I have a web project that needs to merge up up to 100 rows of data into a merge. There are 31 potential merge fields, but normally only 2 or 3 of those fields are used. The final merged document can be up to 500 pages long.
I have a few merge fields that may or may not be HTML and 2 fields that can be image fields.
The merge process seems to take ~5 minutes to complete and I was wondering if there was anything I can do to speed the process.
My code is reasonably simple:
UPDATE:
I noticed another user reported better speeds when they merged documents individually and then combined the individual documents into a single document (Free Support Forum - aspose.com) . I tried this and did infact cut my merge time down to somewhere closer to 3 minutes. This is still longer than I would like though.
New psudo code:
- Would cutting down the number of available merge fields have a noticeable impact on the merge time?
- Would removing the image fields have a noticeable impact on the time?
- Are there any other suggestions that might speed the process up?
My code is reasonably simple:
public static byte[] MergeFields(byte[] sourceDoc, DataTable dt)
{
byte[] rVal;
using (var wordByteStream = new MemoryStream(sourceDoc))
{
Document wordDoc = new Document(wordByteStream);
wordDoc.MailMerge.CleanupOptions = Aspose.Words.Reporting.MailMergeCleanupOptions.RemoveUnusedFields;
wordDoc.MailMerge.FieldMergingCallback = new HandleMerge();
wordDoc.MailMerge.Execute(dt);
rVal = wordDoc.SaveToByteArray();
}
return rVal;
}
public class HandleMerge : IFieldMergingCallback
{
void IFieldMergingCallback.FieldMerging(FieldMergingArgs e)
{
// All merge fields that expect HTML data should be marked with some prefix, e.g. ‘html’.
if (e.FieldValue != null && !string.IsNullOrEmpty(e.FieldValue.ToString()) && e.FieldValue.ToString().ContainsHtml())
{
// Insert the text for this merge field as HTML data, using DocumentBuilder.
DocumentBuilder builder = new DocumentBuilder(e.Document);
builder.MoveToMergeField(e.DocumentFieldName);
builder.InsertHtml((string)e.FieldValue);<span style="color:green;">// The HTML text itself should not be inserted.</span> <span style="color:green;">// We have already inserted it as an HTML.</span> e.Text = <span style="color:#a31515;">""</span>; } } <span style="color:gray;">///</span><span style="color:green;"> </span><span style="color:gray;"><summary></span> <span style="color:gray;">///</span><span style="color:green;"> This is called when mail merge engine encounters Image:XXX merge field in the document.</span> <span style="color:gray;">///</span><span style="color:green;"> You have a chance to return an Image object, file name or a stream that contains the image.</span> <span style="color:gray;">///</span><span style="color:green;"> </span><span style="color:gray;"></summary></span> <span style="color:blue;">void</span> <span style="color:#2b91af;">IFieldMergingCallback</span>.ImageFieldMerging(<span style="color:#2b91af;">ImageFieldMergingArgs</span> e) { <span style="color:green;">// The field value is a byte array, just cast it and create a stream on it.</span> <span style="color:blue;">var</span> bytes = e.FieldValue <span style="color:blue;">as</span> <span style="color:blue;">byte</span>[]; <span style="color:blue;">if</span> (bytes != <span style="color:blue;">null</span>) <span style="color:green;">//it is an image field and the value has bytes!</span> { <span style="color:#2b91af;">MemoryStream</span> imageStream = <span style="color:blue;">new</span> <span style="color:#2b91af;">MemoryStream</span>(bytes); <span style="color:green;">// Now the mail merge engine will retrieve the image from the stream.</span> e.ImageStream = imageStream; } }}
I noticed another user reported better speeds when they merged documents individually and then combined the individual documents into a single document (Free Support Forum - aspose.com) . I tried this and did infact cut my merge time down to somewhere closer to 3 minutes. This is still longer than I would like though.
New psudo code:
foreach (var recommendation in recommendations)
{
file = null;innerFileName = <span style="color:#a31515;">"{0} - {1} Recommendation {2}"</span>.FormatString(recommendation.JudgeRequest.RequesterName, recommendation.JudgeRequest.JudgeName, recommendation.Id.ToString()); innerFileName += <span style="color:#a31515;">".docx"</span>; <span style="color:blue;">var</span> replaceDict = GetReplaceDictionary(recommendation, instSigBytes); <span style="color:blue;">var</span> temp = _Db.RecommendationMergeDocuments.SingleOrDefault(w => w.Id == recommendation.RecommendationMergeDocumentId.Value); <span style="color:blue;">if</span> (temp != <span style="color:blue;">null</span> && temp.TemplateBytes != <span style="color:blue;">null</span>) { <span style="color:blue;">var</span> doc = <span style="color:#2b91af;">AsposeWord</span>.MergeFields(temp.TemplateBytes, replaceDict); recommendation.RecommendationTextDoc = doc; } file = recommendation.RecommendationTextDoc; } <span style="color:blue;">if</span> (file != <span style="color:blue;">null</span> && file.Length > 0) { usedIds.Add(recommendation.Id); filesToMerge.Add(innerFileName, file); } }var c = _Db.SaveChanges();
file = AsposeWord.CombineDocuments(filesToMerge.Values.ToList());// Core.Extensions.ZipArchiveExtensions.CreateZipBytes(filesToMerge);
public static byte[] MergeFields(byte[] sourceDoc, Dictionary<string, object> mergeData)
{
byte[] rVal;
using (var wordByteStream = new MemoryStream(sourceDoc))
{
Document wordDoc = new Document(wordByteStream);
wordDoc.MailMerge.CleanupOptions = Aspose.Words.Reporting.MailMergeCleanupOptions.RemoveUnusedFields;
wordDoc.MailMerge.FieldMergingCallback = new HandleMerge();
wordDoc.MailMerge.Execute(mergeData.Keys.ToArray(), mergeData.Values.ToArray());
rVal = wordDoc.SaveToByteArray();
}
return rVal;
}<span style="color:blue;">public</span> <span style="color:blue;">static</span> <span style="color:blue;">byte</span>[] CombineDocuments(<span style="color:#2b91af;">List</span><<span style="color:blue;">byte</span>[]> docs) { <span style="color:blue;">byte</span>[] rVal; <span style="color:#2b91af;">Document</span> combinedDoc = <span style="color:blue;">null</span>; <span style="color:#2b91af;">Document</span> sourceDoc = <span style="color:blue;">null</span>; <span style="color:blue;">foreach</span> (<span style="color:blue;">var</span> docBytes <span style="color:blue;">in</span> docs) { <span style="color:blue;">using</span> (<span style="color:blue;">var</span> wordByteStream = <span style="color:blue;">new</span> <span style="color:#2b91af;">MemoryStream</span>(docBytes)) { <span style="color:blue;">if</span> (combinedDoc == <span style="color:blue;">null</span>) { combinedDoc = <span style="color:blue;">new</span> <span style="color:#2b91af;">Document</span>(wordByteStream); } <span style="color:blue;">else</span> { sourceDoc = <span style="color:blue;">new</span> <span style="color:#2b91af;">Document</span>(wordByteStream); combinedDoc.AppendDocument(sourceDoc, <span style="color:#2b91af;">ImportFormatMode</span>.KeepSourceFormatting); } } } rVal = combinedDoc.SaveToByteArray(); <span style="color:blue;">return</span> rVal; }</pre><br> </div>