Skip to content

PDFBOX-6251 usecmap cidrange precedence #515

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
pcorless wants to merge 3 commits into
base: trunk
Choose a base branch
from
+371 −95
Open

PDFBOX-6251 usecmap cidrange precedence #515

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
8f23577
PDFBOX-6251: Let a CMap's own cid mappings win over the ones it uses
pcorless Sep 5, 2026
691ae76
PDFBOX-6251 rework parent cmap lookup
pcorless Sep 5, 2026
cd53a45
PDFBOX-6251 rework for -1 or 0 base cid lookups.
pcorless Sep 5, 2026
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
186 changes: 91 additions & 95 deletions fontbox/src/main/java/org/apache/fontbox/cmap/CMap.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,6 +65,9 @@ public class CMap
private final Map<Integer, Map<Integer, Integer>> codeToCid = new HashMap<>();
private final List<CIDRange> codeToCidRanges = new ArrayList<>();

// the CMaps this one inherits from through the usecmap operator, see useCmap
private final List<CMap> parentCMaps = new ArrayList<>();

// inverted map
private final Map <String, byte[]> unicodeToByteCodes = new HashMap<>();

Expand All @@ -80,12 +83,13 @@ public class CMap

/**
* This will tell if this cmap has any CID mappings.
*
*
* @return true If there are any CID mappings, false otherwise.
*/
public boolean hasCIDMappings()
{
return !codeToCid.isEmpty() || !codeToCidRanges.isEmpty();
return !codeToCid.isEmpty() || !codeToCidRanges.isEmpty()
|| parentCMaps.stream().anyMatch(CMap::hasCIDMappings);
}

/**
Expand Down Expand Up @@ -237,22 +241,12 @@ private static int toInt(byte[] data, int dataLen)
*/
public int toCID(byte[] code)
{
if (!hasCIDMappings() || code.length < minCidLength || code.length > maxCidLength)
{
return 0;
}
Integer cid = null;
Map<Integer, Integer> codeToCidMap = codeToCid.get(code.length);
if (codeToCidMap != null)
{
cid = codeToCidMap.get(toInt(code));
}
return cid != null ? cid : toCIDFromRanges(code);
return toCID(toInt(code), code.length);
}

/**
* Returns the CID for the given character code.
*
*
* This method exists for convenience. It may return false values as the origin byte length of the input value is
* unknown and the mapping for some input values aren't unique. <br>
* Example:<br>
Expand All @@ -264,17 +258,15 @@ public int toCID(byte[] code)
*/
public int toCID(int code)
{
if (!hasCIDMappings())
for (int length = minCidLength; length <= maxCidLength; length++)
{
return 0;
}
int cid = 0;
int length = minCidLength;
while (cid == 0 && (length <= maxCidLength))
{
cid = toCID(code, length++);
int cid = findCID(code, length);
if (cid != -1)
{
return cid;
}
}
return cid;
return 0;
}

/**
Expand All @@ -286,57 +278,67 @@ public int toCID(int code)
*/
public int toCID(int code, int length)
{
if (!hasCIDMappings() || length < minCidLength || length > maxCidLength)
{
return 0;
}
Integer cid = null;
Map<Integer, Integer> codeToCidMap = codeToCid.get(length);
if (codeToCidMap != null)
{
cid = codeToCidMap.get(code);
}
return cid != null ? cid : toCIDFromRanges(code, length);
int cid = findCID(code, length);
return cid != -1 ? cid : 0;
}

/**
* Returns the CID for the given character code.
* Returns the CID this CMap, or one of the CMaps it inherits from, maps the given character code
* to, or -1 if none of them maps it. CID 0 is the .notdef glyph and a CMap may map a code to it
* deliberately, so "mapped to 0" has to be told apart from "not mapped" while the usecmap chain
* is walked. The public toCID methods report both as 0.
*
* @param code character code
* @return CID
* @param code character code
* @param length the origin byte length of the code
* @return CID, or -1 if neither this CMap nor any it inherits from maps the code
*/

private int toCIDFromRanges(int code, int length)
private int findCID(int code, int length)
{
for (CIDRange range : codeToCidRanges)
if (length < minCidLength || length > maxCidLength)
{
int ch = range.map(code, length);
if (ch != -1)
return -1;
}
Map<Integer, Integer> codeToCidMap = codeToCid.get(length);
Integer cid = codeToCidMap != null ? codeToCidMap.get(code) : null;
if (cid != null)
{
return cid;
}
int cidFromRange = toCIDFromRanges(code, length);
if (cidFromRange != -1)
{
return cidFromRange;
}
// this CMap doesn't map the code itself, so ask the ones it inherits from
for (CMap parentCMap : parentCMaps)
{
int parentCid = parentCMap.findCID(code, length);
if (parentCid != -1)
{
return ch;
return parentCid;
}
}
return 0;
return -1;
}

/**
* Returns the CID for the given character code.
* Returns the CID, the CID ranges of this CMap map the given character code to.
*
* @param code character code
* @return CID
* @param code character code
* @param length the origin byte length of the code
* @return CID, or -1 if no range covers the code
*/

private int toCIDFromRanges(byte[] code)
private int toCIDFromRanges(int code, int length)
{
for (CIDRange range : codeToCidRanges)
{
int ch = range.map(code);
int ch = range.map(code, length);
if (ch != -1)
{
return ch;
}
}
return 0;
return -1;
}

/**
Expand Down Expand Up @@ -442,11 +444,11 @@ void addCodespaceRange( CodespaceRange range )
maxCodeLength = Math.max(maxCodeLength, range.getCodeLength());
minCodeLength = Math.min(minCodeLength, range.getCodeLength());
}

/**
* Implementation of the usecmap operator. This will
* copy all of the mappings from one cmap to another.
*
*
* @param cmap The cmap to load mappings from.
*/
void useCmap(CMap cmap)
Expand All @@ -459,13 +461,13 @@ void useCmap(CMap cmap)
cmap.charToUnicodeTwoBytes.forEach((k, v) -> unicodeToByteCodes.put(v,
new byte[]{(byte) ((k >>> 8) & 0xFF), (byte) (k & 0xFF)})
);
cmap.charToUnicodeMoreBytes.forEach((k, v) ->
cmap.charToUnicodeMoreBytes.forEach((k, v) ->
{
byte[] bar;
if (k <= 0xFFFFFF)
{
// 3 bytes
bar = new byte[]{(byte) ((k >>> 16) & 0xFF), (byte) ((k >>> 8) & 0xFF),
bar = new byte[]{(byte) ((k >>> 16) & 0xFF), (byte) ((k >>> 8) & 0xFF),
(byte) (k & 0xFF)};
}
else
Expand All @@ -476,15 +478,9 @@ void useCmap(CMap cmap)
}
unicodeToByteCodes.put(v, bar);
});
cmap.codeToCid.forEach((key, value) ->
{
Map<Integer, Integer> existingMapping = codeToCid.putIfAbsent(key, value);
if (existingMapping!=null)
{
existingMapping.putAll(value);
}
});
codeToCidRanges.addAll(cmap.codeToCidRanges);
// The parent is kept, not merged: it is asked only for codes this CMap doesn't map itself,
// so this CMap's own mappings win and a usecmap chain resolves nearest-first. See toCID(int, int).
parentCMaps.add(cmap);
maxCodeLength = Math.max(maxCodeLength, cmap.maxCodeLength);
minCodeLength = Math.min(minCodeLength, cmap.minCodeLength);
maxCidLength = Math.max(maxCidLength, cmap.maxCidLength);
Expand All @@ -495,147 +491,147 @@ void useCmap(CMap cmap)
* Returns the WMode of a CMap.
*
* 0 represents a horizontal and 1 represents a vertical orientation.
*
*
* @return the wmode
*/
public int getWMode()
public int getWMode()
{
return wmode;
}

/**
* Sets the WMode of a CMap.
*
*
* @param newWMode the new WMode.
*/
public void setWMode(int newWMode)
public void setWMode(int newWMode)
{
wmode = newWMode;
}

/**
* Returns the name of the CMap.
*
*
* @return the CMap name.
*/
public String getName()
public String getName()
{
return cmapName;
}

/**
* Sets the name of the CMap.
*
*
* @param name the CMap name.
*/
public void setName(String name)
public void setName(String name)
{
cmapName = name;
}

/**
* Returns the version of the CMap.
*
*
* @return the CMap version.
*/
public String getVersion()
public String getVersion()
{
return cmapVersion;
}

/**
* Sets the version of the CMap.
*
*
* @param version the CMap version.
*/
public void setVersion(String version)
public void setVersion(String version)
{
cmapVersion = version;
}

/**
* Returns the type of the CMap.
*
*
* @return the CMap type.
*/
public int getType()
public int getType()
{
return cmapType;
}

/**
* Sets the type of the CMap.
*
*
* @param type the CMap type.
*/
public void setType(int type)
public void setType(int type)
{
cmapType = type;
}

/**
* Returns the registry of the CIDSystemInfo.
*
*
* @return the registry.
*/
public String getRegistry()
public String getRegistry()
{
return registry;
}

/**
* Sets the registry of the CIDSystemInfo.
*
*
* @param newRegistry the registry.
*/
public void setRegistry(String newRegistry)
public void setRegistry(String newRegistry)
{
registry = newRegistry;
}

/**
* Returns the ordering of the CIDSystemInfo.
*
*
* @return the ordering.
*/
public String getOrdering()
public String getOrdering()
{
return ordering;
}

/**
* Sets the ordering of the CIDSystemInfo.
*
*
* @param newOrdering the ordering.
*/
public void setOrdering(String newOrdering)
public void setOrdering(String newOrdering)
{
ordering = newOrdering;
}

/**
* Returns the supplement of the CIDSystemInfo.
*
*
* @return the supplement.
*/
public int getSupplement()
public int getSupplement()
{
return supplement;
}

/**
* Sets the supplement of the CIDSystemInfo.
*
*
* @param newSupplement the supplement.
*/
public void setSupplement(int newSupplement)
public void setSupplement(int newSupplement)
{
supplement = newSupplement;
}
/**

/**
* Returns the mapping for the space character.
*
*
* @return the mapped code for the space character
*/
public int getSpaceMapping()
Expand Down
Loading