public TextSpan(
int startTextIndex,
int stopTextIndex,
ITextControllerProjectPlugin controller,
object data)
{
StartTextIndex = startTextIndex;
StopTextIndex = stopTextIndex;
Controller = controller;
Data = data;
}
/// <summary>
/// Reads the content file, either from the project reader or the Structure
/// file depending on the persistence settings.
/// </summary>
/// <param name="projectReader">The project reader.</param>
public void Read(XmlReader projectReader)
{
// Figure out which reader we'll be using.
bool createdReader;
XmlReader reader = GetXmlReader(
projectReader, Settings.ContentDataFilename, out createdReader);
// Loop through the resulting file until we get to the end of the
// XML element we care about.
ProjectBlockCollection blocks = Project.Blocks;
bool reachedData = reader.NamespaceURI == XmlConstants.ProjectNamespace &&
reader.LocalName == "content-data";
int blockIndex = 0;
int startTextIndex = 0;
int stopTextIndex = 0;
ITextControllerProjectPlugin plugin = null;
object pluginData = null;
bool inProject = false;
bool inBlocks = false;
while (reader.Read())
{
// Ignore anything outside of our namespace.
if (reader.NamespaceURI != XmlConstants.ProjectNamespace)
{
continue;
}
// Check to see if we're done reading.
if (reader.NodeType == XmlNodeType.EndElement)
{
switch (reader.LocalName)
{
case "content-data":
return;
case "text-span":
// Construct a text span from the gathered properties.
var textSpan = new TextSpan(
startTextIndex, stopTextIndex, plugin, pluginData);
blocks[blockIndex].TextSpans.Add(textSpan);
// Clear out the data to catch any additional errors.
plugin = null;
pluginData = null;
break;
case "block-data":
inBlocks = false;
break;
case "project":
inProject = false;
break;
}
}
// For the rest of this loop, we only deal with begin elements.
if (reader.NodeType != XmlNodeType.Element)
{
continue;
}
// If we haven't reached the Structure, we just cycle through the XML.
if (!reachedData)
{
// Flip the flag if we are starting to read the Structure.
if (reader.NamespaceURI == XmlConstants.ProjectNamespace &&
reader.LocalName == "content-data")
{
reachedData = true;
}
// Continue on since we're done with this if clause.
continue;
}
// We process the remaining elements based on their local name.
switch (reader.LocalName)
{
case "block-data":
inBlocks = true;
break;
case "project":
inProject = true;
break;
case "analyzer":
// Grab the analyzer name and find the associated plugin.
string key = reader.ReadInnerXml();
if (Project.Plugins.Contains(key))
{
var analyzer = Project.Plugins[key] as IBlockAnalyzerProjectPlugin;
if (analyzer != null)
{
blocks[blockIndex].AddAnalysis(analyzer);
}
}
break;
case "block-key":
// Grab the information to identify the block. We can't use
// block key directly so we have an index into the original
// block data and a hash of the text.
blockIndex = Convert.ToInt32(reader["block-index"]);
int textHash = Convert.ToInt32(reader["text-hash"], 16);
// Figure out if we are asking for an index that doesn't exist
// or if the hash doesn't work. If they don't, then stop
// processing entirely.
if (blockIndex >= blocks.Count ||
blocks[blockIndex].Text.GetHashCode() != textHash)
{
return;
}
break;
case "property":
{
var path = new HierarchicalPath(reader["path"]);
string value = reader.ReadString();
if (inBlocks)
{
blocks[blockIndex].Properties[path] = value;
}
if (inProject)
{
Project.Properties[path] = value;
}
break;
}
case "index":
startTextIndex = Convert.ToInt32(reader["start"]);
stopTextIndex = Convert.ToInt32(reader["stop"]);
break;
case "plugin":
string projectKey = reader.ReadString();
plugin = (ITextControllerProjectPlugin)Project.Plugins[projectKey];
break;
}
}
// If we created the reader, close it.
if (createdReader)
{
reader.Close();
reader.Dispose();
}
}
