Guides

Search Text and Extract Range Content

Text handling in the viewer usually falls into two different tasks:

1. search the entire document for a keyword and locate the matched results
2. extract text from a specific page range for summaries, previews, or secondary processing

This page covers both capabilities, but they serve different purposes:

• keyword search is about locating matches and navigating to them
• range-based extraction is about reading text from a defined region

If the goal is to retrieve text that was manually selected inside the viewer, see Get the Selected Content.

Search for keywords across the document

Keyword search is commonly used in the following scenarios:

• locating specific terms inside long documents
• building a search results list
• highlighting results in the current reading view

Create the searcher and iterate through pages

The search flow usually includes these steps:

1. create a result collection
2. get ITextSearcher
3. set the keyword and search options
4. iterate through pages and save matched results

java

List<CPDFTextRange> searchTextInfoList = new ArrayList<>();
ITextSearcher textSearcher = readerView.getTextSearcher();

String keywords = "ComPDF";
textSearcher.setSearchConfig(
    keywords,
    CPDFTextSearcher.PDFSearchOptions.PDFSearchCaseSensitive
);

for (int i = 0; i < document.getPageCount(); i++) {
  CPDFPage page = document.pageAtIndex(i);
  CPDFTextPage textPage = page.getTextPage();
  if (textPage == null || !textPage.isValid()) {
    continue;
  }

  List<CPDFTextRange> searchPageContent = textSearcher.searchKeyword(i);
  if (!searchPageContent.isEmpty()) {
    searchTextInfoList.addAll(searchPageContent);
  }
}

kotlin

val searchTextInfoList = mutableListOf<CPDFTextRange>()
val textSearcher: ITextSearcher = readerView.getTextSearcher()

val keywords = "ComPDF"
textSearcher.setSearchConfig(
  keywords,
  CPDFTextSearcher.PDFSearchOptions.PDFSearchCaseSensitive
)

for (i in 0 until document.pageCount) {
  val page = document.pageAtIndex(i)
  val textPage = page.textPage
  if (textPage == null || !textPage.isValid) {
    continue
  }

  val searchPageContent = textSearcher.searchKeyword(i)
  if (searchPageContent.isNotEmpty()) {
    searchTextInfoList.addAll(searchPageContent)
  }
}

Choose search options

setSearchConfig(...) supports these common options:

Option	Description	Value
PDFSearchCaseInsensitive	Case-insensitive match	0
PDFSearchCaseSensitive	Case-sensitive match	1
PDFSearchMatchWholeWord	Match the whole word	2

Search options should be chosen based on the business goal. Full-text search usually fits case-insensitive matching, while exact term matching is better suited to case-sensitive or whole-word matching.

Read matched text and surrounding context

If the screen must show search summaries, preview snippets, or surrounding text, the matched content can be extracted from CPDFTextRange.

The following example shows how to read the matched text and a short context window.

int pageIndex = 0;
List<CPDFTextRange> searchPageContent = textSearcher.searchKeyword(pageIndex);
if (searchPageContent.isEmpty()) {
    return;
}

CPDFTextRange textRange = searchPageContent.get(0);
CPDFPage page = document.pageAtIndex(pageIndex);
CPDFTextPage textPage = page.getTextPage();
String text = textPage.getText(textRange);

int targetStart = textRange.location - 20;
int length;
if (targetStart > 0) {
  length = textRange.length + 40;
} else {
  length = textRange.length + 40 + targetStart;
  targetStart = 0;
}
CPDFTextRange targetTextRange = new CPDFTextRange(targetStart, length);
String contextText = textPage.getText(targetTextRange);

When handling single-page results, make sure that the CPDFTextRange belongs to the current page's CPDFTextPage to avoid cross-page reads.

Highlight and navigate through search results

When matched results need to be highlighted inside the viewer, the following methods are available.

Highlight a specific result

int pageIndex = 0;
int textRangeIndex = 0;
textSearcher.searchBegin(pageIndex, textRangeIndex);
readerView.invalidateAllChildren();

Go to the previous result

textSearcher.searchBackward();

Go to the next result

textSearcher.searchForward();

End the current search flow

textSearcher.cancelSearch();

Extract text from a page range

In addition to keyword search, CPDFPage and CPDFTextPage also support extracting text from a rectangular region. This is useful for:

• page summary extraction
• custom content analysis
• fixed-area text reading

The following example defines a rectangle in page coordinates, then converts it into the coordinate range used for text extraction on the current page.

java

CPDFPage pdfPage = document.pageAtIndex(0);
CPDFTextPage pdfTextPage = pdfPage.getTextPage();

RectF selectRect = new RectF(0f, 0f, 500f, 500f);
selectRect = pdfPage.convertRectFromPage(
    false,
    pdfPage.getSize().width(),
    pdfPage.getSize().height(),
    selectRect
);

CPDFTextSelection[] textSelectionArr = pdfTextPage.getSelectionsByLineForRect(selectRect);

for (CPDFTextSelection textSelection : textSelectionArr) {
    if (textSelection == null) {
        continue;
    }

    String text = pdfTextPage.getText(textSelection.getTextRange());
}

kotlin

val pdfPage = document.pageAtIndex(0)
val pdfTextPage = pdfPage.textPage

var selectRect = pdfPage.convertRectFromPage(
  false,
  pdfPage.size.width(),
  pdfPage.size.height(),
  RectF(0f, 0f, 500f, 500f)
)
val textSelectionArr = pdfTextPage.getSelectionsByLineForRect(selectRect)

for (textSelection in textSelectionArr) {
  val text = pdfTextPage.getText(textSelection.textRange)
}

This extracts text from a defined rectangular range. It is not the same as retrieving text that was manually selected through a long press in the viewer.

Keep these points in mind

• Full-text search is meant for locating keywords. Range extraction is meant for reading text from a defined area.
• Before reading matched results, check whether the result collection is empty.
• If the screen must render a search result list, it is useful to preserve the mapping between pageIndex and CPDFTextRange.
• If the goal is to retrieve text that was manually selected in the viewer, see Get the Selected Content.

Related pages

• For retrieving manually selected text inside the viewer, see Get the Selected Content.
• For preparing page text for reflow-style rendering, see Text Reflow.