private Tweet[] FillUserTweet(UserTweet result, string content)
{
var matches = Regex.Matches(content, RegexContent, RegexOptions.Multiline | RegexOptions.IgnoreCase);
List <Tweet> tweetList = new List <Tweet>();
try
{
foreach (Match match in matches)
{
Tweet tweet = new Tweet();
int comment;
int.TryParse(match.Groups["Reply"].Value, out comment);
int forward;
int.TryParse(match.Groups["Forward"].Value, out forward);
tweet.Comment = comment;
tweet.Content = TextCleaner.FullClean(match.Groups["Content"].Value);
tweet.Mid = match.Groups["Mid"].Value;
tweet.Forward = forward;
tweet.Source = match.Groups["Source"].Value;
tweet.PubDate = DateTimeParser.Parser(match.Groups["PubDate"].Value) ?? DateTime.MinValue;
tweet.Url = RegexParser.AbsoluteUrl(match.Groups["Url"].Value, result.Url, true);
result.Tweets.Add(tweet);
tweetList.Add(tweet);
}
}
catch {}
return(tweetList.ToArray());
}
private void FillTweetComment(Tweet tweet, SiteEntity site)
{
if (tweet.Comment == 0)
{
return;
}
int currentPage = 1;
string mid = tweet.Mid;
try
{
while (true)
{
string url = string.Format(CommentUrlFormat, mid, currentPage);
var request = BuildRequest(url);
CrawlResponse response = null;
for (int i = 0; i < 5; i++)
{
try
{
response = GeckoRequestProcessor.DoRequest(request, site, null, null);
AggrSum();
}
catch {}
if (response.Status != Enums.CrawlResult.Succ)
{
Logger.Info("访问页面错误:Url = " + response.Url);
}
else
{
break;
}
}
CommentJsonResponse tmpResult =
JsonConvert.DeserializeObject <CommentJsonResponse>(response.Content.Trim("</pre>".ToArray()));
response.Content = HttpUtility.HtmlDecode(tmpResult.data.html);
var pageMatch = Regex.Match(response.Content, RegexCommentPage,
RegexOptions.IgnoreCase | RegexOptions.Multiline);
if (currentPage != 1 &&
(!pageMatch.Success ||
pageMatch.Groups["CurrentPageNum"].Value != currentPage.ToString(CultureInfo.InvariantCulture)))
{
return;
}
//Fill Tweet
var matches = Regex.Matches(response.Content, RegexComment,
RegexOptions.IgnoreCase | RegexOptions.Multiline);
foreach (Match match in matches)
{
Comment comment = new Comment();
comment.Author = match.Groups["Author"].Value;
comment.AuthorUrl = RegexParser.AbsoluteUrl(match.Groups["AuthorUrl"].Value, tweet.Url, true);
comment.Content = TextCleaner.FullClean(match.Groups["Content"].Value);
comment.PubDate = DateTimeParser.Parser(match.Groups["PubDate"].Value) ?? DateTime.MinValue;
tweet.Comments.Add(comment);
}
currentPage++;
}
}
catch {
}
}
