Jon Leech 5300d9f05e Change log for April 15, 2018 Vulkan 1.1.73 spec update:
* Update release number to 73.

Github Issues:

  * Refine swapchain association with surface for slink:VkSwapchainKHR, with
    matching valid usage statements for slink:VkSwapchainCreateInfoKHR and
    discussion following the <<swapchain-wsi-image-create-info>> table
    (public issue 637).
  * Re-remove several valid usage statements from slink:VkImageCreateInfo
    that had previously been removed at the time that
    ename:VK_IMAGE_CREATE_EXTENDED_USAGE_BIT was introduced. These
    statements had incorrectly been restored due to an glitch while merging
    from the old `1.0` branch to the current `master` branch (public issue
    683).

Internal Issues:

  * Fix reference page generation and configure build to generate reference
    pages 1.1 with all extensions, rather than core only, as was the case
    for the 1.0 ref pages (internal issues 484, 1056, 1205).
  * Require that
    slink:VkMemoryDedicatedRequirements::pname:prefersDedicateAllocation is
    ename:VK_TRUE when
    slink:VkMemoryDedicatedRequirements::pname:requiresDedicateAllocation is
    ename:VK_TRUE (internal issue 1222).
  * Fix Ruby extension code so `diff_html` Makefile target works (internal
    issue 1230).
  * Update `genRelease` script to generate 1.1 + all extensions reference
    pages - but not the single-page HTML / PDF versions, which are even
    larger than the API spec (internal issue 1245).

Other Issues:

  * Add missing attributes to `vk.xml` for `VK_ANDROID_native_buffer`.
  * Specify that the slink:VkAttachmentDescription::pname:format member is
    the format of the image *view* that will be used for the attachment.
  * Use core sname:VkPhysicalDeviceFeatures2 in the `structextends` `vk.xml`
    attribute for sname:VkPhysicalDeviceDescriptorIndexingFeaturesEXT and
    sname:VkPhysicalDeviceDescriptorIndexingPropertiesEXT, rather than the
    KHR equivalent it was promoted from.
  * Fix the "`Fragment Input Attachment Interface`" glossary entry to match
    the specification body.
  * Clarify the interaction of sRGB images used as storage or texel buffers
    with <<textures-output-format-conversion, Texel Output Format
    Conversion>>.
  * Moved three valid usage statements from
    slink:VkRenderPassMultiviewCreateInfo up to
    slink:VkRenderPassCreateInfo, and added a new valid usage statement for
    slink:VkRenderPassInputAttachmentAspectCreateInfo.
  * Added valid usage statements for slink:VkBufferMemoryBarrier and
    slink:VkImageMemoryBarrier reflecting the global requirement that
    "`non-sparse resources must be bound to memory before being recorded to
    command`".

New Extensions:
2018-04-15 03:32:07 -07:00

314 lines
12 KiB
Ruby

# Copyright (c) 2016-2018 The Khronos Group Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
require 'asciidoctor/extensions' unless RUBY_ENGINE == 'opal'
include ::Asciidoctor
module Asciidoctor
# Duplicate of "AnyListRx" defined by asciidoctor
# Detects the start of any list item.
#
# NOTE we only have to check as far as the blank character because we know it means non-whitespace follows.
HighlighterAnyListRx = /^(?:#{CG_BLANK}*(?:-|([*.\u2022])\1{0,4}|\d+\.|[a-zA-Z]\.|[IVXivx]+\))#{CG_BLANK}|#{CG_BLANK}*.*?(?::{2,4}|;;)(?:$|#{CG_BLANK})|<?\d+>#{CG_BLANK})/
class ExtensionHighlighterPreprocessorReader < PreprocessorReader
def initialize document, diff_extensions, data = nil, cursor = nil
super(document, data, cursor)
@status_stack = []
@diff_extensions = diff_extensions
@tracking_target = nil
end
# This overrides the default preprocessor reader conditional logic such
# that any extensions which need highlighting and are enabled have their
# ifdefs left intact.
def preprocess_conditional_directive directive, target, delimiter, text
# If we're tracking a target for highlighting already, don't need to do
# additional processing unless we hit the end of that conditional
# section
# NOTE: This will break if for some absurd reason someone nests the same
# conditional inside itself.
if @tracking_target != nil && directive == 'endif' && @tracking_target == target.downcase
@tracking_target = nil
elsif @tracking_target
return super(directive, target, delimiter, text)
end
# If it's an ifdef or ifndef, push the directive onto a stack
# If it's an endif, pop the last one off.
# This is done to apply the next bit of logic to both the start and end
# of an conditional block correctly
status = directive
if directive == 'endif'
status = @status_stack.pop
else
@status_stack.push status
end
# If the status is negative, we need to still include the conditional
# text for the highlighter, so we replace the requirement for the
# extension attribute in question to be not defined with an
# always-undefined attribute, so that it evaluates to true when it needs
# to.
# Undefined attribute is currently just the extension with "_undefined"
# appended to it.
modified_target = target.downcase
if status == 'ifndef'
@diff_extensions.each do | extension |
modified_target.gsub!(extension, extension + '_undefined')
end
end
# Call the original preprocessor
result = super(directive, modified_target, delimiter, text)
# If any of the extensions are in the target, and the conditional text
# isn't flagged to be skipped, return false to prevent the preprocessor
# from removing the line from the processed source.
unless @skipping
@diff_extensions.each do | extension |
if target.downcase.include?(extension)
if directive != 'endif'
@tracking_target = target.downcase
end
return false
end
end
end
return result
end
# Identical to preprocess_conditional_directive, but older versions of
# Asciidoctor used a different name, so this is there to override the same
# method in older versions.
# This is a pure c+p job for awkward inheritance reasons (see use of
# the super() keyword :|)
# At some point, will rewrite to avoid this mess, but this fixes things
# for now without breaking things for anyone.
def preprocess_conditional_inclusion directive, target, delimiter, text
# If we're tracking a target for highlighting already, don't need to do
# additional processing unless we hit the end of that conditional
# section
# NOTE: This will break if for some absurd reason someone nests the same
# conditional inside itself.
if @tracking_target != nil && directive == 'endif' && @tracking_target == target.downcase
@tracking_target = nil
elsif @tracking_target
return super(directive, target, delimiter, text)
end
# If it's an ifdef or ifndef, push the directive onto a stack
# If it's an endif, pop the last one off.
# This is done to apply the next bit of logic to both the start and end
# of an conditional block correctly
status = directive
if directive == 'endif'
status = @status_stack.pop
else
@status_stack.push status
end
# If the status is negative, we need to still include the conditional
# text for the highlighter, so we replace the requirement for the
# extension attribute in question to be not defined with an
# always-undefined attribute, so that it evaluates to true when it needs
# to.
# Undefined attribute is currently just the extension with "_undefined"
# appended to it.
modified_target = target.downcase
if status == 'ifndef'
@diff_extensions.each do | extension |
modified_target.gsub!(extension, extension + '_undefined')
end
end
# Call the original preprocessor
result = super(directive, modified_target, delimiter, text)
# If any of the extensions are in the target, and the conditional text
# isn't flagged to be skipped, return false to prevent the preprocessor
# from removing the line from the processed source.
unless @skipping
@diff_extensions.each do | extension |
if target.downcase.include?(extension)
if directive != 'endif'
@tracking_target = target.downcase
end
return false
end
end
end
return result
end
end
class Highlighter
def initialize
@delimiter_stack = []
@current_anchor = 1
end
def highlight_marks line, previous_line, next_line
if !(line.start_with? 'endif')
# Any intact "ifdefs" are sections added by an extension, and
# "ifndefs" are sections removed.
# Currently don't track *which* extension(s) is/are responsible for
# the addition or removal - though it would be possible to add it.
if line.start_with? 'ifdef'
role = 'added'
else # if line.start_with? 'ifndef'
role = 'removed'
end
# Create an anchor with the current anchor number
anchor = '[[difference' + @current_anchor.to_s + ']]'
# Figure out which markup to use based on the surrounding text
# This is robust enough as far as I can tell, though we may want to do
# something more generic later since currently it relies on the fact
# that if you start inside a list or paragraph, you'll end in the same
# list or paragraph and not cross to other blocks.
# In practice it *might just work* but it also might not.
# May need to consider what to do about this in future - maybe just
# use open blocks for everything?
highlight_delimiter = :inline
if (HighlighterAnyListRx.match(next_line) != nil)
# NOTE: There's a corner case here that should never be hit (famous last words)
# If a line in the middle of a paragraph begins with an asterisk and
# then whitespace, this will think it's a list item and use the
# wrong delimiter.
# That shouldn't be a problem in practice though, it just might look
# a little weird.
highlight_delimiter = :list
elsif previous_line.strip.empty?
highlight_delimiter = :block
end
# Add the delimiter to the stack for the matching 'endif' to consume
@delimiter_stack.push highlight_delimiter
# Add an appropriate method of delimiting the highlighted areas based
# on the surrounding text determined above.
if highlight_delimiter == :block
return ['', anchor, ":role: #{role}", '']
elsif highlight_delimiter == :list
return ['', anchor, "[.#{role}]", '~~~~~~~~~~~~~~~~~~~~', '']
else #if highlight_delimiter == :inline
return [anchor + ' [.' + role + ']##']
end
else # if !(line.start_with? 'endif')
# Increment the anchor when we see a matching endif, and generate a
# link to the next diff section
@current_anchor = @current_anchor + 1
anchor_link = '<<difference' + @current_anchor.to_s + ', =>>>'
# Close the delimited area according to the previously determined
# delimiter
highlight_delimiter = @delimiter_stack.pop
if highlight_delimiter == :block
return [anchor_link, '', ':role:', '']
elsif highlight_delimiter == :list
return [anchor_link, '~~~~~~~~~~~~~~~~~~~~', '']
else #if highlight_delimiter == :inline
return [anchor_link + '##']
end
end
end
end
# Preprocessor hook to iterate over ifdefs to prevent them from affecting asciidoctor's processing.
class ExtensionHighlighterPreprocessor < Extensions::Preprocessor
def process document, reader
# Only attempt to highlight extensions that are also enabled - if one
# isn't, warn about it and skip highlighting that extension.
diff_extensions = document.attributes['diff_extensions'].downcase.split(' ')
actual_diff_extensions = []
diff_extensions.each do | extension |
if document.attributes.has_key?(extension)
actual_diff_extensions << extension
else
puts 'The ' + extension + ' extension is not enabled - changes will not be highlighted.'
end
end
# Create a new reader to return, which leaves extension ifdefs that need highlighting intact beyond the preprocess step.
extension_preprocessor_reader = ExtensionHighlighterPreprocessorReader.new(document, actual_diff_extensions, reader.lines)
highlighter = Highlighter.new
new_lines = []
# Store the old lines so we can reference them in a non-trivial fashion
old_lines = extension_preprocessor_reader.read_lines()
old_lines.each_index do | index |
# Grab the previously processed line
# This is used by the highlighter to figure out if the highlight will
# be inline, or part of a block.
if index > 0
previous_line = old_lines[index - 1]
else
previous_line = ''
end
# Current line to process
line = old_lines[index]
# Grab the next line to process
# This is used by the highlighter to figure out if the highlight is
# between list elements or not - which need special handling.
if index < (old_lines.length - 1)
next_line = old_lines[index + 1]
else
next_line = ''
end
# Highlight any preprocessor directives that were left intact by the
# custom preprocessor reader.
if line.start_with?( 'ifdef::', 'ifndef::', 'endif::')
new_lines += highlighter.highlight_marks(line, previous_line, next_line)
else
new_lines << line
end
end
# Return a new reader after preprocessing - this takes care of creating
# the AST from the new source.
Reader.new(new_lines)
end
end
class AddHighlighterCSS < Extensions::Postprocessor
HighlighterStyleCSS = [
'.added {',
' background-color: lime;',
' border-color: green;',
' padding:1px;',
'}',
'.removed {',
' background-color: pink;',
' border-color: red;',
' padding:1px;',
'}',
'</style>']
def process document, output
output.sub! '</style>', HighlighterStyleCSS.join("\n")
end
end
end