/// <summary> /// Handles META tags that set page encoding /// </summary> /// <param name="oP">HTML parser object that is used for parsing</param> /// <param name="oChunk">Parsed chunk that should contain tag META</param> /// <param name="bEncodingSet">Your own flag that shows whether encoding was already set or not, if set /// once then it should not be changed - this is the logic applied by major browsers</param> /// <returns>True if this was META tag setting Encoding, false otherwise</returns> public static bool HandleMetaEncoding(HtmlParser oP, HtmlChunk oChunk, ref bool bEncodingSet) { if (oChunk.sTag.Length != 4 || oChunk.sTag[0] != 'm' || oChunk.sTag != "meta") { return(false); } if (oChunk.oParams.ContainsKey("http-equiv")) { var sKey = oChunk.oParams["http-equiv"]; // FIXIT: even though this is happening rare I really don't like lower casing stuff // that most likely would not need to be - if you feel bored then rewrite this bit // to make it faster, it is really easy... switch (sKey.ToLower()) { case "content-type": // rare case (appears to work in IE) reported to exist in some pages by Martin Bächtold case "content-category": // we might have charset here that may hint at necessity to decode page // check for possible encoding change // once encoding is set it should not be changed, but you can be damn // sure there are web pages out there that do that!!! if (!bEncodingSet) { // it is possible we have broken META tag without Content part if (oChunk.oParams.ContainsKey("content")) { var sData = oChunk.oParams["content"]; if (oP.SetEncoding(sData)) { // we may need to re-encode title if (!bEncodingSet) { // here you need to reencode any text that you found so far // most likely it will be just TITLE, the rest can be ignored anyway bEncodingSet = true; } } else { // failed to set encoding - most likely encoding string // was incorrect or your machine lacks codepages or something // else - might be good idea to put warning message here } } } return(true); default: break; } } return(false); }
public bool Equals(HtmlChunk other) { return(this.GenerateHtml() == other.GenerateHtml()); }
/// <summary> /// Sets oHTML variable in a chunk to the raw HTML that was parsed for that chunk. /// </summary> /// <param name="oChunk">Chunk returned by ParseNext function, it must belong to the same HtmlParser that /// was initiated with the same HTML data that this chunk belongs to</param> public void SetRawHTML(HtmlChunk oChunk) { // note: this really should have been byte array assigned rather than string // it would be more correct originality-wise oChunk.oHTML = this.oEnc.GetString(this.bHTML, oChunk.iChunkOffset, oChunk.iChunkLength); }