Writing double negates is silly, but it might be hidden by revset alias
if we added such feature.
I made fold_redundant_expression() a separate step from fold_difference()
since I'll probably want to apply the cleanup step before rewriting filter
expressions.
Because a unary negation node '~y' is more primitive than the corresponding
difference node 'x~y', '~y' is easier to deal with while rewriting the tree.
That's the main reason to add RevsetExpression::NotIn node.
As we have a NotIn node, it makes sense to add an operator for that. This
patch reuses '~' token, which I feel intuitive since the other set operators
looks like bitwise ops. Another option is '!'.
The unary '~' operator has the highest precedence among the set operators,
but they are lower than the ranges. This might be counter intuitive, but
useful because a prefix range ':x' can be negated without parens.
Maybe we can remove the redundant infix operator 'x ~ y', but it isn't
decided yet.
Function parameters are processed as local symbols while substituting
alias expression. This isn't as efficient as Mercurial which caches
a tree of fully-expanded function template, but that wouldn't matter in
practice.
Let's acknowledge everyone's contributions by replacing "Google LLC"
in the copyright header by "The Jujutsu Authors". If I understand
correctly, it won't have any legal effect, but maybe it still helps
reduce concerns from contributors (though I haven't heard any
concerns).
Google employees can read about Google's policy at
go/releasing/contributions#copyright.
Follows up c5ed3e14776b. Now change/commit ids are resolved at the same
precedence, which means there are at least three types of ambiguity.
I don't think we would need to discriminate these.
Because the use of the change id is recommended, any operation should abort
if a valid change id happens to match a commit id. We still try the commit
id lookup first as the change id lookup is more costly.
Ambiguous change/commit id is reported as AmbiguousCommitIdPrefix for now.
Maybe we can merge AmbiguousCommit/ChangeIdPrefix errors into one?
Closes#799
Since syntactic information like symbol or function name is lost after
parse(), alias substitution is inserted to the middle of the post-parsing
stage, not after the whole RevsetExpression tree is built. This is the main
difference from Mercurial. Mercurial also caches parsed aliases, but I don't
think that would have a measurable impact.
The CLI will load aliases from config, insert them one by one, and warn if
declaration part is invalid. That's why RevsetAliasesMap is a public struct
and needs to be instantiated by the caller.
I'll add aliases map, substitution stack (to detect recursion), and locals
(for function aliases) there. Fortunately, we can avoid shared mutables
so a copyable struct should be good.
parse_function_argument_to_string() doesn't need a workspace_ctx, but there
should be no reason to explicitly nullify it either.
I'm thinking of adding alias expansion at this stage, and it would be a bit
tedious to pass around mutable context by function parameter. So let's reduce
the number of the intermediate functions.
This also produces a better error message.
The expression 'x ~ empty()' is identical to 'x & file(".")', but more
intuitive.
Note that 'x ~ empty()' is slower than 'x & file(".")' since the negative
intersection isn't optimized right now. I think that can be handled as
follows: 'x ~ filter(f)' -> 'x & filter(!f)' -> 'filter(!f, x)'
The idea is to (ab)use pest::error::Error type to pretty-print error
message with code span. pest::error::Error will be constructed upfront
to erase lifetime from Span<'i>/Position<'i>.
Baby step towards embedding matcher in RevsetExpression. If we had a fileset
language or regex pattern, we would probably want to parse it at this stage
so the syntax error can be reported without evaluation.
Since 'merges()' just filters the candidates set per item, it doesn't need
a candidates argument. Perhaps, 'merges(x)' could be a predicate to select
merge commits within a subgraph 'x', but I don't know if that would be
useful.
Since 'filter(expr)' is identical to 'expr & filter()', it can be rewritten
in order to minimize the candidates set to be scanned.
The implementation is somewhat similar to revsetlang.optimize() of Mercurial,
but I've split the tree rewriting logic to several steps. I think that's good
for maintainability and should help us conclude that a recursion will
eventually terminate.
This helps to match '(filter, _) | (_, filter)' to rewrite the expression
tree. Only one predicate is allowed for now, but I think it can be extended
to internalize 'f(c) & g(c)' as '(g*f)(c)' to eliminate redundant lookup
of commit object.
More workspace-derived parameters will be added, and I don't think wrapping
with Option for each makes sense because all parameters should be available
if workspace exists.
Previously an expression 'foo-bar-' failed to parse because
1. try first rule: 'foo-bar-' matches (identifier_part+ ~ '-')+, but the
trailing '' doesn't match identifier_part+
2. fall back to second rule: 'foo' matches identifier_part+
=> (identifier 'foo')
Instead, we need to consume as much (identifier_part ~ '-' ~ ...) as possible
before falling back to the identifier_part rule.
I think the trailing + of identifier_part+ is redundant, so removed it as
well.
