[gegl] Created tag GEGL_0_3_30

[Øyvind Kolås <ok src gnome org>]

The unsigned tag 'GEGL_0_3_30' was created.

Tagger: Øyvind Kolås <pippin gimp org>
Date: Fri Mar 23 19:01:37 2018 +0100

    0.3.30

Changes since the last tag 'GEGL_0_3_28':

Alan Mortensen (1):
      Updated Danish translation

Alexandre Prokoudine (4):
      Update Russian translation
      Update Russian translation
      ru.po: fix formatting error
      Update Russian translation

Debarshi Ray (3):
      build: Link against -lm, where necessary
      build: Use $MATH_LIB instead of -lm for portability and consistency
      shadows-highlights-correction: Prefer copysignf to SIGN

Dimitris Spingos (Δημήτρης Σπίγγος) (2):
      Update Greek translation
      Update Greek translation

Ell (39):
      transform: remove remaining scan-direction flip optimization
      transform: only rasterize inside the transformed polygon
      transform: don't scale input bounds by level when computing scanline limits
      transform: add near-z property
      transform: remove clip-to-input property
      color-to-alpha: merge back color-to-alpha-plus changes
      Bug 793690 - gegl_buffer_{copy,clear}() don't emit "changed" signal ...
      buffer: remove [un]block_changed_signal(); add ITERATOR_NO_NOTIFY
      buffer: mark non-empty duplicate tiles as dirty
      Bug 793937 - Crash in connection with median blur
      recursive-transform: add new operation to workshop
      color-overlay: add new operation to workshop
      recursive-transform: replace brightness/contrast props with fade-color
      matrix: add license and copyright notice to gegl-matrix.h
      matrix: s/GeglMatrix/GeglMatrix3/ in doc comments
      matrix: add gegl_matrix2_determinant()
      buffer: improve box filtering of linear/cubic samplers
      buffer: more improvements to sampler box filtering
      buffer: fix performance regression due to last commit
      matrix: use non-doc comment for gegl_matrix2_determinant()
      buffer: use a nearest sampler in sampler box-filter
      buffer: improve accuracy of sampler box filter
      buffer: use linear sampler in cubic sampler box filtering
      matrix: add gegl_matrix2_is_scale()
      buffer: special case scaling matrices in _gegl_sampler_box_get()
      recursive-transform: add "fade-opacity" property
      recursive-transform: disconnect unused opacity nodes
      buffer, tests: in access.c, streamline tile alignment; fix unaligned buffer_clear()
      tests: #include <string.h> in test-buffer-unaligned-access.c
      buffer: fix single-column case of gegl_buffer_{get,set}()
      buffer: verify alignment in optimized buffer_set/get paths
      buffer: add optimized 6-byte case to gegl_buffer_iterate_write()
      buffer: improve data extent computation in gegl_buffer_set_pattern()
      buffer: fix single-pixel gegl_buffer_get()
      color-overlay: #include <math.h>
      transform: add OpTransform::get_abyss_policy() virtual function
      Bug 698468 - Add abyss-policy property to scale operations
      transform: don't combine transform ops with different near-z values
      transform: fix gegl_transform_prepare() after last commit

Felipeek (1):
      domain-transform: create new operation

Inaki Larranaga Murgoitio (2):
      Update Basque language
      Update Basque language

Jehan (3):
      Bug 794414: new GeglDistanceMetric enum type.
      gegl, operations: s/GEGL_DISTANCE_FOO/GEGL_DISTANCE_METRIC_FOO/
      operations: update "gegl:value-propagate" description.

Jordi Mas (1):
      Update Catalan translation

Mario Blättermann (1):
      Update German translation

Michael Natterer (1):
      buffer: put the "gegl-debug.h" include back in gegl-buffer.c

Piotr Drąg (7):
      Update Polish translation
      Update POTFILES.in
      Update Polish translation
      Update Polish translation
      Update Polish translation
      Update Polish translation
      Update Polish translation

Rūdolfs Mazurs (1):
      Update Latvian translation

Thomas Manni (1):
      mean-curvature-blur: add new operation to common

Øyvind Kolås (58):
      post release version increment
      gegl: reduce memory fetches in boxfilter u8 nl
      buffer: add GEGL_BUFFER_NEAREST flag for gegl_buffer_get
      gegl: add handling of GEGL_BUFFER_NEAREST also for gegl_node_blit
      algorithms: implement alternatives to mempcy in nearest buffer fetch rescale
      gegl: reduce memory accesses in generic 2x2 downscale
      buffer: implement bilinear dispatch along nearest
      gegl: implement bilinear scaling code paths
      gegl: condense boxfilter_u8_nl a bit with macros
      gegl: condense downscale_2x2_u8_nl with macros
      gegl: bilinear/boxfilter do conversion to float in lut
      buffer: make bilinear interpolation from mipmap be default for scaled get
      tests: update sampling buffer tests for bilinear rather than boxfilter
      gegl: add _FILTER_ prefix to scale algorithm buffer flag
      tests: updates to buffer test results after changes to auto filter
      stress/c2g: add enhance-shadows options, default to off
      gegl: use 12bit precision for buffer filters
      gegl: make buffer filter dispatch code handle half float formats properly
      gegl: make GEGL_ALLOCA_THRESHOLD platform dependent
      buffer: do not ask for zeroed buf for samplers, it takes more time
      buffer: add loop reusing smaller tempbuf for scaled blits.
      buffer: align stack allocation on 16bytes
      buffer: reintroduce zeroing of blit scaling tempbufs
      buffer: trim away superfluous #includes
      gegl: 16 byte align stack allocations in buffer scaling
      buffer: make bilinear buffer interpolation consistent with others
      tests: update scaled buffer tests for bi-change
      workshop: add experimental alpha-inpainting op
      test: update mipmap reference images
      workshop: refactor alpha-inpaint to not have global state
      buffer: improve rowstride handling in chunking of scaled blits
      workshop: split core algorithm out of inpaint.c
      workshop: refactor pixel-duster and add content aware enlarger
      configure.ac: depend on babl 0.1.44
      buffer: remove unused alignment helper
      gegl: remove unused format argument from _gegl_get_required_for_scale
      tools: fix segfault in html generator when using clang
      buffer: rewrite headers to avoid inline function ptr declaration
      pixel-duster: split corpus database is 4096 hashtables
      MAINTAINERS: update
      pixel-duster: make more configurable with ifdefs
      pixel-duster: implement color histogram neighborhood discrimination
      pixel-duster: add another symmetry
      build: do not run tests in perf folder by default
      NEWS: update
      build: force disting of perf folder to pass dist-check
      operations: bulk update of reference hashes
      workshop: make ditto op duplicate with nearest, and update refhash
      operations: more reference hash updates
      whirl-pinch: use GEGL_ABYSS_CLAMP when sampling
      operations: more reference hash updates
      docs: exclude v4l op from gallery generation
      operations: more reference hash updates
      operations: set reference-hash for rgb-clip and saturation
      NEWS: fix inaccurate description of new API
      buffer: add mention of buffer-filter api in gegl_buffer_get
      NEWS: add contributors for this release
      release 0.3.30