2 // System.Web.Compilation.AspTokenizer
5 // Gonzalo Paniagua Javier (gonzalo@ximian.com)
6 // Marek Habersack <mhabersack@novell.com>
8 // (C) 2002,2003 Ximian, Inc (http://www.ximian.com)
9 // (C) 2003-2009 Novell, Inc (http://novell.com)
13 // Permission is hereby granted, free of charge, to any person obtaining
14 // a copy of this software and associated documentation files (the
15 // "Software"), to deal in the Software without restriction, including
16 // without limitation the rights to use, copy, modify, merge, publish,
17 // distribute, sublicense, and/or sell copies of the Software, and to
18 // permit persons to whom the Software is furnished to do so, subject to
19 // the following conditions:
21 // The above copyright notice and this permission notice shall be
22 // included in all copies or substantial portions of the Software.
24 // THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
25 // EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
26 // MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
27 // NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
28 // LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
29 // OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
30 // WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
34 using System.Collections;
37 using System.Security.Cryptography;
39 namespace System.Web.Compilation
43 public const int EOF = 0x0200000;
44 public const int IDENTIFIER = 0x0200001;
45 public const int DIRECTIVE = 0x0200002;
46 public const int ATTVALUE = 0x0200003;
47 public const int TEXT = 0x0200004;
48 public const int DOUBLEDASH = 0x0200005;
49 public const int CLOSING = 0x0200006;
54 const int CHECKSUM_BUF_SIZE = 8192;
58 public readonly string Value;
59 public readonly int Position;
60 public readonly int CurrentToken;
61 public readonly bool InTag;
63 public PutBackItem (string value, int position, int currentToken, bool inTag)
67 CurrentToken = currentToken;
72 static char [] lfcr = new char [] { '\n', '\r' };
75 StringBuilder sb, odds;
81 bool alternatingQuotes;
90 char[] checksum_buf = new char [CHECKSUM_BUF_SIZE];
91 int checksum_buf_pos = -1;
94 get { return checksum; }
97 public AspTokenizer (TextReader reader)
100 sb = new StringBuilder ();
101 odds= new StringBuilder();
103 hasPutBack = inTag = false;
108 get { return verbatim; }
109 set { verbatim = value; }
112 public void put_back ()
114 if (hasPutBack && !inTag)
115 throw new HttpException ("put_back called twice!");
118 if (putBackBuffer == null)
119 putBackBuffer = new Stack ();
122 putBackBuffer.Push (new PutBackItem (val, position, current_token, inTag));
123 position -= val.Length;
126 public int get_token ()
131 pbi = putBackBuffer.Pop () as PutBackItem;
132 string value = pbi.Value;
133 switch (value.Length) {
135 // do nothing, CurrentToken will be used
139 pbi = new PutBackItem (String.Empty, pbi.Position, (int)value [0], false);
143 pbi = new PutBackItem (value, pbi.Position, (int)value [0], false);
147 pbi = putBackBuffer.Pop () as PutBackItem;
149 hasPutBack = putBackBuffer.Count > 0;
150 position = pbi.Position;
153 sb = new StringBuilder (pbi.Value);
154 current_token = pbi.CurrentToken;
156 return current_token;
162 current_token = NextToken ();
163 return current_token;
166 bool is_identifier_start_character (char c)
168 return (Char.IsLetter (c) || c == '_' );
171 bool is_identifier_part_character (char c)
173 return (Char.IsLetterOrDigit (c) || c == '_' || c == '-');
176 void ungetc (int value)
181 // Only '/' passes through here now.
182 // If we ever let \n here, update 'line'
187 void TransformNextBlock (int count, bool final)
189 byte[] input = Encoding.UTF8.GetBytes (checksum_buf, 0, count);
191 if (checksum == null)
192 checksum = MD5.Create ();
195 checksum.TransformFinalBlock (input, 0, input.Length);
197 checksum.TransformBlock (input, 0, input.Length, input, 0);
200 checksum_buf_pos = -1;
203 void UpdateChecksum (int c)
205 bool final = c == -1;
208 if (checksum_buf_pos + 1 >= CHECKSUM_BUF_SIZE)
209 TransformNextBlock (checksum_buf_pos + 1, false);
210 checksum_buf [++checksum_buf_pos] = (char)c;
212 TransformNextBlock (checksum_buf_pos + 1, true);
226 if (c == '\r' && sr.Peek () == '\n') {
245 int ReadAttValue (int start)
250 if (start == '"' || start == '\'') {
254 sb.Append ((char) start);
259 bool inServerTag = false;
260 alternatingQuotes = true;
262 while ((c = sr.Peek ()) != -1) {
263 if (c == '%' && last == '<') {
265 } else if (inServerTag && c == '>' && last == '%') {
267 } else if (!inServerTag) {
268 if (!quoted && c == '/') {
273 } else if (c == '>') {
277 } else if (!quoted && (c == '>' || Char.IsWhiteSpace ((char) c))) {
279 } else if (quoted && c == quoteChar && last != '\\') {
283 } else if (quoted && c == quoteChar) {
284 alternatingQuotes = false;
287 sb.Append ((char) c);
292 return Token.ATTVALUE;
301 while ((c = read_char ()) != -1){
304 sb.Append ((char) c);
308 if (inTag && expectAttrValue && (c == '"' || c == '\''))
309 return ReadAttValue (c);
313 sb.Append ((char) c);
319 sb.Append ((char) c);
323 if (current_token == '<' && "%/!".IndexOf ((char) c) != -1){
324 sb.Append ((char) c);
328 if (inTag && current_token == '%' && "@#=".IndexOf ((char) c) != -1){
329 if (odds.Length == 0 || odds.ToString ().IndexOfAny (lfcr) < 0) {
330 sb.Append ((char) c);
333 sb.Append ((char) c);
337 if (inTag && c == '-' && sr.Peek () == '-'){
340 return Token.DOUBLEDASH;
344 sb.Append ((char) c);
345 while ((c = sr.Peek ()) != -1 && c != '<')
346 sb.Append ((char) read_char ());
348 return (c != -1 || sb.Length > 0) ? Token.TEXT : Token.EOF;
351 if (inTag && current_token == '=' && !Char.IsWhiteSpace ((char) c))
352 return ReadAttValue (c);
354 if (inTag && is_identifier_start_character ((char) c)){
355 sb.Append ((char) c);
356 while ((c = sr.Peek ()) != -1) {
357 if (!is_identifier_part_character ((char) c) && c != ':')
359 sb.Append ((char) read_char ());
362 if (current_token == '@' && Directive.IsDirective (sb.ToString ()))
363 return Token.DIRECTIVE;
365 return Token.IDENTIFIER;
368 if (!Char.IsWhiteSpace ((char) c)) {
369 sb.Append ((char) c);
372 // keep otherwise discarded characters in case we need.
373 odds.Append((char) c);
379 public string Value {
385 val = sb.ToString ();
392 return odds.ToString();
397 get { return inTag; }
398 set { inTag = value; }
401 // Hack for preventing confusion with VB comments (see bug #63451)
402 public bool ExpectAttrValue {
403 get { return expectAttrValue; }
404 set { expectAttrValue = value; }
407 public bool AlternatingQuotes {
408 get { return alternatingQuotes; }
411 public int BeginLine {
412 get { return begline; }
415 public int BeginColumn {
416 get { return begcol; }
423 public int EndColumn {
427 public int Position {
428 get { return position; }