public override T GetSetting <T>(string key)
{
if (ParsedObject.ContainsKey(key))
{
return(ParsedObject.GetValue(key).ToObject <T>());
}
if (DefaultValues.ContainsKey(key))
{
return(DefaultValues[key].ToObject <T>());
}
throw new Exception($"No entry or default entry was found with key {key}.");
}
public async Task <string> GetObjectAsync(string oid, ObjectType expectedType = ObjectType.Blob)
{
string[] split = oid.Split(2);
string obj = await FileExtensions.ReadFileAsync($"{Dir}/objects/{split[0]}/{split[1]}", Encoding);
var po = new ParsedObject(obj);
if (po.Type != expectedType)
{
throw new ArgumentException($"Expected type {expectedType} but found {po.Type}");
}
return(po.Content);
}
private ParsedObject ParseRedditDescription(string sourceURL)
{
ParsedObject parsed = null;
try
{
string queryStr = "http://api.embed.ly/1/extract?key=16357551b6a84e6c88debee64dcd8bf3&maxwidth=500&url=" + HttpUtility.UrlPathEncode(sourceURL);
string result = webClient.DownloadString(queryStr);
parsed = Newtonsoft.Json.JsonConvert.DeserializeObject <ParsedObject>(result);
}
catch (Exception exp)
{
// do nothing
}
return(parsed);
}
private List <ImportRecord> DoRedditGroupImport(string subReddit, bool useTitles = true)
{
List <ImportRecord> newList = new List <ImportRecord>();
string googleParserURL = "http://ajax.googleapis.com/ajax/services/feed/load?v=1.0&num=100&q=";
string finalURL = googleParserURL + HttpUtility.UrlPathEncode(subReddit);
byte[] urlData = webClient.DownloadData(subReddit);
string theData = Encoding.UTF8.GetString(urlData, 0, urlData.Length);
byte[] finalUrlData = webClient.DownloadData(finalURL);
string googleData = Encoding.UTF8.GetString(finalUrlData, 0, finalUrlData.Length);;
GoogleResponse theResult = Newtonsoft.Json.JsonConvert.DeserializeObject <GoogleResponse>(googleData);
XDocument document = XDocument.Parse(theData);
foreach (XElement descendant in document.Descendants("item"))
{
string desc = descendant.Element("description").Value;
string articleURL = GetRedditSourceURL(desc);
ParsedObject parsed = ParseRedditDescription(articleURL);
if (parsed != null)
{
ImportRecord newRec = new ImportRecord();
newRec.Upload = true;
if (useTitles)
{
if (String.IsNullOrEmpty(parsed.title))
{
newRec.Title = descendant.Element("title").Value;
}
else
{
newRec.Title = parsed.title;
}
}
else
{
newRec.Title = null;
}
if (String.IsNullOrEmpty(parsed.description))
{
newRec.Body = descendant.Element("title").Value;
}
else
{
newRec.Body = parsed.description;
}
newRec.MainURL = parsed.url;
newRec.Mature = false;
if ((parsed.images != null) && (parsed.images.Count > 0))
{
newRec.ImageURL = parsed.images[0].url;
newRec.UseImage = true;
}
else
{
newRec.UseImage = false;
}
newList.Add(newRec);
}
}
return(newList);
}
/// <summary>
/// Parse tokens into Modifiers bit flags.
/// </summary>
public static Modifiers Parse(Parser parser, CodeObject parent)
{
// Search the parser's unused list for valid modifier tokens
Modifiers modifiers = 0;
bool hasSandwichedModifiers = false;
bool hasConditionalModifiers = false;
bool needsElseCondition = true;
bool elseIsNotActive = false;
bool foundEndIf = false;
List <ParsedObject> unusedList = parser.Unused;
// First, pre-scan the unused list to check for the special case of conditional directives
// being used on the modifiers. Set a flag if we detect this for later processing below.
for (int i = unusedList.Count - 1; i >= 0; --i)
{
ParsedObject parsedObject = unusedList[i];
if (parsedObject is Token)
{
// Abort if there's a blank line, or if we get a token that isn't a modifier
Token token = (Token)parsedObject;
if (token.NewLines > 1 || !IsModifier(token.Text))
{
break;
}
if (foundEndIf)
{
hasSandwichedModifiers = true;
}
}
else
{
// Abort if we get anything other than the expected #if/#else/#endif chain, or if
// we have any comments before we hit the #endif (going backwards).
CodeObject codeObject = ((UnusedCodeObject)parsedObject).CodeObject;
if (codeObject is EndIfDirective)
{
if (foundEndIf || parsedObject.HasTrailingComments)
{
break;
}
foundEndIf = true;
}
else if (codeObject is ConditionalDirective)
{
if (!foundEndIf)
{
break;
}
if (codeObject is ElseDirective)
{
needsElseCondition = false;
}
else
{
if (((ConditionalDirective)codeObject).IsActive)
{
elseIsNotActive = true;
}
if (codeObject is IfDirective)
{
if (hasSandwichedModifiers)
{
hasConditionalModifiers = true;
}
break;
}
}
}
else
{
break;
}
}
}
// Now, reset and do the real parse scan
bool postDirective = true;
string declarationText = null;
string declarationModifiersText = null;
Token lastUnrecognizedToken = null;
List <ConditionalDirective> inactiveConditions = new List <ConditionalDirective>();
for (int i = unusedList.Count - 1; i >= 0; --i)
{
ParsedObject parsedObject = unusedList[i];
if (parsedObject is Token)
{
// Check for modifier tokens
Token token = (Token)unusedList[i];
Modifiers value;
if (_nameToModifierMap.TryGetValue(token.Text, out value))
{
modifiers |= value;
parent.MoveFormatting(token);
parent.MoveAllComments(token, true);
unusedList.RemoveAt(i);
// If we've passed the conditionals and have prefixed modifiers, add them
// to the skipped text of all of the inactive conditions.
if (!hasConditionalModifiers && inactiveConditions.Count > 0)
{
foreach (ConditionalDirective conditionalDirective in inactiveConditions)
{
conditionalDirective.SkippedText = AsString(value) + conditionalDirective.SkippedText;
}
}
}
else
{
// If the token isn't recognized, continue processing in case there are other unused tokens
// that we do recognize as modifiers (there might be new modifiers added in later C# versions).
// Also, keep track of the last unrecognized token so we can transfer any newlines to it.
lastUnrecognizedToken = (Token)parsedObject;
}
}
else if (hasConditionalModifiers)
{
// Check for compiler directives - specifically, we handle an #endif immediately
// preceeding the parent declaration, allowing for a chain of conditional directives
// used to change the modifiers on the declaration at compile-time. We have already
// verified that we have a valid sequence (above), so we process conditional directives
// backwards here until we get to the starting #if.
UnusedCodeObject unused = (UnusedCodeObject)parsedObject;
if (unused.CodeObject is CompilerDirective)
{
// Store any compiler directives as pre or post annotations on the parent
CompilerDirective compilerDirective = (CompilerDirective)unused.CodeObject;
if (compilerDirective is EndIfDirective)
{
declarationText = parent.AsString();
declarationModifiersText = AsString(modifiers);
}
else if (compilerDirective is ConditionalDirective)
{
ConditionalDirective conditionalDirective = (ConditionalDirective)compilerDirective;
// Create an #else if none existed
if (needsElseCondition)
{
ElseDirective elseDirective = new ElseDirective();
if (elseIsNotActive)
{
elseDirective.SkippedText = declarationText;
inactiveConditions.Add(elseDirective);
}
else
{
postDirective = false;
}
parent.AttachAnnotation(elseDirective, elseIsNotActive ? AnnotationFlags.IsPostfix : AnnotationFlags.None, true);
needsElseCondition = false;
}
// When we find the active condition, switch from post to pre
if (conditionalDirective.IsActive)
{
postDirective = false;
}
else
{
// Update any skipped conditions to reflect the entire declaration
Modifiers orderedModifiers = Parse(conditionalDirective.SkippedText + " " + declarationModifiersText);
conditionalDirective.SkippedText = AsString(orderedModifiers) + declarationText;
inactiveConditions.Add(conditionalDirective);
}
// Stop processing conditionals when we get to the #if
if (compilerDirective is IfDirective)
{
hasConditionalModifiers = false;
}
}
else
{
break; // For now, stop if we find any other directives
}
parent.AttachAnnotation(compilerDirective, postDirective ? AnnotationFlags.IsPostfix : AnnotationFlags.None, true);
parent.MoveComments(unused.LastToken, true);
unusedList.RemoveAt(i);
}
}
else
{
break;
}
}
// If we had any unrecognized tokens, they'll be emitted before the parent object, so if the last one
// doesn't have any newlines, we need to give it the parent's newlines.
if (lastUnrecognizedToken != null)
{
if (lastUnrecognizedToken.NewLines == 0)
{
lastUnrecognizedToken.NewLines = (ushort)parent.NewLines;
}
parent.NewLines = 0;
}
return(modifiers);
}
