FParsec.CSharp
0.1.36
A thin C# wrapper for FParsec.
See the version list below for details.
Install-Package FParsec.CSharp -Version 0.1.36
dotnet add package FParsec.CSharp --version 0.1.36
<PackageReference Include="FParsec.CSharp" Version="0.1.36" />
paket add FParsec.CSharp --version 0.1.36
FParsec.CSharp
FParsec.CSharp is a C# wrapper for the F# package FParsec. FParsec is a parser combinator library with which you can implement parsers declaratively.
Why FParsec.CSharp?
While using FParsec from C# is entirely possible in theory, it is very awkward in practice. Most of FParsec's elegance is lost in translation due to C#'s inferior type inference and its lack of custom operators.
FParsec.CSharp tries to alleviate that by wrapping FParsec's operators as extension functions.
FParsec.CSharp does not try to hide any types from FParsec
or FSharp.Core
--the wrapper is thin and also avoids name collisions. That way you can always fallback to FParsec anytime you need some functionality not (yet) implemented by FParsec.CSharp.
Based on the current implementation it should be easy to extend the wrapper yourself if needed. Pull requests are always welcome!
Getting started
Import the combinators, pre-defined parsers, and helper functions:
using FParsec.CSharp; // extension functions (combinators & helpers)
using static FParsec.CSharp.PrimitivesCS; // combinator functions
using static FParsec.CSharp.CharParsersCS; // pre-defined parsers
Now you can write some parsers:
var p = AnyChar.And(Digit);
var r = p.ParseString("a1");
System.Diagnostics.Debug.Assert(r.Result == ('a', '1'));
Using FParsec.CSharp and FParsec together
In case you need one of FParsec's more specialized parsers you can easily import their namespace:
using static FParsec.CharParsers;
In the example below we are using FParsec.CharParsers.many1Chars2()
. As you can see it integrates seemlessly with FParsec.CSharp:
var first = Letter.Or(CharP('_'));
var rest = Letter.Or(CharP('_')).Or(Digit);
var identifier = many1Chars2(first, rest);
var p = identifier.And(Skip('=')).And(Int);
var r = p.ParseString("my_1st_var=13");
System.Diagnostics.Debug.Assert(r.Result == ("my_1st_var", 13));
FParsec supports parsing inputs with custom user state, wich is reflected by most of its functions taking a user state type variable. FParsec.CSharp however does not support user state, so you will have to specify Microsoft.FSharp.Core.Unit
as the user state type when it can not be inferred:
var p = restOfLine<Unit>(true);
Otherwise you won't be able to use the combinators from FParsec.CSharp on it.
Some of FParsec's parsers take anonymous functions. But since they expect curried FSharpFunc
s they won't accept C# lambdas. FParsec.CSharp comes with a little helper to create FSharpFunc
s from Func
objects:
// convert lambda with factory method
var fsfunc1 = FSharpFunc.From<char, bool>(c => c == 'x' || c == 'y');
// convert Func object with extension method
Func<char, bool> func = c => c == '1' || c == '2';
var fsfunc2 = func.ToFSharpFunc();
var p = manySatisfy<Unit>(fsfunc1).And(manySatisfy<Unit>(fsfunc2));
var r = p.ParseString("xyxyyy212221212");
Examples
Simple JSON
FSharpFunc<CharStream<Unit>, Reply<object>> jvalue = null;
var jnull = StringCI("null").Return((object)null);
var jnum = Int.Map(i => (object)i);
var jbool = StringCI("true").Or(StringCI("false"))
.Map(b => (object)bool.Parse(b));
var quotedString = Between('"', Many(NoneOf("\"")), '"')
.Map(string.Concat);
var jstring = quotedString.Map(s => (object)s);
var arrItems = Many(Rec(() => jvalue), sep: CharP(',').And(WS));
var jarray = Between(CharP('[').And(WS), arrItems, CharP(']'))
.Map(elems => (object)new JArray(elems));
var jidentifier = quotedString;
var jprop = jidentifier.And(WS).And(Skip(':')).And(WS).And(Rec(() => jvalue))
.Map((name, value) => new JProperty(name, value));
var objProps = Many(jprop, sep: CharP(',').And(WS));
var jobject = Between(CharP('{').And(WS), objProps, CharP('}'))
.Map(props => (object)new JObject(props));
jvalue = OneOf(jnum, jbool, jnull, jstring, jarray, jobject).And(WS);
var simpleJsonParser = WS.And(jobject).And(WS).And(EOF).Map(o => (JObject)o);
Simple XML
var nameStart = Letter.Or(CharP('_'));
var nameChar = Letter.Or(Digit).Or(AnyOf("-_."));
var name = nameStart.And(Many(nameChar))
.Map((first, rest) => string.Concat(rest.Prepend(first)));
var quotedString = Between('"', Many(NoneOf("\"")), '"')
.Map(string.Concat);
var attribute = WS1.And(name).And(WS).And(Skip('=')).And(WS).And(quotedString)
.Map((attrName, attrVal) => new XAttribute(attrName, attrVal));
var attributes = Many(Try(attribute));
FSharpFunc<CharStream<Unit>, Reply<XElement>> element = null;
var emptyElement = Between("<", name.And(attributes).And(WS), "/>")
.Map((elName, attrs) => new XElement(elName, attrs));
var openingTag = Between('<', name.And(attributes).And(WS), '>');
FSharpFunc<CharStream<Unit>, Reply<string>> closingTag(string tagName) => Between("</", StringP(tagName).And(WS), ">");
var childElements = Many1(Try(WS.And(Rec(() => element)).And(WS)))
.Map(attrs => (object)attrs);
var text = Many(NoneOf("<"))
.Map(t => (object)string.Concat(t));
var content = childElements.Or(text);
var parentElement = openingTag.And(content).Map(Flat).And(x => closingTag(x.Item1).Return(x))
.Map((elName, elAttrs, elContent) => new XElement(elName, elAttrs, elContent));
element = Try(emptyElement).Or(parentElement);
var simpleXmlParser = element.And(WS).And(EOF);
Glob patterns
var globParser =
Many(OneOf(
Skip('?').Map(NFA.MakeAnyChar),
Skip('*').Map(NFA.MakeAnyChar).Map(NFA.MakeZeroOrMore),
Between('[', AnyChar.And(Skip('-')).And(AnyChar), ']').Map(NFA.MakeCharRange),
Skip('\\').And(AnyOf(@"?*[]\")).Map(NFA.MakeChar),
AnyChar.Map(NFA.MakeChar)))
.And(EOF)
.Map(NFA.Concat)
.Map(proto => proto(new Final()));
This example contructs a non-deterministic finite automaton (NFA) during parsing and can be used for matching:
[Fact] public void CanParseAndMatchGlobPattern() => globParser
.ParseString(@"The * syntax is easy?").Result
.Matches("The glob syntax is easy!")
.ShouldBe(true);
Arithmetic expressions
FParsec.CSharp comes with a builder to construct FParsec.OperatorPrecedenceParser
s:
var exprParser =
WS.And(new OPPBuilder<int, Unit>()
.WithOperators(ops => ops
.AddInfix("+", 10, Associativity.Left, WS, (x, y) => x + y)
.AddInfix("-", 10, Associativity.Left, WS, (x, y) => x - y)
.AddInfix("*", 20, Associativity.Left, WS, (x, y) => x * y)
.AddInfix("/", 20, Associativity.Left, WS, (x, y) => x / y)
.AddPrefix("-", 20, x => -x)
.AddInfix("^", 30, Associativity.Right, WS, (x, y) => (int)Math.Pow(x, y))
.AddPostfix("!", 40, Factorial))
.WithImplicitOperator(20, (x, y) => x * y)
.WithTerms(term => OneOf(
Integer.And(WS),
Between(CharP('(').And(WS), term, CharP(')').And(WS))))
.Build()
.ExpressionParser);
Alternatives
FParsec.CSharp does not wrap all of FParsec's features yet. If you need an all-in-one solution then have a look at the following alternatives:
FParsec.CSharp
FParsec.CSharp is a C# wrapper for the F# package FParsec. FParsec is a parser combinator library with which you can implement parsers declaratively.
Why FParsec.CSharp?
While using FParsec from C# is entirely possible in theory, it is very awkward in practice. Most of FParsec's elegance is lost in translation due to C#'s inferior type inference and its lack of custom operators.
FParsec.CSharp tries to alleviate that by wrapping FParsec's operators as extension functions.
FParsec.CSharp does not try to hide any types from FParsec
or FSharp.Core
--the wrapper is thin and also avoids name collisions. That way you can always fallback to FParsec anytime you need some functionality not (yet) implemented by FParsec.CSharp.
Based on the current implementation it should be easy to extend the wrapper yourself if needed. Pull requests are always welcome!
Getting started
Import the combinators, pre-defined parsers, and helper functions:
using FParsec.CSharp; // extension functions (combinators & helpers)
using static FParsec.CSharp.PrimitivesCS; // combinator functions
using static FParsec.CSharp.CharParsersCS; // pre-defined parsers
Now you can write some parsers:
var p = AnyChar.And(Digit);
var r = p.ParseString("a1");
System.Diagnostics.Debug.Assert(r.Result == ('a', '1'));
Using FParsec.CSharp and FParsec together
In case you need one of FParsec's more specialized parsers you can easily import their namespace:
using static FParsec.CharParsers;
In the example below we are using FParsec.CharParsers.many1Chars2()
. As you can see it integrates seemlessly with FParsec.CSharp:
var first = Letter.Or(CharP('_'));
var rest = Letter.Or(CharP('_')).Or(Digit);
var identifier = many1Chars2(first, rest);
var p = identifier.And(Skip('=')).And(Int);
var r = p.ParseString("my_1st_var=13");
System.Diagnostics.Debug.Assert(r.Result == ("my_1st_var", 13));
FParsec supports parsing inputs with custom user state, wich is reflected by most of its functions taking a user state type variable. FParsec.CSharp however does not support user state, so you will have to specify Microsoft.FSharp.Core.Unit
as the user state type when it can not be inferred:
var p = restOfLine<Unit>(true);
Otherwise you won't be able to use the combinators from FParsec.CSharp on it.
Some of FParsec's parsers take anonymous functions. But since they expect curried FSharpFunc
s they won't accept C# lambdas. FParsec.CSharp comes with a little helper to create FSharpFunc
s from Func
objects:
// convert lambda with factory method
var fsfunc1 = FSharpFunc.From<char, bool>(c => c == 'x' || c == 'y');
// convert Func object with extension method
Func<char, bool> func = c => c == '1' || c == '2';
var fsfunc2 = func.ToFSharpFunc();
var p = manySatisfy<Unit>(fsfunc1).And(manySatisfy<Unit>(fsfunc2));
var r = p.ParseString("xyxyyy212221212");
Examples
Simple JSON
FSharpFunc<CharStream<Unit>, Reply<object>> jvalue = null;
var jnull = StringCI("null").Return((object)null);
var jnum = Int.Map(i => (object)i);
var jbool = StringCI("true").Or(StringCI("false"))
.Map(b => (object)bool.Parse(b));
var quotedString = Between('"', Many(NoneOf("\"")), '"')
.Map(string.Concat);
var jstring = quotedString.Map(s => (object)s);
var arrItems = Many(Rec(() => jvalue), sep: CharP(',').And(WS));
var jarray = Between(CharP('[').And(WS), arrItems, CharP(']'))
.Map(elems => (object)new JArray(elems));
var jidentifier = quotedString;
var jprop = jidentifier.And(WS).And(Skip(':')).And(WS).And(Rec(() => jvalue))
.Map((name, value) => new JProperty(name, value));
var objProps = Many(jprop, sep: CharP(',').And(WS));
var jobject = Between(CharP('{').And(WS), objProps, CharP('}'))
.Map(props => (object)new JObject(props));
jvalue = OneOf(jnum, jbool, jnull, jstring, jarray, jobject).And(WS);
var simpleJsonParser = WS.And(jobject).And(WS).And(EOF).Map(o => (JObject)o);
Simple XML
var nameStart = Letter.Or(CharP('_'));
var nameChar = Letter.Or(Digit).Or(AnyOf("-_."));
var name = nameStart.And(Many(nameChar))
.Map((first, rest) => string.Concat(rest.Prepend(first)));
var quotedString = Between('"', Many(NoneOf("\"")), '"')
.Map(string.Concat);
var attribute = WS1.And(name).And(WS).And(Skip('=')).And(WS).And(quotedString)
.Map((attrName, attrVal) => new XAttribute(attrName, attrVal));
var attributes = Many(Try(attribute));
FSharpFunc<CharStream<Unit>, Reply<XElement>> element = null;
var emptyElement = Between("<", name.And(attributes).And(WS), "/>")
.Map((elName, attrs) => new XElement(elName, attrs));
var openingTag = Between('<', name.And(attributes).And(WS), '>');
FSharpFunc<CharStream<Unit>, Reply<string>> closingTag(string tagName) => Between("</", StringP(tagName).And(WS), ">");
var childElements = Many1(Try(WS.And(Rec(() => element)).And(WS)))
.Map(attrs => (object)attrs);
var text = Many(NoneOf("<"))
.Map(t => (object)string.Concat(t));
var content = childElements.Or(text);
var parentElement = openingTag.And(content).Map(Flat).And(x => closingTag(x.Item1).Return(x))
.Map((elName, elAttrs, elContent) => new XElement(elName, elAttrs, elContent));
element = Try(emptyElement).Or(parentElement);
var simpleXmlParser = element.And(WS).And(EOF);
Glob patterns
var globParser =
Many(OneOf(
Skip('?').Map(NFA.MakeAnyChar),
Skip('*').Map(NFA.MakeAnyChar).Map(NFA.MakeZeroOrMore),
Between('[', AnyChar.And(Skip('-')).And(AnyChar), ']').Map(NFA.MakeCharRange),
Skip('\\').And(AnyOf(@"?*[]\")).Map(NFA.MakeChar),
AnyChar.Map(NFA.MakeChar)))
.And(EOF)
.Map(NFA.Concat)
.Map(proto => proto(new Final()));
This example contructs a non-deterministic finite automaton (NFA) during parsing and can be used for matching:
[Fact] public void CanParseAndMatchGlobPattern() => globParser
.ParseString(@"The * syntax is easy?").Result
.Matches("The glob syntax is easy!")
.ShouldBe(true);
Arithmetic expressions
FParsec.CSharp comes with a builder to construct FParsec.OperatorPrecedenceParser
s:
var exprParser =
WS.And(new OPPBuilder<int, Unit>()
.WithOperators(ops => ops
.AddInfix("+", 10, Associativity.Left, WS, (x, y) => x + y)
.AddInfix("-", 10, Associativity.Left, WS, (x, y) => x - y)
.AddInfix("*", 20, Associativity.Left, WS, (x, y) => x * y)
.AddInfix("/", 20, Associativity.Left, WS, (x, y) => x / y)
.AddPrefix("-", 20, x => -x)
.AddInfix("^", 30, Associativity.Right, WS, (x, y) => (int)Math.Pow(x, y))
.AddPostfix("!", 40, Factorial))
.WithImplicitOperator(20, (x, y) => x * y)
.WithTerms(term => OneOf(
Integer.And(WS),
Between(CharP('(').And(WS), term, CharP(')').And(WS))))
.Build()
.ExpressionParser);
Alternatives
FParsec.CSharp does not wrap all of FParsec's features yet. If you need an all-in-one solution then have a look at the following alternatives:
Release Notes
Move LambdaConvert to namespace FParsec.CSharp.
Dependencies
-
.NETStandard 2.0
- FParsec (>= 1.0.3)
- FSharp.Core (>= 4.6.2)
Used By
NuGet packages
This package is not used by any NuGet packages.
GitHub repositories
This package is not used by any popular GitHub repositories.
Version History
Version | Downloads | Last updated |
---|---|---|
12.1.0 | 39 | 1/1/2021 |
12.0.5 | 39 | 12/30/2020 |
12.0.4 | 160 | 11/14/2020 |
12.0.3 | 162 | 7/10/2020 |
12.0.2 | 224 | 3/6/2020 |
12.0.1 | 109 | 3/6/2020 |
12.0.0 | 184 | 11/12/2019 |
11.0.0 | 174 | 10/3/2019 |
8.3.2 | 185 | 9/1/2019 |
8.3.1 | 195 | 6/23/2019 |
8.3.0 | 180 | 6/23/2019 |
8.2.0 | 173 | 6/18/2019 |
8.1.3 | 215 | 6/12/2019 |
8.1.2 | 203 | 6/11/2019 |
8.1.1 | 195 | 6/11/2019 |
8.1.0 | 194 | 6/11/2019 |
8.0.0 | 194 | 6/11/2019 |
7.1.1 | 200 | 6/10/2019 |
7.1.0 | 204 | 6/10/2019 |
7.0.0 | 199 | 6/10/2019 |
6.4.0 | 218 | 6/6/2019 |
6.3.0 | 228 | 6/6/2019 |
6.2.0 | 220 | 6/4/2019 |
6.1.0 | 217 | 6/4/2019 |
6.0.0 | 217 | 6/4/2019 |
5.12.0 | 225 | 6/3/2019 |
5.11.0 | 212 | 6/2/2019 |
5.10.2 | 207 | 5/31/2019 |
5.10.1 | 214 | 5/31/2019 |
5.9.0 | 214 | 5/28/2019 |
5.8.0 | 196 | 5/28/2019 |
5.7.0 | 196 | 5/28/2019 |
5.6.0 | 208 | 5/27/2019 |
5.5.0 | 198 | 5/27/2019 |
5.4.0 | 211 | 5/26/2019 |
5.3.0 | 201 | 5/25/2019 |
5.2.0 | 212 | 5/25/2019 |
5.1.0 | 207 | 5/25/2019 |
5.0.0 | 211 | 5/25/2019 |
4.5.0 | 210 | 5/24/2019 |
4.4.0 | 208 | 5/24/2019 |
4.3.0 | 202 | 5/24/2019 |
4.2.0 | 214 | 5/24/2019 |
4.1.0 | 198 | 5/24/2019 |
4.0.1 | 203 | 5/23/2019 |
4.0.0 | 198 | 5/23/2019 |
3.6.0 | 209 | 5/23/2019 |
3.5.0 | 202 | 5/23/2019 |
3.4.0 | 211 | 5/22/2019 |
3.3.0 | 216 | 5/22/2019 |
3.2.0 | 216 | 5/21/2019 |
3.1.0 | 231 | 5/21/2019 |
3.0.0 | 214 | 5/21/2019 |
2.17.0 | 206 | 5/21/2019 |
2.16.0 | 204 | 5/19/2019 |
2.15.0 | 216 | 5/19/2019 |
2.11.0 | 217 | 5/19/2019 |
2.10.1 | 212 | 5/19/2019 |
2.10.0 | 209 | 5/18/2019 |
2.9.0 | 214 | 5/18/2019 |
2.8.0 | 206 | 5/18/2019 |
2.7.0 | 200 | 5/18/2019 |
2.6.0 | 214 | 5/17/2019 |
2.5.0 | 219 | 5/17/2019 |
2.4.0 | 205 | 5/17/2019 |
2.3.0 | 210 | 5/17/2019 |
2.2.0 | 217 | 5/17/2019 |
2.1.0 | 210 | 5/17/2019 |
2.0.0 | 212 | 5/16/2019 |
1.1.0 | 216 | 5/16/2019 |
1.0.0 | 217 | 5/15/2019 |
0.1.36 | 222 | 5/10/2019 |
0.1.32 | 221 | 5/5/2019 |
0.1.29 | 225 | 5/3/2019 |
0.1.28 | 221 | 5/3/2019 |
0.1.26 | 218 | 5/3/2019 |
0.1.25 | 315 | 5/2/2019 |
0.1.20 | 227 | 4/29/2019 |
0.1.19 | 228 | 4/29/2019 |
0.1.18 | 215 | 4/29/2019 |
0.1.17 | 218 | 4/29/2019 |
0.1.16 | 209 | 4/28/2019 |
0.1.15 | 211 | 4/28/2019 |
0.1.14 | 222 | 4/28/2019 |
0.1.12 | 238 | 4/28/2019 |
0.1.9 | 242 | 4/28/2019 |
0.1.6 | 241 | 4/27/2019 |