netwerk/streamconv/converters/mozTXTToHTMLConv.h

   1 /* -*- Mode: C; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
   2 /* This Source Code Form is subject to the terms of the Mozilla Public
   3  * License, v. 2.0. If a copy of the MPL was not distributed with this
   4  * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
   5
   6 /**
   7   Description: Currently only functions to enhance plain text with HTML tags.
   8   See mozITXTToHTMLConv. Stream conversion is defunct.
   9 */
  10
  11 #ifndef _mozTXTToHTMLConv_h__
  12 #define _mozTXTToHTMLConv_h__
  13
  14 #include "mozITXTToHTMLConv.h"
  15 #include "nsIThreadRetargetableStreamListener.h"
  16 #include "nsString.h"
  17 #include "nsCOMPtr.h"
  18
  19 class nsIIOService;
  20
  21 class mozTXTToHTMLConv : public mozITXTToHTMLConv {
  22   virtual ~mozTXTToHTMLConv() = default;
  23
  24   //////////////////////////////////////////////////////////
  25  public:
  26   //////////////////////////////////////////////////////////
  27
  28   mozTXTToHTMLConv() = default;
  29   NS_DECL_ISUPPORTS
  30
  31   NS_DECL_MOZITXTTOHTMLCONV
  32   NS_DECL_NSIREQUESTOBSERVER
  33   NS_DECL_NSISTREAMLISTENER
  34   NS_DECL_NSITHREADRETARGETABLESTREAMLISTENER
  35   NS_DECL_NSISTREAMCONVERTER
  36
  37   /**
  38     see mozITXTToHTMLConv::CiteLevelTXT
  39    */
  40   int32_t CiteLevelTXT(const char16_t* line, uint32_t& logLineStart);
  41
  42   //////////////////////////////////////////////////////////
  43  protected:
  44   //////////////////////////////////////////////////////////
  45   nsCOMPtr<nsIIOService>
  46       mIOService;  // for performance reasons, cache the netwerk service...
  47                    /**
  48                      Completes<ul>
  49                      <li>Case 1: mailto: "mozilla@bucksch.org" -> "mailto:mozilla@bucksch.org"
  50                      <li>Case 2: http:   "www.mozilla.org"     -> "http://www.mozilla.org"
  51                      <li>Case 3: ftp:    "ftp.mozilla.org"     -> "ftp://www.mozilla.org"
  52                      </ul>
  53                      It does no check, if the resulting URL is valid.
  54                      @param text (in): abbreviated URL
  55                      @param pos (in): position of "@" (case 1) or first "." (case 2 and 3)
  56                      @return Completed URL at success and empty string at failure
  57                     */
  58   void CompleteAbbreviatedURL(const char16_t* aInString, int32_t aInLength,
  59                               const uint32_t pos, nsString& aOutString);
  60
  61   //////////////////////////////////////////////////////////
  62  private:
  63   //////////////////////////////////////////////////////////
  64
  65   enum LIMTYPE {
  66     LT_IGNORE,     // limitation not checked
  67     LT_DELIMITER,  // not alphanumeric and not rep[0]. End of text is also ok.
  68     LT_ALPHA,      // alpha char
  69     LT_DIGIT
  70   };
  71
  72   /**
  73     @param text (in): the string to search through.<p>
  74            If before = IGNORE,<br>
  75              rep is compared starting at 1. char of text (text[0]),<br>
  76              else starting at 2. char of text (text[1]).
  77            Chars after "after"-delimiter are ignored.
  78     @param rep (in): the string to look for
  79     @param aRepLen (in): the number of bytes in the string to look for
  80     @param before (in): limitation before rep
  81     @param after (in): limitation after rep
  82     @return true, if rep is found and limitation spec is met or rep is empty
  83   */
  84   bool ItMatchesDelimited(const char16_t* aInString, int32_t aInLength,
  85                           const char16_t* rep, int32_t aRepLen, LIMTYPE before,
  86                           LIMTYPE after);
  87
  88   /**
  89     @param see ItMatchesDelimited
  90     @return Number of ItMatchesDelimited in text
  91   */
  92   uint32_t NumberOfMatches(const char16_t* aInString, int32_t aInStringLength,
  93                            const char16_t* rep, int32_t aRepLen, LIMTYPE before,
  94                            LIMTYPE after);
  95
  96   /**
  97     Currently only changes "<", ">" and "&". All others stay as they are.<p>
  98     "Char" in function name to avoid side effects with nsString(ch)
  99     constructors.
 100     @param ch (in)
 101     @param aStringToAppendto (out) - the string to append the escaped
 102                                      string to.
 103     @param inAttribute (in) - will escape quotes, too (which is
 104                               only needed for attribute values)
 105   */
 106   void EscapeChar(const char16_t ch, nsAString& aStringToAppendto,
 107                   bool inAttribute);
 108
 109   /**
 110     See EscapeChar. Escapes the string in place.
 111   */
 112   void EscapeStr(nsString& aInString, bool inAttribute);
 113
 114   /**
 115     Currently only reverts "<", ">" and "&". All others stay as they are.<p>
 116     @param aInString (in) HTML string
 117     @param aStartPos (in) start index into the buffer
 118     @param aLength (in) length of the buffer
 119     @param aOutString (out) unescaped buffer
 120   */
 121   void UnescapeStr(const char16_t* aInString, int32_t aStartPos,
 122                    int32_t aLength, nsString& aOutString);
 123
 124   /**
 125     <em>Note</em>: I use different strategies to pass context between the
 126     functions (full text and pos vs. cutted text and col0, glphyTextLen vs.
 127     replaceBefore/-After). It makes some sense, but is hard to understand
 128     (maintain) :-(.
 129   */
 130
 131   /**
 132     <p><em>Note:</em> replaceBefore + replaceAfter + 1 (for char at pos) chars
 133     in text should be replaced by outputHTML.</p>
 134     <p><em>Note:</em> This function should be able to process a URL on multiple
 135     lines, but currently, ScanForURLs is called for every line, so it can't.</p>
 136     @param text (in): includes possibly a URL
 137     @param pos (in): position in text, where either ":", "." or "@" are found
 138     @param whathasbeendone (in): What the calling ScanTXT did/has to do with the
 139                 (not-linkified) text, i.e. usually the "whattodo" parameter.
 140                 (Needed to calculate replaceBefore.) NOT what will be done with
 141                 the content of the link.
 142     @param outputHTML (out): URL with HTML-a tag
 143     @param replaceBefore (out): Number of chars of URL before pos
 144     @param replaceAfter (out): Number of chars of URL after pos
 145     @return URL found
 146   */
 147   bool FindURL(const char16_t* aInString, int32_t aInLength, const uint32_t pos,
 148                const uint32_t whathasbeendone, nsString& outputHTML,
 149                int32_t& replaceBefore, int32_t& replaceAfter);
 150
 151   enum modetype {
 152     unknown,
 153     RFC1738,    /* Check, if RFC1738, APPENDIX compliant,
 154                    like "<URL:http://www.mozilla.org>". */
 155     RFC2396E,   /* RFC2396, APPENDIX E allows anglebrackets (like
 156                    "<http://www.mozilla.org>") (without "URL:") or
 157                    quotation marks(like ""http://www.mozilla.org"").
 158                    Also allow email addresses without scheme,
 159                    e.g. "<mozilla@bucksch.org>" */
 160     freetext,   /* assume heading scheme
 161                    with "[a-zA-Z][a-zA-Z0-9+\-\.]*:" like "news:"
 162                    (see RFC2396, Section 3.1).
 163                    Certain characters (see code) or any whitespace
 164                    (including linebreaks) end the URL.
 165                    Other certain (punctation) characters (see code)
 166                    at the end are stripped off. */
 167     abbreviated /* Similar to freetext, but without scheme, e.g.
 168                    "www.mozilla.org", "ftp.mozilla.org" and
 169                    "mozilla@bucksch.org". */
 170                 /* RFC1738 and RFC2396E type URLs may use multiple lines,
 171                    whitespace is stripped. Special characters like ")" stay intact.*/
 172   };
 173
 174   /**
 175    * @param text (in), pos (in): see FindURL
 176    * @param check (in): Start must be conform with this mode
 177    * @param start (out): Position in text, where URL (including brackets or
 178    *             similar) starts
 179    * @return |check|-conform start has been found
 180    */
 181   bool FindURLStart(const char16_t* aInString, int32_t aInLength,
 182                     const uint32_t pos, const modetype check, uint32_t& start);
 183
 184   /**
 185    * @param text (in), pos (in): see FindURL
 186    * @param check (in): End must be conform with this mode
 187    * @param start (in): see FindURLStart
 188    * @param end (out): Similar to |start| param of FindURLStart
 189    * @return |check|-conform end has been found
 190    */
 191   bool FindURLEnd(const char16_t* aInString, int32_t aInStringLength,
 192                   const uint32_t pos, const modetype check,
 193                   const uint32_t start, uint32_t& end);
 194
 195   /**
 196    * @param text (in), pos (in), whathasbeendone (in): see FindURL
 197    * @param check (in): Current mode
 198    * @param start (in), end (in): see FindURLEnd
 199    * @param txtURL (out): Guessed (raw) URL.
 200    *             Without whitespace, but not completed.
 201    * @param desc (out): Link as shown to the user, but already escaped.
 202    *             Should be placed between the <a> and </a> tags.
 203    * @param replaceBefore(out), replaceAfter (out): see FindURL
 204    */
 205   void CalculateURLBoundaries(const char16_t* aInString,
 206                               int32_t aInStringLength, const uint32_t pos,
 207                               const uint32_t whathasbeendone,
 208                               const modetype check, const uint32_t start,
 209                               const uint32_t end, nsString& txtURL,
 210                               nsString& desc, int32_t& replaceBefore,
 211                               int32_t& replaceAfter);
 212
 213   /**
 214    * @param txtURL (in), desc (in): see CalculateURLBoundaries
 215    * @param outputHTML (out): see FindURL
 216    * @return A valid URL could be found (and creation of HTML successful)
 217    */
 218   bool CheckURLAndCreateHTML(const nsString& txtURL, const nsString& desc,
 219                              const modetype mode, nsString& outputHTML);
 220
 221   /**
 222     @param text (in): line of text possibly with tagTXT.<p>
 223                 if col0 is true,
 224                   starting with tagTXT<br>
 225                 else
 226                   starting one char before tagTXT
 227     @param col0 (in): tagTXT is on the beginning of the line (or paragraph).
 228                 open must be 0 then.
 229     @param tagTXT (in): Tag in plaintext to search for, e.g. "*"
 230     @param aTagTxtLen (in): length of tagTXT.
 231     @param tagHTML (in): HTML-Tag to replace tagTXT with,
 232                 without "<" and ">", e.g. "strong"
 233     @param attributeHTML (in): HTML-attribute to add to opening tagHTML,
 234                 e.g. "class=txt_star"
 235     @param aOutString: string to APPEND the converted html into
 236     @param open (in/out): Number of currently open tags of type tagHTML
 237     @return Conversion succeeded
 238   */
 239   bool StructPhraseHit(const char16_t* aInString, int32_t aInStringLength,
 240                        bool col0, const char16_t* tagTXT, int32_t aTagTxtLen,
 241                        const char* tagHTML, const char* attributeHTML,
 242                        nsAString& aOutString, uint32_t& openTags);
 243
 244   /**
 245     @param text (in), col0 (in): see GlyphHit
 246     @param tagTXT (in): Smily, see also StructPhraseHit
 247     @param imageName (in): the basename of the file that contains the image for
 248                            this smilie
 249     @param outputHTML (out): new string containing the html for the smily
 250     @param glyphTextLen (out): see GlyphHit
 251   */
 252   bool SmilyHit(const char16_t* aInString, int32_t aLength, bool col0,
 253                 const char* tagTXT, const nsString& imageName,
 254                 nsString& outputHTML, int32_t& glyphTextLen);
 255
 256   /**
 257     Checks, if we can replace some chars at the start of line with prettier HTML
 258     code.<p>
 259     If success is reported, replace the first glyphTextLen chars with outputHTML
 260
 261     @param text (in): line of text possibly with Glyph.<p>
 262                 If col0 is true,
 263                   starting with Glyph <br><!-- (br not part of text) -->
 264                 else
 265                   starting one char before Glyph
 266     @param col0 (in): text starts at the beginning of the line (or paragraph)
 267     @param aOutString (out): APPENDS html for the glyph to this string
 268     @param glyphTextLen (out): Length of original text to replace
 269     @return see StructPhraseHit
 270   */
 271   bool GlyphHit(const char16_t* aInString, int32_t aInLength, bool col0,
 272                 nsAString& aOutputString, int32_t& glyphTextLen);
 273
 274   /**
 275     Check if a given url should be linkified.
 276     @param aURL (in): url to be checked on.
 277   */
 278   bool ShouldLinkify(const nsCString& aURL);
 279 };
 280
 281 // It's said, that Win32 and Mac don't like static const members
 282 const int32_t mozTXTToHTMLConv_lastMode = 4;
 283 // Needed (only) by mozTXTToHTMLConv::FindURL
 284 const int32_t mozTXTToHTMLConv_numberOfModes = 4;  // dito; unknown not counted
 285
 286 #endif