bookmaker can now generate SkColor.h
This required adding support for #define, typedef,
global constexpr, and fixing various bugs, like
forward declared structs between comments.
Docs-Preview: https://skia.org/?cl=131260
Bug: skia:6898
TBR=caryclark@google.com
Change-Id: I6bee0c6f5c3a6820b04472a318abde8a2523dbbb
Reviewed-on: https://skia-review.googlesource.com/131260
Reviewed-by: Cary Clark <caryclark@skia.org>
Commit-Queue: Cary Clark <caryclark@skia.org>
Auto-Submit: Cary Clark <caryclark@skia.org>
turn off the example that generates the failing fiddle
Docs-Preview: https://skia.org/?cl=132401
Bug: skia:8051
NOTRY=true
Change-Id: Ic018dbf9ba4f3b7d948660ca2946291ec37f3555
Reviewed-on: https://skia-review.googlesource.com/132401
Commit-Queue: Cary Clark <caryclark@skia.org>
Auto-Submit: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
- formalize how aliases and substitutions work together
- add constexpr, #define, typedef support
- check for correct description order
- write short enum, struct members
R=caryclark@google.com
Docs-Preview: https://skia.org/?cl=129455
Bug: skia:6898
Change-Id: Id60fc2ed02f38a7ba4e5cad5ef493d8c674e6183
Reviewed-on: https://skia-review.googlesource.com/129455
Commit-Queue: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
Auto-Submit: Cary Clark <caryclark@skia.org>
- add links to types within methods
- add check to see that all references and definitions match
- add style to tables to make them easier to read
- use https everywhere
- remove trailing spaces
- move overview inside class
- split class and struct in summary tables
- add missing #Line
- clean up SkImageInfo constant documentation
- work on SkColor documentation
- allow common phrases to take different parameters
- add more flexibility to generated tables
- tighten token parent requirements
- generalize deprecated and example interfaces
- detect inner constructors
R=caryclark@google.com
Docs-Preview: https://skia.org/?cl=121799
Bug: skia:6898
Change-Id: Ia75a23740b80259460916890b310e2a9f024962a
Reviewed-on: https://skia-review.googlesource.com/121799
Commit-Queue: Cary Clark <caryclark@skia.org>
Auto-Submit: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
typedef SkImageInfo Info saves a little typing at the cost of
obfuscating the parameter type. It's only used in a couple of
places so remove it.
The documentation catalog has gotten a bit stale, so update
that as well.
Docs-Preview: https://skia.org/?cl=128400
Bug: skia:6898
Change-Id: I621743a2b2d9a78d5f62742501eebf5f4327efa0
Reviewed-on: https://skia-review.googlesource.com/128400
Commit-Queue: Cary Clark <caryclark@skia.org>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Auto-Submit: Cary Clark <caryclark@skia.org>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Anonymous enums play havoc with documentation;
there's no name to refer to. It may be that all
enums may either be named or replaced with constexpr
without breaking anything. Try replacing all
anonymous enums in include/core to see what happens.
This names SkCanvas::SaveLayerFlagsSet but leaves
SkCanvas::SaveLayerFlags as a uint32_t, to reduce
risk as compared to review.skia.org/123584.
There's also some chance that external linkage will
break if some client refers to anonymous enum in a way
that could require its address: see
https://stackoverflow.com/questions/22867654/enum-vs-constexpr-for-actual-static-constants-inside-classes
(This CL does not require definitions + declarations)
Brought bookmaker docs in line with this change.
This also tripped over missing code in bookmaker
handling constexpr so added that as well.
R=reed@google.com,bsalomon@google.com
Docs-Preview: https://skia.org/?cl=123920
Docs-Preview: https://skia.org/?cl=123584
Bug: skia:6898
Change-Id: I14a342edcfd59e139ef9e4501f562417c4c60391
Reviewed-on: https://skia-review.googlesource.com/123920
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Cary Clark <caryclark@skia.org>
The generated web files have fallen
slightly behind the bookmaker masters,
so bring them up to date.
TBR=caryclark@google.com
No-Try: true
Docs-Preview: https://skia.org/?cl=123460
Bug: skia:6898
Change-Id: Ia630c923ed8784960e85b4ae6edb661c9113d31b
Reviewed-on: https://skia-review.googlesource.com/123460
Commit-Queue: Cary Clark <caryclark@skia.org>
Auto-Submit: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
SkColor.h uses #define liberally, and has many global symbols,
two things bookmaker hasn't seen in other includes.
Revised .h -> .bmh converter to work with SkColor.h as well
as updating how .bmh indices are built.
Generated SkColor_Reference.bmh for globals, and
SkColor4f_Reference.bmh for class.
Other than the existing comments, this doesn't update the
documentation or add new examples.
Docs-Preview: https://skia.org/?cl=118985TBR=caryclark@google.com
Bug: skia:6898
Change-Id: I5978257ee0e51319823efbe8dfc467a08c99ffe0
Reviewed-on: https://skia-review.googlesource.com/118985
Commit-Queue: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
gm bug7792 had wrong dimensions and clipped out half the draws
bm SkRect_Reference.bmh referred to SkIPoint::center.. which
has been removed.
Docs-Preview: https://skia.org/?cl=120640
Bug: skia:6898
Change-Id: Id4c59c82005e1060d0ca933d9a3650fe4f121264
Reviewed-on: https://skia-review.googlesource.com/120640
Commit-Queue: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
Check to see if a method is inside a struct or class
before increasing its indentation. Also, add missing
description.
Docs-Preview: https://skia.org/?cl=118381
Bug: skia:6898
Change-Id: I9402442a89dafc96d87bf69f157dcfa40cdf6e2b
Reviewed-on: https://skia-review.googlesource.com/118381
Commit-Queue: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
- More rigorous symbol identification creates additional
links, and finds spurious errors like unnecessary
parentheses.
- Fixed code to allow preprocessor directives in
#Code blocks (still more to do).
- Fixed examples waiting on fiddle updates.
- Make SkImageInfo officially documented (update in status.json).
- Fixed enum indentation.
Docs-Preview: https://skia.org/?cl=115656
Bug: skia:6898
Change-Id: Ifcf5cfd3f6c03b3b83155c898b2b035a2f3d443c
Reviewed-on: https://skia-review.googlesource.com/115656
Commit-Queue: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
Newly added MakeRenderTarget doesn't have
an example yet, so just mark it as such;
and remove the doxygen style comments.
MakeFormBackendRenderTarget was missing
TBR=robertphillips@google.com
Docs-Preview: https://skia.org/?cl=117629
NOTRY=true
Bug: skia:6898
Change-Id: Icf6fb7195503b54c96cbf3a5785c2c4b3c0f9d8d
Reviewed-on: https://skia-review.googlesource.com/117629
Commit-Queue: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
fix a couple of fiddles with changed output
rewrite some SkSurface image makers
move towards replacing SkImageInfo.h
R=caryclark@google.com
Docs-Preview: https://skia.org/?cl=115163
Bug: skia:6898
Change-Id: Id9895b051cd457336c30250a69cf95359d1d34d7
Reviewed-on: https://skia-review.googlesource.com/115163
Commit-Queue: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
If a method in an include is marked deprecated, make sure that
the documentation is marked deprecated.
It's OK for the documentation to mark something deprecated that
is not marked as such in the includes since the documentation may
be ahead of the includes.
Fix a couple of mistakes found around deprecated methods.
Docs-Preview: https://skia.org/?cl=114184TBR=caryclark@google.com
Bug: skia:6898
Change-Id: I2bb4c293d7bf28e5d12f9ae01b7be49ce48b9ee4
Reviewed-on: https://skia-review.googlesource.com/114184
Commit-Queue: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
more imageinfo text
rewrote many examples to fix newly exposed compiler warnings
marked a couple of YUV method bodies as deprecated
also cleaned up line endings to use the linux style
R=bsalomon@google.comTBR=caryclark@google.com
Docs-Preview: https://skia.org/?cl=112302
Bug: skia:6898
Change-Id: I89626a27353aa84526f9b9475d927bd0e9d8f0d5
Reviewed-on: https://skia-review.googlesource.com/112302
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Cary Clark <caryclark@google.com>
SkDumpCanvas was replaced by SkDebugCanvas as an undefined reference,
but SkDumpCanvas was not added to undocumented.bmh and caused bookmaker
to generate a broken version of site/user/api/SkCanva_Reference.md
The Skia-Commit-Bot detected that the SkCanvas_Reference.md was
changed, and checked it in, causing the online version of SkCanvas
like https://skia.org/user/api/SkCanvas_Reference#SkCanvas_readPixels
to be truncated.
I assume there is a mistake in bookmaker where it wrote an error
to stdout but failed to abort the tool when the *.md file was partially
written. I'll look into that.
This CL adds SkDebugCanvas to undocumented.bmh to fix the documentation.
TBR=brianosman@google.com,rmistry@google.com
Docs-Preview: https://skia.org/?cl=113700
Bug: skia:6898
Change-Id: Ib4e32eea3e56c178d0016f8ae392e28aab5ffafa
Reviewed-on: https://skia-review.googlesource.com/113700
Commit-Queue: Cary Clark <caryclark@skia.org>
Reviewed-by: Cary Clark <caryclark@skia.org>
This is a reland of 78cb579f33
This time, lowp stages are controlled by !defined(JUMPER_IS_SCALAR), not
by defined(__clang__). The two are usually the same, except when we opt
Clang builds into JUMPER_IS_SCALAR artificially.
Some Google3 builds use compilers old enough that they barf when
compiling our NEON code. It's conceivably also possible to define
JUMPER_IS_SCALAR yourself, but I don't think anyone does that.
Original change's description:
> Reland "make SkJumper stages normal Skia code"
>
> This is a reland of 22e536e3a1
>
> Now with fixed #include paths in SkRasterPipeline_opts.h,
> and -ffp-contract=fast for the :hsw target to minimize
> diffs on non-Windows Clang AVX2/AVX-512 bots.
>
> Original change's description:
> > make SkJumper stages normal Skia code
> >
> > Enough clients are using Clang now that we can say, use Clang to build
> > if you want these software pipeline stages to go fast.
> >
> > This lets us drop the offline build aspect of SkJumper stages, instead
> > building as part of Skia using the SkOpts framework.
> >
> > I think everything should work, except I've (temporarily) removed
> > AVX-512 support. I will put this back in a follow up.
> >
> > I have had to drop Windows down to __vectorcall and our narrower
> > stage calling convention that keeps the d-registers on the stack.
> > I tried forcing sysv_abi, but that crashed Clang. :/
> >
> > Added a TODO to up the same narrower stage calling convention
> > for lowp stages... we just *don't* today, for no good reason.
> >
> > Change-Id: Iaaa792ffe4deab3508d2dc5d0008c163c24b3383
> > Reviewed-on: https://skia-review.googlesource.com/110641
> > Commit-Queue: Mike Klein <mtklein@chromium.org>
> > Reviewed-by: Herb Derby <herb@google.com>
> > Reviewed-by: Florin Malita <fmalita@chromium.org>
>
> Change-Id: I44f2c03d33958e3807747e40904b6351957dd448
> Reviewed-on: https://skia-review.googlesource.com/112742
> Reviewed-by: Mike Klein <mtklein@chromium.org>
Change-Id: I3d71197d4bbb19ca4a94961a97fa2e54d5cbfb0d
Reviewed-on: https://skia-review.googlesource.com/112744
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This reverts commit 78cb579f33.
Reason for revert: lowp should be controlled by defined(JUMPER_IS_SCALAR), not defined(__clang__). So close.
Original change's description:
> Reland "make SkJumper stages normal Skia code"
>
> This is a reland of 22e536e3a1
>
> Now with fixed #include paths in SkRasterPipeline_opts.h,
> and -ffp-contract=fast for the :hsw target to minimize
> diffs on non-Windows Clang AVX2/AVX-512 bots.
>
> Original change's description:
> > make SkJumper stages normal Skia code
> >
> > Enough clients are using Clang now that we can say, use Clang to build
> > if you want these software pipeline stages to go fast.
> >
> > This lets us drop the offline build aspect of SkJumper stages, instead
> > building as part of Skia using the SkOpts framework.
> >
> > I think everything should work, except I've (temporarily) removed
> > AVX-512 support. I will put this back in a follow up.
> >
> > I have had to drop Windows down to __vectorcall and our narrower
> > stage calling convention that keeps the d-registers on the stack.
> > I tried forcing sysv_abi, but that crashed Clang. :/
> >
> > Added a TODO to up the same narrower stage calling convention
> > for lowp stages... we just *don't* today, for no good reason.
> >
> > Change-Id: Iaaa792ffe4deab3508d2dc5d0008c163c24b3383
> > Reviewed-on: https://skia-review.googlesource.com/110641
> > Commit-Queue: Mike Klein <mtklein@chromium.org>
> > Reviewed-by: Herb Derby <herb@google.com>
> > Reviewed-by: Florin Malita <fmalita@chromium.org>
>
> Change-Id: I44f2c03d33958e3807747e40904b6351957dd448
> Reviewed-on: https://skia-review.googlesource.com/112742
> Reviewed-by: Mike Klein <mtklein@chromium.org>
TBR=mtklein@chromium.org,herb@google.com,fmalita@chromium.org
Change-Id: Ie64da98f5187d44e03c0ce05d7cb189d4a6e6663
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/112743
Reviewed-by: Mike Klein <mtklein@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This is a reland of 22e536e3a1
Now with fixed #include paths in SkRasterPipeline_opts.h,
and -ffp-contract=fast for the :hsw target to minimize
diffs on non-Windows Clang AVX2/AVX-512 bots.
Original change's description:
> make SkJumper stages normal Skia code
>
> Enough clients are using Clang now that we can say, use Clang to build
> if you want these software pipeline stages to go fast.
>
> This lets us drop the offline build aspect of SkJumper stages, instead
> building as part of Skia using the SkOpts framework.
>
> I think everything should work, except I've (temporarily) removed
> AVX-512 support. I will put this back in a follow up.
>
> I have had to drop Windows down to __vectorcall and our narrower
> stage calling convention that keeps the d-registers on the stack.
> I tried forcing sysv_abi, but that crashed Clang. :/
>
> Added a TODO to up the same narrower stage calling convention
> for lowp stages... we just *don't* today, for no good reason.
>
> Change-Id: Iaaa792ffe4deab3508d2dc5d0008c163c24b3383
> Reviewed-on: https://skia-review.googlesource.com/110641
> Commit-Queue: Mike Klein <mtklein@chromium.org>
> Reviewed-by: Herb Derby <herb@google.com>
> Reviewed-by: Florin Malita <fmalita@chromium.org>
Change-Id: I44f2c03d33958e3807747e40904b6351957dd448
Reviewed-on: https://skia-review.googlesource.com/112742
Reviewed-by: Mike Klein <mtklein@chromium.org>
This reverts commit 22e536e3a1.
Reason for revert: wrong include path :/
Original change's description:
> make SkJumper stages normal Skia code
>
> Enough clients are using Clang now that we can say, use Clang to build
> if you want these software pipeline stages to go fast.
>
> This lets us drop the offline build aspect of SkJumper stages, instead
> building as part of Skia using the SkOpts framework.
>
> I think everything should work, except I've (temporarily) removed
> AVX-512 support. I will put this back in a follow up.
>
> I have had to drop Windows down to __vectorcall and our narrower
> stage calling convention that keeps the d-registers on the stack.
> I tried forcing sysv_abi, but that crashed Clang. :/
>
> Added a TODO to up the same narrower stage calling convention
> for lowp stages... we just *don't* today, for no good reason.
>
> Change-Id: Iaaa792ffe4deab3508d2dc5d0008c163c24b3383
> Reviewed-on: https://skia-review.googlesource.com/110641
> Commit-Queue: Mike Klein <mtklein@chromium.org>
> Reviewed-by: Herb Derby <herb@google.com>
> Reviewed-by: Florin Malita <fmalita@chromium.org>
TBR=mtklein@chromium.org,herb@google.com,fmalita@chromium.org
Change-Id: I2bdc709c80cdfa6b13ff24e024b3721bef887f46
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/112741
Reviewed-by: Mike Klein <mtklein@chromium.org>
Commit-Queue: Mike Klein <mtklein@chromium.org>
Enough clients are using Clang now that we can say, use Clang to build
if you want these software pipeline stages to go fast.
This lets us drop the offline build aspect of SkJumper stages, instead
building as part of Skia using the SkOpts framework.
I think everything should work, except I've (temporarily) removed
AVX-512 support. I will put this back in a follow up.
I have had to drop Windows down to __vectorcall and our narrower
stage calling convention that keeps the d-registers on the stack.
I tried forcing sysv_abi, but that crashed Clang. :/
Added a TODO to up the same narrower stage calling convention
for lowp stages... we just *don't* today, for no good reason.
Change-Id: Iaaa792ffe4deab3508d2dc5d0008c163c24b3383
Reviewed-on: https://skia-review.googlesource.com/110641
Commit-Queue: Mike Klein <mtklein@chromium.org>
Reviewed-by: Herb Derby <herb@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
This adds the ability to define long phrases
in one place and refer to those phrases in
many places.
Bookmaker has new syntax to support phrase substitution.
When it encounters
#some_phrase_reference#
It substitutes the body of
#PhraseDef some_phrase_reference
text to substitute when encountering the phrase
##
The phrase label must start with a lowercase letter,
and be bracketed by single hash marks, without spaces
between the label and the hash marks.
Docs-Preview: https://skia.org/?cl=111224TBR=caryclark@google.com
Bug: skia:6898
Change-Id: I12c57d916ccedbd86b421377d117399150ada72a
Reviewed-on: https://skia-review.googlesource.com/111224
Reviewed-by: Cary Clark <caryclark@skia.org>
Commit-Queue: Cary Clark <caryclark@skia.org>
