Regular expressions in C # that use Python syntax.

Reprint Please specify source: Http://blog.csdn.net/zxsean

Because of the project requirements, it is now necessary to use Python's regular in C #, and the first approach is IronPython.

But compiling on iOS doesn't work. Fortunately IronPython is an open source project, so get the code to start research.

Because the functionality I need is simple, the final code just determines if the incoming string matches the Python regular expression I passed in.

Parse some of the code directly using:

/* **************************************************************************** * * Copyright (c) Microsoft  Corporation. * * This source code was subject to terms and conditions of the Microsoft public * License. A copy of the license can is found in the license.html file at the * root of this distribution. If You cannot locate the Microsoft public * License, please send a email to [email protected] by using this SOURC E * code in any fashion, you is agreeing to being bound by the terms of the * Microsoft public License. * * You must not remove the this notice, or any other, from this software. * * ***************************************************************************/using System;using System.Text; Using system.text.regularexpressions;///<summary>///python regular parsing//by zealotsean///</summary>public    Static class pythonregex{#region CONSTANTS//short forms//public Static Object I = 0x02;    public static Object L = 0x04; public static Object M = 0x08;    public static Object S = 0x10;    public static Object U = 0x20;    public static Object X = 0x40;    Long forms public static Object IGNORECASE = 0x02;    public static Object LOCALE = 0x04;    public static Object MULTILINE = 0x08;    public static Object dotall = 0x10;    public static Object UNICODE = 0x20;    public static Object VERBOSE = 0x40; #endregion//<summary>///compile a regular expression////</summary>//<param name= "_pattern" ></param&    Gt <returns></returns> public static Python_pattern Compile (string _pattern) {return new Python    _pattern (_pattern, 0, true);    public static bool Ismatched (string _pattern, String _str) {return ismatched (_pattern, _str, 0);  } public static bool Ismatched (string _pattern, string _str, int flags) {return new Python_pattern (_pattern,    Flags). ismatched (_STR); }//<summary>//Compiled Reg-ex pattern///</summary&Gt        public class Python_pattern {internal Regex m_re;        Internal Parsedregex m_pre; Public Python_pattern (String pattern): This (pattern, 0) {} public Python_pattern (string pattern, int flags): This (pattern, flags, false) {} public Python_pattern (string pattern            , int flags, BOOL compiled) {M_pre = Preparseregex (pattern);            RegexOptions opts = flagstooption (flags); This.m_re = new Regex (m_pre. Pattern, opts | (compiled?)        RegexOptions.Compiled:RegexOptions.None)); }///<summary>///Incoming string lookup matches regular expression///</summary>//<param name= "_str" >& lt;/param>//<returns></returns> public bool Ismatched (string _str) {RET Urn M_re. Match (_STR).        Success; The public string Pattern {get {return m_pre.     Userpattern;       }}} private static RegexOptions flagstooption (int flags) {RegexOptions opts = Regexoption        S.none;        if (Flags & (int) IGNORECASE)! = 0) opts |= regexoptions.ignorecase;        if (Flags & (int) MULTILINE)! = 0) opts |= regexoptions.multiline;        if ((Flags & (int) LOCALE) = = 0) opts &= (~regexoptions.cultureinvariant);        if (Flags & (int) dotall)! = 0) opts |= regexoptions.singleline;        if (Flags & (int) VERBOSE)! = 0) opts |= regexoptions.ignorepatternwhitespace;    return opts; } internal class Parsedregex {public Parsedregex (string pattern) {this.        Userpattern = pattern;        } public string Userpattern;        public string Pattern;    Public RegexOptions Options = regexoptions.cultureinvariant; }///<summary>//preparses A regular expression text returning a Parsedregex class//So can be used    For further regular expressions. /// Pass in a python regular expression that returns a C # available format//</summary> private static Parsedregex Preparseregex (string pattern) {        Parsedregex res = new Parsedregex (pattern);        String Newpattern;        int cur = 0, nameindex;        int curgroup = 0;        BOOL Containsnamedgroup = false;        for (;;) {Nameindex = pattern.            IndexOf ("(", cur);                if (Nameindex > 0 && pattern[nameindex-1] = = ' \ \ ') {int curindex = nameIndex-2;                int backslashcount = 1;                    while (curindex >= 0 && pattern[curindex] = = ' \ \ ') {backslashcount++;                curindex--;                }//Odd number of the back slashes, which is a optional//paren that we should ignore.                    if (Backslashcount & 0x01)! = 0) {cur++;                Continue }} IF (Nameindex = =-1) break; if (Nameindex = = pattern.            LENGTH-1) break;                    Switch (Pattern[++nameindex]) {case '? '://EXTENSION syntax if (Nameindex = = pattern.                    LENGTH-1) {return null;                            } switch (Pattern[++nameindex]) {case ' P ':                            Named regex,. NET doesn ' t expect the P so we ' ll remove it; Also, once we see a named group i.e.? P then we need to start artificially//naming all unnamed groups from and on---the IS to g  Et around the fact that//the CLR RegEx support orders all the unnamed groups before all the                            Named//groups, even if the named groups is before the unnamed ones in the pattern; The ArtificiaL naming preserves the order of the groups and thus the order of//The matches if (Nameindex + 1 < pattern. Length && Pattern[nameindex + 1] = = ' = ') {//Match what Ever was previously matched by the named Group//Remove the (? p= pattern = pattern.                                Remove (NameIndex-2, 4); Pattern = pattern.                                Insert (nameIndex-2, "\\\\k<");                                int tmpindex = Nameindex; while (Tmpindex < pattern.                                Length && pattern[tmpindex]! = ') ' tmpindex++; if (Tmpindex = = pattern.                                Length) {return null; } pattern = pattern. Substring (0, Tmpindex) + ">" + pattern.                            Substring (Tmpindex + 1);                                } else {containsnamedgroup = true; Pattern = pattern.                            Remove (Nameindex, 1);                        } break; Case ' I ': Res. Options |= regexoptions.ignorecase;                        Break Case ' L ': Res. Options &= ~ (regexoptions.cultureinvariant);                        Break Case ' m ': Res. Options |= Regexoptions.multiline;                        Break Case ' s ': Res. Options |= Regexoptions.singleline;                        Break                        Case ' u ': break; Case ' x ': Res. Options |= Regexoptions.ignorepatternwhitespace;                        Break Case ': ': break; non-capturing case ' = ': break; Look ahead assertion case ' < ': break;          Positive look behind assertion              Case '! ': break; Negative look ahead assertion case ' # ': break;                        Inline comment case ' ('://yes/no If group exists, we don ' t                            Default: {return null;                }} break;                    Default://Just another group curgroup++;                        if (containsnamedgroup) {//need to name this unnamed group Pattern = pattern.                    Insert (Nameindex, "? <named" + getrandomstring () + ">");            } break;        } cur = nameindex;        } cur = 0;        for (;;) {Nameindex = pattern.            IndexOf (' \ \ ', cur); if (Nameindex = =-1 | | nameindex = = pattern.            LENGTH-1) break; char Curchar = PATtern[++nameindex]; Switch (Curchar) {case ' x ': Case ' u ': Case ' a ': CA                Se ' B ': Case ' e ': Case ' F ': Case ' n ': Case ' R ':                Case ' t ': Case ' V ': Case ' C ': Case ' s ': Case ' W ':                Case ' W ': Case ' P ': Case ' P ': Case ' S ': Case ' d ':                    Case ' D ': Case ' Z '://Known escape sequences, leave escaped.                Break                    Case ' \ \ '://escaping a \ cur + = 2;                Break                    Default:System.Globalization.UnicodeCategory Charclass = char.getunicodecategory (Curchar); Switch (charclass) {//recognized word characterS, always unescape. Case System.Globalization.UnicodeCategory.ModifierLetter:case System.Globalization.UnicodeCategory .                        Lowercaseletter:case System.Globalization.UnicodeCategory.UppercaseLetter: Case System.Globalization.UnicodeCategory.TitlecaseLetter:case System.Globalization.UnicodeCategor Y.otherletter:case System.Globalization.UnicodeCategory.LetterNumber:case System.Globalization.UnicodeCategory.OtherNumber:case System.Globalization.UnicodeCategory.Connect Orpunctuation:pattern = pattern.                            Remove (nameIndex-1, 1);                        Break Case System.Globalization.UnicodeCategory.DecimalDigitNumber://actually don ' t want to Unesca PE ' \1 ', ' \2 ' etc. which is references to groups BreaK            } break;        } cur++; } res.        pattern = pattern;    return res;    } static Random R = new Random (DateTime.Now.Millisecond); private static string getrandomstring () {return R.next (INT32.MAXVALUE/2, Int32.MaxValue).    ToString (); }}

Copyright NOTICE: This article for Bo Master original article, without Bo Master permission not reproduced.

Regular expressions in C # that use Python syntax.