1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/netwerk/streamconv/converters/mozTXTToHTMLConv.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,294 @@
1.4 +/* -*- Mode: C; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
1.5 +/* This Source Code Form is subject to the terms of the Mozilla Public
1.6 + * License, v. 2.0. If a copy of the MPL was not distributed with this
1.7 + * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
1.8 +
1.9 +/**
1.10 + Description: Currently only functions to enhance plain text with HTML tags. See mozITXTToHTMLConv. Stream conversion is defunct.
1.11 +*/
1.12 +
1.13 +#ifndef _mozTXTToHTMLConv_h__
1.14 +#define _mozTXTToHTMLConv_h__
1.15 +
1.16 +#include "mozITXTToHTMLConv.h"
1.17 +#include "nsString.h"
1.18 +#include "nsCOMPtr.h"
1.19 +
1.20 +class nsIIOService;
1.21 +
1.22 +class mozTXTToHTMLConv : public mozITXTToHTMLConv
1.23 +{
1.24 +
1.25 +
1.26 +//////////////////////////////////////////////////////////
1.27 +public:
1.28 +//////////////////////////////////////////////////////////
1.29 +
1.30 + mozTXTToHTMLConv();
1.31 + virtual ~mozTXTToHTMLConv();
1.32 + NS_DECL_ISUPPORTS
1.33 +
1.34 + NS_DECL_MOZITXTTOHTMLCONV
1.35 + NS_DECL_NSIREQUESTOBSERVER
1.36 + NS_DECL_NSISTREAMLISTENER
1.37 + NS_DECL_NSISTREAMCONVERTER
1.38 +
1.39 +/**
1.40 + see mozITXTToHTMLConv::ScanTXT
1.41 + */
1.42 + void ScanTXT(const char16_t * aInString, int32_t aInStringLength, uint32_t whattodo, nsString& aOutString);
1.43 +
1.44 +/**
1.45 + see mozITXTToHTMLConv::ScanHTML. We will modify aInString potentially...
1.46 + */
1.47 + void ScanHTML(nsString& aInString, uint32_t whattodo, nsString &aOutString);
1.48 +
1.49 +/**
1.50 + see mozITXTToHTMLConv::CiteLevelTXT
1.51 + */
1.52 + int32_t CiteLevelTXT(const char16_t * line,uint32_t& logLineStart);
1.53 +
1.54 +
1.55 +//////////////////////////////////////////////////////////
1.56 +protected:
1.57 +//////////////////////////////////////////////////////////
1.58 + nsCOMPtr<nsIIOService> mIOService; // for performance reasons, cache the netwerk service...
1.59 +/**
1.60 + Completes<ul>
1.61 + <li>Case 1: mailto: "mozilla@bucksch.org" -> "mailto:mozilla@bucksch.org"
1.62 + <li>Case 2: http: "www.mozilla.org" -> "http://www.mozilla.org"
1.63 + <li>Case 3: ftp: "ftp.mozilla.org" -> "ftp://www.mozilla.org"
1.64 + </ul>
1.65 + It does no check, if the resulting URL is valid.
1.66 + @param text (in): abbreviated URL
1.67 + @param pos (in): position of "@" (case 1) or first "." (case 2 and 3)
1.68 + @return Completed URL at success and empty string at failure
1.69 + */
1.70 + void CompleteAbbreviatedURL(const char16_t * aInString, int32_t aInLength,
1.71 + const uint32_t pos, nsString& aOutString);
1.72 +
1.73 +
1.74 +//////////////////////////////////////////////////////////
1.75 +private:
1.76 +//////////////////////////////////////////////////////////
1.77 +
1.78 + enum LIMTYPE
1.79 + {
1.80 + LT_IGNORE, // limitation not checked
1.81 + LT_DELIMITER, // not alphanumeric and not rep[0]. End of text is also ok.
1.82 + LT_ALPHA, // alpha char
1.83 + LT_DIGIT
1.84 + };
1.85 +
1.86 +/**
1.87 + @param text (in): the string to search through.<p>
1.88 + If before = IGNORE,<br>
1.89 + rep is compared starting at 1. char of text (text[0]),<br>
1.90 + else starting at 2. char of text (text[1]).
1.91 + Chars after "after"-delimiter are ignored.
1.92 + @param rep (in): the string to look for
1.93 + @param aRepLen (in): the number of bytes in the string to look for
1.94 + @param before (in): limitation before rep
1.95 + @param after (in): limitation after rep
1.96 + @return true, if rep is found and limitation spec is met or rep is empty
1.97 +*/
1.98 + bool ItMatchesDelimited(const char16_t * aInString, int32_t aInLength,
1.99 + const char16_t * rep, int32_t aRepLen, LIMTYPE before, LIMTYPE after);
1.100 +
1.101 +/**
1.102 + @param see ItMatchesDelimited
1.103 + @return Number of ItMatchesDelimited in text
1.104 +*/
1.105 + uint32_t NumberOfMatches(const char16_t * aInString, int32_t aInStringLength,
1.106 + const char16_t* rep, int32_t aRepLen, LIMTYPE before, LIMTYPE after);
1.107 +
1.108 +/**
1.109 + Currently only changes "<", ">" and "&". All others stay as they are.<p>
1.110 + "Char" in function name to avoid side effects with nsString(ch)
1.111 + constructors.
1.112 + @param ch (in)
1.113 + @param aStringToAppendto (out) - the string to append the escaped
1.114 + string to.
1.115 + @param inAttribute (in) - will escape quotes, too (which is
1.116 + only needed for attribute values)
1.117 +*/
1.118 + void EscapeChar(const char16_t ch, nsString& aStringToAppendto,
1.119 + bool inAttribute);
1.120 +
1.121 +/**
1.122 + See EscapeChar. Escapes the string in place.
1.123 +*/
1.124 + void EscapeStr(nsString& aInString, bool inAttribute);
1.125 +
1.126 +/**
1.127 + Currently only reverts "<", ">" and "&". All others stay as they are.<p>
1.128 + @param aInString (in) HTML string
1.129 + @param aStartPos (in) start index into the buffer
1.130 + @param aLength (in) length of the buffer
1.131 + @param aOutString (out) unescaped buffer
1.132 +*/
1.133 + void UnescapeStr(const char16_t * aInString, int32_t aStartPos,
1.134 + int32_t aLength, nsString& aOutString);
1.135 +
1.136 +/**
1.137 + <em>Note</em>: I use different strategies to pass context between the
1.138 + functions (full text and pos vs. cutted text and col0, glphyTextLen vs.
1.139 + replaceBefore/-After). It makes some sense, but is hard to understand
1.140 + (maintain) :-(.
1.141 +*/
1.142 +
1.143 +/**
1.144 + <p><em>Note:</em> replaceBefore + replaceAfter + 1 (for char at pos) chars
1.145 + in text should be replaced by outputHTML.</p>
1.146 + <p><em>Note:</em> This function should be able to process a URL on multiple
1.147 + lines, but currently, ScanForURLs is called for every line, so it can't.</p>
1.148 + @param text (in): includes possibly a URL
1.149 + @param pos (in): position in text, where either ":", "." or "@" are found
1.150 + @param whathasbeendone (in): What the calling ScanTXT did/has to do with the
1.151 + (not-linkified) text, i.e. usually the "whattodo" parameter.
1.152 + (Needed to calculate replaceBefore.) NOT what will be done with
1.153 + the content of the link.
1.154 + @param outputHTML (out): URL with HTML-a tag
1.155 + @param replaceBefore (out): Number of chars of URL before pos
1.156 + @param replaceAfter (out): Number of chars of URL after pos
1.157 + @return URL found
1.158 +*/
1.159 + bool FindURL(const char16_t * aInString, int32_t aInLength, const uint32_t pos,
1.160 + const uint32_t whathasbeendone,
1.161 + nsString& outputHTML, int32_t& replaceBefore, int32_t& replaceAfter);
1.162 +
1.163 + enum modetype {
1.164 + unknown,
1.165 + RFC1738, /* Check, if RFC1738, APPENDIX compliant,
1.166 + like "<URL:http://www.mozilla.org>". */
1.167 + RFC2396E, /* RFC2396, APPENDIX E allows anglebrackets (like
1.168 + "<http://www.mozilla.org>") (without "URL:") or
1.169 + quotation marks(like ""http://www.mozilla.org"").
1.170 + Also allow email addresses without scheme,
1.171 + e.g. "<mozilla@bucksch.org>" */
1.172 + freetext, /* assume heading scheme
1.173 + with "[a-zA-Z][a-zA-Z0-9+\-\.]*:" like "news:"
1.174 + (see RFC2396, Section 3.1).
1.175 + Certain characters (see code) or any whitespace
1.176 + (including linebreaks) end the URL.
1.177 + Other certain (punctation) characters (see code)
1.178 + at the end are stripped off. */
1.179 + abbreviated /* Similar to freetext, but without scheme, e.g.
1.180 + "www.mozilla.org", "ftp.mozilla.org" and
1.181 + "mozilla@bucksch.org". */
1.182 + /* RFC1738 and RFC2396E type URLs may use multiple lines,
1.183 + whitespace is stripped. Special characters like ")" stay intact.*/
1.184 + };
1.185 +
1.186 +/**
1.187 + * @param text (in), pos (in): see FindURL
1.188 + * @param check (in): Start must be conform with this mode
1.189 + * @param start (out): Position in text, where URL (including brackets or
1.190 + * similar) starts
1.191 + * @return |check|-conform start has been found
1.192 + */
1.193 + bool FindURLStart(const char16_t * aInString, int32_t aInLength, const uint32_t pos,
1.194 + const modetype check, uint32_t& start);
1.195 +
1.196 +/**
1.197 + * @param text (in), pos (in): see FindURL
1.198 + * @param check (in): End must be conform with this mode
1.199 + * @param start (in): see FindURLStart
1.200 + * @param end (out): Similar to |start| param of FindURLStart
1.201 + * @return |check|-conform end has been found
1.202 + */
1.203 + bool FindURLEnd(const char16_t * aInString, int32_t aInStringLength, const uint32_t pos,
1.204 + const modetype check, const uint32_t start, uint32_t& end);
1.205 +
1.206 +/**
1.207 + * @param text (in), pos (in), whathasbeendone (in): see FindURL
1.208 + * @param check (in): Current mode
1.209 + * @param start (in), end (in): see FindURLEnd
1.210 + * @param txtURL (out): Guessed (raw) URL.
1.211 + * Without whitespace, but not completed.
1.212 + * @param desc (out): Link as shown to the user, but already escaped.
1.213 + * Should be placed between the <a> and </a> tags.
1.214 + * @param replaceBefore(out), replaceAfter (out): see FindURL
1.215 + */
1.216 + void CalculateURLBoundaries(const char16_t * aInString, int32_t aInStringLength,
1.217 + const uint32_t pos, const uint32_t whathasbeendone,
1.218 + const modetype check, const uint32_t start, const uint32_t end,
1.219 + nsString& txtURL, nsString& desc,
1.220 + int32_t& replaceBefore, int32_t& replaceAfter);
1.221 +
1.222 +/**
1.223 + * @param txtURL (in), desc (in): see CalculateURLBoundaries
1.224 + * @param outputHTML (out): see FindURL
1.225 + * @return A valid URL could be found (and creation of HTML successful)
1.226 + */
1.227 + bool CheckURLAndCreateHTML(
1.228 + const nsString& txtURL, const nsString& desc, const modetype mode,
1.229 + nsString& outputHTML);
1.230 +
1.231 +/**
1.232 + @param text (in): line of text possibly with tagTXT.<p>
1.233 + if col0 is true,
1.234 + starting with tagTXT<br>
1.235 + else
1.236 + starting one char before tagTXT
1.237 + @param col0 (in): tagTXT is on the beginning of the line (or paragraph).
1.238 + open must be 0 then.
1.239 + @param tagTXT (in): Tag in plaintext to search for, e.g. "*"
1.240 + @param aTagTxtLen (in): length of tagTXT.
1.241 + @param tagHTML (in): HTML-Tag to replace tagTXT with,
1.242 + without "<" and ">", e.g. "strong"
1.243 + @param attributeHTML (in): HTML-attribute to add to opening tagHTML,
1.244 + e.g. "class=txt_star"
1.245 + @param aOutString: string to APPEND the converted html into
1.246 + @param open (in/out): Number of currently open tags of type tagHTML
1.247 + @return Conversion succeeded
1.248 +*/
1.249 + bool StructPhraseHit(const char16_t * aInString, int32_t aInStringLength, bool col0,
1.250 + const char16_t* tagTXT,
1.251 + int32_t aTagTxtLen,
1.252 + const char* tagHTML, const char* attributeHTML,
1.253 + nsString& aOutputString, uint32_t& openTags);
1.254 +
1.255 +/**
1.256 + @param text (in), col0 (in): see GlyphHit
1.257 + @param tagTXT (in): Smily, see also StructPhraseHit
1.258 + @param imageName (in): the basename of the file that contains the image for this smilie
1.259 + @param outputHTML (out): new string containing the html for the smily
1.260 + @param glyphTextLen (out): see GlyphHit
1.261 +*/
1.262 + bool
1.263 + SmilyHit(const char16_t * aInString, int32_t aLength, bool col0,
1.264 + const char* tagTXT, const char* imageName,
1.265 + nsString& outputHTML, int32_t& glyphTextLen);
1.266 +
1.267 +/**
1.268 + Checks, if we can replace some chars at the start of line with prettier HTML
1.269 + code.<p>
1.270 + If success is reported, replace the first glyphTextLen chars with outputHTML
1.271 +
1.272 + @param text (in): line of text possibly with Glyph.<p>
1.273 + If col0 is true,
1.274 + starting with Glyph <br><!-- (br not part of text) -->
1.275 + else
1.276 + starting one char before Glyph
1.277 + @param col0 (in): text starts at the beginning of the line (or paragraph)
1.278 + @param aOutString (out): APPENDS html for the glyph to this string
1.279 + @param glyphTextLen (out): Length of original text to replace
1.280 + @return see StructPhraseHit
1.281 +*/
1.282 + bool GlyphHit(const char16_t * aInString, int32_t aInLength, bool col0,
1.283 + nsString& aOutString, int32_t& glyphTextLen);
1.284 +
1.285 +/**
1.286 + Check if a given url should be linkified.
1.287 + @param aURL (in): url to be checked on.
1.288 +*/
1.289 + bool ShouldLinkify(const nsCString& aURL);
1.290 +};
1.291 +
1.292 +// It's said, that Win32 and Mac don't like static const members
1.293 +const int32_t mozTXTToHTMLConv_lastMode = 4;
1.294 + // Needed (only) by mozTXTToHTMLConv::FindURL
1.295 +const int32_t mozTXTToHTMLConv_numberOfModes = 4; // dito; unknown not counted
1.296 +
1.297 +#endif
