]> gitweb.factorcode.org Git - factor.git/log
factor.git
6 years agoFactor: Rename <enum> to <enumerated> to not confuse with ENUM:s
Doug Coleman [Mon, 3 Jul 2017 19:00:44 +0000 (12:00 -0700)]
Factor: Rename <enum> to <enumerated> to not confuse with ENUM:s

6 years agoldcache: fixing article title
Björn Lindqvist [Mon, 3 Jul 2017 16:48:39 +0000 (18:48 +0200)]
ldcache: fixing article title

6 years agoio.standard-paths.unix: ofc /bin/ps should be included...
Björn Lindqvist [Sun, 2 Jul 2017 22:56:55 +0000 (00:56 +0200)]
io.standard-paths.unix: ofc /bin/ps should be included...

6 years agoldcache: new vocab for reading and parsing the /etc/ld.so.cache file
Björn Lindqvist [Sun, 2 Jul 2017 22:47:40 +0000 (00:47 +0200)]
ldcache: new vocab for reading and parsing the /etc/ld.so.cache file

6 years agonumber-game: need to flush the output
Björn Lindqvist [Sun, 2 Jul 2017 22:13:06 +0000 (00:13 +0200)]
number-game: need to flush the output

6 years agofullscreen: looks like the vocab belongs in the windows hierarchy
Björn Lindqvist [Fri, 30 Jun 2017 04:54:29 +0000 (06:54 +0200)]
fullscreen: looks like the vocab belongs in the windows hierarchy

6 years agoio.standard-paths.unix: ifconfig isnt installed everywhere, but ps is
Björn Lindqvist [Sun, 2 Jul 2017 15:17:46 +0000 (17:17 +0200)]
io.standard-paths.unix: ifconfig isnt installed everywhere, but ps is

6 years agochecksums,openssl.libcrypto: make it work with openssl 1.1
Björn Lindqvist [Sat, 1 Jul 2017 20:19:42 +0000 (22:19 +0200)]
checksums,openssl.libcrypto: make it work with openssl 1.1

6 years agoopenssl: docs
Björn Lindqvist [Sat, 1 Jul 2017 11:50:29 +0000 (13:50 +0200)]
openssl: docs

6 years agoopenssl.*,io.sockets.*: support for openssl 1.1
Björn Lindqvist [Sat, 1 Jul 2017 11:41:22 +0000 (13:41 +0200)]
openssl.*,io.sockets.*: support for openssl 1.1

In the new version, a lot of c function names have changed. So a new
global ssl-new-api? is needed to know if the new or old names should be
used.

6 years agoopenssl.*: adding functions that are different in libbsl 1.1.x
Björn Lindqvist [Sat, 1 Jul 2017 07:25:09 +0000 (09:25 +0200)]
openssl.*: adding functions that are different in libbsl 1.1.x

6 years agoui.tools: setting the initial tool dim as a multiple of the default font
Björn Lindqvist [Wed, 28 Jun 2017 12:25:38 +0000 (14:25 +0200)]
ui.tools: setting the initial tool dim as a multiple of the default font

6 years agoVM: removing init_globals(), calling init_mvm() instead
Björn Lindqvist [Tue, 23 Aug 2016 13:25:19 +0000 (15:25 +0200)]
VM: removing init_globals(), calling init_mvm() instead

6 years agoFUEL: make the req-id always a 5-digit number.. prettier that way
Björn Lindqvist [Fri, 30 Jun 2017 00:31:53 +0000 (02:31 +0200)]
FUEL: make the req-id always a 5-digit number.. prettier that way

6 years agoFUEL: improved logging
Björn Lindqvist [Fri, 30 Jun 2017 00:27:42 +0000 (02:27 +0200)]
FUEL: improved logging

point goes to end of buffer and a timestamp is added

6 years agoBuild: let's remove the NO_UI variable
Björn Lindqvist [Thu, 29 Jun 2017 22:24:24 +0000 (00:24 +0200)]
Build: let's remove the NO_UI variable

That option has no effect on the build just confuses people because they
think Factor can be built with/without the UI. The runtime never depends
on any ui libraries.

6 years agoformatting.tests: conditioning some tests to pass on windows
Björn Lindqvist [Wed, 28 Jun 2017 09:17:15 +0000 (11:17 +0200)]
formatting.tests: conditioning some tests to pass on windows

6 years agomath.parser: fill argument must not be an empty byte-array
Björn Lindqvist [Wed, 28 Jun 2017 09:01:32 +0000 (11:01 +0200)]
math.parser: fill argument must not be an empty byte-array

6 years agoui: more docs and tests
Björn Lindqvist [Tue, 27 Jun 2017 21:29:52 +0000 (23:29 +0200)]
ui: more docs and tests

6 years agoui: the ui-thread variable isn't needed
Björn Lindqvist [Tue, 27 Jun 2017 21:15:41 +0000 (23:15 +0200)]
ui: the ui-thread variable isn't needed

6 years agoui: better handling of cascading errors
Björn Lindqvist [Tue, 27 Jun 2017 21:02:25 +0000 (23:02 +0200)]
ui: better handling of cascading errors

The second update-ui is wrapped in a recover so that if it fails
again, the UI system is stopped. That should fix the cascade problem
in #1827.

6 years agoFUEL: can use the factor-on-vocab function instead
Björn Lindqvist [Tue, 27 Jun 2017 17:17:52 +0000 (19:17 +0200)]
FUEL: can use the factor-on-vocab function instead

6 years agoRevert "compiler.cfg.gvn.avail: use ?infimum"
John Benediktsson [Tue, 27 Jun 2017 17:03:41 +0000 (10:03 -0700)]
Revert "compiler.cfg.gvn.avail: use ?infimum"

This reverts commit e6ce2eb45546494fd3503e3390294f1e8522c2db.

6 years agocompiler.cfg.gvn.avail: use ?infimum
Alexander Iljin [Thu, 22 Jun 2017 10:49:01 +0000 (13:49 +0300)]
compiler.cfg.gvn.avail: use ?infimum

6 years agoVM: removing the annotations to please musl (#1836)
Björn Lindqvist [Mon, 26 Jun 2017 15:43:14 +0000 (17:43 +0200)]
VM: removing the annotations to please musl (#1836)

6 years agohelp: fix to make the orphan articles index page render in fuel-help
Björn Lindqvist [Mon, 26 Jun 2017 13:57:58 +0000 (15:57 +0200)]
help: fix to make the orphan articles index page render in fuel-help

6 years agofuel.help: fix for a small help bug on vocabs without words in them
Björn Lindqvist [Mon, 26 Jun 2017 13:53:54 +0000 (15:53 +0200)]
fuel.help: fix for a small help bug on vocabs without words in them

6 years agoFUEL: confirmation prompt for cleaning browsing history seem overkill
Björn Lindqvist [Mon, 26 Jun 2017 12:51:11 +0000 (14:51 +0200)]
FUEL: confirmation prompt for cleaning browsing history seem overkill

6 years agoFUEL: need to specify used vocabs
Björn Lindqvist [Mon, 26 Jun 2017 03:39:26 +0000 (05:39 +0200)]
FUEL: need to specify used vocabs

6 years agofuel.eval: fixing the manifest gunk more
Björn Lindqvist [Mon, 26 Jun 2017 01:46:07 +0000 (03:46 +0200)]
fuel.eval: fixing the manifest gunk more

Some FUEL features doesn't ship a correct usings list so you need to use
the current manifests vocabs.

6 years agoFUEL: this looks unused
Björn Lindqvist [Mon, 26 Jun 2017 01:14:04 +0000 (03:14 +0200)]
FUEL: this looks unused

6 years agofuel.eval: refactoring to remove the globals eval-error and eval-output
Björn Lindqvist [Mon, 26 Jun 2017 00:03:41 +0000 (02:03 +0200)]
fuel.eval: refactoring to remove the globals eval-error and eval-output

6 years agofuel.eval: better manifest handling in eval-in-context
Björn Lindqvist [Sun, 25 Jun 2017 23:39:28 +0000 (01:39 +0200)]
fuel.eval: better manifest handling in eval-in-context

6 years agosequences.extras: add ?supremum and ?infimum
Alexander Iljin [Thu, 22 Jun 2017 10:32:46 +0000 (13:32 +0300)]
sequences.extras: add ?supremum and ?infimum

6 years agoFUEL: workaround for a bug caused by session.el
Björn Lindqvist [Sun, 25 Jun 2017 18:06:45 +0000 (20:06 +0200)]
FUEL: workaround for a bug caused by session.el

6 years agoVM: we can save the process by throwing a normal error here
Björn Lindqvist [Sat, 24 Jun 2017 23:43:31 +0000 (01:43 +0200)]
VM: we can save the process by throwing a normal error here

6 years agoVM: merging the profiling_sample and profiling_sample_count classes
Björn Lindqvist [Sat, 24 Jun 2017 23:15:40 +0000 (01:15 +0200)]
VM: merging the profiling_sample and profiling_sample_count classes

This refactoring makes the code a bit simpler.

6 years agofile-picker.linux: fixing open-file-dialog and save-file-dialog
Björn Lindqvist [Sat, 24 Jun 2017 21:49:07 +0000 (23:49 +0200)]
file-picker.linux: fixing open-file-dialog and save-file-dialog

I had to use utf8 string>alien over utf8 encode because the latter
doesnt null-terminate the string.

6 years agoio.sockets.secure.openssl,openssl.libssl: ENUM: over syntax
Björn Lindqvist [Sat, 24 Jun 2017 14:46:00 +0000 (16:46 +0200)]
io.sockets.secure.openssl,openssl.libssl: ENUM: over syntax

The enum does the same thing as the custom syntax did and is a little
easier to understand.

6 years agocore/: bunch of more docs
Björn Lindqvist [Sat, 24 Jun 2017 14:42:59 +0000 (16:42 +0200)]
core/: bunch of more docs

6 years agobasis/,extra/: removing some unnecessary EXCLUDE declarations
Björn Lindqvist [Sat, 24 Jun 2017 14:38:34 +0000 (16:38 +0200)]
basis/,extra/: removing some unnecessary EXCLUDE declarations

6 years agotools.trace: we can surely hardcode the excluded vocab list
Björn Lindqvist [Fri, 23 Jun 2017 22:14:52 +0000 (00:14 +0200)]
tools.trace: we can surely hardcode the excluded vocab list

6 years agopixel-format-docs: add a missing space 1837/head
Alexander Iljin [Tue, 20 Jun 2017 22:15:38 +0000 (01:15 +0300)]
pixel-format-docs: add a missing space

6 years agosytax-docs: mention that numbers may contain commas
Alexander Iljin [Tue, 20 Jun 2017 18:51:14 +0000 (21:51 +0300)]
sytax-docs: mention that numbers may contain commas

6 years agomodels-docs: update documentation
Alexander Iljin [Tue, 20 Jun 2017 18:50:23 +0000 (21:50 +0300)]
models-docs: update documentation

6 years agoui.gadgets.panes-docs: add a missing parenthesis
Alexander Iljin [Thu, 15 Jun 2017 22:37:21 +0000 (01:37 +0300)]
ui.gadgets.panes-docs: add a missing parenthesis

6 years agosequences-docs: add short help for the words longest and shortest
Alexander Iljin [Sun, 11 Jun 2017 20:14:40 +0000 (23:14 +0300)]
sequences-docs: add short help for the words longest and shortest

6 years agoui.gadgets-docs: fix wording
Alexander Iljin [Wed, 7 Jun 2017 19:33:43 +0000 (22:33 +0300)]
ui.gadgets-docs: fix wording

6 years agomath.parser: allow 0B, 0O, and 0X number prefixes.
John Benediktsson [Thu, 15 Jun 2017 22:53:01 +0000 (15:53 -0700)]
math.parser: allow 0B, 0O, and 0X number prefixes.

6 years agoui.gadgets.line-support: even though 1/.0 parses, change to 1/0.
John Benediktsson [Mon, 12 Jun 2017 14:28:33 +0000 (07:28 -0700)]
ui.gadgets.line-support: even though 1/.0 parses, change to 1/0.

6 years agoFactor.app: update copyright.
John Benediktsson [Sun, 11 Jun 2017 21:16:53 +0000 (14:16 -0700)]
Factor.app: update copyright.

6 years agoprettyprint: make sure float number-base is tested.
John Benediktsson [Sun, 11 Jun 2017 20:45:22 +0000 (13:45 -0700)]
prettyprint: make sure float number-base is tested.

6 years agoprettyprint: remove tests for float .b and .o.
John Benediktsson [Sun, 11 Jun 2017 20:43:02 +0000 (13:43 -0700)]
prettyprint: remove tests for float .b and .o.

6 years agoprettyprint: fix .b, .o, and .h for negative numbers.
John Benediktsson [Sun, 11 Jun 2017 17:57:01 +0000 (10:57 -0700)]
prettyprint: fix .b, .o, and .h for negative numbers.

Also makes these useful for nested printing in different number bases.

6 years agoprettyprint.backend: allow octal and binary float printing.
John Benediktsson [Sun, 11 Jun 2017 17:54:13 +0000 (10:54 -0700)]
prettyprint.backend: allow octal and binary float printing.

6 years agoprettyprint.sections: don't create empty strings with zero indent.
John Benediktsson [Sun, 11 Jun 2017 17:53:55 +0000 (10:53 -0700)]
prettyprint.sections: don't create empty strings with zero indent.

6 years agoui.tools.inspector: improve performance for large arrays and hashtables.
John Benediktsson [Sun, 11 Jun 2017 15:59:45 +0000 (08:59 -0700)]
ui.tools.inspector: improve performance for large arrays and hashtables.

Using trick suggested by @AlexIljin, we compute the longest key and then
set the column widths to be { length-of-longest-key 0 }.

6 years agoformatting: revert change to %d, %o, %b, %x for now.
John Benediktsson [Sat, 10 Jun 2017 20:31:26 +0000 (13:31 -0700)]
formatting: revert change to %d, %o, %b, %x for now.

We should have a format character in the future to format using our Factor
binary, octal, and hex parser.  Maybe using the "alternate form" format
character like Python does? And if we do that, maybe print the 0b, 0o, 0x
prefixes? TBD.

6 years agomath.functions: fix another test impacted by float truncate change.
John Benediktsson [Fri, 9 Jun 2017 18:28:21 +0000 (11:28 -0700)]
math.functions: fix another test impacted by float truncate change.

6 years agoodbc: Minor style fixes and add an error function.
Doug Coleman [Fri, 9 Jun 2017 02:05:03 +0000 (21:05 -0500)]
odbc: Minor style fixes and add an error function.

6 years agobase64: Fix base64 for strings that don't end in equal signs.
Doug Coleman [Fri, 9 Jun 2017 02:02:13 +0000 (21:02 -0500)]
base64: Fix base64 for strings that don't end in equal signs.

Some uses of base64 in the wild do not have trailing ==, such as JWT (JSON Web Token) strings.

6 years agomath.functions: make -0.3 truncate return 0.0 instead of -0.0.
John Benediktsson [Thu, 8 Jun 2017 18:27:48 +0000 (11:27 -0700)]
math.functions: make -0.3 truncate return 0.0 instead of -0.0.

6 years agoformatting: fix bug when using %e and the number rounds up in magnitude
Jon Harper [Sun, 9 Apr 2017 18:29:11 +0000 (20:29 +0200)]
formatting: fix bug when using %e and the number rounds up in magnitude

99/10 0 format-scientific-simple was outputting 1.0e0 because
the mantissa was rounded up, so it had one extra character, and
the exponent was wrong.

6 years agoformatting: change %.0f and %.0e to remove the radix character. It's more standard.
Jon Harper [Sun, 9 Apr 2017 18:26:39 +0000 (20:26 +0200)]
formatting: change %.0f and %.0e to remove the radix character. It's more standard.

6 years agoformatting: make these tests actually test the expected code path
Jon Harper [Sun, 9 Apr 2017 18:25:28 +0000 (20:25 +0200)]
formatting: make these tests actually test the expected code path

6 years agoformatting: don't force the presence of ".0" for %f and %e when precision is 0
Jon Harper [Sun, 26 Feb 2017 15:55:49 +0000 (16:55 +0100)]
formatting: don't force the presence of ".0" for %f and %e when precision is 0

It's misleading because you can think you have more precision than you really
do when looking at the output. So instead of "1.0", we can format it as "1" or
"1.".  And instead of "1.0e+00" we can format "1e+00" or "1.e+00".  In C,
printf does not print the radix character, it's prettier. But since the factor
parser accepts both styles, and keeping the radix character gives a stronger sense
that the number is approximate, I prefer to keep the radix character..

6 years agoformatting-docs, document ' ' as an alternative to '+' for positive numbers sign...
Jon Harper [Sun, 26 Feb 2017 15:14:52 +0000 (16:14 +0100)]
formatting-docs, document ' ' as an alternative to '+' for positive numbers sign prefixing

6 years agoformatting, allow uppercase exponent for bin floats in base 2 and 8
Jon Harper [Sun, 26 Feb 2017 14:54:35 +0000 (15:54 +0100)]
formatting, allow uppercase exponent for bin floats in base 2 and 8

6 years agoformatting, make %d, %o, %b work for ratios and floats
Jon Harper [Sun, 26 Feb 2017 14:50:38 +0000 (15:50 +0100)]
formatting, make %d, %o, %b work for ratios and floats

6 years agoformatting, fix %f and %e for ratios and integers
Jon Harper [Fri, 24 Feb 2017 16:01:45 +0000 (17:01 +0100)]
formatting, fix %f and %e for ratios and integers

6 years agoformatting: fix missing '+' for %+e with negative exponent
Jon Harper [Sun, 26 Feb 2017 14:15:47 +0000 (15:15 +0100)]
formatting: fix missing '+' for %+e with negative exponent

6 years agomove math.extras:round-to-even to math.functions to use in basis:formatting
Jon Harper [Fri, 24 Feb 2017 16:00:07 +0000 (17:00 +0100)]
move math.extras:round-to-even to math.functions to use in basis:formatting

Also add round-to-odd in case it is needed.
Also change float rounding to be independent of the current rounding mode

6 years agoadd basis.math.functions.integer-logs: exact integer logarithms
Jon Harper [Fri, 24 Feb 2017 12:14:00 +0000 (13:14 +0100)]
add basis.math.functions.integer-logs: exact integer logarithms

6 years agomath.functions, speed up truncate for floats
Jon Harper [Sat, 25 Feb 2017 15:36:36 +0000 (16:36 +0100)]
math.functions, speed up truncate for floats

6 years agomath.functions: fix truncate and round to output -0.0 for negative floats near 0
Jon Harper [Sat, 25 Feb 2017 14:34:38 +0000 (15:34 +0100)]
math.functions: fix truncate and round to output -0.0 for negative floats near 0

6 years agomath.functions, fix round for ratios between -1/2 and -1
Jon Harper [Sat, 25 Feb 2017 13:37:23 +0000 (14:37 +0100)]
math.functions, fix round for ratios between -1/2 and -1

it was outputting 1 instead of -1

6 years agoio.directories.windows: add size slot to windows-directory-entry tuple
Alexander Iljin [Mon, 5 Jun 2017 10:18:22 +0000 (13:18 +0300)]
io.directories.windows: add size slot to windows-directory-entry tuple

6 years agoAdd .gitattributes to enforce EOL policy for factor sources
Alexander Iljin [Mon, 5 Jun 2017 20:02:52 +0000 (23:02 +0300)]
Add .gitattributes to enforce EOL policy for factor sources

6 years agohelp.lint.spaces: new vocab
Alexander Iljin [Mon, 5 Jun 2017 17:17:57 +0000 (20:17 +0300)]
help.lint.spaces: new vocab

6 years ago*-docs: replace double spaces with single spaces
Alexander Iljin [Mon, 5 Jun 2017 17:13:54 +0000 (20:13 +0300)]
*-docs: replace double spaces with single spaces

Code examples and other places where multiple spaces were used to align
text were left intact.

6 years agotest runner: non-zero exit code in case of failed tests
Robert Vollmert [Fri, 5 May 2017 16:11:33 +0000 (18:11 +0200)]
test runner: non-zero exit code in case of failed tests

This makes it easier to detect unit test failures from the
command line.

6 years agoclasses.intersection: Fix typo in commented-out unit-test.
Doug Coleman [Sun, 4 Jun 2017 21:14:22 +0000 (16:14 -0500)]
classes.intersection: Fix typo in commented-out unit-test.

6 years agounmaintained: New repo here: https://github.com/factor/factor-unmaintained
Doug Coleman [Sun, 4 Jun 2017 20:39:41 +0000 (15:39 -0500)]
unmaintained: New repo here: https://github.com/factor/factor-unmaintained

It's confusing to ripgrep through unmaintained/ and this eliminates all
incompatible/outdated code in the Factor repository.

Please contribute the worthy code back to Factor and remove it from factor-unmaintained.

6 years agotools.test: Make the flag public. Finish porting tester changes to fuzzer. unmaintained
Doug Coleman [Sun, 4 Jun 2017 19:47:35 +0000 (14:47 -0500)]
tools.test: Make the flag public. Finish porting tester changes to fuzzer.

6 years agoclasses.singleton: Update another test.
Doug Coleman [Sun, 4 Jun 2017 17:57:38 +0000 (12:57 -0500)]
classes.singleton: Update another test.

6 years agotools.deploy: Change unit-test to long-unit-test for deploy tests.
Doug Coleman [Sun, 4 Jun 2017 17:55:58 +0000 (12:55 -0500)]
tools.deploy: Change unit-test to long-unit-test for deploy tests.

long-unit-tests are enabled by default.

6 years agotools.test: Add long-unit-test and refactor a bit.
Doug Coleman [Sun, 4 Jun 2017 17:30:51 +0000 (12:30 -0500)]
tools.test: Add long-unit-test and refactor a bit.

6 years agosingleton: Move definitions to top level in tests.
Doug Coleman [Sun, 4 Jun 2017 15:41:26 +0000 (10:41 -0500)]
singleton: Move definitions to top level in tests.

6 years agoNmakefile: add the /largeaddressaware linker option for 32-bit builds
Alexander Iljin [Sat, 3 Jun 2017 16:52:37 +0000 (19:52 +0300)]
Nmakefile: add the /largeaddressaware linker option for 32-bit builds

This somewhat mitigates issue #1828 by allowing more memory to be allocated
for the 32-bit apps on some systems. Has no effect on 32-bit Windows XP.

6 years agocuda.devices: Utility word to print all cuda devices.
Doug Coleman [Sun, 4 Jun 2017 01:59:51 +0000 (20:59 -0500)]
cuda.devices: Utility word to print all cuda devices.

6 years agokernel: Fix the docs for curry/compose a bit.
Doug Coleman [Sat, 3 Jun 2017 07:10:06 +0000 (02:10 -0500)]
kernel: Fix the docs for curry/compose a bit.

6 years agocompiler.tree.propagation.call-effect: I deleted an important method on compose on...
Doug Coleman [Sat, 3 Jun 2017 04:31:35 +0000 (23:31 -0500)]
compiler.tree.propagation.call-effect: I deleted an important method on compose on accident.

6 years agosequences: Remove start for real in favor of subseq-start.
Doug Coleman [Fri, 2 Jun 2017 22:40:13 +0000 (17:40 -0500)]
sequences: Remove start for real in favor of subseq-start.

Update a usage of start that was masked by the ALIAS:

Compatibility/migration is a .99 feature.

6 years agofactor: last iota -> <iota> fix.
Doug Coleman [Fri, 2 Jun 2017 22:39:40 +0000 (17:39 -0500)]
factor: last iota -> <iota> fix.

6 years agocore/basis: Rename tuples kernel:curry -> curried, kernel:compose -> composed.
Doug Coleman [Fri, 2 Jun 2017 22:02:08 +0000 (17:02 -0500)]
core/basis: Rename tuples kernel:curry -> curried, kernel:compose -> composed.

Also rename the stack-checker curried -> curried-effect, composed -> composed-effect.

6 years agowords: Remove dummy ``word`` word and comment. Totally bogus.
Doug Coleman [Fri, 2 Jun 2017 21:58:14 +0000 (16:58 -0500)]
words: Remove dummy ``word`` word and comment. Totally bogus.

6 years agoiota: Fix second iota -> <iota> in line several times.
Doug Coleman [Fri, 2 Jun 2017 20:38:05 +0000 (15:38 -0500)]
iota: Fix second iota -> <iota> in line several times.

6 years agocore: Rename iota to <iota> so we can have TUPLE: iota ... ; instead of TUPLE: iota...
Doug Coleman [Thu, 1 Jun 2017 21:59:35 +0000 (16:59 -0500)]
core: Rename iota to <iota> so we can have TUPLE: iota ... ; instead of TUPLE: iota-tuple ... ;

6 years agoui.backend.cocoa.views: Use ?-> to avoid objc lookup errors if backingScaleFactor...
Doug Coleman [Thu, 1 Jun 2017 17:27:33 +0000 (12:27 -0500)]
ui.backend.cocoa.views: Use ?-> to avoid objc lookup errors if backingScaleFactor selector does not exist.

6 years agococoa: Allow ?-> syntax for methods that might not exist.
Doug Coleman [Thu, 1 Jun 2017 16:55:08 +0000 (11:55 -0500)]
cocoa: Allow ?-> syntax for methods that might not exist.

If a method doesn't exist we need to provide a signature so the stack is balanced. This should also allow deploying from macOS versions that do not contain methods to ones that do. This is an alternative to asking a class if it provides a selector.