Changes in TIFF v4.0.0¶
v4.0.0 (tag Release-v4-0-0)
Master Download Site
Master HTTP Site
This document describes the changes made to the software between the previous and current versions (see above). If you don't find something listed here, then it was not done in this timeframe, or it was not considered important enough to be mentioned. Please consult the ChangeLog file in the source package for full change details. The following information is located here:
BigTIFF support changes:
optionsparameter in the
TIFFClientOpen()funcs has been extended. When creating new files, you can add option
4to specify you want to create a ClassicTIFF file, though that is the default and the option is not strictly necessary. (As such, old calling code will continue to function and create ClassicTIFF files.) Or you can add option
8to specify you want to create a BigTIFF file instead. This new option is also reflected in some of the tools we already upgraded. For instance, you can use the
-8option on tiffcp to have tiffcp produce BigTIFF files instead of the default ClassicTIFF. (Whilst an additional option is provided for version selection when creating new files, no such option is necessary when reading TIFF files. LibTiff reads ClassicTIFF and BigTIFF both, and the application does not need to be aware which TIFF version an opened file is.)
Although the tag count in BigTIFF is 64bit, we restricted the count in the implementation to a much more reasonable size. This is necessary in current implementation, because all tag data gets read automatically in the IFD reading stage, so if there's half a dozen private tags with multiple gigabytes of data that causes considerable overhead even if the application level is never interested in these tags. Our choice to ignore tags with data longer then a certain sanity value is much needed as things stand. We also recommend to step away from writing tiles that are 8 kilobyte in their uncompressed form, or writing single-line strips, in really big files, resulting in mega's of tiles or strips. It's much more efficient to choose bigger tile or strip sizes, up to several megabyte if needed, and have a few kilo of tiles or strips instead.
Although it's rare, some application code does directly access file offsets. Some of these are automatically upgraded because they used the
toff_ttype, others need to be aware that the datatype changed and need to start using
uint64. This impacts access to tags like the
EXIFIFD tag, for example, or the
SubIfdstag, or to
TileOffsets, the return type of functions like
TIFFCurrentDirOffset(), and a parameter type to functions like
Although it's rare, some application code does use structures like
TIFFDirEntrythat used to be an exact binary representation of TIFF structures. These need to change. The old TIFFHeader structure is replaced by the new TIFFHeaderClassic, TIFFHeaderBig, and TIFFHeaderCommon structures that are an exact binary representation of the ClassicTIFF and BigTIFF header, and of the part that is common to both. There is no new equivalent for the old
TIFFDirEntrystructure (or more precisely, there is still a
TIFFDirEntrystructure, but it is changed, moved to library-private definition, and no longer an exact binary representation of the tag structure of either TIFF version).
Sizer functions, like
TIFFScanlineSize()and the like, return a
tmsize_t()is defined as
int32on 32bit machines, and
int64on 64bit machines, and as such it is meant to represent signed memory sizes). This is because we figure 98% of the calling code uses the return value as sizes in allocations and the like. So, any overflow that is theoretically possible with BigTIFF when LibTiff is running on a 32bit system, is best detected inside the sizer functions and it is best to return a type that makes sense as a memory size. If your calling code is the exception and is interested in actual file size, you best use the newer
TIFFScanlineSize64()function that returns an
These TIFF tags require a 64-bit type as an argument in libtiff 4.0.0:
Other important backward incompatible changes in the public API:
_TIFFRewriteField()and moved out from the public interface (from
tiffiop.h). Type of its
countparameter changed from
TIFFMergeFieldInfo()returns non-void result now. It returns 0 if successful and -1 if failed. Though this is now obsoleted function and should not be used in new programs. Use the new tag extension scheme instead.
TIFFFieldWithName()functions now return pointer to
TIFFFieldconstant object instead of
TIFFIgnoreSenseenumeration have been removed. They was unused and never been used properly. Should be unneeded for high-level applications.
TIFFTagValuestructure removed from the public
tif_dir.hand not accessible anymore. It should be unneeded for high-level applications.
Software configuration changes¶
Updated autotools: Autoconf 2.68, Automake 1.11.1, libtool 2.4.
Enabled support for Automake silent build rules (
Enabled support for Automake colorized and parallel tests.
Added detection of 64-bit integer types since libtiff 4.0 requires use of 64-bit signed and unsigned integer types.
Libtiff now provides a more comprehensive test suite with over 72 tests, which may be executed on Unix-like systems, or under Microsoft Windows using MinGW/MSYS or Cygwin.
--disable-lzmaconfigure option to disable use of liblzma.
--enable-defer-strile-loadconfigure option to enable experimental deferred strip/tile offset/size loading. May cause some extremely sophisticated uses of libtiff to fail.
--enable-chunky-strip-readconfigure option to enable experimental enable reading large strips in chunks in
Now always uses WIN32 native I/O functions for Microsoft Windows except for under Cygwin.
Now provides a pkg-config support file (libtiff-4.pc).
Patches/fixes made to stable libtiff (v3.9.X) are also applied to 4.0.0. There are too many to list here. See the distribution
ChangeLogfor a detailed change list.
There is considerable change in some files like
tif_dirwrite(). These changes don't impact backwards compatibility, they are mostly a clean rewrite that does allow BigTIFF support as well as somewhat more robust reading of the unexpected already and will also serve future API extension but does not impact current API or functionality in a negative way that you need to know about.
Although there is still a functional definition for types like
tstrip_t(strip index number), etc, we recommend against using these in newer code. We have learned that it is next to impossible to use these consistently and make real abstraction of the binary format of these types. Instead, at a certain level we always end up doing casts anyway, and taking the exact binary format into account, so these types are nothing but dangerously misleading and obfuscating. You do not need to update calling code that uses them, as 99.9% of such code will continue to work. But we recommend against using them in newer calling code, and we started replacing them with binary clear types like
uint32and such in the library.
We do use and will continue to use one functional type that is an exception to the above rule, being
tmsize_t. This is a signed memory size type, i.e. it is
int32on 32bit machines, or
int64on 64bit machines.
Optionally support LZMA compression via TIFF tag 34925. tiffcp supports compression levels similar to
-c zip:p9for setting the LZMA compression parameters.
Optionally defer the load of strip/tile offset and size tags for optimized scanning of directories. Enabled with the
--enable-defer-strile-loadconfigure option (
Optionally enable experimental support for reading big strips in chunks. Enabled with the
-sdswitches to allow operation on a particular directory, not just the first.
Contributed software changes¶