public void ProcessWordlist(ref CardsStream stream, List<string> wordlist, string labels, bool crossreferenceFlag)
{
//Разбираем исходные страницы. Возможно, удастся ускорить работу засчёт распараллеливания.
//Собираем ссылки в теле статьи и добавляем в список.
//downloadPages(ref stream, GetArticleCrossrefenceLinkList(), labels);
//Работаем с ссылками, находящиеся в блоке Search Results, в т.ч. и на само word)
var crossreferenceLinks = new Hashtable();
foreach (String word in wordlist)
{
var page = (new HtmlWeb()).Load(searchPath + word);
ParsePage(ref stream, page, word, labels);
Console.WriteLine("{0}", word);
reportStream.Write("Success. Page was parsed. Link: " + word + "\n");
//Берём все слова из блока search results
var linksNodes = page.DocumentNode.SelectNodes("//div[@class='entrylist']/ul/li/a");
if (linksNodes != null)
{
foreach (HtmlNode link in linksNodes)
{
var url = getCleanUrl(link);
if (crossreferenceLinks.ContainsKey(url) == false)
{
crossreferenceLinks.Add(url, false);
}
}
}
var updatedWordList = new Hashtable();
foreach (string link in crossreferenceLinks.Keys)
{
if ((bool)crossreferenceLinks[link] == true)
{
continue;
}
var tranformedWordRegExp = new Regex("^(" + word.Replace(' ', '-') + "_\\d+)$", RegexOptions.IgnoreCase);
if (crossreferenceFlag == true || tranformedWordRegExp.Match(link).Success)
{
ParsePage(ref stream, (new HtmlWeb()).Load(searchPath + link), link, labels);
Console.WriteLine("{0}", link);
updatedWordList.Add(link, true);
reportStream.Write("Success. Page was parsed. Link: " + link + "\n");
}
}
foreach (DictionaryEntry update in updatedWordList)
{
crossreferenceLinks[update.Key] = update.Value;
}
}
reportStream.Write("Total: " + count_ + "\n");
reportStream.Save();
}
void ParsePage(ref CardsStream stream, HtmlDocument document, string link, string userLabels)
{
var examples = document.DocumentNode.SelectNodes("//div[@class='EXAMPLES']");
if (examples == null)
{
reportStream.Write("Failure. Examples was not found. Link: " + link + "\n");
return;
}
var gbrTranscription = document.DocumentNode.SelectSingleNode("//span[@class='PRON']");
var word = getText(document.DocumentNode.SelectSingleNode("//span[@class='BASE']"));
var wordLabel = word.Replace(' ', '-');
userLabels = (userLabels == "") ? "macmillan " + wordLabel : "macmillan " + wordLabel + " " + userLabels;
foreach (HtmlNode example in examples)
{
var card = new Card();
var parentNode = example.ParentNode;
//Getting a structure
var structure = example.SelectSingleNode("strong");
if (structure == null)
{
//DIV#SENSE_BODY -> DIV -> LI
structure = parentNode.ParentNode.ParentNode.SelectSingleNode("div/h2/span[@class='BASE']");
}
card.Sentence = getText(example.SelectSingleNode("p[@id='EXAMPLE']"));
card.Definition = getText(parentNode.SelectSingleNode("span[@class='DEFINITION']"));
card.GbrTranscription = getText(gbrTranscription).Replace("/", "");
card.SimpleStructure = getText(structure).Replace(":", "");
if (card.Definition == "")
{
reportStream.Write("Failure. Definition was not found. Link: " + link + " Example: '" + card.Sentence + "'\n");
}
if (card.Sentence == "")
{
reportStream.Write("Failure. Example was not found. Link: " + link + "\n");
}
var outStr = card.Sentence + "\t" + card.Interpretation +
"\t" + word + "\t" + card.GbrTranscription +
"\t" + card.UsaTranscription + "\t" +
card.SimpleStructure + "\t" + card.Definition +
"\t" + userLabels + "\n";
stream.Write(outStr);
count_++;
}
}
void ParsePage(ref CardsStream stream, HtmlDocument document, string link, string userLabels)
{
var examples = document.DocumentNode.SelectNodes("//div[@class='EXAMPLES']");
if (examples == null)
{
reportStream.Write("Failure. Examples was not found. Link: " + link + "\n");
return;
}
var gbrTranscription = document.DocumentNode.SelectSingleNode("//span[@class='PRON']");
var word = getText(document.DocumentNode.SelectSingleNode("//span[@class='BASE']"));
var wordLabel = word.Replace(' ', '-');
userLabels = (userLabels == "") ? "macmillan " + wordLabel : "macmillan " + wordLabel + " " + userLabels;
foreach (HtmlNode example in examples)
{
var card = new Card();
var parentNode = example.ParentNode;
//Getting a structure
var structure = example.SelectSingleNode("strong");
if (structure == null)
{
//DIV#SENSE_BODY -> DIV -> LI
structure = parentNode.ParentNode.ParentNode.SelectSingleNode("div/h2/span[@class='BASE']");
}
card.Sentence = getText(example.SelectSingleNode("p[@id='EXAMPLE']"));
card.Definition = getText(parentNode.SelectSingleNode("span[@class='DEFINITION']"));
card.GbrTranscription = getText(gbrTranscription).Replace("/", "");
card.SimpleStructure = getText(structure).Replace(":", "");
if (card.Definition == "")
{
reportStream.Write("Failure. Definition was not found. Link: " + link + " Example: '" + card.Sentence + "'\n");
}
if (card.Sentence == "")
{
reportStream.Write("Failure. Example was not found. Link: " + link + "\n");
}
var outStr = card.Sentence + "\t" + card.Interpretation +
"\t" + word + "\t" + card.GbrTranscription +
"\t" + card.UsaTranscription + "\t" +
card.SimpleStructure + "\t" + card.Definition +
"\t" + userLabels + "\n";
stream.Write(outStr);
count_++;
}
}
public void ProcessWordlist(ref CardsStream stream, List<string> wordlist, string labels, int limit)
{
labels = (labels == "") ? "lingvo " + labels : "lingvo " + labels + " ";
foreach (var word in wordlist)
{
var offset = 0;
reportStream.Write("Success. Page was parsed. Word: " + word + "\n");
Console.WriteLine("{0}", word);
while (offset < limit)
{
var page = new HtmlDocument();
page.LoadHtml(DownloadExamples(word, offset));
var examples = page.DocumentNode.SelectNodes("//tr[@class='item first']");
if (examples == null)
{
reportStream.Write("Failure. Examples was not found. Word: " + word + "\n");
break;
}
stream.Write(Parse(examples, labels + word));
count_ += examples.Count;
examples = page.DocumentNode.SelectNodes("//tr[@class='item']");
if (examples == null)
{
break;
}
stream.Write(Parse(examples, labels + word));
count_ += examples.Count;
offset += 5;
}
}
reportStream.Write("Total: " + count_ + "\n");
reportStream.Save();
return;
}
public void ProcessWordlist(ref CardsStream stream, List <string> wordlist, string labels, int limit)
{
labels = (labels == "") ? "lingvo " + labels : "lingvo " + labels + " ";
foreach (var word in wordlist)
{
var offset = 0;
reportStream.Write("Success. Page was parsed. Word: " + word + "\n");
Console.WriteLine("{0}", word);
while (offset < limit)
{
var page = new HtmlDocument();
page.LoadHtml(DownloadExamples(word, offset));
var examples = page.DocumentNode.SelectNodes("//tr[@class='item first']");
if (examples == null)
{
reportStream.Write("Failure. Examples was not found. Word: " + word + "\n");
break;
}
stream.Write(Parse(examples, labels + word));
count_ += examples.Count;
examples = page.DocumentNode.SelectNodes("//tr[@class='item']");
if (examples == null)
{
break;
}
stream.Write(Parse(examples, labels + word));
count_ += examples.Count;
offset += 5;
}
}
reportStream.Write("Total: " + count_ + "\n");
reportStream.Save();
return;
}
void ParsePage(ref CardsStream stream, HtmlDocument document, string link, string userLabels)
{
var examples = document.DocumentNode.SelectNodes("//span[@class='x-g']");
if (examples == null)
{
reportStream.Write("Failure. Examples was not found. Link: " + link + "\n");
return;
}
//Transcription
var usaTranscription = document.DocumentNode.SelectSingleNode("//span[@class='y']");
var gbrTranscription = document.DocumentNode.SelectSingleNode("//span[@class='i']");
//Word "name"
var word = getText(document.DocumentNode.SelectSingleNode("//h2[@class='h']"));
//Label
var wordLabel = word.Replace(' ', '-');
userLabels = (userLabels == "") ? "oald8 " + wordLabel : "oald8 " + wordLabel + " " + userLabels;
foreach (HtmlNode example in examples)
{
var card = new Card();
var parentNode = example.ParentNode;
//Getting a structure
var structure1 = example.SelectSingleNode("span[@class='cf']");
if (GetName(parentNode) == "pv-g")
{
var structure2 = parentNode.SelectSingleNode("h4[@class='pv']");
if (structure2 != null)
{
card.Structure.Add(structure2.InnerText);
}
if (structure1 != null)
{
card.Structure.Add(structure1.InnerText);
}
}
else if (GetName(parentNode) == "n-g" && GetName(parentNode.ParentNode) == "pv-g")
{
var structure2 = parentNode.SelectSingleNode("span[@class='vs-g']");
var structure3 = parentNode.ParentNode.SelectSingleNode("h4[@class='pv']");
if (structure3 != null)
{
card.Structure.Add(structure3.InnerText);
}
if (structure2 != null)
{
card.Structure.Add(structure2.InnerText);
}
if (structure1 != null)
{
card.Structure.Add(structure1.InnerText);
}
}
else if (GetName(parentNode) == "n-g")
{
var structure2 = parentNode.SelectSingleNode("span[@class='cf']");
if (structure2 != null)
{
card.Structure.Add(structure2.InnerText);
}
if (structure1 != null)
{
card.Structure.Add(structure1.InnerText);
}
}
else if (GetName(parentNode) == "id-g")
{
var structure2 = parentNode.SelectSingleNode("h4[@class='id']");
if (structure2 != null)
{
card.Structure.Add(structure2.InnerText);
}
if (structure1 != null)
{
card.Structure.Add(structure1.InnerText);
}
}
else if (GetName(parentNode) == "h-g")
{
var structure2 = parentNode.SelectSingleNode("span[@class='cf']");
if (structure2 != null)
{
card.Structure.Add(structure2.InnerText);
}
if (structure1 != null)
{
card.Structure.Add(structure1.InnerText);
}
}
//Getting a definition
HtmlNode definition = null;
if (GetName(parentNode) == "id-g" || GetName(parentNode) == "h-g")
{
definition = parentNode.SelectSingleNode("div[@class='def_block']");
}
if (definition == null)
{
definition = parentNode;
}
var temp = definition.SelectSingleNode("span[@class='ud']");
if (temp == null)
{
definition = definition.SelectSingleNode("span[@class='d']");
}
else
{
definition = temp;
}
//An example itself
card.Interpretation = getText(example.SelectSingleNode("span[@class='x']"));
card.Sentence = (new Regex(" \\(=.*?\\)")).Replace(card.Interpretation, "");
card.Definition = getText(definition);
card.Definition = card.Definition.Replace(" ", " ");
card.UsaTranscription = getText(usaTranscription);
card.GbrTranscription = getText(gbrTranscription);
if (card.Interpretation == card.Sentence)
{
card.Interpretation = "";
}
if (card.Definition == "")
{
reportStream.Write("Failure. Definition was not found. Link: " + link + " Example: '" + card.Sentence + "'\n");
}
if (card.Sentence == "")
{
reportStream.Write("Failure. Example was not found. Link: " + link + "\n");
}
var outStr = card.Sentence + "\t" + card.Interpretation +
"\t" + word + "\t" + card.GbrTranscription +
"\t" + card.UsaTranscription + "\t" +
PrintList(card.Structure) + "\t" + card.Definition +
"\t" + userLabels + "\n";
stream.Write(outStr);
count_++;
}
}
public void ProcessWordlist(ref CardsStream stream, List<string> wordlist, string userLabels, string domain, int limit)
{
var step_ = (limit < 48) ? limit : 48;
userLabels = "vocabulary_com " + userLabels;
foreach (String word in wordlist)
{
var offset = 0;
var step = step_;
var wordLabel = word.Replace(' ', '-');
var primaryDefinitions = String.Empty;
var fullDefinitions = String.Empty;
var page = (new HtmlWeb()).Load(definitionsPath + word).DocumentNode;
Console.WriteLine("{0}", word);
//Primary
var definitionsNodes = page.SelectNodes("//div[@class='def']");
if (definitionsNodes != null)
{
foreach (var p in definitionsNodes)
{
var href = "#s" + (new Regex(@"quickDef(\d+)")).Replace(p.Attributes["id"].Value, "$1");
primaryDefinitions += "<i><b>" + getText(page.SelectSingleNode("//a[@href='" + href +"']")) + "</i></b> " + p.InnerText + "<br/>";
}
}
definitionsNodes = page.SelectNodes("//div[@class='def selected']");
if (definitionsNodes != null)
{
foreach (var p in definitionsNodes)
{
var href = "#s" + (new Regex(@"quickDef(\d+)")).Replace(p.Attributes["id"].Value, "$1");
primaryDefinitions += "<i><b>" + getText(page.SelectSingleNode("//a[@href='" + href + "']")) + "</i></b> " + p.InnerText + "<br/>";
}
}
if (primaryDefinitions.Length == 0)
{
definitionsNodes = page.SelectNodes("//h3[@class='definition']");
if (definitionsNodes != null)
{
foreach (var f in definitionsNodes)
{
var partOfSpeech = f.SelectSingleNode("a");
var currentDefinition = new Regex("[\t\r\n]").Replace(getText(f), "");
currentDefinition = (new Regex(@"^\w+\s+(.*?)$")).Replace(currentDefinition, "$1");
primaryDefinitions += "<i><b>" + getText(partOfSpeech) + "</i></b> " + currentDefinition + "<br/>";
}
primaryDefinitions = primaryDefinitions.Substring(0, primaryDefinitions.Length - 5);
}
}
do
{
if (limit - offset < step)
{
step = limit - offset;
}
if (step <= 0)
{
break;
}
try
{
var json_ = System.Text.UTF8Encoding.ASCII.GetString(client.DownloadData(examplesPath + word + "&maxResults=" + step + "&startOffset=" + offset + "&filter=0&domain=" + domain));
json_ = (new Regex(@"\$d\((.*?)\)")).Replace(json_, @"""$1""");
var json = JObject.Parse(json_);
if (offset == 0)
{
var hits = (int)json.SelectToken("result.totalHits");
reportStream.Write("Success. Word: " + word + " Hits: " + hits + "\n");
Console.WriteLine("Hits: {0}", hits);
if (hits == 0)
{
reportStream.Write("Failure. Examples was not found. Word: " + word + "\n");
}
}
Console.WriteLine("Processed: {0}", offset);
offset += step;
foreach (var _ in json.SelectToken("result.sentences"))
{
count_++;
var example = new StringBuilder()
.Append(SafeTrim((string)_.SelectToken("sentence")))
.Append("%%!!%%")
.Append(SafeTrim((string)_.SelectToken("sentence")))
.Append("%%!!%%")
.Append(word)
.Append("%%!!%%")
.Append(primaryDefinitions)
.Append("%%!!%%")
//.Append(fullDefinitions)
//.Append("%%!!%%")
.Append(wordLabel)
.Append(" ")
.Append(userLabels);
stream.Write((new Regex("[\t\n\r]").Replace(example.ToString(), "")).Replace("%%!!%%", "\t") + "\n");
}
}
catch
{
continue;
}
}
while (step > 0);
Console.WriteLine("Processed: {0}\n", offset);
}
reportStream.Write("Total: " + count_ + "\n");
reportStream.Save();
}
public void ProcessWordlist(ref CardsStream stream, List <string> wordlist, string labels, bool crossreferenceFlag)
{
//Разбираем исходные страницы. Возможно, удастся ускорить работу засчёт распараллеливания.
//Собираем ссылки в теле статьи и добавляем в список.
//downloadPages(ref stream, GetArticleCrossrefenceLinkList(), labels);
//Работаем с ссылками, находящиеся в блоке Search Results, в т.ч. и на само word)
var crossreferenceLinks = new Hashtable();
foreach (String word in wordlist)
{
var page = (new HtmlWeb()).Load(searchPath + word);
ParsePage(ref stream, page, word, labels);
Console.WriteLine("{0}", word);
reportStream.Write("Success. Page was parsed. Link: " + word + "\n");
//Берём все слова из блока search results
var linksNodes = page.DocumentNode.SelectNodes("//div[@class='entrylist']/ul/li/a");
if (linksNodes != null)
{
foreach (HtmlNode link in linksNodes)
{
var url = getCleanUrl(link);
if (crossreferenceLinks.ContainsKey(url) == false)
{
crossreferenceLinks.Add(url, false);
}
}
}
var updatedWordList = new Hashtable();
foreach (string link in crossreferenceLinks.Keys)
{
if ((bool)crossreferenceLinks[link] == true)
{
continue;
}
var tranformedWordRegExp = new Regex("^(" + word.Replace(' ', '-') + "_\\d+)$", RegexOptions.IgnoreCase);
if (crossreferenceFlag == true || tranformedWordRegExp.Match(link).Success)
{
ParsePage(ref stream, (new HtmlWeb()).Load(searchPath + link), link, labels);
Console.WriteLine("{0}", link);
updatedWordList.Add(link, true);
reportStream.Write("Success. Page was parsed. Link: " + link + "\n");
}
}
foreach (DictionaryEntry update in updatedWordList)
{
crossreferenceLinks[update.Key] = update.Value;
}
}
reportStream.Write("Total: " + count_ + "\n");
reportStream.Save();
}
public static void Main(string[] args)
{
ParserMask useParser = ParserMask.None;
var inputPath = "";
var outputPath = String.Empty;
var wordlist = new List <String>();
var labels = "";
var relatedFlag = false;
var examplesLimit = Int32.MaxValue;
var vocabularyComDomain = String.Empty;
CardsStream output;
StreamReader input;
for (int i = 0; i < args.Length; i++)
{
if (args[i] == "-oald8")
{
outputPath = "oald8";
useParser |= ParserMask.Oald8;
}
else if (args[i] == "-macmillan")
{
outputPath = "macmillan";
useParser |= ParserMask.Macmillan;
}
else if (args[i] == "-vocabcom")
{
outputPath = "vocabulary.com";
useParser |= ParserMask.VocabularyCom;
}
else if (args[i] == "-lingvo")
{
outputPath = "lingvo";
useParser |= ParserMask.LingvoRu;
}
else if (args[i] == "-l" && i + 1 < args.Length)
{
labels = (args[i + 1]).Trim();
i++;
}
else if (args[i] == "-limit" && i + 1 < args.Length)
{
examplesLimit = Convert.ToInt32((args[i + 1]).Trim());
i++;
}
else if (args[i] == "-vdomain" && i + 1 < args.Length)
{
vocabularyComDomain = (args[i + 1]).Trim();
i++;
}
else if (args[i] == "-p" && i + 1 < args.Length)
{
inputPath = (args[i + 1]).Trim();
i++;
}
else if (args[i] == "-related")
{
relatedFlag = true;
}
}
if (inputPath == "")
{
Console.Write("Where's a listname, uh? You should enter -p <path_to_you_wordlist>");
Console.ReadKey();
return;
}
else
{
try
{
input = new StreamReader(inputPath);
while (input.EndOfStream == false)
{
//
var word = (new Regex("[^- 0-9a-zA-Z']+")).Replace(input.ReadLine(), "").Trim();
//
if (wordlist.Contains(word) == false)
{
wordlist.Add(word);
}
}
input.Close();
}
catch (FileNotFoundException e)
{
Console.Write("Wrong path: {0}", inputPath);
}
catch (DirectoryNotFoundException e)
{
Console.Write("Wrong directory: {0}", inputPath);
}
}
if (labels.Length != 0)
{
outputPath = "./" + outputPath + " " + labels;
}
else
{
outputPath = "./" + outputPath;
}
outputPath += " " + DateTime.Now.ToString("yyyy.MM.dd HH-mm-ss") + ".txt";
output = new CardsStream(outputPath, 100000);
if ((useParser & ParserMask.Oald8) != ParserMask.None)
{
var parser = new Oald8();
parser.ProcessWordlist(ref output, wordlist, labels, relatedFlag);
Console.WriteLine("\nCount: {0}\n", parser.count);
}
else if ((useParser & ParserMask.Macmillan) != ParserMask.None)
{
var parser = new Macmillan();
parser.ProcessWordlist(ref output, wordlist, labels, relatedFlag);
Console.WriteLine("\nCount: {0}\n", parser.count);
}
else if ((useParser & ParserMask.VocabularyCom) != ParserMask.None)
{
var parser = new VocabularyCom();
parser.ProcessWordlist(ref output, wordlist, labels, vocabularyComDomain, examplesLimit);
Console.WriteLine("\nCount: {0}\n", parser.count);
}
else if ((useParser & ParserMask.LingvoRu) != ParserMask.None)
{
var parser = new LingvoRu();
parser.ProcessWordlist(ref output, wordlist, labels, examplesLimit);
Console.WriteLine("\nCount: {0}\n", parser.count);
}
if (useParser != ParserMask.None)
{
output.Save();
}
}
public static void Main(string[] args)
{
ParserMask useParser = ParserMask.None;
var inputPath = "";
var outputPath = String.Empty;
var wordlist = new List<String>();
var labels = "";
var relatedFlag = false;
var examplesLimit = Int32.MaxValue;
var vocabularyComDomain = String.Empty;
CardsStream output;
StreamReader input;
for (int i = 0; i < args.Length; i++)
{
if (args[i] == "-oald8")
{
outputPath = "oald8";
useParser |= ParserMask.Oald8;
}
else if (args[i] == "-macmillan")
{
outputPath = "macmillan";
useParser |= ParserMask.Macmillan;
}
else if (args[i] == "-vocabcom")
{
outputPath = "vocabulary.com";
useParser |= ParserMask.VocabularyCom;
}
else if (args[i] == "-lingvo")
{
outputPath = "lingvo";
useParser |= ParserMask.LingvoRu;
}
else if (args[i] == "-l" && i + 1 < args.Length)
{
labels = (args[i + 1]).Trim();
i++;
}
else if (args[i] == "-limit" && i + 1 < args.Length)
{
examplesLimit = Convert.ToInt32((args[i + 1]).Trim());
i++;
}
else if (args[i] == "-vdomain" && i + 1 < args.Length)
{
vocabularyComDomain = (args[i + 1]).Trim();
i++;
}
else if (args[i] == "-p" && i + 1 < args.Length)
{
inputPath = (args[i + 1]).Trim();
i++;
}
else if (args[i] == "-related")
{
relatedFlag = true;
}
}
if (inputPath == "")
{
Console.Write("Where's a listname, uh? You should enter -p <path_to_you_wordlist>");
Console.ReadKey();
return;
}
else
{
try
{
input = new StreamReader(inputPath);
while (input.EndOfStream == false)
{
var word = (new Regex("[^- 0-9a-zA-Z']+")).Replace(input.ReadLine(), "").Trim();
if (wordlist.Contains(word) == false)
{
wordlist.Add(word);
}
}
input.Close();
}
catch (FileNotFoundException e)
{
Console.Write("Wrong path: {0}", inputPath);
}
catch (DirectoryNotFoundException e)
{
Console.Write("Wrong directory: {0}", inputPath);
}
}
if (labels.Length != 0)
{
outputPath = "./" + outputPath + " " + labels;
}
else
{
outputPath = "./" + outputPath;
}
outputPath += " " + DateTime.Now.ToString("yyyy.MM.dd HH-mm-ss") + ".txt";
output = new CardsStream(outputPath, 100000);
if ((useParser & ParserMask.Oald8) != ParserMask.None)
{
var parser = new Oald8();
parser.ProcessWordlist(ref output, wordlist, labels, relatedFlag);
Console.WriteLine("\nCount: {0}\n", parser.count);
}
else if ((useParser & ParserMask.Macmillan) != ParserMask.None)
{
var parser = new Macmillan();
parser.ProcessWordlist(ref output, wordlist, labels, relatedFlag);
Console.WriteLine("\nCount: {0}\n", parser.count);
}
else if ((useParser & ParserMask.VocabularyCom) != ParserMask.None)
{
var parser = new VocabularyCom();
parser.ProcessWordlist(ref output, wordlist, labels, vocabularyComDomain, examplesLimit);
Console.WriteLine("\nCount: {0}\n", parser.count);
}
else if ((useParser & ParserMask.LingvoRu) != ParserMask.None)
{
var parser = new LingvoRu();
parser.ProcessWordlist(ref output, wordlist, labels, examplesLimit);
Console.WriteLine("\nCount: {0}\n", parser.count);
}
if (useParser != ParserMask.None)
{
output.Save();
}
}
public void ProcessWordlist(ref CardsStream stream, List <string> wordlist, string userLabels, string domain, int limit)
{
var step_ = (limit < 48) ? limit : 48;
userLabels = "vocabulary_com " + userLabels;
foreach (String word in wordlist)
{
var offset = 0;
var step = step_;
var wordLabel = word.Replace(' ', '-');
var primaryDefinitions = String.Empty;
var fullDefinitions = String.Empty;
var page = (new HtmlWeb()).Load(definitionsPath + word).DocumentNode;
Console.WriteLine("{0}", word);
//Primary
var definitionsNodes = page.SelectNodes("//div[@class='def']");
if (definitionsNodes != null)
{
foreach (var p in definitionsNodes)
{
var href = "#s" + (new Regex(@"quickDef(\d+)")).Replace(p.Attributes["id"].Value, "$1");
primaryDefinitions += "<i><b>" + getText(page.SelectSingleNode("//a[@href='" + href + "']")) + "</i></b> " + p.InnerText + "<br/>";
}
}
definitionsNodes = page.SelectNodes("//div[@class='def selected']");
if (definitionsNodes != null)
{
foreach (var p in definitionsNodes)
{
var href = "#s" + (new Regex(@"quickDef(\d+)")).Replace(p.Attributes["id"].Value, "$1");
primaryDefinitions += "<i><b>" + getText(page.SelectSingleNode("//a[@href='" + href + "']")) + "</i></b> " + p.InnerText + "<br/>";
}
}
if (primaryDefinitions.Length == 0)
{
definitionsNodes = page.SelectNodes("//h3[@class='definition']");
if (definitionsNodes != null)
{
foreach (var f in definitionsNodes)
{
var partOfSpeech = f.SelectSingleNode("a");
var currentDefinition = new Regex("[\t\r\n]").Replace(getText(f), "");
currentDefinition = (new Regex(@"^\w+\s+(.*?)$")).Replace(currentDefinition, "$1");
primaryDefinitions += "<i><b>" + getText(partOfSpeech) + "</i></b> " + currentDefinition + "<br/>";
}
primaryDefinitions = primaryDefinitions.Substring(0, primaryDefinitions.Length - 5);
}
}
do
{
if (limit - offset < step)
{
step = limit - offset;
}
if (step <= 0)
{
break;
}
try
{
var json_ = System.Text.UTF8Encoding.ASCII.GetString(client.DownloadData(examplesPath + word + "&maxResults=" + step + "&startOffset=" + offset + "&filter=0&domain=" + domain));
json_ = (new Regex(@"\$d\((.*?)\)")).Replace(json_, @"""$1""");
var json = JObject.Parse(json_);
if (offset == 0)
{
var hits = (int)json.SelectToken("result.totalHits");
reportStream.Write("Success. Word: " + word + " Hits: " + hits + "\n");
Console.WriteLine("Hits: {0}", hits);
if (hits == 0)
{
reportStream.Write("Failure. Examples was not found. Word: " + word + "\n");
}
}
Console.WriteLine("Processed: {0}", offset);
offset += step;
foreach (var _ in json.SelectToken("result.sentences"))
{
count_++;
var example = new StringBuilder()
.Append(SafeTrim((string)_.SelectToken("sentence")))
.Append("%%!!%%")
.Append(SafeTrim((string)_.SelectToken("sentence")))
.Append("%%!!%%")
.Append(word)
.Append("%%!!%%")
.Append(primaryDefinitions)
.Append("%%!!%%")
//.Append(fullDefinitions)
//.Append("%%!!%%")
.Append(wordLabel)
.Append(" ")
.Append(userLabels);
stream.Write((new Regex("[\t\n\r]").Replace(example.ToString(), "")).Replace("%%!!%%", "\t") + "\n");
}
}
catch
{
continue;
}
}while (step > 0);
Console.WriteLine("Processed: {0}\n", offset);
}
reportStream.Write("Total: " + count_ + "\n");
reportStream.Save();
}
void ParsePage(ref CardsStream stream, HtmlDocument document, string link, string userLabels)
{
var examples = document.DocumentNode.SelectNodes("//span[@class='x-g']");
if (examples == null)
{
reportStream.Write("Failure. Examples was not found. Link: " + link + "\n");
return;
}
//Transcription
var usaTranscription = document.DocumentNode.SelectSingleNode("//span[@class='y']");
var gbrTranscription = document.DocumentNode.SelectSingleNode("//span[@class='i']");
//Word "name"
var word = getText(document.DocumentNode.SelectSingleNode("//h2[@class='h']"));
//Label
var wordLabel = word.Replace(' ', '-');
userLabels = (userLabels == "") ? "oald8 " + wordLabel : "oald8 " + wordLabel + " " + userLabels;
foreach (HtmlNode example in examples)
{
var card = new Card();
var parentNode = example.ParentNode;
//Getting a structure
var structure1 = example.SelectSingleNode("span[@class='cf']");
if (GetName(parentNode) == "pv-g")
{
var structure2 = parentNode.SelectSingleNode("h4[@class='pv']");
if (structure2 != null) card.Structure.Add(structure2.InnerText);
if (structure1 != null) card.Structure.Add(structure1.InnerText);
}
else if (GetName(parentNode) == "n-g" && GetName(parentNode.ParentNode) == "pv-g")
{
var structure2 = parentNode.SelectSingleNode("span[@class='vs-g']");
var structure3 = parentNode.ParentNode.SelectSingleNode("h4[@class='pv']");
if (structure3 != null) card.Structure.Add(structure3.InnerText);
if (structure2 != null) card.Structure.Add(structure2.InnerText);
if (structure1 != null) card.Structure.Add(structure1.InnerText);
}
else if (GetName(parentNode) == "n-g")
{
var structure2 = parentNode.SelectSingleNode("span[@class='cf']");
if (structure2 != null) card.Structure.Add(structure2.InnerText);
if (structure1 != null) card.Structure.Add(structure1.InnerText);
}
else if (GetName(parentNode) == "id-g")
{
var structure2 = parentNode.SelectSingleNode("h4[@class='id']");
if (structure2 != null) card.Structure.Add(structure2.InnerText);
if (structure1 != null) card.Structure.Add(structure1.InnerText);
}
else if (GetName(parentNode) == "h-g")
{
var structure2 = parentNode.SelectSingleNode("span[@class='cf']");
if (structure2 != null) card.Structure.Add(structure2.InnerText);
if (structure1 != null) card.Structure.Add(structure1.InnerText);
}
//Getting a definition
HtmlNode definition = null;
if (GetName(parentNode) == "id-g" || GetName(parentNode) == "h-g")
{
definition = parentNode.SelectSingleNode("div[@class='def_block']");
}
if (definition == null)
{
definition = parentNode;
}
var temp = definition.SelectSingleNode("span[@class='ud']");
if (temp == null)
{
definition = definition.SelectSingleNode("span[@class='d']");
}
else {
definition = temp;
}
//An example itself
card.Interpretation = getText(example.SelectSingleNode("span[@class='x']"));
card.Sentence = (new Regex(" \\(=.*?\\)")).Replace(card.Interpretation, "");
card.Definition = getText(definition);
card.Definition = card.Definition.Replace(" ", " ");
card.UsaTranscription = getText(usaTranscription);
card.GbrTranscription = getText(gbrTranscription);
if (card.Interpretation == card.Sentence)
{
card.Interpretation = "";
}
if (card.Definition == "")
{
reportStream.Write("Failure. Definition was not found. Link: " + link + " Example: '" + card.Sentence + "'\n");
}
if (card.Sentence == "")
{
reportStream.Write("Failure. Example was not found. Link: " + link + "\n");
}
var outStr = card.Sentence + "\t" + card.Interpretation +
"\t" + word + "\t" + card.GbrTranscription +
"\t" + card.UsaTranscription + "\t" +
PrintList(card.Structure) + "\t" + card.Definition +
"\t" + userLabels + "\n";
stream.Write(outStr);
count_++;
}
}
public void ProcessWordlist(ref CardsStream stream, List<string> wordlist, string labels, bool crossreferenceFlag)
{
//Разбираем исходные страницы. Возможно, удастся ускорить работу засчёт распараллеливания.
//Собираем ссылки в теле статьи и добавляем в список.
//downloadPages(ref stream, GetArticleCrossrefenceLinkList(), labels);
//Работаем с ссылками, находящиеся в блоке Search Results, в т.ч. и на само word)
var crossreferenceLinks = new Hashtable();
string currentPageLink;
foreach(String word in wordlist)
{
var page = (new HtmlWeb()).Load(searchPath + word);
//Определяем, на какой по какой ссылке находится наше слово
var currentPageLink_ = page.DocumentNode.SelectSingleNode("//li[@class='currentpage']/a");
currentPageLink = (currentPageLink_ != null) ? getCleanUrl(currentPageLink_) : "";
//Берём все слова из блока search results
var linksNodes = page.DocumentNode.SelectNodes("//div[@id='relatedentries']/ul/li/a");
if (linksNodes != null)
{
foreach (HtmlNode link in linksNodes)
{
var url = getCleanUrl(link);
if (crossreferenceLinks.ContainsKey(url) == false)
{
crossreferenceLinks.Add(url, false);
}
}
}
else
{
reportStream.Write("Failure. Page not found. Link: " + word + "\n");
}
//Всё нужное получили, начинаем обрабатывать searchCrossreferenceLinkList
var updatedWordList = new Hashtable();
foreach (string link in crossreferenceLinks.Keys)
{
if ((bool)crossreferenceLinks[link] == true)
{
continue;
}
var tranformedWordRegExp = new Regex("^(" + word.Replace(' ', '-') + "_\\d+)$", RegexOptions.IgnoreCase);
//Первую страницу мы уже скачали.
if (currentPageLink == link)
{
ParsePage(ref stream, page, word, labels);
}
else if (tranformedWordRegExp.Match(link).Success)
{
ParsePage(ref stream, (new HtmlWeb()).Load(searchPath + link), link, labels);
}
else if (crossreferenceFlag == true)
{
ParsePage(ref stream, (new HtmlWeb()).Load(searchPath + link), link, labels);
}
if (currentPageLink == link || tranformedWordRegExp.Match(link).Success || crossreferenceFlag == true)
{
Console.WriteLine("{0}", link);
updatedWordList.Add(link, true);
reportStream.Write("Success. Page was parsed. Link: " + link + "\n");
}
}
foreach (DictionaryEntry update in updatedWordList)
{
crossreferenceLinks[update.Key] = update.Value;
}
}
reportStream.Write("Total: " + count_ + "\n");
reportStream.Save();
}
public void ProcessWordlist(ref CardsStream stream, List <string> wordlist, string labels, bool crossreferenceFlag)
{
//Разбираем исходные страницы. Возможно, удастся ускорить работу засчёт распараллеливания.
//Собираем ссылки в теле статьи и добавляем в список.
//downloadPages(ref stream, GetArticleCrossrefenceLinkList(), labels);
//Работаем с ссылками, находящиеся в блоке Search Results, в т.ч. и на само word)
var crossreferenceLinks = new Hashtable();
string currentPageLink;
foreach (String word in wordlist)
{
var page = (new HtmlWeb()).Load(searchPath + word);
//Определяем, на какой по какой ссылке находится наше слово
var currentPageLink_ = page.DocumentNode.SelectSingleNode("//li[@class='currentpage']/a");
currentPageLink = (currentPageLink_ != null) ? getCleanUrl(currentPageLink_) : "";
//Берём все слова из блока search results
var linksNodes = page.DocumentNode.SelectNodes("//div[@id='relatedentries']/ul/li/a");
if (linksNodes != null)
{
foreach (HtmlNode link in linksNodes)
{
var url = getCleanUrl(link);
if (crossreferenceLinks.ContainsKey(url) == false)
{
crossreferenceLinks.Add(url, false);
}
}
}
else
{
reportStream.Write("Failure. Page not found. Link: " + word + "\n");
}
//Всё нужное получили, начинаем обрабатывать searchCrossreferenceLinkList
var updatedWordList = new Hashtable();
foreach (string link in crossreferenceLinks.Keys)
{
if ((bool)crossreferenceLinks[link] == true)
{
continue;
}
var tranformedWordRegExp = new Regex("^(" + word.Replace(' ', '-') + "_\\d+)$", RegexOptions.IgnoreCase);
//Первую страницу мы уже скачали.
if (currentPageLink == link)
{
ParsePage(ref stream, page, word, labels);
}
else if (tranformedWordRegExp.Match(link).Success)
{
ParsePage(ref stream, (new HtmlWeb()).Load(searchPath + link), link, labels);
}
else if (crossreferenceFlag == true)
{
ParsePage(ref stream, (new HtmlWeb()).Load(searchPath + link), link, labels);
}
if (currentPageLink == link || tranformedWordRegExp.Match(link).Success || crossreferenceFlag == true)
{
Console.WriteLine("{0}", link);
updatedWordList.Add(link, true);
reportStream.Write("Success. Page was parsed. Link: " + link + "\n");
}
}
foreach (DictionaryEntry update in updatedWordList)
{
crossreferenceLinks[update.Key] = update.Value;
}
}
reportStream.Write("Total: " + count_ + "\n");
reportStream.Save();
}