private bool TryGetHashOfOneNotePage(string pageId, out string hash)
{
hash = "";
// The OneNote page consists of XML-like markups.
// Though the innerText is identical, it is common to have different 'objectID' and 'lastModifiedTime' attributes.
// These differences would cause a complete different hash value even if the contents are the same.
// Therefore, I will ignore those attributes by extracting 'innerText' and calculate a hash value without those attributes.
string pageContents = "";
bool success = onenoteApplication.TryGetPageContent(pageId, out pageContents);
if (success)
{
System.Xml.XmlDocument pageXmlContents = new System.Xml.XmlDocument();
try
{
pageXmlContents.LoadXml(pageContents);
}
catch (System.Exception exception)
{
etc.LoggerHelper.LogException(exception);
return(false);
}
bool successCalculateHash = TryCalculateHashOf(pageXmlContents.InnerText, out hash);
if (successCalculateHash)
{
return(true);
}
else
{
etc.LoggerHelper.LogWarn("Unable to calculate a hash, pageId:{0}", pageId);
return(false);
}
}
else
{
etc.LoggerHelper.LogWarn("Unable to get a page content, pageId:{0}", pageId);
return(false);
}
}
public Tuple <bool, string> ScanOneNotePages(IProgress <Tuple <int, int, int, string> > progress, System.Threading.CancellationToken cancellationToken)
{
Tuple <bool, string> resultUpdatePageInfos = UpdatePageInfos();
if (resultUpdatePageInfos.Item1 == false)
{
return(resultUpdatePageInfos);
}
else
{
etc.LoggerHelper.LogInfo("Found {0} pages.", pageInfos.Count);
int statCountReadSuccess = 0;
int statCountReadFailed = 0;
int statCountTotal = pageInfos.Count;
string statPageTitle = null;
progress.Report(Tuple.Create(statCountReadSuccess, statCountReadFailed, statCountTotal, statPageTitle));
foreach (KeyValuePair <string, OneNotePageInfo> elem in pageInfos)
{
if (cancellationToken.IsCancellationRequested)
{
OnCancelled.Invoke();
break;
}
string pageId = elem.Key;
OneNotePageInfo pageInfo = elem.Value;
string pageContent = "";
bool successHash = false;
pageInfo.HashValueForInnerText = null;
if (onenoteApplication.TryGetPageContent(elem.Key, out pageContent))
{
try
{
System.Xml.XmlDocument pageContentXml = new System.Xml.XmlDocument();
pageContentXml.LoadXml(pageContent);
/*
* Though the page contents are identical, it is quite common to see different 'objectID' and 'lastModified' attributes.
* This difference results in a completely different hash value, which cannot be detected by other duplicate remove software.
* By simply taking 'innerText' of the internal XML-like format, those attributes will be ignored.
* [!] The underlying assumption is that a user cannot modify the attributes directly.
*/
if (TryCalculateHashValue(pageContentXml.InnerText, out string hashValue))
{
successHash = true;
pageInfo.HashValueForInnerText = hashValue;
}
else
{
etc.LoggerHelper.LogWarn("Failed to calculate hash for the page ({0}).", pageId);
}
}
catch (System.Exception exception)
{
etc.LoggerHelper.LogUnexpectedException(exception);
}
}
statPageTitle = pageInfo.PageTitle;
if (successHash)
{
statCountReadSuccess += 1;
}
else
{
statCountReadFailed += 1;
etc.LoggerHelper.LogWarn("Failed to retrieve the content of the page ({0}).", pageId);
}
progress.Report(Tuple.Create(statCountReadSuccess, statCountReadFailed, statCountTotal, statPageTitle));
}
}
return(Tuple.Create(true, ""));
}
