Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions src/main/java/org/verapdf/as/warnings/StringWarnings.java
Original file line number Diff line number Diff line change
@@ -0,0 +1,7 @@
package org.verapdf.as.warnings;

public class StringWarnings {
public static final String NOT_ASCII_LETTER = "Text string language escape sequence contains not ASCII letter";
public static final String INVALID_LANGUAGE_ESCAPE_SEQUENCE_LENGTH = "Text string language escape sequence has invalid length";
public static final String NOT_SUPPORTED_UTF16LE_ENCODING = "String object uses encoding UTF16-LE not supported by PDF";
}
71 changes: 70 additions & 1 deletion src/main/java/org/verapdf/cos/COSString.java
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@
*/
package org.verapdf.cos;

import org.verapdf.as.warnings.StringWarnings;
import org.verapdf.cos.filters.COSFilterASCIIHexEncode;
import org.verapdf.cos.visitor.ICOSVisitor;
import org.verapdf.cos.visitor.IVisitor;
Expand Down Expand Up @@ -111,7 +112,7 @@ public String getString() {
return new String(value, 2, value.length - 2, StandardCharsets.UTF_16BE);
}
if ((value[0] & 0xFF) == 0xFF && (value[1] & 0xFF) == 0xFE) {
LOGGER.log(Level.WARNING, "String object uses encoding UTF16-LE not supported by PDF");
LOGGER.log(Level.WARNING, StringWarnings.NOT_SUPPORTED_UTF16LE_ENCODING);
}
}
if (value.length >= 3) {
Expand Down Expand Up @@ -221,17 +222,85 @@ public String getASCIIString() {
public boolean isTextString() {
if (value.length > 2) {
if ((value[0] & 0xFF) == 0xFE && (value[1] & 0xFF) == 0xFF) {
checkUTF16BEEscapeSequence(value);
return true;
}
}
if (value.length > 3) {
if ((value[0] & 0xFF) == 0xEF && (value[1] & 0xFF) == 0xBB && (value[2] & 0xFF) == 0xBF) {
checkUTF8EscapeSequence(value);
return true;
}
}
return PDFDocEncoding.isPDFDocEncodingString(value);
}

private boolean checkUTF16BEEscapeSequence(byte[] value) {
for (int i = 0; i < value.length; i++) {
if (i + 1 < value.length && value[i] == 0x00 && value[i + 1] == 0x1B) {
if (i + 5 < value.length && value[i + 4] == 0x00 && value[i + 5] == 0x1B) {
if (isASCIILetter(value[i + 2]) && isASCIILetter(value[i + 3])) {
i += 5;
continue;
} else {
LOGGER.log(Level.WARNING, StringWarnings.NOT_ASCII_LETTER);
return false;
}
}

if (i + 7 < value.length && value[i + 6] == 0x00 && value[i + 7] == 0x1B) {
if (isASCIILetter(value[i + 2]) && isASCIILetter(value[i + 3]) &&
isASCIILetter(value[i + 4]) && isASCIILetter(value[i + 5])) {
i += 7;
continue;
} else {
LOGGER.log(Level.WARNING, StringWarnings.NOT_ASCII_LETTER);
return false;
}
}
LOGGER.log(Level.WARNING,StringWarnings.INVALID_LANGUAGE_ESCAPE_SEQUENCE_LENGTH);
return false;
}
}
return true;
}

private boolean checkUTF8EscapeSequence(byte[] value) {
for (int i = 0; i < value.length; i++) {
if (value[i] == 0x1B) {
if (i + 3 < value.length && value[i + 3] == 0x1B) {
if (isASCIILetter(value[i + 1]) && isASCIILetter(value[i + 2])) {
i += 3;
continue;
} else {
LOGGER.log(Level.WARNING, StringWarnings.NOT_ASCII_LETTER);
return false;
}
}

if (i + 5 < value.length && value[i + 5] == 0x1B) {
if (isASCIILetter(value[i + 1]) && isASCIILetter(value[i + 2]) &&
isASCIILetter(value[i + 3]) && isASCIILetter(value[i + 4])) {
i += 5;
continue;
} else {
LOGGER.log(Level.WARNING, StringWarnings.NOT_ASCII_LETTER);
return false;
}
}

LOGGER.log(Level.WARNING,StringWarnings.INVALID_LANGUAGE_ESCAPE_SEQUENCE_LENGTH);
return false;
}
}
return true;
}

private boolean isASCIILetter(byte b) {
int c = b & 0xFF;
return (c >= 0x41 && c <= 0x5A) || (c >= 0x61 && c <= 0x7A);
}

protected String toLitString() {
StringBuilder result = new StringBuilder();
result.append('(');
Expand Down