matio @1.5.3_1: update to 1.5.4

[Schamschula (Marius Schamschula)]

matio has been updated to version 1.5.4. New this version:

• Added support of reading sparse array from v4 MAT files
• Added support of writing sparse array to v4 MAT files
• Added support of writing character array to v4 MAT files
• Added support of autogen.sh on Darwin
• Enabled C++ compilation
• Updated snprintf.c to C99 snprintf v1.1
• Increased testsuite coverage
• Fixed bug writing complex array to v4 MAT files
• Fixed bug writing cell containing a struct with fields but empty array to compressed v5 MAT files
• Fixed bug reading hyperslabs of cells and struct fields from compressed v5 MAT files (broken by v1.5.3)
• Fixed bug writing empty character array to uncompressed v5 MAT files
• Fixed bug reading cells and struct fields from HDF5 MAT files
• Fixed bug calculating array sizes of structs in Mat_VarGetSize
• Fixed bug duplicating sparse array in Mat_VarDuplicate
• Fixed testsuite if zlib/hdf5 dependencies are missing
• Several other fixes, especially for memory leaks

Note: there are ports that need to be rev-bumped. I my case vips was rebuilt.

[kurthindenburg (Kurt Hindenburg)]

Cc Me!

[kurthindenburg (Kurt Hindenburg)]

that's a lot of ports that appear to need revbumped

 grep port:matio * -Ri | grep Portfile  | wc -l       
      31

[ryandesign (Ryan Carsten Schmidt)]

In matio 1.5.3 we had:

$ otool -L /opt/local/lib/libmatio.dylib | head -n2
/opt/local/lib/libmatio.dylib:
	/opt/local/lib/libmatio.2.dylib (compatibility version 4.0.0, current version 4.0.0)

In matio 1.5.4 we have:

$ otool -L /opt/local/lib/libmatio.dylib | head -n2
/opt/local/lib/libmatio.dylib:
	/opt/local/lib/libmatio.0.dylib (compatibility version 1.0.0, current version 1.0.0)

So from matio 1.5.3 to 1.5.4, the library's major version number has decreased from 2 to 0 and its minor version number has decreased from 4.0.0 to 1.0.0. That certainly looks like a mistake that should be reported to the developers of matio so that they can correct it. The library major version number should never decrease.

[Schamschula (Marius Schamschula)]

In the meantime matio has been updated two more times. The issue in comment 3 has been addressed:

diff -u Portfile.orig Portfile > Portfile-matio.diff
thor:ports marius$ otool -L /opt/local/lib/libmatio.dylib | head -n2
/opt/local/lib/libmatio.dylib:
	/opt/local/lib/libmatio.4.dylib (compatibility version 5.0.0, current version 5.2.0)

Updates since version 1.5.4:

Changes in 1.5.6 (23 February 2016)

• Fixed bug writing array to v4 MAT files on big endian architecture
• Fixed bug writing header with more than 127 characters to v5 or HDF5 MAT files

Changes in 1.5.5 (19 February 2016)

• Added support of reading sparse array from v4 MAT files
• Added support of writing sparse array to v4 MAT files
• Added support of writing character array to v4 MAT files
• Added support of autogen.sh on Darwin
• Enabled C++ compilation
• Updated snprintf.c to C99 snprintf v1.1
• Increased testsuite coverage
• Fixed bug writing complex array to v4 MAT files
• Fixed bug writing cell containing a struct with fields but empty array to compressed v5 MAT files
• Fixed bug reading hyperslabs of cells and struct fields from compressed v5 MAT files (broken by v1.5.3)
• Fixed bug writing empty character array to uncompressed v5 MAT files
• Fixed bug reading cells and struct fields from HDF5 MAT files
• Fixed bug printing array dimensions in Mat_VarPrint on MinGW
• Fixed bug calculating array sizes of structs in Mat_VarGetSize
• Fixed bug duplicating sparse array in Mat_VarDuplicate
• Fixed testsuite if zlib/hdf5 dependencies are missing
• Several other fixes, especially for memory leaks

[ryandesign (Ryan Carsten Schmidt)]

Ok great! Now with 1.5.6 the major library version is 4 and the minor library version is 5.2.0, which are greater numbers than in 1.5.3, so that's good. Since the major library version is still different though, we'll still need to revbump everything linking with libmatio.

[ryandesign (Ryan Carsten Schmidt)]

Replying to khindenburg@…:

that's a lot of ports that appear to need revbumped

Looks like it's only vips (builds fine), kst (fails for a different reason; see #50764), and the py-bobs which I haven't tried yet.

[Schamschula (Marius Schamschula)]

Updated patch file due to #51074

[Schamschula (Marius Schamschula)]

Committed in r148199 with revision bump to all dependents.