Struct rustc_span::span_encoding::Span [−][src]
Expand description
A compressed span.
Whereas SpanData
is 12 bytes, which is a bit too big to stick everywhere, Span
is a form that only takes up 8 bytes, with less space for the length and
context. The vast majority (99.9%+) of SpanData
instances will fit within
those 8 bytes; any SpanData
whose fields don’t fit into a Span
are
stored in a separate interner table, and the Span
will index into that
table. Interning is rare enough that the cost is low, but common enough
that the code is exercised regularly.
An earlier version of this code used only 4 bytes for Span
, but that was
slower because only 80–90% of spans could be stored inline (even less in
very large crates) and so the interner was used a lot more.
Inline (compressed) format:
span.base_or_index == span_data.lo
span.len_or_tag == len == span_data.hi - span_data.lo
(must be<= MAX_LEN
)span.ctxt == span_data.ctxt
(must be<= MAX_CTXT
)
Interned format:
span.base_or_index == index
(indexes into the interner table)span.len_or_tag == LEN_TAG
(high bit set, all other bits are zero)span.ctxt == 0
The inline form uses 0 for the tag value (rather than 1) so that we don’t need to mask out the tag bit when getting the length, and so that the dummy span can be all zeroes.
Notes about the choice of field sizes:
base
is 32 bits in bothSpan
andSpanData
, which means thatbase
values never cause interning. The number of bits needed forbase
depends on the crate size. 32 bits allows up to 4 GiB of code in a crate.len
is 15 bits inSpan
(a u16, minus 1 bit for the tag) and 32 bits inSpanData
, which means that largelen
values will cause interning. The number of bits needed forlen
does not depend on the crate size. The most common numbers of bits forlen
are from 0 to 7, with a peak usually at 3 or 4, and then it drops off quickly from 8 onwards. 15 bits is enough for 99.99%+ of cases, but larger values (sometimes 20+ bits) might occur dozens of times in a typical crate.ctxt
is 16 bits inSpan
and 32 bits inSpanData
, which means that largectxt
values will cause interning. The number of bits needed forctxt
values depend partly on the crate size and partly on the form of the code. No crates inrustc-perf
need more than 15 bits forctxt
, but larger crates might need more than 16 bits.
In order to reliably use parented spans in incremental compilation,
the dependency to the parent definition’s span. This is performed
using the callback SPAN_TRACK
to access the query engine.
Fields
base_or_index: u32
len_or_tag: u16
ctxt_or_zero: u16
Implementations
Creates a fresh expansion with given properties. Expansions are normally created by macros, but in some cases expansions are created for other compiler-generated code to set per-span properties like allowed unstable features. The returned span belongs to the created expansion and has the new properties, but its location is inherited from the current span.
pub fn fresh_expansion_with_transparency(
self,
expn_data: ExpnData,
transparency: Transparency,
ctx: impl HashStableContext
) -> Span
pub fn mark_with_reason(
self,
allow_internal_unstable: Option<Lrc<[Symbol]>>,
reason: DesugaringKind,
edition: Edition,
ctx: impl HashStableContext
) -> Span
pub fn mark_with_reason(
self,
allow_internal_unstable: Option<Lrc<[Symbol]>>,
reason: DesugaringKind,
edition: Edition,
ctx: impl HashStableContext
) -> Span
Reuses the span but adds information like the kind of the desugaring and features that are allowed inside this span.
Internal function to translate between an encoded span and the expanded representation. This function must not be used outside the incremental engine.
Returns true
if this span comes from a macro or desugaring.
Returns true
if span
originates in a derive-macro’s expansion.
Returns a new span representing an empty span at the beginning of this span.
Returns a new span representing an empty span at the end of this span.
Returns self
if self
is not the dummy span, and other
otherwise.
Returns true
if the spans are equal with regards to the source text.
Use this instead of ==
when either span could be generated code,
and you only care that they point to the same bytes of source text.
Returns Some(span)
, where the start is trimmed by the end of other
.
Returns the source span – this is either the supplied span, or the span for the macro callsite that expanded to it.
The Span
for the tokens in the previous macro expansion from which self
was generated,
if any.
Walk down the expansion ancestors to find a span that’s contained within outer
.
Returns the source callee.
Returns None
if the supplied span has no expansion trace,
else returns the ExpnData
for the macro definition
corresponding to the source callsite.
Checks if a span is “internal” to a macro in which #[unstable]
items can be used (that is, a macro marked with
#[allow_internal_unstable]
).
Checks if this span arises from a compiler desugaring of kind kind
.
Returns the compiler desugaring that created this span, or None
if this span is not from a desugaring.
Checks if a span is “internal” to a macro in which unsafe
can be used without triggering the unsafe_code
lint.
Returns a Span
that would enclose both self
and end
.
____ ___
self lorem ipsum end
^^^^^^^^^^^^^^^^^^^^
Returns a Span
between the end of self
to the beginning of end
.
____ ___
self lorem ipsum end
^^^^^^^^^^^^^
Returns a Span
from the beginning of self
until the beginning of end
.
____ ___
self lorem ipsum end
^^^^^^^^^^^^^^^^^
Equivalent of Span::def_site
from the proc macro API,
except that the location is taken from the self
span.
Equivalent of Span::call_site
from the proc macro API,
except that the location is taken from the self
span.
Equivalent of Span::mixed_site
from the proc macro API,
except that the location is taken from the self
span.
Produces a span with the same location as self
and context produced by a macro with the
given ID and transparency, assuming that macro was defined directly and not produced by
some other macro (which is the case for built-in and procedural macros).
Trait Implementations
Hashes a span in a stable way. We can’t directly hash the span’s BytePos
fields (that would be similar to hashing pointers, since those are just
offsets into the SourceMap
). Instead, we hash the (file name, line, column)
triple, which stays the same even if the containing SourceFile
has moved
within the SourceMap
.
Also note that we are hashing byte offsets for the column, not unicode codepoint offsets. For the purpose of the hash that’s sufficient. Also, hashing filenames is expensive so we avoid doing it twice when the span starts and ends in the same file, which is almost always the case.
This method returns an ordering between self
and other
values if one exists. Read more
This method tests less than (for self
and other
) and is used by the <
operator. Read more
This method tests less than or equal to (for self
and other
) and is used by the <=
operator. Read more
This method tests greater than (for self
and other
) and is used by the >
operator. Read more
Auto Trait Implementations
Blanket Implementations
Mutably borrows from an owned value. Read more
Layout
Note: Most layout information is completely unstable and may even differ between compilations. The only exception is types with certain repr(...)
attributes. Please see the Rust Reference’s “Type Layout” chapter for details on type layout guarantees.
Size: 8 bytes