnoreCase -Not
+        New-RegEx -Atomic -Or @(
+            New-RegEx -Pattern "protected\s{1,}internal"
+            New-RegEx -Pattern "protected"
+            New-RegEx -Pattern "private\s{1,}protected"
+            New-RegEx -Pattern "internal"
+            New-RegEx -Pattern "private"
+            New-RegEx -Pattern "public"            
+        ) -Optional
+    ) -Comment "An Optional Access Modifier" |
+    New-RegEx -CharacterClass Whitespace -Min 1 |
+    New-RegEx -NoCapture @(
+        New-RegEx -Modifier IgnoreCase -Not | 
+        New-RegEx -Pattern class | 
+        New-RegEx -CharacterClass Whitespace -Min 1
+    ) -Comment "Followed by 'class'" |
+    New-RegEx -Name ClassName @(   
+        New-Regex -LiteralCharacter _ -CharacterClass Letter |    
+            New-RegEx -CharacterClass PunctuationConnector, Letter, Digit, MarkNonSpacing -Min 0
+    ) -Comment "Followed by an identifier"|
+    New-RegEx -Optional -Pattern @(
+        New-RegEx -CharacterClass Whitespace -Min 1 |
+        New-RegEx -LiteralCharacter ':' -Name IsInheriting |
+        New-RegEx -CharacterClass Whitespace -Min 1
+    ) -NoCapture -Comment "Followed by an optional colon" |
+    New-RegEx -If IsInheriting -Then @(
+        New-RegEx -NoCapture @(
+            New-RegEx -CharacterClass Whitespace -Min 0
+            New-RegEx -Name Inherits @(
+                New-Regex -LiteralCharacter _ -CharacterClass Letter |    
+                    New-RegEx -CharacterClass PunctuationConnector, Letter, Digit, MarkNonSpacing -Min 0
+            ) |
+            New-RegEx -CharacterClass Whitespace -Min 0 |
+            New-RegEx -LiteralCharacter ',' -Min 0
+        ) -Min 1        
+    ) -Else '' -Description "If a colon was present, followed by a number of identifiers" |
+    New-RegEx -CharacterClass Whitespace -Min 0 -Comment "Followed by whitespace and balanced curly brackets" |
+    New-RegEx -Pattern ?<BalancedCurlyBracket> -Name ClassBody |
+    Set-Content -Path (Join-Path $myRoot $myName) -Encoding UTF8 -PassThru
diff --git a/RegEx/CSharp/Class.regex.txt b/RegEx/CSharp/Class.regex.txt
@@ -0,0 +1,23 @@
+# Matches a CSharp class
+(?<AccessModifier>(?:(?-i) (?>
+  protected\s{1,}internal  |
+  protected  |
+  private\s{1,}protected  |
+  internal  |
+  private  |
+  public)?))                                                                                           # An Optional Access Modifier
+\s{1,}(?:(?-i)class\s{1,})                                                                             # Followed by 'class'
+(?<ClassName>[\p{L}_][\p{Pc}\p{L}\d\p{Mn}]{0,})                                                        # Followed by an identifier
+(?:\s{1,}(?<IsInheriting>\:)\s{1,})?                                                                   # Followed by an optional colon
+# If a colon was present, followed by a number of identifiers
+(?(IsInheriting)((?:(?:\s{0,} (?<Inherits>[\p{L}_][\p{Pc}\p{L}\d\p{Mn}]{0,})\s{0,}\,{0,})){1,}))\s{0,} # Followed by whitespace and balanced curly brackets
+(?<ClassBody>(?<BalancedCurlyBracket>
+\{                                                                                                     # An open {
+(?>                                                                                                    # Followed by...
+    [^\{\}]+|                                                                                          # any number of non-bracket character OR
+    \{(?<Depth>)|                                                                                      # an open curly bracket (in which case increment depth) OR
+    \}(?<-Depth>)                                                                                      # a closed curly bracket (in which case decrement depth)
+)*?(?(Depth)(?!))                                                                                      # until depth is 0.
+\}                                                                                                     # followed by a }
+)
+)
diff --git a/RegEx/CSharp/Identifier.regex.source.ps1 b/RegEx/CSharp/Identifier.regex.source.ps1
@@ -0,0 +1,6 @@
+$myName = ($MyInvocation.MyCommand.ScriptBlock.File | Split-Path -Leaf) -replace '\.source', '' -replace '\.ps1', '.txt'
+$myRoot = $MyInvocation.MyCommand.ScriptBlock.File | Split-Path
+New-RegEx -Description 'Matches a CSharp identifier' | 
+    New-Regex -Comment "Must start with a letter or underscore" -LiteralCharacter _ -CharacterClass Letter |    
+    New-RegEx -CharacterClass PunctuationConnector, Letter, Digit, MarkNonSpacing -Description "Followed by any number of letters, digitis, nonspacing marks, and connector punctation" -Min 0|
+    Set-Content -Path (Join-Path $myRoot $myName) -Encoding UTF8 -PassThru
diff --git a/RegEx/CSharp/Identifier.regex.txt b/RegEx/CSharp/Identifier.regex.txt
@@ -0,0 +1,4 @@
+# Matches a CSharp identifier
+[\p{L}_] # Must start with a letter or underscore
+# Followed by any number of letters, digitis, nonspacing marks, and connector punctation
+[\p{Pc}\p{L}\d\p{Mn}]{0,}
diff --git a/RegEx/CSharp/README.md b/RegEx/CSharp/README.md
@@ -1,9 +1,11 @@
-This directory contains regular expressions to help parse C#.
+This directory contains regular expressions to help parse C# code.
 
 
-|Name                                      |Description                   |Source                              |
-|------------------------------------------|------------------------------|------------------------------------|
-|[?<CSharp_Namespace>](Namespace.regex.txt)|Matches a CSharp namespace    |[source](Namespace.regex.source.ps1)|
-|[?<CSharp_Using>](Using.regex.txt)        |Matches a CSharp using keyword|
+|Name                                        |Description                   |Source                               |
+|--------------------------------------------|------------------------------|-------------------------------------|
+|[?<CSharp_Class>](Class.regex.txt)          |Matches a CSharp class        |[source](Class.regex.source.ps1)     |
+|[?<CSharp_Identifier>](Identifier.regex.txt)|Matches a CSharp identifier   |[source](Identifier.regex.source.ps1)|
+|[?<CSharp_Namespace>](Namespace.regex.txt)  |Matches a CSharp namespace    |[source](Namespace.regex.source.ps1) |
+|[?<CSharp_Using>](Using.regex.txt)          |Matches a CSharp using keyword|
 
 
diff --git a/RegEx/CSharp/README.ps1.md b/RegEx/CSharp/README.ps1.md
@@ -1,4 +1,4 @@
-This directory contains regular expressions to help parse C#.
+This directory contains regular expressions to help parse C# code.
 
 ~~~PipeScript{
     Import-Module ../../Irregular.psd1 -Global

diff --git a/RegEx/Digits.regex.txt b/RegEx/Digits.regex.txt
@@ -1,2 +1,2 @@
-# Repeated Digits
+# Repeated Digits
 \d+
diff --git a/RegEx/Git/DiffHeader.regex.source.ps1 b/RegEx/Git/DiffHeader.regex.source.ps1
@@ -10,9 +10,9 @@ Write-RegEx -Pattern 'diff' -StartAnchor LineStart  -Comment 'diff line start' |
     Write-Regex -CharacterClass Whitespace -Comment Whitespace |
     Write-RegEx -Name To -CharacterClass NonWhitespace -Repeat -Comment To |
     Write-RegEx -Name ExtendedHeader -Until (
-        Write-RegEx 'index' -StartAnchor LineStart
-    ) -Comment "Until 'index'" |
-    Write-RegEx 'index' -StartAnchor LineStart |
+        Write-RegEx -Pattern 'index\s{1}' -StartAnchor LineStart
+    ) -Comment "Until 'index'"  <#  |
+    Write-RegEx -Pattern 'index' |
     Write-RegEx -CharacterClass Whitespace -Comment 'Index Line Start' |
     Write-RegEx -Name FromHash -CharacterClass Word -Repeat -Comment FromHash |
     Write-RegEx -LiteralCharacter '.' -Min 2 -Max 2 -Comment DotDot |
@@ -29,6 +29,6 @@ Write-RegEx -Pattern 'diff' -StartAnchor LineStart  -Comment 'diff line start' |
     Write-RegEx -LiteralCharacter '+' -Min 3 -Max 3 |
     Write-RegEx -CharacterClass Whitespace -Comment PlusLineAndWhitespace |
     Write-RegEx -Name ToUnified -CharacterClass NonWhitespace -Repeat -Comment ToUnified |
-    Write-RegEx -CharacterClass Whitespace |
+    Write-RegEx -CharacterClass Whitespace #> |
     Set-Content -Path (Join-Path $myRoot $myName)
 
diff --git a/RegEx/Git/DiffHeader.regex.txt b/RegEx/Git/DiffHeader.regex.txt
@@ -1,18 +1,8 @@
 # Matches Header information from the output of git diff
-(?m)^diff                                     # diff line start
-\s--git\s                                     # whitespace, then --git, then whitespace
-(?<From>\S+)                                  # From
-\s                                            # Whitespace
-(?<To>\S+)                                    # To
-(?<ExtendedHeader>(?:.|\s){0,}?(?=\z|^index)) # Until 'index'
-^index\s                                      # Index Line Start
-(?<FromHash>\w+)                              # FromHash
-\.{2,2}                                       # DotDot
-(?<ToHash>\w+)                                # ToHash
-\s(?<Mode>\w+)                                # FileMode
-(?:.|\s){0,}?(?=\z|^---)                      # UntilDashLine
-\-{3,3}\s                                     # DashLineAndWhitespace
-(?<FromUnified>\S+)                           # FromUnified
-\s+\+{3,3}\s                                  # PlusLineAndWhitespace
-(?<ToUnified>\S+)                             # ToUnified
-\s
+(?m)^diff                                          # diff line start
+\s--git\s                                          # whitespace, then --git, then whitespace
+(?<From>\S+)                                       # From
+\s                                                 # Whitespace
+(?<To>\S+)                                         # To
+(?<ExtendedHeader>(?:.|\s){0,}?(?=\z|^index\s{1})) # Until 'index'
+