Commit Graph

5421 Commits

Author SHA1 Message Date
John Cupitt 5860126c58 memory option forces memory load
make memory option to loaders force memory load ... previously, it only
switched between memory and disc

see https://github.com/jcupitt/libvips/issues/972
2018-05-11 14:45:15 +01:00
John Cupitt be4803677d limit memory use on svg load
keep the max memuse down a bit, see https://github.com/jcupitt/libvips/pull/969
2018-05-10 10:50:18 +01:00
John Cupitt d3c1becb2b
Merge pull request #969 from lovell/svgload-cache
Reduce SVG re-renders with taller cache tiles
2018-05-10 10:42:14 +01:00
Lovell Fuller 76bc64ca3c Reduce SVG re-renders with larger cache tiles 2018-05-09 20:20:01 +01:00
John Cupitt c3552ad21e note dzsave angle restrictions
see https://github.com/jcupitt/libvips/issues/968
2018-05-09 16:06:20 +01:00
John Cupitt 258edf3325
Merge pull request #963 from lhecker/master
Fixed configure not picking up JPEG_CFLAGS
2018-05-07 16:59:55 +01:00
Leonard Hecker 70d7dd75d2 Fixed configure not picking up JPEG_CFLAGS 2018-05-07 16:16:37 +02:00
John Cupitt 5649847836 make vips_cache_print_nolock() static
not needed in public API
2018-05-06 17:15:35 +01:00
John Cupitt e1ed5c3735
Merge pull request #960 from lhecker/master
Fixed deadlock in vips_cache_drop_all if vips__cache_dump is true
2018-05-06 17:13:51 +01:00
Leonard Hecker 80a0a24cfa Fixed deadlock in vips_cache_drop_all if vips__cache_dump is true
Both vips_cache_drop_all and vips_cache_print acquire vips_cache_lock
on entry, but vips_cache_drop_all calls vips_cache_print internally if
vips__cache_dump is true, leading to a deadlock situation.
This commit solves the issue using one further level of indirection
and a new vips_cache_print_nolock function.
2018-05-06 12:47:41 +02:00
John Cupitt 88b3644592
Merge pull request #959 from felixbuenemann/fix-png-filters
Fix wrong value for VIPS_FOREIGN_PNG_FILTER_ALL
2018-05-04 07:59:01 +01:00
Felix Bünemann d3439f4d49 Fix wrong value for VIPS_FOREIGN_PNG_FILTER_ALL
The value is derived by oring all the filter flags:

    0x08 | 0x10 | 0x20 | 0x40 | 0x80 == 0xf8

The value was also checked against PNG_ALL_FILTERS in png.h.
2018-05-02 21:50:08 +02:00
John Cupitt e496df598b working! 2018-05-01 15:50:48 +01:00
John Cupitt 9ac45f42e3 needs testing still 2018-04-30 17:38:32 +01:00
John Cupitt b752d2e442
Merge branch 'master' into region-shrink-method 2018-04-27 15:58:24 +01:00
John Cupitt dd07d6ab7d note g_free() for write_to_memory result
see https://github.com/jcupitt/libvips/issues/929
2018-04-27 11:34:56 +01:00
John Cupitt 4d683559aa add a fontfile option to text
use @fontfile to give a font to load ... note that this just loads the
font, you'll need to set it with @font as well

see https://github.com/jcupitt/libvips/issues/950
2018-04-26 21:03:40 +01:00
John Cupitt 6ee7789095 tiny docs update 2018-04-26 16:28:09 +01:00
John Cupitt 6952c57ca0 better temp filename handling
- make access() fail only if we are certain the file does not exist
- remove the g_mkstemp() from vips__temp_name()

should help selinux

see https://github.com/jcupitt/libvips/pull/930
2018-04-26 16:14:12 +01:00
John Cupitt 9714dcdee3 test depth as well in magickload
We were testing width / height / bands when verifying that it was OK to
load a toilet-roll image ... of course we should test depth as well.
2018-04-26 13:57:33 +01:00
John Cupitt 3a9d322574 magick7 tests for all frames equal before load
it used to crash if pages changed in size
2018-04-26 13:45:50 +01:00
John Cupitt 05fd7d9301 oop typo 2018-04-26 11:49:47 +01:00
John Cupitt da6f4fd043 add n-pages metadata item
tiff, magick and pdf load now attach an n-pages metadata item recording
the number of pages in the orginal file

see https://github.com/jcupitt/libvips/issues/953
2018-04-26 08:55:31 +01:00
John Cupitt 4f22e8d1dc add jpeg-chroma-subsample to jpegload
The jpeg loader now sets the field jpeg-chroma-subsample to record
chroma subsample.

See https://github.com/jcupitt/libvips/issues/954
2018-04-25 16:04:20 +01:00
John Cupitt 38ea5937ca more help for graphicsmagick
but still not quite there
2018-04-24 17:14:51 +01:00
John Cupitt c410dc2db8 add format option to magickload
Some magick coders (eg. ICO) don't sniff the filetype from the data, so
when you try to load from a string, imagemagick is unable to pick the
right decode path.

Add a @format option so callers can hint the filetype.

see https://github.com/jcupitt/pyvips/issues/39
2018-04-24 15:06:52 +01:00
John Cupitt a40382fe50 Merge branch '8.6' 2018-04-23 15:18:17 +01:00
John Cupitt 18c50c3231 typo in configure.ac 2018-04-23 15:17:53 +01:00
John Cupitt 12cbe7c2e4 some work on fixing GM
graphicsmagick + magicksave is a bit tricky :(
2018-04-23 13:44:56 +01:00
John Cupitt b8ff9c2069 remove progressive pdfload advice
does not apply to pdfium, and hardly to poppler
2018-04-13 10:15:41 +01:00
John Cupitt 1dbd03eefe update for new pdfium package 2018-04-12 13:10:30 +01:00
John Cupitt fbcdb58f7b remove precise from travis 2018-04-11 19:36:18 +01:00
John Cupitt c0569d25bf final fixups
- add buffer load
- test under valgrind
- code sharing with the poppler pdf loader
2018-04-11 18:53:38 +01:00
John Cupitt 3d696b9c1e Merge branch 'master' into pdfium-experiment 2018-04-11 18:00:33 +01:00
John Cupitt 5b55dd8ed4 Merge branch '8.6' 2018-04-11 09:38:59 +01:00
John Cupitt 24b146790f oop reorder crop to come after cache
or we'll write beyond the buffer end
2018-04-10 15:39:51 +01:00
John Cupitt d1dd41a21f strict round down on jpeg shrink
libjpeg rounds up on shrink-on-load. In some cases this can leave a dark
line along the right and bottom edge, since it only contains (for
example) 1/4 of a pixel of data.

This change adds a crop after jpeg load so that only complete pixels are
output.

See https://github.com/lovell/sharp/issues/1185
2018-04-10 15:18:18 +01:00
John Cupitt f1f2a9de55 fix a warning if tiff is disabled 2018-04-10 15:14:09 +01:00
John Cupitt 15abbbe986 set "interlaced" for jpg and png
set "interlaced" for interlaced png and jpg images
2018-04-09 17:14:38 +01:00
John Cupitt 06802e4110 use O_TMPFILE, if available
If O_TNMPFILE is available, use it. This is a linux extension that
creates an unlinked file, so it'll be closed by the system when the last
associated fd is closed.

see https://github.com/jcupitt/libvips/pull/930
2018-04-09 13:27:57 +01:00
Karthik Karanth 8d04dab0db remove unnecessary args in composite2 2018-04-09 13:14:03 +05:30
Karthik Karanth d06f420763 use vips_embed to position images 2018-04-09 12:57:21 +05:30
John Cupitt 2a82744a10 working at a basic level
TODO

- more code sharing with pdfload.c, eg.  vips_foreign_load_pdf_is_a_buffer()
  and get_flags etc.
- could share the page layout code too
- make pdf.c with base stuff in?
- what about filename encodings
- test transparency
- new_from_buffer needs doing
2018-04-08 18:04:06 +01:00
John Cupitt f3842dcc4b update cpp example
thanks fangqiao

see https://github.com/jcupitt/libvips/issues/932
2018-04-08 11:44:15 +01:00
John Cupitt 05456125b6 builds, links and mostly runs 2018-04-07 17:36:52 +01:00
Karthik Karanth e406015fe8 add args 2018-04-07 13:57:27 +05:30
John Cupitt 56fe608cfb start adding pdfium load 2018-04-06 09:29:45 +01:00
John Cupitt 83d4ee0b0b update find_pdfium for latest packaging
hey ho ouch
2018-04-05 12:10:43 +01:00
John Cupitt 67916e8ead Merge branch 'master' into pdfium-experiment 2018-04-05 11:17:48 +01:00
John Cupitt 1e647a2af4 Merge branch '8.6' 2018-04-04 17:46:36 +01:00