Skip to content

Fix serialization of TagSet#2279

Closed
dalen wants to merge 14 commits intopuppetlabs:masterfrom
dalen:tagset_serialization
Closed

Fix serialization of TagSet#2279
dalen wants to merge 14 commits intopuppetlabs:masterfrom
dalen:tagset_serialization

Conversation

@dalen
Copy link
Contributor

@dalen dalen commented Jan 21, 2014

The TagSet class didn't include FormatSupport and didn't have a
to_data_hash method, so it couldn't be serialized to msgpack.

jpartlow and others added 10 commits January 13, 2014 12:17
This test was accidentally checking that the scheduled task was still in
place in the Windows scheduled task list. Because the
Win32::TaskScheduler#delete call seems to be asynchronous, as long as
the test for scheduled task happened fast enough, it was still seen as
present and the test 'passed'.  In fact, puppet resource was scheduling
the task to be removed, and it does get removed.  This test began to fail
when the vms slowed down enough that the broken verification step began
to fail because it was no longer executing fast enough to see the
transient task prior to its deletion completing.

This change corrects the verification step to verify by asserting that
the schtasks.exe list command failed (did not show the deleted task).
And it wraps it in a timeout to allow it a few seconds for the scheduled
tasks state to finalize.
The documentation of the each function used deprecated syntax.
…eduled-task-delete-acceptance

(maint) Fix acceptance test for scheduled_task deletion
It seems that calling String#encode('UTF-16LE') sometimes adds only a single
NULL terminator in both ruby 1.9 and 2.0 on Windows. However, a wide string
is supposed to have two NULL terminators.

For reasons unknown, versions prior to 2012 seem to accept wide strings with
only a single NULL terminator, but in 2012 and 2012R2 the string is often times
contains a non-zero high byte, which causes Windows to read past the single
terminator, and attempt to access a path that does not exist.
Puppet::Util::Windows::File.replace_file now calls FFI to
replace files in place on Windows.
Puppet::Util::Windows::File.wide_string has now been moved to
Puppet::Util::Windows::String.wide_string.

This moves wide_string to lib/puppet/ubilt/windows/string.rb,
adds specs concerning different encodings.

In addition, we've removed Windows::API::WideString usages from
everywhere it was found as it is known to cause issues with Windows 2012.
Puppet::Util::Windows::Process.lookup_privilege_value in particular was
changed to use LookupPrivilegeValueA (ANSI) as we control the privilege name
being passed to it.

Paired with Josh Cooper <josh@puppetlabs.com>
This commit simplifies the process for appending a NULL terminator to
the wide string.
* pr/2261:
  (PUP-1389) Simplify logic for appending extra NULL
  (PUP-1389) ReplaceFile to FFI / move wide_string
  (PUP-1389) Ensure wide strings have double null terminators
Previously, we were appending a NULL to the wide encoded string to work around
a ruby bug fixed in 2.1.

However, we were calling `String#strip`, which makes a copy of the string, and
in the process only preserves the first NULL.

This commit embeds a NULL within the string itself, e.g. "bob\0". Since the
NULL is part of the string data, ruby will preserve it.

Also, `String#+` cannot combine strings with different encodings,
e.g. "a".encode('UTF-16LE') + "\0" results in:

    incompatible character encodings: UTF-16LE and US-ASCII

which is why we need to encode the terminator. The actual message depends on
the default encoding, e.g. US-ASCII, UTF-8.
@puppetcla
Copy link

CLA signed by all contributors.

Jeff McCune and others added 4 commits January 21, 2014 22:32
Without this patch we're getting the following error on Ruby 2.1.0:
`can't modify frozen Symbol (Puppet::Error)`.  This problem is caused by
our monkey patching which attempts to cache Symbol#to_proc return values
instead of creating new instances on every call.

As it turns out, the caching optimization is not necessary on Ruby 1.9.3
and later because Symbol#to_proc implements its own caching system in
these versions.

Here is the full error this patch avoids by not monkey patching
Symbol#to_proc on versions of Ruby prior to 1.9.3.

    bundle exec rspec -fd spec/integration/agent/logging_spec.rb
    /workspace/serious/src/puppet/lib/puppet/util/monkey_patches.rb:79:in `to_proc': Could not autoload puppet/type/component: can't modify frozen Symbol (Puppet::Error)
      from /workspace/serious/src/puppet/lib/puppet/transaction/event.rb:15:in `<class:Event>'
      from /workspace/serious/src/puppet/lib/puppet/transaction/event.rb:8:in `<top (required)>'
      from /workspace/serious/src/puppet/lib/puppet/transaction.rb:13:in `require'
      from /workspace/serious/src/puppet/lib/puppet/transaction.rb:13:in `<class:Transaction>'
      from /workspace/serious/src/puppet/lib/puppet/transaction.rb:11:in `<top (required)>'
      from /workspace/serious/src/puppet/lib/puppet/type/component.rb:4:in `require'
      from /workspace/serious/src/puppet/lib/puppet/type/component.rb:4:in `<top (required)>'
      from /workspace/serious/src/puppet/lib/puppet/util/autoload.rb:61:in `load'
      from /workspace/serious/src/puppet/lib/puppet/util/autoload.rb:61:in `load_file'
      from /workspace/serious/src/puppet/lib/puppet/util/autoload.rb:202:in `load'
      from /workspace/serious/src/puppet/lib/puppet/metatype/manager.rb:159:in `type'
      from /workspace/serious/src/puppet/lib/puppet/type.rb:2466:in `<top (required)>'
      from /workspace/serious/src/puppet/lib/puppet.rb:175:in `require'
      from /workspace/serious/src/puppet/lib/puppet.rb:175:in `<top (required)>'
      from /workspace/serious/src/puppet/spec/spec_helper.rb:16:in `require'
      from /workspace/serious/src/puppet/spec/spec_helper.rb:16:in `<top (required)>'
      from /workspace/serious/src/puppet/spec/integration/agent/logging_spec.rb:2:in `require'
      from /workspace/serious/src/puppet/spec/integration/agent/logging_spec.rb:2:in `<top (required)>'
      from /workspace/serious/src/puppet/.bundle/gems/ruby/2.1.0/gems/rspec-core-2.11.1/lib/rspec/core/configuration.rb:780:in `load'
      from /workspace/serious/src/puppet/.bundle/gems/ruby/2.1.0/gems/rspec-core-2.11.1/lib/rspec/core/configuration.rb:780:in `block in load_spec_files'
      from /workspace/serious/src/puppet/.bundle/gems/ruby/2.1.0/gems/rspec-core-2.11.1/lib/rspec/core/configuration.rb:780:in `map'
      from /workspace/serious/src/puppet/.bundle/gems/ruby/2.1.0/gems/rspec-core-2.11.1/lib/rspec/core/configuration.rb:780:in `load_spec_files'
      from /workspace/serious/src/puppet/.bundle/gems/ruby/2.1.0/gems/rspec-core-2.11.1/lib/rspec/core/command_line.rb:22:in `run'
      from /workspace/serious/src/puppet/.bundle/gems/ruby/2.1.0/gems/rspec-core-2.11.1/lib/rspec/core/runner.rb:69:in `run'
      from /workspace/serious/src/puppet/.bundle/gems/ruby/2.1.0/gems/rspec-core-2.11.1/lib/rspec/core/runner.rb:8:in `block in autorun'
Ruby 2.1.0 includes RDoc 4.1.0 which added a requirement that
subclasses of ClassModule implement 'add_prefix'. This requirement
was added in this commit:

ruby/rdoc@1701151

The rdoc docs for other classes mention that aref_prefix should be
implemented by subclasses, but generally a base class implementation is
provided.  The rdoc doc for ClassModule hasn't been updated to reflect
the new requirement.

This patch simply implements the 'aref_prefix' method.
The TagSet class didn't include FormatSupport and didn't have a
to_data_hash method, so it couldn't be serialized to msgpack.
@dalen
Copy link
Contributor Author

dalen commented Jan 22, 2014

reopened against stable: #2287

@dalen dalen closed this Jan 22, 2014
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

9 participants