improve code formatting and enhance readability in auto-redact.html, RedactionService, TextDecodingHelper, TextEncodingHelper, and TextFinder

Signed-off-by: Balázs Szücs <bszucs1209@gmail.com>

app/core/src/main/java/stirling/software/SPDF/pdf/TextFinder.java

@@ -3,8 +3,6 @@ package stirling.software.SPDF.pdf;
 import java.io.IOException;
 import java.util.ArrayList;
 import java.util.List;
-import java.util.regex.Matcher;
-import java.util.regex.Pattern;
 
 import org.apache.pdfbox.pdmodel.PDPage;
 import org.apache.pdfbox.text.PDFTextStripper;
@@ -71,26 +69,33 @@ public class TextFinder extends PDFTextStripper {
             super.endPage(page);
             return;
         }
-        String regex = this.useRegex ? processedSearchTerm : "\\Q" + processedSearchTerm + "\\E";
+        // Build patterns using unified utility for consistency
-        if (this.wholeWordSearch) {
+        List<java.util.regex.Pattern> patterns =
-            if (processedSearchTerm.length() == 1
+                stirling.software.SPDF.utils.text.TextFinderUtils.createOptimizedSearchPatterns(
-                    && Character.isDigit(processedSearchTerm.charAt(0))) {
+                        java.util.Collections.singleton(processedSearchTerm),
-                regex = "(?<![\\w])(?<!\\d[\\.,])" + regex + "(?![\\w])(?![\\.,]\\d)";
+                        this.useRegex,
-            } else if (processedSearchTerm.length() == 1) {
+                        this.wholeWordSearch);
-                regex = "(?<![\\w])" + regex + "(?![\\w])";
+        java.util.regex.Matcher matcher = null;
-            } else {
+        java.util.regex.Pattern activePattern = null;
-                regex = "\\b" + regex + "\\b";
+        for (java.util.regex.Pattern p : patterns) {
+            matcher = p.matcher(text);
+            if (matcher
+                    .find()) { // prime by checking has at least one match; we will re-iterate below
+                activePattern = p;
+                break;
             }
         }
-
+        if (activePattern == null) {
-        Pattern pattern = Pattern.compile(regex, Pattern.CASE_INSENSITIVE | Pattern.UNICODE_CASE);
+            super.endPage(page);
-        Matcher matcher = pattern.matcher(text);
+            return;
+        }
+        matcher = activePattern.matcher(text);
 
         log.debug(
-                "Searching for '{}' in page {} with regex '{}' (wholeWord: {}, useRegex: {})",
+                "Searching for '{}' in page {} with pattern '{}' (wholeWord: {}, useRegex: {})",
                 processedSearchTerm,
                 getCurrentPageNo(),
-                regex,
+                activePattern,
                 wholeWordSearch,
                 useRegex);
 

app/core/src/main/java/stirling/software/SPDF/service/RedactionService.java

@@ -86,7 +86,7 @@ public class RedactionService {
     private static final Set<String> TEXT_SHOWING_OPERATORS = Set.of("Tj", "TJ", "'", "\"");
     private static final COSString EMPTY_COS_STRING = new COSString("");
     private static final int MAX_SWEEPS = 3;
-    private static final Pattern PATTERN = Pattern.compile(".*(hoepap|temp|generated).*");
+    private static final Pattern PATTERN = Pattern.compile(".*(placeholder|temp|generated).*");
     private boolean aggressiveMode = false;
     private Map<Integer, List<AggressiveSegMatch>> aggressiveSegMatches = null;
     private final CustomPDFDocumentFactory pdfDocumentFactory;
@@ -2409,11 +2409,9 @@ public class RedactionService {
                                 textSegments.indexOf(task.segment), Collections.emptyList());
 
                 if (task.segment.tokenIndex >= newTokens.size()) {
-
                     continue;
                 }
                 if (task.segment.getText() == null || task.segment.getText().isEmpty()) {
-
                     continue;
                 }
 
@@ -2430,19 +2428,22 @@ public class RedactionService {
 
     private static String extractStringWithFallbacks(COSString cosString, PDFont font) {
         if (cosString == null) return "";
-
         try {
+            // Prefer font-guided decoding for correctness
+            if (font != null) {
+                String enhanced =
+                        TextDecodingHelper.decodeCharactersEnhanced(font, cosString.getBytes());
+                if (enhanced != null && !isGibberish(enhanced)) return enhanced;
+            }
+            // Fallback to COSString raw string if it seems valid
             String text = cosString.getString();
             if (!text.trim().isEmpty() && !isGibberish(text)) return text;
-
+            // Fallback: try basic font-based extraction
             if (font != null) {
                 String fontBasedText = tryFontBasedExtraction(cosString, font);
                 if (fontBasedText != null && !isGibberish(fontBasedText)) return fontBasedText;
             }
-
+            // Last resort: sanitized raw
-            String encodingFallback = tryEncodingFallbacks(cosString);
-            if (encodingFallback != null && !isGibberish(encodingFallback)) return encodingFallback;
-
             return sanitizeText(text);
         } catch (Exception e) {
             return "\uFFFD";

app/core/src/main/java/stirling/software/SPDF/utils/text/TextDecodingHelper.java

@@ -123,9 +123,8 @@ public class TextDecodingHelper {
                 }
             } catch (Exception ignored) {
             }
-            if (ch == null || !isPrintable(ch)) {
+            if (ch == null) {
-                // Handle problematic character codes specifically
+                return null; // fail fast if undecodable via font tables
-                ch = "<EFBFBD>";
             }
             out.append(ch);
             i += consumed;
@@ -250,16 +249,8 @@ public class TextDecodingHelper {
     }
 
     public String handleProblematicCharacterCode(int code, PDFont font) {
-        if (code >= PROBLEMATIC_CODE_LOWER_BOUND && code <= PROBLEMATIC_CODE_UPPER_BOUND) {
+        // For correctness, avoid speculative remapping. Return replacement char only when needed.
-            int adjustedCode = code - PROBLEMATIC_CODE_LOWER_BOUND;
+        return "\uFFFD";
-            if (adjustedCode >= ASCII_LOWER_BOUND) {
-                return String.valueOf((char) adjustedCode);
-            }
-            if (font != null && font.getName() != null && font.getName().contains("+")) {
-                return mapSubsetCharacter(adjustedCode);
-            }
-        }
-        return "<EFBFBD>";
     }
 
     public String mapSubsetCharacter(int code) {
@@ -267,7 +258,8 @@ public class TextDecodingHelper {
             return String.valueOf((char) code);
         }
         if (code >= EXTENDED_ASCII_LOWER_BOUND && code <= EXTENDED_ASCII_UPPER_BOUND) {
-            return String.valueOf((char) (code - 128));
+            // Do not alter code point arbitrarily; extended ASCII maps directly for correctness.
+            return String.valueOf((char) code);
         }
         return null;
     }

app/core/src/main/java/stirling/software/SPDF/utils/text/TextEncodingHelper.java

@@ -20,82 +20,46 @@ public class TextEncodingHelper {
         if (font == null || text == null) {
             return false;
         }
-
         if (text.isEmpty()) {
             return true;
         }
-
+        // Strict: every code point must be encodable by the font
+        for (int i = 0; i < text.length(); ) {
+            int cp = text.codePointAt(i);
+            String ch = new String(Character.toChars(cp));
             try {
-            byte[] encoded = font.encode(text);
+                byte[] encoded = font.encode(ch);
-            if (encoded.length > 0) {
+                if (encoded == null || encoded.length == 0) {
+                    return false;
+                }
+            } catch (Exception ex) {
+                return false;
+            }
+            i += Character.charCount(cp);
+        }
         return true;
     }
-        } catch (Exception e) {
-        }
-
-        return validateAsCodePointArray(font, text);
-    }
 
     private boolean validateAsCodePointArray(PDFont font, String text) {
         if (text == null || text.isEmpty()) {
             return true;
         }
-
-        int totalCodePoints = 0;
-        int successfulCodePoints = 0;
-
         for (int i = 0; i < text.length(); ) {
             int codePoint = text.codePointAt(i);
             String charStr = new String(Character.toChars(codePoint));
-            totalCodePoints++;
-
             try {
                 byte[] charEncoded = font.encode(charStr);
-                if (charEncoded.length > 0) {
+                if (charEncoded == null || charEncoded.length == 0) {
-                    try {
+                    return false;
-                        float charWidth = font.getStringWidth(charStr);
-                        if (charWidth >= 0) {
-                            successfulCodePoints++;
                 }
             } catch (Exception e) {
-                        try {
+                return false;
-                            if (canDecodeCharacter(font, charStr)) {
-                                successfulCodePoints++;
             }
-                        } catch (Exception e2) {
-                        }
-                    }
-                } else {
-                    try {
-                        if (canDecodeCharacter(font, charStr)) {
-                            successfulCodePoints++;
-                        }
-                    } catch (Exception e) {
-                    }
-                }
-            } catch (Exception e) {
-                try {
-                    if (canDecodeCharacter(font, charStr)) {
-                        successfulCodePoints++;
-                    }
-                } catch (Exception e2) {
-                    if (isBasicCharacter(codePoint)) {
-                        successfulCodePoints++;
-                    }
-                }
-            }
-
             i += Character.charCount(codePoint);
         }
-
-        if (totalCodePoints == 0) {
         return true;
     }
 
-        double successRate = (double) successfulCodePoints / totalCodePoints;
-        return successRate >= 0.1;
-    }
-
     private boolean canDecodeCharacter(PDFont font, String charStr) {
         if (font == null || charStr == null || charStr.isEmpty()) {
             return false;
@@ -128,27 +92,18 @@ public class TextEncodingHelper {
         if (font == null || text == null) {
             return false;
         }
-
         if (text.isEmpty()) {
             return true;
         }
-
+        // Strict: removable only if we can encode every codepoint and measure width
-        if (isSimpleCharacter(text)) {
+        if (!canEncodeCharacters(font, text)) return false;
         try {
-                font.encode(text);
             font.getStringWidth(text);
             return true;
         } catch (Exception e) {
-                try {
-                    return canHandleText(font, text);
-                } catch (Exception e2) {
             return false;
         }
     }
-        }
-
-        return isTextFullyRemovable(font, text);
-    }
 
     private boolean canHandleText(PDFont font, String text) {
         if (font == null || text == null) {
@@ -197,70 +152,16 @@ public class TextEncodingHelper {
     }
 
     public boolean isTextFullyRemovable(PDFont font, String text) {
-        if (font == null || text == null) {
+        if (font == null || text == null) return false;
-            return false;
+        if (text.isEmpty()) return true;
-        }
+        if (!canEncodeCharacters(font, text)) return false;
-
-        if (text.isEmpty()) {
-            return true;
-        }
-
-        try {
-            if (!canEncodeCharacters(font, text)) {
-                return false;
-            }
-
         try {
             float width = font.getStringWidth(text);
-                if (width < 0) {
+            return width >= 0;
-                    return false;
-                }
-            } catch (Exception e) {
-                try {
-                    if (!canCalculateTextWidth(font, text)) {
-                        return false;
-                    }
-                } catch (Exception e2) {
-                    return false;
-                }
-            }
-
-            try {
-                if (font.getFontDescriptor() == null) {
-                    try {
-                        return canHandleWithoutDescriptor(font, text);
         } catch (Exception e) {
             return false;
         }
     }
-            } catch (Exception e) {
-                try {
-                    return canHandleWithoutDescriptor(font, text);
-                } catch (Exception e2) {
-                    return false;
-                }
-            }
-
-            try {
-                font.getFontDescriptor().getFontBoundingBox();
-            } catch (Exception e) {
-                try {
-                    return canHandleWithoutBoundingBox(font, text);
-                } catch (Exception e2) {
-                    return false;
-                }
-            }
-
-            return true;
-
-        } catch (Exception e) {
-            try {
-                return canHandleText(font, text);
-            } catch (Exception e2) {
-                return false;
-            }
-        }
-    }
 
     private boolean canCalculateTextWidth(PDFont font, String text) {
         if (font == null || text == null) {
@@ -381,45 +282,19 @@ public class TextEncodingHelper {
     }
 
     public boolean fontSupportsCharacter(PDFont font, String character) {
-        if (font == null || character == null) {
+        if (font == null || character == null) return false;
-            return false;
+        if (character.isEmpty()) return true;
-        }
-
-        if (character.isEmpty()) {
-            return true;
-        }
-
-        try {
-            byte[] encoded = font.encode(character);
-            if (encoded.length > 0) {
-                try {
-                    float width = font.getStringWidth(character);
-                    if (width >= 0) {
-                        return true;
-                    }
-                } catch (Exception e) {
-                }
-                return true;
-            }
-        } catch (Exception e) {
-        }
-
-        try {
-            if (canDecodeCharacter(font, character)) {
-                return true;
-            }
-        } catch (Exception e) {
-        }
-
         for (int i = 0; i < character.length(); ) {
-            int codePoint = character.codePointAt(i);
+            int cp = character.codePointAt(i);
-            if (isBasicCharacter(codePoint)) {
+            String ch = new String(Character.toChars(cp));
-                i += Character.charCount(codePoint);
+            try {
-                continue;
+                byte[] encoded = font.encode(ch);
-            }
+                if (encoded == null || encoded.length == 0) return false;
+            } catch (Exception e) {
                 return false;
             }
-
+            i += Character.charCount(cp);
+        }
         return true;
     }
 

app/core/src/main/resources/templates/security/auto-redact.html

@@ -115,12 +115,15 @@
                 <label class="form-check-label" for="keepLayout" th:text="#{autoRedact.keepLayoutLabel}">Remove Text & Cover (Preserve Layout)</label>
                 <small class="form-text text-muted d-block mt-1" id="keep-desc" th:text="#{autoRedact.keepLayoutDescription}">Removes the underlying text and places a redaction box in its place, preserving the document's original layout.</small>
               </div>
-
-                            <div class="form-check">
-                                <input aria-describedby="guarantee-desc" class="form-check-input" id="guaranteeRedaction" name="convertPDFToImage" type="checkbox">
-                                <label class="form-check-label" for="guaranteeRedaction" th:text="#{autoRedact.pdfImageLabel}">PDF image</label>
-                                <small class="form-text text-muted d-block mt-1" id="guarantee-desc" th:text="#{autoRedact.pdfImageDescription}">For maximum security, uses an image-based method to ensure text is unrecoverable. May slightly affect document quality.</small>
             </div>
+
+            <!-- PDF Image checkbox - moved outside redaction-options-group to use plain styling -->
+            <div class="mb-3">
+              <div class="form-check">
+                <input aria-describedby="guarantee-desc" class="form-check-input" id="convertPDFToImage" name="convertPDFToImage" type="checkbox">
+                <label class="form-check-label" for="convertPDFToImage" th:text="#{autoRedact.convertPDFToImageLabel}">Convert PDF to PDF-Image</label>
+              </div>
+              <small class="form-text text-muted d-block mt-1" id="guarantee-desc" th:text="#{autoRedact.pdfImageDescription}">For maximum security, uses an image-based method to ensure text is unrecoverable. May slightly affect document quality.</small>
             </div>
 
             <br>
@@ -154,8 +157,8 @@
               <label class="form-label" for="languages">OCR Languages</label>
               <div id="languages">
                 <div class="form-check" th:each="language, iterStat : ${languages}">
-                                    <input onchange="handleLangSelection()" required th:checked="${language == 'eng'}" th:id="${'language-' + language}" th:name="languages" th:value="${language}" type="checkbox" />
+                  <input class="form-check-input" onchange="handleLangSelection()" required th:checked="${language == 'eng'}" th:id="${'language-' + language}" th:name="languages" th:value="${language}" type="checkbox" />
-                     <label th:attr="data-lang-code=${language}, data-lang-name=#{'lang.' + language}"
+                  <label class="form-check-label" th:attr="data-lang-code=${language}, data-lang-name=#{'lang.' + language}"
                          th:for="${'language-' + language}"
                          th:text="${language}"></label>
                 </div>
@@ -273,7 +276,7 @@
   document.addEventListener('DOMContentLoaded', function () {
     const redactionModeRadios = document.querySelectorAll('input[name="redactionMode"]');
     const aggressiveModeHidden = document.getElementById('aggressiveMode');
-        const guaranteeRedactionCheckbox = document.getElementById('guaranteeRedaction');
+    const convertPDFToImageCheckbox = document.getElementById('convertPDFToImage');
     const defaultColor = document.getElementById('defaultColor');
 
     function updateMode() {
@@ -285,10 +288,10 @@
         // Handle PDF image checkbox based on selection
         if (selectedMode.value === 'visual') {
           // Visual mode automatically enables PDF image for maximum security
-                    guaranteeRedactionCheckbox.checked = true;
+          convertPDFToImageCheckbox.checked = true;
         } else {
           // Delete Text and Keep Layout modes disable PDF image
-                    guaranteeRedactionCheckbox.checked = false;
+          convertPDFToImageCheckbox.checked = false;
         }
 
         // Highlight selected card