Search document collections with GroupDocs.Search Leave feedback
On this page
GroupDocs.Search is the one API in the suite that works on a collection rather than a single file. You build an index over a folder of documents once, then query it as often as you like — the index does the reading, so queries stay fast no matter how large the corpus grows.
That makes it a two-phase API: index, then search. The entry point is Index, not a “searcher”.
NoteThe code on this page is complete and ready to run against GroupDocs.Total for Java. Downloadable output samples for these examples are being added — in the meantime, the GroupDocs.Search for Java documentation carries the full set of runnable examples.
The index lives in a folder of its own, separate from the documents. add reads a folder of documents into it; search queries what has been indexed.
import com.groupdocs.search.Index;
import com.groupdocs.search.results.FoundDocument;
import com.groupdocs.search.results.SearchResult;
import java.io.File;
// The index is a folder the library owns; keep it out of your documents folder
try (Index index = new Index("search-build-index/index")) {
// Index every supported document in the folder
index.add("archive");
SearchResult result = index.search("throughput");
System.out.println("Documents found: " + result.getDocumentCount());
System.out.println("Total occurrences: " + result.getOccurrenceCount());
for (int i = 0; i < result.getDocumentCount(); i++) {
FoundDocument document = result.getFoundDocument(i);
// getFilePath() is absolute; the file name is the useful part here
String fileName = new File(document.getDocumentInfo().getFilePath()).getName();
System.out.println(" " + fileName + " (" + document.getOccurrenceCount() + " occurrences)");
}
}
This example indexes the archive folder — three working documents from the engagement: discovery-findings.docx, steering-minutes.docx and phase-two-proposal.pdf. Put them in a folder named archive next to your program.
Exact matching fails on scanned documents and typos. Fuzzy search accepts a similarity level — how many character edits away a word may be and still count as a hit.
import com.groupdocs.search.Index;
import com.groupdocs.search.options.SearchOptions;
import com.groupdocs.search.options.TableDiscreteFunction;
import com.groupdocs.search.results.SearchResult;
try (Index index = new Index("search-fuzzy/index")) {
index.add("archive");
SearchOptions options = new SearchOptions();
options.getFuzzySearch().setEnabled(true);
options.getFuzzySearch().setFuzzyAlgorithm(new TableDiscreteFunction(3));
// Matches "throughput" even when spelled "througput" or "throughtput"
SearchResult result = index.search("throughput", options);
System.out.println("Documents found: " + result.getDocumentCount());
System.out.println("Total occurrences: " + result.getOccurrenceCount());
}
This example indexes the archive folder — three working documents from the engagement: discovery-findings.docx, steering-minutes.docx and phase-two-proposal.pdf. Put them in a folder named archive next to your program.
Query text supports AND, OR and NOT, so you can express the same conditions a user would type into a search box. All three archived documents mention throughput; only the phase-two proposal mentions a forecast, so this query returns the other two.
import com.groupdocs.search.Index;
import com.groupdocs.search.results.FoundDocument;
import com.groupdocs.search.results.SearchResult;
import java.io.File;
try (Index index = new Index("search-boolean/index")) {
index.add("archive");
SearchResult result = index.search("throughput AND NOT forecast");
System.out.println("Documents found: " + result.getDocumentCount());
for (int i = 0; i < result.getDocumentCount(); i++) {
FoundDocument document = result.getFoundDocument(i);
System.out.println(" " + new File(document.getDocumentInfo().getFilePath()).getName());
}
}
This example indexes the archive folder — three working documents from the engagement: discovery-findings.docx, steering-minutes.docx and phase-two-proposal.pdf. Put them in a folder named archive next to your program.
NoteAn index is durable. PointIndexat an existing index folder and it reopens what is already there rather than rebuilding — calladdagain only when documents change, andupdateto refresh ones that were modified.
GroupDocs.Search also offers phrase, wildcard, regex, faceted and date-range queries, highlights matches in the source document, indexes asynchronously with progress events, runs OCR over scanned pages, supports synonyms, homophones and stop words, and can distribute an index across a search network.
- GroupDocs.Search for Java documentation — the full product guide
- Supported file formats
- API reference
Was this page helpful?
Any additional feedback you'd like to share with us?
Please tell us how we can improve this page.
Thank you for your feedback!
We value your opinion. Your feedback will help us improve our documentation.
