private static void BuildDocument(List<Source> sources, WordprocessingDocument output)
{
if (RelationshipMarkup == null)
RelationshipMarkup = new Dictionary<XName, XName[]>()
{
//{ button, new [] { image }},
{ A.blip, new [] { R.embed, R.link }},
{ A.hlinkClick, new [] { R.id }},
{ A.relIds, new [] { R.cs, R.dm, R.lo, R.qs }},
//{ a14:imgLayer, new [] { R.embed }},
//{ ax:ocx, new [] { R.id }},
{ C.chart, new [] { R.id }},
{ C.externalData, new [] { R.id }},
{ C.userShapes, new [] { R.id }},
{ DGM.relIds, new [] { R.cs, R.dm, R.lo, R.qs }},
{ O.OLEObject, new [] { R.id }},
{ VML.fill, new [] { R.id }},
{ VML.imagedata, new [] { R.href, R.id, R.pict }},
{ VML.stroke, new [] { R.id }},
{ W.altChunk, new [] { R.id }},
{ W.attachedTemplate, new [] { R.id }},
{ W.control, new [] { R.id }},
{ W.dataSource, new [] { R.id }},
{ W.embedBold, new [] { R.id }},
{ W.embedBoldItalic, new [] { R.id }},
{ W.embedItalic, new [] { R.id }},
{ W.embedRegular, new [] { R.id }},
{ W.footerReference, new [] { R.id }},
{ W.headerReference, new [] { R.id }},
{ W.headerSource, new [] { R.id }},
{ W.hyperlink, new [] { R.id }},
{ W.printerSettings, new [] { R.id }},
{ W.recipientData, new [] { R.id }}, // Mail merge, not required
{ W.saveThroughXslt, new [] { R.id }},
{ W.sourceFileName, new [] { R.id }}, // Framesets, not required
{ W.src, new [] { R.id }}, // Mail merge, not required
{ W.subDoc, new [] { R.id }}, // Sub documents, not required
//{ w14:contentPart, new [] { R.id }},
{ WNE.toolbarData, new [] { R.id }},
};
// This list is used to eliminate duplicate images
List<ImageData> images = new List<ImageData>();
XDocument mainPart = output.MainDocumentPart.GetXDocument();
mainPart.Declaration.Standalone = "yes";
mainPart.Declaration.Encoding = "UTF-8";
mainPart.Root.ReplaceWith(
new XElement(W.document, NamespaceAttributes,
new XElement(W.body)));
if (sources.Count > 0)
{
using (OpenXmlMemoryStreamDocument streamDoc = new OpenXmlMemoryStreamDocument(sources[0].WmlDocument))
using (WordprocessingDocument doc = streamDoc.GetWordprocessingDocument())
{
CopyStartingParts(doc, output, images);
}
int sourceNum = 0;
foreach (Source source in sources)
{
if (source.InsertId != null)
{
while (true)
{
XDocument mainXDoc = output.MainDocumentPart.GetXDocument();
if (!mainXDoc.Descendants(PtOpenXml.Insert).Any(d => (string)d.Attribute(PtOpenXml.Id) == source.InsertId))
break;
using (OpenXmlMemoryStreamDocument streamDoc = new OpenXmlMemoryStreamDocument(source.WmlDocument))
using (WordprocessingDocument doc = streamDoc.GetWordprocessingDocument())
{
#if TestForUnsupportedDocuments
// throws exceptions if a document contains unsupported content
TestForUnsupportedDocument(doc, sources.IndexOf(source));
#endif
List<XElement> contents = doc.MainDocumentPart.GetXDocument()
.Root
.Element(W.body)
.Elements()
.Skip(source.Start)
.Take(source.Count)
.ToList();
try
{
AppendDocument(doc, output, contents, source.KeepSections, source.InsertId, images);
}
catch (DocumentBuilderInternalException dbie)
{
if (dbie.Message.Contains("{0}"))
throw new DocumentBuilderException(string.Format(dbie.Message, sourceNum));
else
throw dbie;
}
}
}
}
else
{
using (OpenXmlMemoryStreamDocument streamDoc = new OpenXmlMemoryStreamDocument(source.WmlDocument))
using (WordprocessingDocument doc = streamDoc.GetWordprocessingDocument())
{
#if TestForUnsupportedDocuments
// throws exceptions if a document contains unsupported content
TestForUnsupportedDocument(doc, sources.IndexOf(source));
#endif
List<XElement> contents = doc.MainDocumentPart.GetXDocument()
.Root
.Element(W.body)
.Elements()
.Skip(source.Start)
.Take(source.Count)
.ToList();
try
{
AppendDocument(doc, output, contents, source.KeepSections, null, images);
}
catch (DocumentBuilderInternalException dbie)
{
if (dbie.Message.Contains("{0}"))
throw new DocumentBuilderException(string.Format(dbie.Message, sourceNum));
else
throw dbie;
}
}
}
++sourceNum;
}
if (!sources.Any(s => s.KeepSections))
{
using (OpenXmlMemoryStreamDocument streamDoc = new OpenXmlMemoryStreamDocument(sources[0].WmlDocument))
using (WordprocessingDocument doc = streamDoc.GetWordprocessingDocument())
{
var sectPr = doc.MainDocumentPart.GetXDocument().Root.Element(W.body)
.Elements().Last();
if (sectPr.Name == W.sectPr)
{
AddSectionAndDependencies(doc, output, sectPr, images);
output.MainDocumentPart.GetXDocument().Root.Element(W.body).Add(sectPr);
}
}
}
else
FixUpSectionProperties(output);
}
foreach (var part in output.GetAllParts())
if (part.Annotation<XDocument>() != null)
part.PutXDocument();
}
private static void BuildDocument(List<Source> sources, WordprocessingDocument output)
{
if (RelationshipMarkup == null)
RelationshipMarkup = new Dictionary<XName, XName[]>()
{
//{ button, new [] { image }},
{ A.blip, new [] { R.embed, R.link }},
{ A.hlinkClick, new [] { R.id }},
{ A.relIds, new [] { R.cs, R.dm, R.lo, R.qs }},
//{ a14:imgLayer, new [] { R.embed }},
//{ ax:ocx, new [] { R.id }},
{ C.chart, new [] { R.id }},
{ C.externalData, new [] { R.id }},
{ C.userShapes, new [] { R.id }},
{ DGM.relIds, new [] { R.cs, R.dm, R.lo, R.qs }},
{ O.OLEObject, new [] { R.id }},
{ VML.fill, new [] { R.id }},
{ VML.imagedata, new [] { R.href, R.id, R.pict }},
{ VML.stroke, new [] { R.id }},
{ W.altChunk, new [] { R.id }},
{ W.attachedTemplate, new [] { R.id }},
{ W.control, new [] { R.id }},
{ W.dataSource, new [] { R.id }},
{ W.embedBold, new [] { R.id }},
{ W.embedBoldItalic, new [] { R.id }},
{ W.embedItalic, new [] { R.id }},
{ W.embedRegular, new [] { R.id }},
{ W.footerReference, new [] { R.id }},
{ W.headerReference, new [] { R.id }},
{ W.headerSource, new [] { R.id }},
{ W.hyperlink, new [] { R.id }},
{ W.printerSettings, new [] { R.id }},
{ W.recipientData, new [] { R.id }}, // Mail merge, not required
{ W.saveThroughXslt, new [] { R.id }},
{ W.sourceFileName, new [] { R.id }}, // Framesets, not required
{ W.src, new [] { R.id }}, // Mail merge, not required
{ W.subDoc, new [] { R.id }}, // Sub documents, not required
//{ w14:contentPart, new [] { R.id }},
{ WNE.toolbarData, new [] { R.id }},
};
// This list is used to eliminate duplicate images
List<ImageData> images = new List<ImageData>();
XDocument mainPart = output.MainDocumentPart.GetXDocument();
mainPart.Declaration.Standalone = "yes";
mainPart.Declaration.Encoding = "UTF-8";
mainPart.Root.ReplaceWith(
new XElement(W.document, NamespaceAttributes,
new XElement(W.body)));
if (sources.Count > 0)
{
using (OpenXmlMemoryStreamDocument streamDoc = new OpenXmlMemoryStreamDocument(sources[0].WmlDocument))
using (WordprocessingDocument doc = streamDoc.GetWordprocessingDocument())
{
CopyStartingParts(doc, output, images);
}
int sourceNum = 0;
foreach (Source source in sources)
{
if (source.InsertId != null)
{
while (true)
{
XDocument mainXDoc = output.MainDocumentPart.GetXDocument();
if (!mainXDoc.Descendants(PtOpenXml.Insert).Any(d => (string)d.Attribute(PtOpenXml.Id) == source.InsertId))
break;
using (OpenXmlMemoryStreamDocument streamDoc = new OpenXmlMemoryStreamDocument(source.WmlDocument))
using (WordprocessingDocument doc = streamDoc.GetWordprocessingDocument())
{
#if TestForUnsupportedDocuments
// throws exceptions if a document contains unsupported content
TestForUnsupportedDocument(doc, sources.IndexOf(source));
#endif
if (source.KeepSections && source.DiscardHeadersAndFootersInKeptSections)
RemoveHeadersAndFootersFromSections(doc);
else if (source.KeepSections)
ProcessSectionsForLinkToPreviousHeadersAndFooters(doc);
List<XElement> contents = doc.MainDocumentPart.GetXDocument()
.Root
.Element(W.body)
.Elements()
.Skip(source.Start)
.Take(source.Count)
.ToList();
try
{
AppendDocument(doc, output, contents, source.KeepSections, source.InsertId, images);
}
catch (DocumentBuilderInternalException dbie)
{
if (dbie.Message.Contains("{0}"))
throw new DocumentBuilderException(string.Format(dbie.Message, sourceNum));
else
throw dbie;
}
}
}
}
else
{
using (OpenXmlMemoryStreamDocument streamDoc = new OpenXmlMemoryStreamDocument(source.WmlDocument))
using (WordprocessingDocument doc = streamDoc.GetWordprocessingDocument())
{
#if TestForUnsupportedDocuments
// throws exceptions if a document contains unsupported content
TestForUnsupportedDocument(doc, sources.IndexOf(source));
#endif
if (source.KeepSections && source.DiscardHeadersAndFootersInKeptSections)
RemoveHeadersAndFootersFromSections(doc);
else if (source.KeepSections)
ProcessSectionsForLinkToPreviousHeadersAndFooters(doc);
List<XElement> contents = doc.MainDocumentPart.GetXDocument()
.Root
.Element(W.body)
.Elements()
.Skip(source.Start)
.Take(source.Count)
.ToList();
try
{
AppendDocument(doc, output, contents, source.KeepSections, null, images);
}
catch (DocumentBuilderInternalException dbie)
{
if (dbie.Message.Contains("{0}"))
throw new DocumentBuilderException(string.Format(dbie.Message, sourceNum));
else
throw dbie;
}
}
}
++sourceNum;
}
if (!sources.Any(s => s.KeepSections))
{
using (OpenXmlMemoryStreamDocument streamDoc = new OpenXmlMemoryStreamDocument(sources[0].WmlDocument))
using (WordprocessingDocument doc = streamDoc.GetWordprocessingDocument())
{
var sectPr = doc.MainDocumentPart.GetXDocument().Root.Element(W.body)
.Elements().Last();
if (sectPr.Name == W.sectPr)
{
AddSectionAndDependencies(doc, output, sectPr, images);
output.MainDocumentPart.GetXDocument().Root.Element(W.body).Add(sectPr);
}
}
}
else
{
FixUpSectionProperties(output);
// Any sectPr elements that do not have headers and footers should take their headers and footers from the *next* section,
// i.e. from the running section.
var mxd = output.MainDocumentPart.GetXDocument();
var sections = mxd.Descendants(W.sectPr).Reverse().ToList();
CachedHeaderFooter[] cachedHeaderFooter = new[]
{
new CachedHeaderFooter() { Ref = W.headerReference, Type = "first" },
new CachedHeaderFooter() { Ref = W.headerReference, Type = "even" },
new CachedHeaderFooter() { Ref = W.headerReference, Type = "default" },
new CachedHeaderFooter() { Ref = W.footerReference, Type = "first" },
new CachedHeaderFooter() { Ref = W.footerReference, Type = "even" },
new CachedHeaderFooter() { Ref = W.footerReference, Type = "default" },
};
bool firstSection = true;
foreach (var sect in sections)
{
if (firstSection)
{
foreach (var hf in cachedHeaderFooter)
{
var referenceElement = sect.Elements(hf.Ref).FirstOrDefault(z => (string)z.Attribute(W.type) == hf.Type);
if (referenceElement != null)
hf.CachedPartRid = (string)referenceElement.Attribute(R.id);
}
firstSection = false;
continue;
}
else
{
CopyOrCacheHeaderOrFooter(output, cachedHeaderFooter, sect, W.headerReference, "first");
CopyOrCacheHeaderOrFooter(output, cachedHeaderFooter, sect, W.headerReference, "even");
CopyOrCacheHeaderOrFooter(output, cachedHeaderFooter, sect, W.headerReference, "default");
CopyOrCacheHeaderOrFooter(output, cachedHeaderFooter, sect, W.footerReference, "first");
CopyOrCacheHeaderOrFooter(output, cachedHeaderFooter, sect, W.footerReference, "even");
CopyOrCacheHeaderOrFooter(output, cachedHeaderFooter, sect, W.footerReference, "default");
}
}
}
}
foreach (var part in output.GetAllParts())
if (part.Annotation<XDocument>() != null)
part.PutXDocument();
}
private static bool ValidateWordprocessingDocument(WordprocessingDocument wDoc, List<XElement> metrics, List<string> notes, Dictionary<XName, int> metricCountDictionary)
{
bool valid = ValidateAgainstSpecificVersion(wDoc, metrics, DocumentFormat.OpenXml.FileFormatVersions.Office2007, H.SdkValidationError2007);
valid |= ValidateAgainstSpecificVersion(wDoc, metrics, DocumentFormat.OpenXml.FileFormatVersions.Office2010, H.SdkValidationError2010);
#if !NET35
valid |= ValidateAgainstSpecificVersion(wDoc, metrics, DocumentFormat.OpenXml.FileFormatVersions.Office2013, H.SdkValidationError2013);
#endif
int elementCount = 0;
int paragraphCount = 0;
int textCount = 0;
foreach (var part in wDoc.ContentParts())
{
XDocument xDoc = part.GetXDocument();
foreach (var e in xDoc.Descendants())
{
if (e.Name == W.txbxContent)
IncrementMetric(metricCountDictionary, H.TextBox);
else if (e.Name == W.sdt)
IncrementMetric(metricCountDictionary, H.ContentControl);
else if (e.Name == W.customXml)
IncrementMetric(metricCountDictionary, H.CustomXmlMarkup);
else if (e.Name == W.fldChar)
IncrementMetric(metricCountDictionary, H.ComplexField);
else if (e.Name == W.fldSimple)
IncrementMetric(metricCountDictionary, H.SimpleField);
else if (e.Name == W.altChunk)
IncrementMetric(metricCountDictionary, H.AltChunk);
else if (e.Name == W.tbl)
IncrementMetric(metricCountDictionary, H.Table);
else if (e.Name == W.hyperlink)
IncrementMetric(metricCountDictionary, H.Hyperlink);
else if (e.Name == W.framePr)
IncrementMetric(metricCountDictionary, H.LegacyFrame);
else if (e.Name == W.control)
IncrementMetric(metricCountDictionary, H.ActiveX);
else if (e.Name == W.subDoc)
IncrementMetric(metricCountDictionary, H.SubDocument);
else if (e.Name == VML.imagedata || e.Name == VML.fill || e.Name == VML.stroke || e.Name == A.blip)
{
var relId = (string)e.Attribute(R.embed);
if (relId != null)
ValidateImageExists(part, relId, metricCountDictionary);
relId = (string)e.Attribute(R.pict);
if (relId != null)
ValidateImageExists(part, relId, metricCountDictionary);
relId = (string)e.Attribute(R.id);
if (relId != null)
ValidateImageExists(part, relId, metricCountDictionary);
}
if (part.Uri == wDoc.MainDocumentPart.Uri)
{
elementCount++;
if (e.Name == W.p)
paragraphCount++;
if (e.Name == W.t)
textCount += ((string)e).Length;
}
}
}
foreach (var item in metricCountDictionary)
{
metrics.Add(
new XElement(item.Key, new XAttribute(H.Val, item.Value)));
}
metrics.Add(new XElement(H.ElementCount, new XAttribute(H.Val, elementCount)));
metrics.Add(new XElement(H.AverageParagraphLength, new XAttribute(H.Val, (int)((double)textCount / (double)paragraphCount))));
if (RevisionAccepter.HasTrackedRevisions(wDoc))
metrics.Add(new XElement(H.RevisionTracking, new XAttribute(H.Val, true)));
if (wDoc.GetAllParts().Any(part => part.ContentType == "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"))
metrics.Add(new XElement(H.EmbeddedXlsx, new XAttribute(H.Val, true)));
NumberingFormatListAssembly(wDoc, metrics);
XDocument wxDoc = wDoc.MainDocumentPart.GetXDocument();
foreach (var d in wxDoc.Descendants())
{
if (d.Name == W.saveThroughXslt)
{
string rid = (string)d.Attribute(R.id);
var tempExternalRelationship = wDoc
.MainDocumentPart
.DocumentSettingsPart
.ExternalRelationships
.FirstOrDefault(h => h.Id == rid);
if (tempExternalRelationship == null)
metrics.Add(new XElement(H.InvalidSaveThroughXslt, new XAttribute(H.Val, true)));
valid = false;
}
else if (d.Name == W.trackRevisions)
metrics.Add(new XElement(H.TrackRevisionsEnabled, new XAttribute(H.Val, true)));
else if (d.Name == W.documentProtection)
metrics.Add(new XElement(H.DocumentProtection, new XAttribute(H.Val, true)));
}
FontAndCharSetAnalysis(wDoc, metrics, notes);
return valid;
}
private static XElement GetMetricsForWmlPart(WordprocessingDocument noTrackedDocument, OpenXmlPart part, MetricsGetterSettings settings)
{
XElement contentControls = null;
if (part is MainDocumentPart ||
part is HeaderPart ||
part is FooterPart ||
part is FootnotesPart ||
part is EndnotesPart)
{
var noTrackedPart = noTrackedDocument.GetAllParts().FirstOrDefault(p => p.Uri == part.Uri);
if (noTrackedPart == null)
throw new OpenXmlPowerToolsException("Internal error");
var xd = noTrackedPart.GetXDocument();
contentControls = (XElement)GetContentControlsTransform(xd.Root, settings);
if (!contentControls.HasElements)
contentControls = null;
}
var partMetrics = new XElement(H.Part,
new XAttribute(H.ContentType, part.ContentType),
new XAttribute(H.Uri, part.Uri.ToString()),
contentControls);
if (partMetrics.HasElements)
return partMetrics;
return null;
}
private static XElement GetWmlMetrics(string fileName, bool invalidHyperlink, WordprocessingDocument document, WordprocessingDocument noTrackedDocument, MetricsGetterSettings settings)
{
var parts = new XElement(H.Parts,
document.GetAllParts().Select(part =>
{
return GetMetricsForWmlPart(noTrackedDocument, part, settings);
}));
if (!parts.HasElements)
parts = null;
var metrics = new XElement(H.Metrics,
new XAttribute(H.FileName, fileName),
new XAttribute(H.FileType, "WordprocessingML"),
GetStyleHierarchy(document),
GetMiscWmlMetrics(document, noTrackedDocument, invalidHyperlink),
parts);
return metrics;
}
