| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
| |
Some of the nodes classes are missing either one or many of the common comparison methods #hash, #eql? and #==.
This makes comparision and working with the ast sometimes painful, as equality or operations likes array differences (which uses a hash behind the scene) produces unexpected results.
A test has been added that ensures that every descendants of Node:
* have all 3 methods
* that all 3 methods were defined from the same class
* that the class defining all 3 is also a descendant of Node, to avoid the default ones that rely on identity only
|
|
|
|
|
|
|
|
|
|
| |
We need `value` to have a writer for `StatementCache` to work when
prepared statements are disabled. This is something I'd like to revert
eventually, either by disabling that form of caching in that case or
re-introducing something like the old `Bind` collector.
The addition of `nil?` is to make `IS NULL` be inserted correctly,
similar to what we already do with quoted and casted nodes
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is part of a greater refactoring to have the `BindParam` nodes hold
onto their values. We want to generally keep the AST decoupled from what
you're actually doing with those values, but ultimately the usage of
`BindParam` is almost identical to how you'd use `Casted` or `Quoted`.
Forcing consumers of Arel's API to maintain the bind values separately
from the AST makes manipulating the AST essentially impossible, as you
would need to perform a full walk of the AST to determine whether a
given node contains bind parameters, and which value it maps to.
By storing the value on the bind parameter directly, we can collect them
in another AST pass (realistically it'll be part of the same pass that
performs SQL construction for performance reasons). This will
dramatically simplify AST manipulation for Rails or any other consumers
that work with bind params.
As part of this change I've removed the `BindVisitor`, which appears to
be dead code, and had tests break from this change.
|
| |
|
|
|
|
|
|
|
| |
This reverts commit 6d105c7f891a14316eab47dfff3bf1b94f3204e7, reversing
changes made to 437aa3a4bb8ad4f3f4eba299dbb1112852f9c7ac.
This broke Active Record when the values are sql literals
|
| |
|
| |
|
|
|
|
| |
There is no @class variable.
|
|
|
|
| |
Visitor
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Individual commit messages included below ***
* Preliminary support for bitwise operations as infix operators. Tests to follow.
* Added bitwise xor, shift left and shift right operators
* Fixed the BitwiseOr class so it uses the :| operator instead of :&
* All the methods for the bitwise operators in the Arel::Math module now wrap them up in Arel::Nodes::Grouping so the operation becomes isolated like addition and subtraction
* Preliminary set of tests for the new operators
* Updated README with examples of bitwise operations
* Added a new UnaryOperation class which is a riff on the InfixOperation class
* Added tests for UnaryOperation (inspired by InfixOperation tests)
* Added the bitwise not (~) operator as a UnaryOperation
* Added tests for the bitwise not operator
* Added documentation for the bitwise not operator
* Updated gemspec using `rake arel.gemspec`
|
| |
|
| |
|
|\
| |
| | |
Add OrderPredications back into Nodes::Function
|
| |
| |
| |
| | |
Expression)
|
| |
| |
| |
| |
| |
| |
| | |
Explicitly declare if this is case sensitive or not
currently postgres assumes case insensitive regexp
no other databases support regexps
|
|/
|
|
|
|
|
| |
Explicitly declare if this is case sensitive or not
most implementation assume case sensitive
postgres assumes case insensitive
|
| |
|
|
|
|
|
| |
Conflicts:
lib/arel/visitors/to_sql.rb
|
|
|
|
|
|
| |
These two clauses have nearly identical semantics with regards to how
they would be constructed as an AST. It doesn't make sense for their
interfaces to be separate.
|
|
|
|
|
| |
It is impossible to test equality of things constructing trees with bind
params otherwise.
|
|
|
|
|
|
|
|
|
| |
This will allow most consuming code to avoid the deprecation introduced
in 008445d6fd5f825d9b445ac75a7be67f0f7ab52c. The only code which will be
affected is code that is building the `Arel::Table` object manually,
rather than calling `arel_table` on an Active Record class. Hopefully
this case will be rare enough that we don't need to introduce any
additional APIs to work around it.
|
|
|
|
|
|
| |
This constructor parameter was unused for everything except the
convenience methods `to_sql` and `where_sql`. We can pass the engine
into those methods directly.
|
|
|
|
|
|
|
|
|
|
|
|
| |
This removes the need for us to do the re-ordering by walking the AST in
ActiveRecord. We're using a block to communicate with the collector,
since the collector needs to be the thing which knows about the index,
while the visitor is the thing that needs to know the syntax. The
BindParam needs to know about neither of these things, so it's been
changed to stop being a subclass of SqlLiteral
I could also see an alternative implementation using format strings if
for some reason blocks cause a problem.
|
| |
|
| |
|
| |
|
|
|
|
| |
Fixes https://github.com/rails/rails/issues/16913
|
| |
|
| |
|
| |
|
| |
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
* master:
remove order_clauses since we do not use it
fix whitespace and unsupported method args
Add Regexp and NotRegexp nodes for PostgreSQL
Revert "Merge pull request #253 from corrupt952/master"
flatten object.children in visit_Arel_Node_And
Added right and full outer joins
Conflicts:
lib/arel/visitors/to_sql.rb
lib/arel/visitors/visitor.rb
|
| | |
|
| | |
|
|/ |
|
| |
|
|
|
|
|
| |
If we add the casting node to the ast at build time, then we can avoid
doing the lookup at visit time.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
People are often trying to use ARel nodes inside ActiveRecord, and when
they do so, lots of things can break, because ActiveRecord relies on
Array#uniq and sometimes hash key equality to handle values that end up
in wheres, havings, etc. By implementing equality for all the nodes, we
should hopefully be able to prevent any nodes (even nodes containing
other nodes) from failing an equality check they should otherwise pass,
and alleviate many of these errors.
Fixes #130
|
|
|
|
|
| |
Eventually #engine should go away, but until that time, this means that
Table and Nodes::TableAlias can be used more interchangeably.
|
|
|
|
|
|
|
|
|
|
| |
It is untested. There is no `@frame` variable. Presumably it is supposed
to be `@framing`, but changing that shows that some of the tests *are*
setting frame twice.
I don't see why this level of strictness is necessary. If someone
disagrees, they should add a test for this behaviour and make the other
tests pass.
|
|
|
|
|
|
| |
Also, removed unused ordering.rb file, since it is identical
to the one being created in unary.rb already, and isn't
required anywhere.
|
| |
|
| |
|
| |
|
| |
|