24/1/20 started 8.10.0

- more conformat IIIF output from dzsave [regisrob]

- add @id to dzsave to set IIIF id property [regisrob]

- add max and min to region shrink [rgluskin]

- allow \ as an escape character in vips_break_token() [akemrir]

- tiffsave has a "depth" param to set max pyr depth

- libtiff LOGLUV images load and save as libvips XYZ

- add gifload_source, csvload_source, csvsave_target, matrixload_source,

matrixsave_source, pdfload_source, heifload_source, heifsave_target

- revise vipsthumbnail flags

- add VIPS_LEAK env var

- add vips_pipe_read_limit_set(), --vips-pipe-read-limit,

VIPS_PIPE_READ_LIMIT

- revise gifload to fix BACKGROUND and PREVIOUS dispose [alon-ne]

- add subsample_mode, deprecate no_subsample in jpegsave [Elad-Laufer]

- add vips_isdirf()

- add PAGENUMBER support to tiff write [jclavoie-jive]

- add all to smartcrop

- flood fill could stop half-way for some very complex shapes

- better handling of unaligned reads in multipage tiffs [petoor]

- mark old --delete option to vipsthumbnail as deprecated [UweOhse]

- png save with a bad ICC profile just gives a warning

- add "premultipled" option to vips_affine(), clarified vips_resize()

behaviour with alpha channels

24/4/20 started 8.9.3

- better iiif tile naming [IllyaMoskvin]

31/1/19 started 8.9.2

- fix a deadlock with --vips-leak [DarthSim]

- better gifload behaviour for DISPOSAL_UNSPECIFIED [DarthSim]

- ban ppm max_value < 0

- add fuzz corpus to dist

- detect read errors correctly in source_sniff

- fix regression in autorot [malomalo]

- thumbnail on HEIC images could select the thumbnail incorrectly under some

size modes [ZorinArsenij]

20/6/19 started 8.9.1

- don't use the new source loaders for new_from_file or new_from_buffer, it

will break the loader priority system

- fix thumbnail autorot [janko]

- fix a warning with magicksave with no delay array [chregu]

- fix a race in tiled tiff load [kleisauke]

- better imagemagick init [LebronCurry]

- lock for metadata changes [jcupitt]

20/6/19 started 8.9.0

- add vips_image_get/set_array_int()

- disable webp alpha output if all frame fill the canvas and are solid

- support arrays of delays for animated images [deftomat]

- add "unlimited" flag to svgload

- disable webp alpha output if all frames fill the canvas and are solid

- add "compression" option to heifsave [lovell]

- support webp and zstd compression in tiff

- loaders use "minimise" to close input files earlier

- integrate support for oss-fuzz [omira-sch]

- add vips_switch() / vips_case() ... fast many-way ifthenelse

- better const handling for arithmetic operators fixes comparisons against out

of range values

- sharpen restores input colourspace

- handle alpha in heifload / heifsave [meyermarcel]

- add @interpretation and @format to rawload

- nifti load/save uses double for all floating point metadata

- add vips_error_buffer_copy()

- add VipsSource and VipsTarget: a universal IO class for loaders and savers

- jpeg, png, tiff (though not tiffsave), rad, svg, ppm and webp use the

new IO class

- rewritten ppm load/save is faster and uses less memory

- add @no_strip option to dzsave [kalozka1]

- add iiif layout to dzsave

- fix use of resolution-unit metadata on tiff save [kayarre]

- support TIFF CIELAB images with alpha [angelmixu]

- support TIFF with premultiplied alpha in any band

- block metadata changes on shared images [pvdz]

- RGB and sRGB are synonmous

17/9/19 started 8.8.4

- improve compatibility with older imagemagick versions

- remove realpath, since it can fail on systems with grsec

31/8/19 started 8.8.3

- revert sharpen restoring the input colourspace

- xres/yres tiffsave params were in pixels/cm [f--f]

9/7/19 started 8.8.2

- better early shutdown in readers

- don't attempt to save large XMP to jpeg [tnextday]

- always fetch HEIC metadata from the main image [zhoux2016]

- fix loop in malformed ppm [Kyle-Kyle]

- better support for PNGs with long comment names

- fix build with GM

- add locks for pdfium load

- fix build with MSVC

- fix a problem with shinkv tail processing [angelmixu]

- fix a read one byte beyond buffer bug in jpegload

- make GIF parsing less strict

- better feof() handling in GIF load

- clip coding and interpretation on vips image read

- check image bounds for GIF load

- prevent over-pre-shrink in thumbnail [kleisauke]

- fix sharpen with sigma 0.5 [2h4dl]

- sharpen restores input colourspace

- verify bands/format for coded images

- improve data_length handling for jpeg metadata

24/5/19 started 8.8.1

- improve realpath() use on older libc

- better magickload error messages

- more consistent behaviour for page-height metadata

- fix for composite with many small images and some combinations of blend modes

- fix memleak in tiff pyr save to memory [scossu]

- istiff attempts to read the first directory rather than just testing the

magic number [przemyslawpluta]

- much faster ismagick() [jcupitt]

- better behaviour for vips_region_fetch() if request lies partly ouside image

- remove 256 band limit in arithmetic.c [erdmann]

- disable Orc if building with CET [lovell]

- fix vipsthumbnail with pyr tiff [kleisauke]

- text autofit could occasionally terminate early [levmorozov]

- fewer warnings on tiffload [chregu]

- vips_resize() breaks aspect ratio and limits shrink to prevent <1px

dimensions [lovell]

21/9/18 started 8.8.0

- much faster smartcrop [lovell]

- add low/high to smartcrop [jcupitt]

- add XMP support to png read/write [jcupitt]

- deprecate thumbnail auto_rotate, add no_rotate [jcupitt]

- implement thumbnail shrink-on-load for openslide images [jcupitt]

- add animated webp support [jcupitt]

- revise vips_cast() to improve behaviour with uint images [erdmann]

- add bandand()/or()/eor() to cplusplus binding [clcaalu]

- implement shrink-on-load for tiff pyramids [jcupitt]

- added vips_image_set_blob_copy() [jcupitt]

- don't stop composite on first non-transparent image [felixbuenemann, GDmac]

- add vips_rect_overlapsrect()

- composite is much faster at positioning subimages

- stop tiff pyr layers if width or height drop to 1 [gvincke]

- dzsave has a new skip_blanks option

- add vips_CMYK2XYZ() and vips_XYZ2CMYK(), plus associated routes

- include cmyk and srgb fallback profiles

- add vips_profile_load() and use it everywhere

- fix race in temp filename creation [lhecker]

- add @reduction_effort param to webpsave [lovell]

- add @option_string param to thumbnail_buffer [kleisauke]

- add XMP, IPCT, ICC, EXIF etc. support to magickload/magicksave

- much lower memuse for gifload

- tilecache speedups

- add vips_heifload(), vips_heifsave()

- add heif thumbnail support to vips_thumbnail()

- free threadpool earlier, reducing mem growth for some long-running

processes [jtorresfabra]

- add vips_region_fetch() / _width() / _height() for language bindings

- vips_text() supports justification

- move vips_image_set_kill() and iskilled() to the public API

- dzsave to szi sets suffix correctly [martinweihrauch]

- dzsave szi writes "scan-properties.xml"

- add vips_image_(get|set)_image()

- add openslideload option to attach all associated images as metadata

- dzsave to szi will write all associated images

- remove old c++ and python interfaces

- vipsthumbnail can thumbnail animated and multipage images

- deprecate webpload @shrink, use @scale instead

31/3/19 started 8.7.5

- better buffer sizing in tiff reader [omira-sch]

4/1/19 started 8.7.4

- magickload with magick6 API did not chain exceptions correctly causing a

memory leak under some conditions [kleisauke]

- zero memory on allocate to prevent write of uninitialized memory under some

error conditions [Balint Varga-Perke]

21/11/18 started 8.7.3

- fix infinite loop for autofit with non-scaleable font

- mapim was not offsetting by window offset [erdmann]

- better rounding for scale [kleisauke]

- fix a memleak in magick6load [kleisauke]

21/11/18 started 8.7.2

- more info output for temp files to help diagnose problems

- vips_text() could set the wrong DPI

- vips_text() leaked in autofit mode

23/9/18 started 8.7.1

- update function list in docs [janko-m]

- test for g_str_to_ascii() [jcupitt]

- fix temp file open on Windows and fallback on linux [lovell]

23/12/17 started 8.7.0

- add magicksave, save image with libMagick [dlemstra]

- remove jpeg thumbnail from EXIF if "jpeg-thumbnail-data" has been removed by

user

- hough_line scales width to 0 - 180, not 0 - 360

- hough_line is 4x faster

- hough_circle is 2x faster

- add vips_sobel() and vips_canny() edge detectors

- add vips_rotate() ... a convenience method for vips_similarity()

- svgload was missing is_a [lovell]

- better header sniffing for small files

- drop incompatible ICC profiles before save

- better hasalpha rules

- create funcs always make MULTIBAND (ie. no alpha)

- use O_TMPFILE, if available [Alexander--]

- set "interlaced=1" for interlaced JPG and PNG images

- add PDFium PDF loader

- jpegload adds a jpeg-chroma-subsample field with eg. 4:4:4 for no

chrominance subsampling.

- tiffload, pdfload, magickload set VIPS_META_N_PAGES "n-pages" metadata item

- add fontfile option to vips_text() [fangqiao]

- add vips_transpose3d() -- swap major dimensions in a volumetric image

- remove vips7 stuff from default API ... you must now #include it explicitly

- added vips_argument_get_id() to fix derived classes on win32 [angelmixu]

- fix compile with MSVC 2017 [angelmixu]

- pdfload has a option for background

- vips7 C++ interface defaults off

- make members, getters and operators "const" in cpp API

- composite has params for x/y position of sub-images [medakk]

- add Mitchell kernel

- pyramid builders have a choice of 2x2 shrinkers [harukizaemon]

- add `palette` option to pngsave [felixbuenemann]

- add basic nifti load/save support

- support writing string-valued fields via libexif

- paste in the test suite from pyvips

- get EXIF tag names from tag plus ifd [@Nan619]

- escape ASCII control characters in XML

- magickload now sniffs some file types itself

- update radiance load from upstream

- add region_shrink to tiffsave

- mapim could fail for float index images with coordinates out of int range

- scale openexr alpha to 0 - 255

- close input earlier, when we can [kleisauke]

- add vips_object_get_args() for language bindings [kleisauke]

12/3/18 started 8.6.4

- better fitting of fonts with overhanging edges [Adrià]

- revise C++ example [fangqiao]

- strict round down on jpeg shrink on load [davidwood]

- configure test for g++ 7.2 and composite.cpp

- don't Ping in magickload, too unreliable

- ensure WebP can add metadata when compiled with libwebpmux [lovell]

- improve accuracy of vector path convolution [felixbuenemann]

12/2/18 started 8.6.3

- use pkg-config to find libjpeg, if we can

- better clean of output image in vips_image_write() fixes a crash

writing twice to memory

- better rounding behaviour in convolution means we hit the vector path more

often

- fix a crash if a delayed load failed [gsharpsh00ter]

- icc_import attaches the fallback profile if it used it

5/1/18 started 8.6.2

- vips_sink_screen() keeps a ref to the input image ... stops a rare race

- fix a minor accidental ABI break in 8.6.0 -> 8.6.1 [remicollet]

- fix read of plane-separate TIFFs with large strips [remicollet]

- fix a C++ warning in composite.cpp [lovell]

- remove number of images limit in composite

- composite allows 1 mode ... reused for all joins

- fix race in vips_sink() for seq read

10/12/17 started 8.6.1

- fix mmap window new/free cycling

- fix some compiler warnings

- remove the 64-image limit on bandary operations

- better version date [bmwiedemann]

- bump wrapper script version [bgilbert]

- fix a memleak on error during jpeg buffer write [lovell]

- fix misspelling of IPTC as IPCT [lovell]

- seq could be set on small images opened in random-access mode [aferrero2707]

- fix small memleak in dzsave [lovell]

- small speedup for rgb->g [lovell]

15/4/17 started 8.6.0

- supports fits images with leading non-image HDUs, thanks benepo

- add vips_image_new_from_image() and vips_image_new_from_image1() ... make a

constant image

- add new_from_image() to Python as well

- slight change to cpp new_from_image() to match py/C behaviour

- vips_conv(), vips_compass(), vips_convsep() default to FLOAT precision

- add FORCE resize mode to break aspect ratio

- add vips_thumbnail_image()

- better prefix guessing on Windows, thanks tumagonx

- savers support a "page_height" option for multipage save

- rename 'disc' as 'memory' and default off

- add vips_find_trim(), search for non-background areas

- remove lcms1 support, it had bitrotted

- `join` tagged as seq

- support tiffsave_buffer for pyramids, thanks bubba

- thumbnail and vipsthumbnail have an option for rendering intent, thanks

kleisauke

- set file create time on Windows, thanks dlong500

- remove python tests ... moved to pyvips test suite

- vips7 and vips8 python bindings default to off ... use the new pyvips

binding instead

- better svgload: larger output, handle missing width/height, thanks lovell

- add vips_gravity() ... embed, but with direction rather than position

- vips_text() can autofit text to a box, thanks gargsms

- add vips_composite() / vips_composite2(): merge a set of images with

a set of blend modes

- better gobject-introspection annotations, thanks astavale

- vips_image_write() severs all links between images, when it can ... thanks

Warren and Nakilon

- vector path for convolution is more accurate and can handle larger masks

- linear and cubic kernels for reduce are higher quality

- added vips_value_set_blob_free()

- "--size Nx" to vipsthumbnail was broken, thanks jrochkind

- fix build with gcc 7

- add vips_fill_nearest() ... fill pixels with nearest colour

- add VIPS_COMBINE_MIN, a new combining mode for vips_compass()

- vips_hist_find_indexed() now has a @combine parameter

- vips_affine() and vips_similarity() have a "background" parameter

- fix nasty jaggies on the edges of affine output, thanks chregu

- add gif-delay, gif-comment and gif-loop metadata

- add dispose handling to gifload

- dzsave outputs extra right and bottom overlap-only tiles, for closer spec

adherence

- deprecate the "centre" option for vips_resize(): it's now always on

- setting the EXIF data block automatically sets other image tags

- add "extend" option to affine; resize uses it to stop black edges

29/8/17 started 8.5.9

- make --fail stop jpeg read on any libjpeg warning, thanks @mceachen

- don't build enumtypes so often, removing perl as a compile dependency

- fix a crash with heavy use of draw operations from language bindings,

thanks @Nakilon

2/8/17 started 8.5.8

- fix transparency detection in merge, thanks Haida

- define env var VIPS_WARNING to hide warning messages

- shut down dzsave output earlier to help mark-sweep bindings

- fix webp thumbnail upscale, thanks Kleis

9/6/17 started 8.5.7

- better smartcrop

- transform cmyk->rgb automatically on write if there's an embedded profile

and the saver does not support cmyk

- fix DPI mixup in svgload ... we were writing images about 20% too large,

thanks Fosk

19/5/17 started 8.5.6

- tiff read with start page > 0 could break edge tiles or strips

- raise b64 limit to allow for huge profiles (thanks jaume)

- fix error return in blob save (thanks jaume)

- tag vipsprofile as py2 (thanks ioquatix)

- don't cache thumbnail (thanks tomasc)

23/4/17 started 8.5.5

- doc polishing

- more improvements for truncated PNG files, thanks juyunsang

- improve corrupted jpg handling, thanks juyunsang

- fix small test suite issues on os x

23/4/17 started 8.5.4

- don't depend on image width when setting n_lines, thanks kleisauke

7/4/17 started 8.5.3

- more link fixing in docs

- revise cache sizing again to help out of order errors under heavy load, thanks

kleisauke

25/3/17 started 8.5.2

- better behaviour for truncated PNG files, thanks Yury

- missing proto for vips_tiffsave_buffer(), thanks greut

- move some docs from the wiki and blog into core libvips docs

- add support for markdown in docs

25/3/17 started 8.5.1

- init more classes earlier, thanks David

13/10/16 started 8.5.0

- rewritten buffer system is safer and frees memory earlier

- added tiff save to buffer

- added dzsave save to buffer (zip only)

- revise header get/set functions

- better vipsheader behaviour with complex field types

- added vips_image_hasalpha()

- added vips_thumbnail() / vips_thumbnail_buffer()

- webpload/webpsave read and write icc, xmp, exif metadata

- better >4gb detect for zip dzsave output [Felix Bünemann]

- all loaders have a @fail option, meaning fail on first warning, though it

only does anything for jpg and csv

- add vips_image_get_fields() to help bindings

- add tiff multi-page read/write

- add VIPS_META_PAGE_HEIGHT metadata

- IM6/IM7 magickload supports page/n/page-height, all_frames deprecated

- gifload supports n/page-height

- added #defines for VIPS_SONAME, VIPS_LIBRARY_CURRENT, VIPS_LIBRARY_REVISION,

VIPS_LIBRARY_AGE

- better support for bscale / bzero in fits images

- deprecate vips_warn() / vips_info(); use g_warning() / g_info() instead

- vipsthumbnail supports much fancier geometry strings, thanks tomasc

- vips_thumbnail() has new @size option

- fix --vips-cache-max etc.

- add compute reordering, plus some new API to support it:

vips_reorder_margin_hint() and vips_reorder_prepare_many(), thanks

aferrero2707

- kick load operations from cache on read error, thanks gaillard

- fix return from C++ assignment operator overloads (+=, -= etc)

- add @max_slope to vips_hist_local() to implement CLAHE, thanks hunter-87

- vips_gaussnoise() pixels are reproducible on recalc, thanks MvGulik

- max/min sort values by y and x coordinate

- tiff read uses libtiff scanline API if rows-per-strip is large

- vips_region_shrink() knows about alpha, helps dzsave and tiffsave

- use expat, not libxml, for XML load ... removes a required dependency, since

we get expat as part of glib

- new sequential mode infrastructure is faster and more flexible

- add vips_smartcrop(), based on sharp's smartcropper

- vipsthumbnail has a --smartcrop option

- added vips_rot90() etc. convenience functions

- fix vips_resize() bug when hscale and vscale were very different

8/12/16 started 8.4.5

- allow libgsf-1.14.26 to help centos, thanks tdiprima

11/11/16 started 8.4.4

- fix crash in vips.exe arg parsing on Windows, thanks Yury

18/10/16 started 8.4.3

- fix error detection in gif_close, thanks aaron42net

- fix tiny threading memleak

- improve compatibility with very old glib, see #548

27/9/16 started 8.4.2

- small doc improvements

- fix error message for metadata fetch type mismatch

- resolve a race condition in thread shutdown, thanks Lovell

1/5/16 started 8.4

- many more wepsave options [Felix Bünemann]

- added quant_table option to wepsave [Felix Bünemann]

- added @n option to pdfload, thanks andris

- dzsave won't write empty tiles in google mode, thanks bverem, perog,

felixbuenemann

- allow nested [] in CLI args

- restore BandFmt on unpremultiply in vipsthumbnail

- better python detection and build [Felix Bünemann]

- max-alpha defaults to 65535 for RGB16/GREY16

- added radsave_buffer [Henri Chain]

- support tiff orientation tag

- autorotate option for tiff load

- tiffsave converts for jpg if jpg compression is turned on

- tiffsave supports --strip

- conversions to GREY16 could lock

- free pixel buffers on image close as well as thread exit ... stops main

thread buffers clogging up the system

- dzsave can write compressed zips [Felix Bünemann]

- vips_image_write() only refs the input when it has to ... makes it easier to

combine many images in bounded memory

- VImage::write() implementation was missing

- VImage::write() return value changed from void to VImage to help chaining

- added C++ arithmetic assignment overloads, += etc.

- VImage::ifthenelse() with double args was missing =0 on options

- better accuracy for reducev with smarter multiplication

- better quality for vips_resize() with linear/cubic kernels

- pyvips8 can create new metadata

- better upsizing with vips_resize()

- add imagemagick v7 support, thanks sachinwalia2k8

- added vips_worley(), vips_perlin() noise generators

- added vips_convf(), vips_convi(), vips_convasep(), vips_conva() ...

im_conv*() functions rewritten as classes

- vips_convsep() calls vips_convasep() for the approximate case

- new fixed-point vector path for convi is up to about 2x faster

- gif loader can make 1, 2, 3, or 4 bands depending on file contents

- support --strip for pngsave

- add svgz support [Felix Bünemann]

- rename boostrap.sh -> autogen.sh to help snapcraft

- support unicode filenames on Windows

- added VIPS_ROUND as well as VIPS_RINT

- resize/reduce*/shrink*/affine now round output size to nearest rather than

rounding down, thanks ioquatix

- better support for tile overlaps in google maps mode in dzsave

- dzsave puts vips-properties.xml in the main dir for gm and zoomify layouts

- resize and reduce have @centre option for centre convention downsampling

- vipsthumbnail uses centre convention to better match imagemagick

_ add vips_foreign_get_suffixes()

19/8/16 started 8.3.4

- better transparency handling in gifload, thanks diegocsandrim

30/7/16 started 8.3.3

- fix performance regression in 8.3.2, thanks Lovell

- yet more robust vips file reading

18/5/16 started 8.3.2

- more robust vips image reading

- more robust tiff read [Matt Richards]

15/4/16 started 8.3.1

- rename vips wrapper script, it was still vips-8.2, thanks Benjamin

- export C++ operator overloads for MSVC linking [Lovell]

- fix magickload @page with GraphicsMagick

- add giflib5 support

- allow resize >1 on one axis, <1 on the other

- vips_resize has an optional @kernel argument

- fix giflib4 detection [felixbuenemann]

29/1/16 started 8.3

- add vips_reduce*() ... a fast path for affine downsize

- vips_resize() uses vips_reduce() with lanczos3

- bicubic is better on 32-bit int images

- add pdfload, svgload, gifload for PDF, SVG and GIF rendering

- vipsthumbnail knows about pdfload and svgload

- added @page param to magickload

- matload is more specific (thanks bithive)

- lower mem use for progressive jpg decode

- sharpen has a new @sigma param, @radius is deprecated

- sharpen allows a much greater range of parameters

- sharpen defaults now suitable for screen output

- better handling of deprecated args in python

- much better handling of arrayimage command-line args

- faster hist_find (Lovell Fuller)

- webpload has a @shrink parameter for shrink-on-load

- vipsthumbnail knows about webp shrink-on-load

- better behaviour for vips_cast() shift of non-int types (thanks apacheark)

- python .bandrank() now works like .bandjoin()

- vipsthumbnail --interpolator and --sharpen switches are deprecated

- switches to disable PPM, Rad and Analyze support

- added VIPS_COUNT_PIXELS(), overcomputation tracking

- @out_format in vips_system() can contain [options]

- webpsave_buffer no longer ignores @lossless, thanks aaron42net

- float tiff tagged as scRGB to match photoshop convention, thanks Murat

- better jpeg autorot, thanks otto

24/3/16 started 8.2.4

- fix nohalo and vsqbs interpolators, thanks Rafael

27/1/16 started 8.2.3

- fix a crash with SPARC byte-order labq vips images

- fix parsing of filenames containing brackets, thanks shilpi230

- fix hist_entropy (lovell)

- small fixes to radiance load

12/1/16 started 8.2.2

- changes to ease compiling C++ binding with MSVC [Lovell Fuller]

- reorder file tests to put slow loaders last

- ifthenelse needs less C stack during eval

- better rounding in bilinear interpolator

- fix to "make check" in non-C locales [felixbuenemann]

- use compiler builtins isnan, isinf, fabs, fmin, fmax, ceil, floor when

possible [Lovell Fuller]

- tune vips_shrinkh(), 30% faster [Lovell Fuller]

- remove SEQ hint from vips_subsample(), fixes cli performance [erdmann]

- fix double free on attach ICC profile from file in tiff write [erdmann]

- use g_assert_not_reached()

- better vips-from-C docs

- remove Duff from im_conv() / im_conv_f() for a 25% speedup [Lovell Fuller]

1/1/16 started 8.2.1

- add a compat stub [Benjamin Gilbert]

- python bandjoin is now just an instance function

- small doc improvements

- small vips7 C++ improvement

- remove exception specifications from vips8 C++ interface [Lovell Fuller]

- VImage::get_typeof() now returns GType

7/10/15 started 8.2.0

- added im_bufmagick2vips(), a vips7 wrapper for magick load from buffer

- fetch unset property now returns default value rather than warning

- many more const declarations to help gobject-introspection

- rewritten vips_shrink() is 2x faster, much lower memuse, now handles complex

- only allow [] for filename options

- add memory.h to Python API .. makes tracked highwater visible

- added bandjoin_const to add constant bands to an image

- better alpha handling for tiff write, thanks sadaqatullahn

- better cache sizing for vips_resize()

- sizealike / formatalike / bandsalike elide completely if they can for

a x2 saving in C stack use in many cases

- added vips_mapim() ... resample with an index image, plus test

- try to improve vips_resize() quality a little more

- vips_resize() can do non-square resizes

- dzsave removes tile metadata by default, thanks Benjamin

- jpeg strip option removes a little more, thanks Benjamin

- added vips_image_new_from_memory_copy()

- improve vips_sink_screen() stability under heavy load

- added vips_arrayjoin()

- Python x.bandjoin(y) is now x.ibandjoin(y), sorry

- faster and lower-mem TIFF read

- faster bilinear interpolator

- TIFF loads and saves IMAGEDESCRIPTION

- add --properties flag to tiffsave

- dzsave defaults changed: now writes 256x256 jpegs for non-edge tiles, thanks

Daniel

7/5/15 started 8.1.1

- oop, vips-8.0 wrapper script should be vips-8.1, thanks Danilo

- fix vips7 pathname parsing on windows, thanks Lovell

7/5/15 starteld 8.1.0

- add vips_premultiply(), vips_unpremultiply()

- change the alpha range rules for vips_flatten() to match vips_premultiply()

- vipsthumbnail uses vips_resize() rather than its own code

- vipsthumbnail uses vips_premultiply() for better alpha quality

- added bandand() bandor() bandeor() convenience funcs to Python

- oops, base64 encode could pad by up to two zero bytes

- added VipsRefString as a thing that gi bindings can unpack

- support "with Vips.Image as i:" in Python

- try to support DOS CSV and PPM files on linux

- add vips_byteswap(), remove byteswap option from vips_copy()

- add vips_bandfold()/vips_bandunfold()

- dzsave supports zip output > 4gb, thanks benjamin

- add support for HSV colourspace [Jonas Øgaard]

- skip oversized markers in jpeg write

- jpeg exif tags saved as name rather than title

- can now set any jpeg exif tag, not just modify existing tags

- add vips_hist_entropy()

- vips_log(), vips_log10() are zero-avoiding

- better overlap handling for dzsave, thanks robclouth

- add @spacing option to vips_text()

- tiff loads and saves IPCT and Photoshop data

7/5/15 started 8.0.3

- dzsave and tif pyr write could fail for some image dimensions, thanks Jonas

4/5/15 started 8.0.2

- fix a refcount error in C++ wrapper, thanks huskier

- better C++ api test

- test suite improvements

- remove a couple of stray header decls, thanks benjamin

25/4/15 started 8.0.1

- fix some compiler warnings

- work around a glib bug that can cause segv under load

- add some notes on threading to the docs

- better leak reporting

11/2/15 started 8.0

- remove old doc stuff, lots of doc improvements

- add fliphor(), flipver(), rot90(), rot180(), rot270(), median(), dilate(),

erode() convenience methods to Python and C++

- python: use [] to index and slice image bands, use () to get a point

- c++: use [] to band index, () returns a vector<double>

- add shift option to cast

- sRGB2scRGB and scRGB2sRGB scale 16-bit alpha to and from 8-bit

- add magickload_buffer() [mcuelenaere]

- add vips_foreign_is_a_buffer() [mcuelenaere]

- added test_foreign.py, plus more test images

- added vips_region_shrink(), fast x2 shrinker

- rewritten tiff writer is about 3 - 4x faster at making pyramids

- jpg, magick, png, tiff readers now use only 1 fd per input image

- added vips_info_set(), vips_progress_set(), vips_profile_set() ... bindings

can now support all the vips command-line options if they wish

- better conversion to greyscale, thanks bkw

- add vips_image_copy_memory(), improves stability with heavy threading

- jpegsave supports new mozjpeg features [lovell]

- add vips_vipsload(), vips_vipssave() ... why not

26/3/15 started 7.42.4

- im_maxpos_avg() avoids NaN

- small tiffsave doc improvements

- better thresholding for tiffsave "squash" mode

- add @miniswhite mode to tiffsave

6/2/15 started 7.42.3

- bump version for back-compat ABI change

- added vips_image_memory(), an alias for vips_image_new_memory()

- improvements to configure for python

- remove --disable-cxx configure flag

- python imageize preserves interpretation

- fix dzsave as a target format

30/12/14 started 7.42.2

- C++ required output params were broken, thanks Lovell

- remove VImage::scale() to get scale from header, it clashed with the

VipsScale operator, thanks Lovell

- allow c++ set enum from string

- display param default and range in usage

- better docs

- more tests

- renamed VIPS_FOREIGN_DZ_DEPTH_1 as VIPS_FOREIGN_DZ_DEPTH_ONE etc. to help

bindings

- vipsthumbnail will return an error code if one or more conversions failed

- disable chroma subsample in jpeg-tiff if Q >= 90

- try to handle tiffs with old-style 8-bit colormaps

- rename vipsthumbnail -o as -f, -o stays as a hidden flag

- fix some small leaks

- faster openslide load, thanks Benjamin

- add VInterpolate class to cplusplus binding, thanks Lovell

- add lower-level operation cache access

- turn on leak testing in test suite

- don't use isnormal() to test for crazy FP numbers, thanks Murat

- much faster RGB16 -> sRGB path

24/12/14 started 7.42.1

- add gobject-2.0 to Requires: in vips and vips-cpp .pc files

- bump soname

- fix VipsBlob read

- remove "future" dependency, thanks bgilbert

4/11/14 started 7.42.0

- better default resolution for png load

- better pbm (one bit) load, better pfm (float) load/save

- added pbm (one bit) save

- changed vips_gaussblur() parameters, sorry

- add .szi as a dzsave zip synonym

- support tiff XMP metadata

- support @density arg to magickload [Lovell]

- support python3.4 and python2.7 in new python binding

- vips_gaussmat() and vips_logmat() are now int by default, to match

vips_conv(), and use @precision, not @integer

- added --disable-pyvips8 switch to turn off new py binding

- "vips thing" exit status can now be used to test for optional components

25/7/14 started 7.41.0

- start working on --disable-deprecated

- fix pngload with libpng >1.6.1

- add vips_resize()

- return of vips_init(), but just for bindings

- revised type.c to make it more binding-friendly

- add @background arg to save: the colour to flatten against

- add VIPS_ARGUMENT_MODIFY flag ... used for draw_circle etc, meaning an op

which modifies its argument

- rename VIPS_OPERATION_RELATIONAL_NOTEQUAL as VIPS_OPERATION_RELATIONAL_NOTEQ

for consistency

- python vips8 binding

- python vips8 test suite: test_arithmetic.py, test_colour.py,

test_conversion.py

- move zoomify ImageProperties file, now a better match to the offical tool

- rename VIPS_ANGLE_180 as VIPS_ANGLE_D180 etc. to help python

- remove cimg support, we have a gmic plugin now instead

- add support for vips8 plugins

- added "autorotate" option to jpeg load

- added autorot operator

- added @filter option to pngsave (Lovell)

- C++ vips8 binding

27/10/14 started 7.40.12

- better tiff detection (Lovell)

- fix memleak in sharpen (Lovell)

- fix profile handling in XYZ2scRGB, thanks Lovell

8/10/14 started 7.40.11

- rework extra band handling for colour functions

- set interpretation of matlut output more carefully

8/9/14 started 7.40.10

- icc_import and icc_transform checks the input profile for compatibility

with the image, thanks James

- try to make vips_thread_shutdown() optional

8/9/14 started 7.40.9

- support jfif resunit "none"

- support GRAY as an input and output ICC space

- fix a read loop with setjmp() in png read, if the png file is broken

- fix vipsthumbnail with both input cmyk and output rgb profiles specified

- vipsthumbnail retries with specified input profile if embedded profile is

broken

- add @profile option to pngsave, matching tiff and jpeg

- fix a race in the operation cache [Lovell]

8/9/14 started 7.40.8

- fix configure on rhel6 [Lovell]

- mono <-> rgb converters were not handling extra bands, thanks James

- support jpeg resunit "none"

21/8/14 started 7.40.7

- width and height were swapped in matlab load

- set interpretation more carefully on matlab load

- fix memleak in tilecache [Lovell]

- fix memleak in VipsArray [Lovell]

- fix memleak in webp load from buffer [Lovell]

- fix memleak in png save to buffer [Lovell]

- make out of order read in png and jpg a fatal error

- add thread_shutdown() to C++ (and thereby to python)

12/8/14 started 7.40.6

- more doc fixes

- fix similarity rotate+scale, thanks Topochicho

- fix 16-bit PNG save, thanks John

- fix dzsave date on Windows, thanks John

- fix vipsthumbnail on many-core systems, thanks James

25/7/14 started 7.40.5

- fix a race in im_maxpos_avg()

- limit n_thr on tiny images

- don't exit() on memleak detected, just warn

- add "autocrop" option to openslide load

- argh fix affine, again, there were sometimes black bars with nohalo and the

vips8 interface

- pngsave in interlaced mode makes a copy of the image, so it's always seq

- vipsthumbnail shrinks to 1/2 window_size

- vipsthumbnail has an anti-alias filter between shrink and affine

- vipsthumbnail defaults to bicubic

- better rounding behaviour for fixed-point bicubic reduces noise

- fix pngload with libpng >=1.6.11

- fix colour for openslide read associated

4/7/14 started 7.40.4

- fix vips_rawsave_fd(), thanks aferrero2707

- fix im_point()

- vips_scale() now does round to nearest to avoid rounding errors

- improve im_openout() compat macro

- more vips7 compatibility fixes, thanks steve

- more robust vips_system()

- add webp support to vips7

30/6/14 started 7.40.3

- fix interlaced thumbnails in vipsthumbnail, thanks lovell

- fix use of "header" in benchmark/, thanks David

- fits save now caches the image before write, so it's top-to-bottom

- add --properties argument to dzsave, thanks bgilbert, jhenriksen

25/6/14 started 7.40.2

- dzsave write to zip stops at 4gb, thanks bgilbert

- improve short option name handling, thanks bgilbert

- added --enable-docs configure option to help freebsd

- removed a bash-ism from configure to help freebsd

- don't assume GType fits in an int to help freebsd

24/6/14 started 7.40.1

- revise man.1 pages

- fix vips_guess_prefix()

23/6/14 started 7.40.0

- version bump

- renamed "header" as "vipsheader" and "edvips" as "vipsedit"

21/1/14 started 7.39.0

- auto-decode for (almost) all operations, see vips_image_decode()

- background render thread cleans up and quits neatly

- colourspace has a source_space option

- operations can be tagged as "deprecated"

- redo im_draw_circle(), im_draw_flood(), im_draw_line(), im_draw_mask(),

im_draw_image(), im_draw_rect(), im_draw_point(), im_read_point(),

im_draw_smudge(), im_label_regions() as classes

- better rounding in vips_flatten()

- VipsStatistic operations are sequential

- vipsthumbnail has --rotate auto-rotate option

- removed embedded thumbnail reader from vipsthumbnail: embedded thumbnails

are too unlike the main image

- fix to vipsthumbnail --crop, thanks Alessandro

- add vips_sum()

- add vips_hough base class and vips_hough_line()

- add "mode" param to vips_draw_image()

- add vips_hough_circle()

- reduce default cache size to 1,000 operations

- added "postbuild" signal

- vips_system() now supports many input images and you can change image

argument order

- support 16-bit palette TIFFs, plus palette TIFFs can have an alpha

- libgsf-1 is now an optional dependency

- dzsave can directly write a ZIP file

- add ".vips" as an alternative suffix for vips files

- added vips_tiffload_buffer()

- added vips_image_new_from_buffer(), vips_image_write_to_buffer()

- added vips_object_set_from_string()

- added @container option to dzsave

- support 1/2/4 bit palette tiff images with alpha

- vips_system() now uses g_spawn_command_line_sync()

- added im_tile_cache_random() to help nip2

- added hough_circle() to vips7 C++ API

- added Travis CI config, thanks Lovell

- im_*merge(), im_*mosaic(), im_match*(), im_global_balance*(), im_remosaic(),

im_*mosaic1(), im_*merge1() redone as classes

- better filename tracking for globalbalance

- revised vips8 image load/save API, now simpler and more logical

- operations emit "invalidate" if any of their input images invalidate

- operation cache drops ops on invalidate

- skipahead is back, thanks to a new threadpool tweak

6/3/14 started 7.38.6

- grey ramp minimum was wrong

- vipsthumbnail --crop could fail for very non-square images, thanks

Alessandro

- fix a crash in vips_rawsave(), thanks Andrea

- updated German translation, thanks Chris

- fix coordinate error in affine, thanks ferryfax

24/2/14 started 7.38.5

- jpeg load from buffer could write to input, thanks Lovell

- fix webpload from buffer, thanks Lovell

- vips_sequential() could fail under heavy load

- remove support for seq mode read for operations like extract

13/2/14 started 7.38.4

- --sharpen=none option to vipsthumbnail was broken, thanks ferryfax

- more locking on property create and lookup to help very-threaded systems,

thanks Nick

22/1/14 started 7.38.3

- undeprecate VIPS_MASK_IDEAL_HIGHPASS and friends, ruby-vips was using them,

thanks ahacking

22/1/14 started 7.38.2

- auto RAD decode for affine

- falsecolour was not working for some image types

- foreign memory buffer images did not have the right dhint, broke

command-line falsecolour on sequential images

- support many Radiance readers active at once

- add secret "rgbjpeg" flag to vips_tiffsave() to help IIP

19/1/14 started 7.38.1

- bump soname, thanks benjamin

- better conversion to and from scrgb/xyz for rad (hdr)

- fix --interpolate flag to vipsthumbnail, thanks Lovell

18/1/14 started 7.38.0

- version bump

19/10/13 started 7.37.0

- redone im_rotate_*mask45(), im_gauss_*mask*(), im_log_*mask(), im_dilate(),

im_erode(), im_rank_image(), im_compass(), im_linedet(), im_gradient(),

im_convsep(), im_convsep_f(), im_fastcor(), im_spcor(), im_sharpen()

as classes

- im_gradcor() deprecated

- vips_init() now does some ABI compat checking, though this change requires

an ABI break

- add "interlace" option to vips_jpegsave()

- remove vips_image_copy_fields() and vips_demand_hint() and add

vips_image_pipeline() to do both jobs

- vipsthumbnail allows non-square bounding boxes, thanks seth

- add vips_matrixprint()

- add @point subsample mode to vips_subsample()

- im_contrast_surface() deprecated: it was slower than calling conv a few

times

- radiance load supports sequential read

- rewritten radiance decode is much faster

- add vips_crop(), a synonym for vips_extract_area()

- rename vips_gammacorrect() as vips_gamma(), now takes 1 / exp

- vips_gamma() works for any format

- add --linear mode to vipsthumbnail

- support XYZ as a PCS for vips_icc_import() and vips_icc_export()

- add --strip option to jpegsave

- added vips_gaussblur() convenience function

- added --vips-profile, records and dumps thread timing and memory use info

- added vipsprofile, visualises --vips-profile output

- auto-vectorization-friendly inner loops

- added vips::init() and vips::shutdown() to C++ API

- reuse pixel buffers on sharing to reduce mem cycling

- conv is SMALLTILE, huge mem use saving on wide images

- vipsthumbnail has a --crop option

- remove video4linux1 code, it was useless on all modern linuxes

- redone freq filter builders as classes

- redone im_fwfft(), im_invfft(), im_freqflt(), im_disp_ps(), im_fractsurf(),

im_phasecor() as classes

- vips_colourspace() allows B_W, GREY16, RGB16 as source / target

- added vips_thread_shutdown(), thanks Lovell

- vips_linear() has a uchar output mode

- redone im_cntlines(), im_rank() as classes

- move im_zerox() to deprecated, it wasm't very useful

9/1/14 started 7.36.6

- fix some clang compiler warnings

20/11/13 started 7.36.5

- better cache sizing in unbuffered sequential mode

- allow larger tile_size in dzsave

- remove use of PATH_MAX to help gnu hurd

- fix vips_hist_match()

15/11/13 started 7.36.4

- improve compat with im_init_world()

18/10/13 started 7.36.3

- fix compiler warnings in ubuntu 13.10

- reverse similarity rotation direction to match the convention used

elsewhere in vips

- fix blocked caching of sequential load operations

- fix cache flags

- fix --delete option to vipsthumbnail

10/10/13 started 7.36.2

- better jpeg startup

- rename jpeg bool type to reduce confusion between libraries

3/10/13 started 7.36.1

- fix to help OS X build

3/10/13 started 7.36.0

- version bump

1/7/13 started 7.35.0

- added vips_matrixload() and vips_matrixsave(), load and save vips mat format

- rename image arrays as image matrices ... INTERPRETATION_ARRAY ->

INTERPRETATION_MATRIX etc.

- rewrite im_buildlut(), im_identity*(), im_maplut(), im_falsecolour(),

im_gammacorrect(), im_histgr(), im_histcum(), im_histnorm(), im_heq(),

im_histnD(), im_histindexed(), im_histspec(), im_invertlut(), im_lhisteq(),

im_stdif(), im_project(), im_profile(), im_tone_build*(), im_mpercent*(),

im_ismonotonic() as classes

- vips_hist_local(), vips_stdif() do any number of bands

- thin vips8 wrapper for im_histplot()

- added vips_error_freeze() / vips_error_thaw()

- used freeze() / thaw() to stop file format sniffers logging spurious errors

- vipsthumbnail uses embedded jpg thumbnails if it can

- rename vips_diag() as vips_info(), add --vips-info flag

- deprecate im_hsp()

- added vips_webpload(), vips_webpload_buffer(), vips_webpsave(),

vips_webpsave_buffer(), vips_webpsave_mime()

- tiff reader allows separate planes for strip read

- tiff reader and writer allow many more formats, eg. 32-bit int, complex, etc.

- tiff reader and writer allow any number of bands

- added vips_image_new_matrixv()

- dzsave basename param now called filename, so you can use .dz as a

destination (basename is still there but deprecated)

- new _UNBUFFERED sequential mode saves memory in some important cases

- vips_conv() is a simple wrapper over the old convolution functions

- new optimize_coding param for jpeg write produces optimal Huffman tables,

thanks Lovell