tame/tamer/src/xir/reader.rs

590 lines
21 KiB
Rust
Raw Normal View History

// XIR reader
//
// Copyright (C) 2014-2022 Ryan Specialty Group, LLC.
//
// This file is part of TAME.
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
//! Parse XML files into a XIR [`Token`] stream.
//!
//! This uses [`quick_xml`] as the parser.
tamer: xir: Introduce {Ele,Open,Close}Span This isn't conceptally all that significant of a change, but there was a lot of modify to get it working. I would generally separate this into a commit for the implementation and another commit for the integration, but I decided to keep things together. This serves a role similar to AttrSpan---this allows deriving a span representing the element name from a span representing the entire XIR token. This will provide more useful context for errors---including the tag delimiter(s) means that we care about the fact that an element is in that position (as opposed to some other type of node) within the context of an error. However, if we are expecting an element but take issue with the element name itself, we want to place emphasis on that instead. This also starts to consider the issue of span contexts---a blob of detached data that is `Span` is useful for error context, but it's not useful for manipulation or deriving additional information. For that, we need to encode additional context, and this is an attempt at that. I am interested in the concept of providing Spans that are guaranteed to actually make sense---that are instantiated and manipulated with APIs that ensure consistency. But such a thing buys us very little, practically speaking, over what I have now for TAMER, and so I don't expect to actually implement that for this project; I'll leave that for a personal project. TAMER's already take a lot of my personal interests and it can cause me a lot of grief sometimes (with regards to letting my aspirations cause me more work). DEV-7145
2022-06-24 13:51:49 -04:00
use super::{
error::SpanlessError, CloseSpan, DefaultEscaper, Error, Escaper, OpenSpan,
Token,
};
use crate::{
parse::{ParseError, Parsed, ParsedObject, ParsedResult},
span::Context,
sym::{st::raw::WS_EMPTY, GlobalSymbolInternBytes},
};
use quick_xml::{
self,
events::{
attributes::Attributes, BytesDecl, BytesStart, Event as QuickXmlEvent,
},
Error as QuickXmlError,
};
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
use std::{borrow::Cow, collections::VecDeque, io::BufRead, result};
pub type Result<T> = result::Result<T, Error>;
/// Parse XML into a XIR [`Token`] stream.
///
/// This reader is intended to be used as an [`Iterator`].
///
/// The underlying reader produces events in chunks that are far too
/// large for XIR,
/// so most [`Token`]s retrieved via this call are buffered.
/// Parsing takes place when that buffer is exhausted and the next event
/// is requested from the underlying reader
/// (see [`XmlXirReader::refill_buf`]).
/// Errors can only occur during parsing,
/// and will never occur on buffered tokens.
///
/// [`None`] is returned only on EOF,
/// not on error.
pub struct XmlXirReader<'s, B, S = DefaultEscaper>
where
B: BufRead,
S: Escaper,
{
/// Inner parser.
reader: quick_xml::Reader<B>,
/// Parsing context for reader.
ctx: Context,
/// Buffer for [`quick_xml::Reader`].
readbuf: Vec<u8>,
/// [`Token`] buffer populated upon receiving a new event from
/// `reader`.
///
/// This buffer serves [`Iterator::next`] requests until it is
/// depleted,
/// after which [`XmlXirReader::refill_buf`] requests another token
/// from `reader`.
tokbuf: VecDeque<Token>,
/// System for unescaping string data.
escaper: &'s S,
}
impl<'s, B: BufRead, S: Escaper> XmlXirReader<'s, B, S> {
pub fn new(reader: B, escaper: &'s S, ctx: Context) -> Self {
let mut reader = quick_xml::Reader::from_reader(reader);
// XIR must support mismatched tags so that we are able to represent
// and reconstruct malformed inputs.
// XIRT will handle mismatch errors itself.
reader.check_end_names(false);
Self {
reader,
ctx,
readbuf: Vec::new(),
// This capacity is largely arbitrary,
// but [`Token`]s are small enough that it likely does not
// matter much.
tokbuf: VecDeque::with_capacity(32),
escaper,
}
}
/// Parse using the underlying [`quick_xml::Reader`] and populate the
/// [`Token`] buffer.
///
/// This is intended to be invoked once the buffer has been depleted by
/// [`XmlXirReader::next`].
pub fn refill_buf(&mut self) -> Option<Result<Token>> {
// Clear any previous buffer to free unneeded data.
self.tokbuf.clear();
tamer: xir::reader::XmlXirReader::refill_buf: Clear read buffer This was done in the old reader many months ago, but I somehow forgot to do it here (or forgot to). The new reader was using substantially more memory. Here's how this change affects the memory profile for one of our systems (output from `ms_print`): Before: MB 79.75^ # | # | # @ | @@@@ # @ | @@@ # @@ | @@@ @@@#@ @@@@@ | @@@ @@ #@@@@@@@@@@ | @@@@@@ @@@@ #@@@@@@@@@@ | @@ @@ @@@ @@ @ @@ #@@@@@@@@@@ | @@ @@ @@@ @@@@@ @@ #@@@@@@@@@@ | @@@@@ @@@ @@@@@@ @@ #@@@@@@@@@@ | @@@@@ @@@ @@@@@@ @@ #@@@@@@@@@@ | @@ @@@@@ @@@ @@@@@@ @@ #@@@@@@@@@@ | @ @@ @@ @ @@@@@@ @@@ @@@@@@ @@ #@@@@@@@@@@ | @ @ @@@ @@ @@@ @@@@@@ @@@ @@@@@@ @@ #@@@@@@@@@@ | @ @@@@ @@@@@@@@@@@@@@@@@@@@@ @@@@@@@@@@ @@@ @@@@@@ @@ #@@@@@@@@@@ | @@@ @@@@@@ @@@@@@@@@ @@@@@ @@@@@ @@ @@@@@@@ @@@ @@@@@@ @@ #@@@@@@@@@@ | @@@ @ @@@@ @@@@@@@@@ @@@@@ @@@@@ @@ @@@@@@@ @@@ @@@@@@ @@ #@@@@@@@@@@ | @@@ @@@ @@@@ @@@@@@@@@ @@@@@ @@@@@ @@ @@@@@@@ @@@ @@@@@@ @@ #@@@@@@@@@@ | @@@ @@@ @@@@ @@@@@@@@@ @@@@@ @@@@@ @@ @@@@@@@ @@@ @@@@@@ @@ #@@@@@@@@@@ 0 +----------------------------------------------------------------------->Gi 0 15.20 After: MB 63.25^ # | # | @@@@@@@@@#@ | @@@@@@ @@#@ | @@@@@@ @@#@ | @@@@@@ @@#@ | @@@@@@ @@#@ | @@@@@@@@@@@@ @@#@ | @@@@@@@@@ @@ @@@@@@ @@#@ | @@@@@@@@ @@@ @@@ @@ @@@@@@ @@#@ | @@@@@ @ @@@ @@@ @@ @@@@@@ @@#@ | @@@@@ @ @@@ @@@ @@ @@@@@@ @@#@ | @@@@@@ @ @@@ @@@ @@ @@@@@@ @@#@ | @@@@@@ @ @@@ @@@ @@ @@@@@@ @@#@ | @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ @ @@@ @@@ @@ @@@@@@ @@#@ | @@@@@@@@@@@@@ @@@@@@@@ @@@@@@@@@@@@@@@ @ @@@ @@@ @@ @@@@@@ @@#@ | @@@@@@@@@@@@@@@ @@@@@@@@ @@@@@@@@@@@@@@@ @ @@@ @@@ @@ @@@@@@ @@#@ | @@@@@@@@@@@@@@@@@ @@@@@@@@ @@@@@@@@@@@@@@@ @ @@@ @@@ @@ @@@@@@ @@#@ | @@@@@@@@@@@@@@@@@@@@ @@@@@@@@ @@@@@@@@@@@@@@@ @ @@@ @@@ @@ @@@@@@ @@#@ | @@@@@@@@@@@@@@@@@@@@ @@@@@@@@ @@@@@@@@@@@@@@@ @ @@@ @@@ @@ @@@@@@ @@#@ 0 +----------------------------------------------------------------------->Gi 0 15.20 The bottom graph is virtually identical to the memory profile of the old reader, just with the exception that it's interning a bit more data than before, because we're reading more comprehensively. That's (potentially) the subject of future changes. DEV-12038
2022-04-06 11:50:07 -04:00
self.readbuf.clear();
let ctx = self.ctx;
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
let prev_pos = self.reader.buffer_position();
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
match self.reader.read_event(&mut self.readbuf) {
// TODO: To provide better spans and error messages,
// we need to map specific types of errors.
// But we don't encounter much of anything here with how we make
// use of quick-xml.
Err(inner) => Some(Err({
let span = ctx.span_or_zz(prev_pos, 0);
SpanlessError::from(inner).with_span(span)
})),
Ok(ev) => match ev {
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
// This is the only time we'll consider the iterator to be
// done.
QuickXmlEvent::Eof => None,
QuickXmlEvent::Empty(ele) => Some(
Self::parse_element_open(
&self.escaper,
&mut self.tokbuf,
ele,
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
prev_pos,
ctx,
true,
)
.and_then(|open| {
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
let new_pos = self.reader.buffer_position();
// `<tag ... />`
// ||
let span = ctx.span_or_zz(new_pos - 2, 2);
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
// Tag is self-closing, but this does not yet
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
// handle whitespace before the `/`
// (as indicated in the span above).
tamer: xir: Introduce {Ele,Open,Close}Span This isn't conceptally all that significant of a change, but there was a lot of modify to get it working. I would generally separate this into a commit for the implementation and another commit for the integration, but I decided to keep things together. This serves a role similar to AttrSpan---this allows deriving a span representing the element name from a span representing the entire XIR token. This will provide more useful context for errors---including the tag delimiter(s) means that we care about the fact that an element is in that position (as opposed to some other type of node) within the context of an error. However, if we are expecting an element but take issue with the element name itself, we want to place emphasis on that instead. This also starts to consider the issue of span contexts---a blob of detached data that is `Span` is useful for error context, but it's not useful for manipulation or deriving additional information. For that, we need to encode additional context, and this is an attempt at that. I am interested in the concept of providing Spans that are guaranteed to actually make sense---that are instantiated and manipulated with APIs that ensure consistency. But such a thing buys us very little, practically speaking, over what I have now for TAMER, and so I don't expect to actually implement that for this project; I'll leave that for a personal project. TAMER's already take a lot of my personal interests and it can cause me a lot of grief sometimes (with regards to letting my aspirations cause me more work). DEV-7145
2022-06-24 13:51:49 -04:00
self.tokbuf.push_front(Token::Close(
None,
CloseSpan::empty(span),
));
Ok(open)
}),
),
QuickXmlEvent::Start(ele) => Some(Self::parse_element_open(
&self.escaper,
&mut self.tokbuf,
ele,
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
prev_pos,
ctx,
false,
)),
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
QuickXmlEvent::End(ele) => Some({
// Only whitespace is permitted following the element
// name,
// so we can simply take the delta of the buffer pos.
//
// </foo >
// [------] name + '<' + '/' + " >"
tamer: xir: Introduce {Ele,Open,Close}Span This isn't conceptally all that significant of a change, but there was a lot of modify to get it working. I would generally separate this into a commit for the implementation and another commit for the integration, but I decided to keep things together. This serves a role similar to AttrSpan---this allows deriving a span representing the element name from a span representing the entire XIR token. This will provide more useful context for errors---including the tag delimiter(s) means that we care about the fact that an element is in that position (as opposed to some other type of node) within the context of an error. However, if we are expecting an element but take issue with the element name itself, we want to place emphasis on that instead. This also starts to consider the issue of span contexts---a blob of detached data that is `Span` is useful for error context, but it's not useful for manipulation or deriving additional information. For that, we need to encode additional context, and this is an attempt at that. I am interested in the concept of providing Spans that are guaranteed to actually make sense---that are instantiated and manipulated with APIs that ensure consistency. But such a thing buys us very little, practically speaking, over what I have now for TAMER, and so I don't expect to actually implement that for this project; I'll leave that for a personal project. TAMER's already take a lot of my personal interests and it can cause me a lot of grief sometimes (with regards to letting my aspirations cause me more work). DEV-7145
2022-06-24 13:51:49 -04:00
let pos_delta = self.reader.buffer_position() - prev_pos;
let span = ctx.span_or_zz(prev_pos, pos_delta);
let name_len = ele.name().len();
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
ele.name()
.try_into()
.map_err(Error::from_with_span(span))
tamer: xir: Introduce {Ele,Open,Close}Span This isn't conceptally all that significant of a change, but there was a lot of modify to get it working. I would generally separate this into a commit for the implementation and another commit for the integration, but I decided to keep things together. This serves a role similar to AttrSpan---this allows deriving a span representing the element name from a span representing the entire XIR token. This will provide more useful context for errors---including the tag delimiter(s) means that we care about the fact that an element is in that position (as opposed to some other type of node) within the context of an error. However, if we are expecting an element but take issue with the element name itself, we want to place emphasis on that instead. This also starts to consider the issue of span contexts---a blob of detached data that is `Span` is useful for error context, but it's not useful for manipulation or deriving additional information. For that, we need to encode additional context, and this is an attempt at that. I am interested in the concept of providing Spans that are guaranteed to actually make sense---that are instantiated and manipulated with APIs that ensure consistency. But such a thing buys us very little, practically speaking, over what I have now for TAMER, and so I don't expect to actually implement that for this project; I'll leave that for a personal project. TAMER's already take a lot of my personal interests and it can cause me a lot of grief sometimes (with regards to letting my aspirations cause me more work). DEV-7145
2022-06-24 13:51:49 -04:00
.and_then(|qname| {
Ok(Token::Close(
Some(qname),
CloseSpan(
span,
name_len.try_into().unwrap_or(0),
),
))
})
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
}),
// quick_xml emits a useless text event if the first byte is
// a '<'.
QuickXmlEvent::Text(bytes) if bytes.escaped().is_empty() => {
self.refill_buf()
}
// quick_xml _escapes_ the unescaped CData before handing it
// off to us,
// which is a complete waste since we'd just have to
// unescape it again.
QuickXmlEvent::CData(bytes) => todo!("CData: {:?}", bytes),
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
QuickXmlEvent::Text(bytes) => Some({
// <text>foo bar</text>
// [-----]
let span = ctx.span_or_zz(prev_pos, bytes.len());
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
bytes
.intern_utf8()
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
.map_err(Into::into)
.and_then(|sym| self.escaper.unescape(sym))
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
.map_err(Error::from_with_span(span))
.and_then(|unesc| Ok(Token::Text(unesc, span)))
}),
// Comments are _not_ returned escaped.
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
QuickXmlEvent::Comment(bytes) => Some({
// <!-- foo -->
// [----------] " foo " + "<!--" + "-->"
let span = ctx.span_or_zz(prev_pos, bytes.len() + 7);
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
bytes
.intern_utf8()
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
.map_err(Error::from_with_span(span))
.and_then(|comment| Ok(Token::Comment(comment, span)))
}),
// TODO: This must appear in the prologue.
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
QuickXmlEvent::Decl(decl) => {
match Self::validate_decl(&decl, prev_pos, ctx) {
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
Err(x) => Some(Err(x)),
Ok(()) => self.refill_buf(),
}
}
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
// We do not support processor instructions or doctypes.
// TODO: Convert this into an error/warning?
// Previously `xml-stylesheet` was present in some older
// source files and may linger for a bit after cleanup.
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
QuickXmlEvent::PI(..) | QuickXmlEvent::DocType(..) => {
self.refill_buf()
}
},
}
}
/// Validate an that an XML declaration contains expected values.
///
/// A declaration looks like `<?xml version="1.0" encoding="utf-8"?>`,
/// where `@encoding` is optional but `@version` is not.
/// It may also contain `@standalone`,
/// but we do not check for that.
///
/// We expect version 1.0 and UTF-8 encoding.
/// Failing when these expectations are voilated helps to ensure that
/// people unfamiliar with the system do not have expectations that
/// are going to be unmet,
/// which may result in subtle (or even serious) problems.
fn validate_decl(decl: &BytesDecl, pos: usize, ctx: Context) -> Result<()> {
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
// Starts after `<?`, which we want to include.
let decl_ptr = decl.as_ptr() as usize - 2 + pos;
// Fallback span that covers the entire declaration.
let decl_span = ctx.span_or_zz(pos, decl.len() + 4);
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
let ver =
&decl.version().map_err(Error::from_with_span(decl_span))?[..];
// NB: `quick-xml` docs state that `version` returns the quotes,
// but it does not.
if ver != b"1.0" {
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
// <?xml version="X.Y"?>
// [-]
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
let ver_pos = (ver.as_ptr() as usize) - decl_ptr;
let span = ctx.span_or_zz(ver_pos, ver.len());
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
Err(Error::UnsupportedXmlVersion(
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
ver.intern_utf8().map_err(Error::from_with_span(span))?,
span,
))?
}
if let Some(enc) = decl.encoding() {
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
match &enc.map_err(Error::from_with_span(decl_span))?[..] {
b"utf-8" | b"UTF-8" => (),
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
invalid => {
let enc_pos = (invalid.as_ptr() as usize) - decl_ptr;
let span = ctx.span_or_zz(enc_pos, invalid.len());
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
Err(Error::UnsupportedEncoding(
invalid
.intern_utf8()
.map_err(Error::from_with_span(span))?,
span,
))?
}
}
}
Ok(())
}
/// Parse opening element and its attributes into a XIR [`Token`]
/// stream.
///
/// The opening element is returned rather than being added to the token
/// buffer,
/// since the intent is to provide that token immediately.
fn parse_element_open(
escaper: &'s S,
tokbuf: &mut VecDeque<Token>,
ele: BytesStart,
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
pos: usize,
ctx: Context,
empty_tag: bool,
) -> Result<Token> {
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
// Starts after the opening tag `<`, so adjust.
let addr = ele.as_ptr() as usize - 1;
let len = ele.name().len();
match ele.name().last() {
None => {
// TODO: QName should be self-validating. Move this.
return Err(Error::InvalidQName(
WS_EMPTY,
// <>
// | where QName should be
ctx.span_or_zz(pos + 1, 0),
));
}
// Quick-and-dirty guess as to whether they may have missed the
// element name and included an attribute instead,
// which quick-xml does not check for.
Some(b'"' | b'\'') => {
return Err({
// <foo="bar" ...>
// [-------]
let span = ctx.span_or_zz(pos + 1, len);
Error::InvalidQName(
ele.name()
.intern_utf8()
.map_err(Error::from_with_span(span))?,
span,
)
});
}
_ => (),
};
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
// `ele` contains every byte up to the [self-]closing tag.
ele.name()
.try_into()
.map_err(Error::from_with_span(ctx.span_or_zz(pos + 1, len)))
.and_then(|qname| {
// The whitespace check is to handle input like this:
// <foo />
// ^ whitespace making `attributes_raw().len` > 0
let has_attrs = ele
.attributes_raw()
.iter()
tamer: Xirf::Text refinement This teaches XIRF to optionally refine Text into RefinedText, which determines whether the given SymbolId represents entirely whitespace. This is something I've been putting off for some time, but now that I'm parsing source language for NIR, it is necessary, in that we can only permit whitespace Text nodes in certain contexts. The idea is to capture the most common whitespace as preinterned symbols. Note that this heuristic ought to be determined from scanning a codebase, which I haven't done yet; this is just an initial list. The fallback is to look up the string associated with the SymbolId and perform a linear scan, aborting on the first non-whitespace character. This combination of checks should be sufficiently performant for now considering that this is only being run on source files, which really are not all that large. (They become large when template-expanded.) I'll optimize further if I notice it show up during profiling. This also frees XIR itself from being concerned by Whitespace. Initially I had used quick-xml's whitespace trimming, but it messed up my span calculations, and those were a pain in the ass to implement to begin with, since I had to resort to pointer arithmetic. I'd rather avoid tweaking it. tameld will not check for whitespace, since it's not important---xmlo files, if malformed, are the fault of the compiler; we can ignore text nodes except in the context of code fragments, where they are never whitespace (unless that's also a compiler bug). Onward and yonward. DEV-7145
2022-07-27 15:49:38 -04:00
.find(|b| !is_xml_whitespace_u8(**b))
.is_some();
// The tail is anything following the last byte of the QName
// in a non-empty tag with no attributes.
// For example:
// <foo > <foo> <foo bar="baz">
// ~~~~ tail ~ tail (no tail)
let tail = if has_attrs {
let found = Self::parse_attrs(
escaper,
tokbuf,
ele.attributes(),
addr - pos, // offset relative to _beginning_ of buf
pos,
ctx,
)?;
// Given this input, quick-xml ignores the bytes entirely:
// <foo bar>
// [--] missing `="value"`
if !found {
return Err(Error::AttrValueExpected(
None,
ctx.span_or_zz(pos + ele.len() + 1, 0),
));
}
// No tail because of attributes.
0
} else {
match empty_tag {
// Empty tag cannot have a tail.
true => 0,
// The "attributes" buffer represents whitespace,
// so the tail is the number of bytes of
// whitespace plus the closing '>' tag delimiter.
Revert "tamer: xir: Initial re-introduction of AttrEnd" This reverts commit b973d36862a4a2aaf53fb0b25fba01b57e5a7463. Alright, I'm getting sick of fighting with myself on this. But rather than just removing the last commit, I'm going to keep it around, so that my thoughts are clearly documented for my future quarrels with myself. Firstly: this added more overhead than I wanted it to. While it wasn't significant, it did add 100--150ms to one of our largest systems, up from ~2.8s, which seems a bit much for a token that's really just meant to make life easier for the parser. Further, it seems that all I've managed to do is push my original problem to a different layer---this started as a means to resolve having to emit both an object and an error simultaneously in the case where aggregate attribute parsing has completed, but we encounter an error on the next token (e.g. an unexpected element). But XIRF, if it's missing AttrEnd, should throw an error, but should also recover. Recovery is easy---just assume that it was present---_but then we don't emit a XIRF `AttrEnd` token_, which is necessary for downstream systems. So we'd need to either: (a) emit both a token and an error; or (b) panic. But if we're doing (a), then the need for `AttrEnd` goes away, because it solves the original problem (though the other concerns of the previous commit still stand). (b) is not ideal at all, even though the missing token does represent an internal system error; it's not something the user can correct. But, given that it's something that the user cannot correct, doesn't that imply that it's an awkward thing to include in the token stream? So back to `AttrEnd` being an awkward PITA to have. So, given (a), I'll just do that: errors will become more of a "hey, this error just occurred, but I'm trying to recover---here's an object that you should use if you choose to continue parsing, but it may or may not be what you're looking for; proceed with caution". That flips the original script: I imagined having external systems feed recovery tokens, but this encapsulates recovery within the parser, which really is more appropriate, though less flexible than having an omniscient external recovery system; such a monolith was always an awkward concept and would be difficult to implement cleanly. This can also potentially be implemented as a generalization of the Dead state change that allowed an object to be emitted alongside the lookahead/error. Anyway, back to where I was...I'm sure I'll look back on this in the future shaking my head, reflecting on how naive I was. DEV-7145
2022-06-29 11:02:18 -04:00
false => ele.attributes_raw().len() + 1,
}
};
// <tag ... /> <tag/>
// [--] name + '<' [--] name + '<'
//
// <tag >...</tag> <tag ...>...</tag>
// [-----] name + '<' + " >" [--] name + '<'
// ~~~ tail
let span = ctx.span_or_zz(pos, len + 1 + tail);
// The first token will be immediately returned
// via the Iterator.
tamer: xir: Introduce {Ele,Open,Close}Span This isn't conceptally all that significant of a change, but there was a lot of modify to get it working. I would generally separate this into a commit for the implementation and another commit for the integration, but I decided to keep things together. This serves a role similar to AttrSpan---this allows deriving a span representing the element name from a span representing the entire XIR token. This will provide more useful context for errors---including the tag delimiter(s) means that we care about the fact that an element is in that position (as opposed to some other type of node) within the context of an error. However, if we are expecting an element but take issue with the element name itself, we want to place emphasis on that instead. This also starts to consider the issue of span contexts---a blob of detached data that is `Span` is useful for error context, but it's not useful for manipulation or deriving additional information. For that, we need to encode additional context, and this is an attempt at that. I am interested in the concept of providing Spans that are guaranteed to actually make sense---that are instantiated and manipulated with APIs that ensure consistency. But such a thing buys us very little, practically speaking, over what I have now for TAMER, and so I don't expect to actually implement that for this project; I'll leave that for a personal project. TAMER's already take a lot of my personal interests and it can cause me a lot of grief sometimes (with regards to letting my aspirations cause me more work). DEV-7145
2022-06-24 13:51:49 -04:00
Ok(Token::Open(
qname,
OpenSpan(span, len.try_into().unwrap_or(0)),
))
})
}
/// Parse attributes into a XIR [`Token`] stream.
///
/// The order of attributes will be maintained.
///
/// This does not yet handle whitespace between attributes,
/// or around `=`.
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
///
/// Note About Pointer Arithmetic
/// =============================
/// `ele_ptr` is expected to be a pointer to the buffer containing the
/// bytes read from the source file.
/// Attributes reference this buffer,
/// so we can use pointer arithmetic to determine the offset within
/// the buffer relative to the node.
/// This works because the underlying buffer is a `Vec`,
/// which is contiguous in memory.
///
/// However, since this is a `Vec`,
/// it is important that the address be retrieved _after_ quick-xml
/// read events,
/// otherwise the buffer may be expanded and will be reallocated.
fn parse_attrs<'a>(
escaper: &'s S,
tokbuf: &mut VecDeque<Token>,
mut attrs: Attributes<'a>,
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
ele_ptr: usize,
ele_pos: usize,
ctx: Context,
) -> Result<bool> {
let mut found = false;
// Disable checks to allow duplicate attributes;
// XIR does not enforce this,
// because it needs to accommodate semantically invalid XML for
// later analysis.
for result in attrs.with_checks(false) {
found = true;
let attr = result.map_err(|e| match e {
QuickXmlError::NoEqAfterName(pos) => {
// TODO: quick-xml doesn't give us the name,
// but we should discover it.
Error::AttrValueExpected(
None,
ctx.span_or_zz(ele_pos + pos, 0),
)
}
QuickXmlError::UnquotedValue(pos) => {
// TODO: name and span length
Error::AttrValueUnquoted(
None,
ctx.span_or_zz(ele_pos + pos, 0),
)
}
// fallback
e => Error::from_with_span(ctx.span_or_zz(ele_pos, 0))(e),
})?;
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
let keyoffset = attr.key.as_ptr() as usize;
let name_offset = keyoffset - ele_ptr;
// Accommodates zero-length values (e.g. `key=""`) with a
// zero-length span at the location the value _would_ be.
let valoffset = match attr.value {
Cow::Borrowed(b) => b.as_ptr() as usize,
// This should never happen since we have a reference to the
// underlying buffer.
Cow::Owned(_) => unreachable!(
"internal error: unexpected owned attribute value"
),
};
let value_offset = valoffset - ele_ptr;
let span_name = ctx.span_or_zz(name_offset, attr.key.len());
let span_value = ctx.span_or_zz(value_offset, attr.value.len());
tamer: xir::XirString: WIP implementation (likely going away) I'm not fond of this implementation, which is why it's not fully completed. I wanted to commit this for future reference, and take the opportunity to explain why I don't like it. First: this task started as an idea to implement a third variant to AttrValue and friends that indicates that a value is fixed, in the sense of a fixed-point function: escaped or unescaped, its value is the same. This would allow us to skip wasteful escape/unescape operations. In doing so, it became obvious that there's no need to leak this information through the API, and indeed, no part of the system should care. When we read XML, it should be unescaped, and when we write, it should be escaped. The reason that this didn't quite happen to begin with was an optimization: I'll be creating an echo writer in place of the current filesystem-based copy in tamec shortly, and this would allow streaming XIR directly from the reader to the writer without any unescaping or re-escaping. When we unescape, we know the value that it came from, so we could simply store both symbols---they're 32-bit, so it results in a nicely compressed 64-bit value, so it's essentially cost-free, as long as we accept the expense of internment. This is `XirString`. Then, when we want to escape or unescape, we first check to see whether a symbol already exists and, if so, use it. While this works well for echoing streams, it won't work all that well in practice: the unescaped SymbolId will be taken and the XirString discarded, since nothing after XIR should be coupled with it. Then, when we later construct a XIR stream for writting, XirString will no longer be available and our previously known escape is lost, so the writer will have to re-escape. Further, if we look at XirString's generic for the XirStringEscaper---it uses phantom, which hints that maybe it's not in the best place. Indeed, I've already acknowledged that only a reader unescapes and only a writer escapes, and that the rest of the system works with normal (unescaped) values, so only readers and writers should be part of this process. I also already acknowledged that XirString would be lost and only the unescaped SymbolId would be used. So what's the point of XirString, then, if it won't be a useful optimization beyond the temporary echo writer? Instead, we can take the XirStringWriter and implement two caches on that: mapping SymbolId from escaped->unescaped and vice-versa. These can be simple vectors, since SymbolId is a 32-bit value we will not have much wasted space for symbols that never get read or written. We could even optimize for preinterned symbols using markers, though I'll probably not do so, and I'll explain why later. If we do _that_, we get even _better_ optimizations through caching that _will_ apply in the general case (so, not just for echo), and we're able to ditch XirString entirely and simply use a SymbolId. This makes for a much more friendly API that isn't leaking implementation details, though it _does_ put an onus on the caller to pass the encoder to both the reader and the writer, _if_ it wants to take advantage of a cache. But that burden is not significant (and is, again, optional if we don't want it). So, that'll be the next step.
2021-11-10 09:42:18 -05:00
// The name must be parsed as a QName.
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
let name = attr
.key
.try_into()
.map_err(Error::from_with_span(span_name))?;
tamer: xir::XirString: WIP implementation (likely going away) I'm not fond of this implementation, which is why it's not fully completed. I wanted to commit this for future reference, and take the opportunity to explain why I don't like it. First: this task started as an idea to implement a third variant to AttrValue and friends that indicates that a value is fixed, in the sense of a fixed-point function: escaped or unescaped, its value is the same. This would allow us to skip wasteful escape/unescape operations. In doing so, it became obvious that there's no need to leak this information through the API, and indeed, no part of the system should care. When we read XML, it should be unescaped, and when we write, it should be escaped. The reason that this didn't quite happen to begin with was an optimization: I'll be creating an echo writer in place of the current filesystem-based copy in tamec shortly, and this would allow streaming XIR directly from the reader to the writer without any unescaping or re-escaping. When we unescape, we know the value that it came from, so we could simply store both symbols---they're 32-bit, so it results in a nicely compressed 64-bit value, so it's essentially cost-free, as long as we accept the expense of internment. This is `XirString`. Then, when we want to escape or unescape, we first check to see whether a symbol already exists and, if so, use it. While this works well for echoing streams, it won't work all that well in practice: the unescaped SymbolId will be taken and the XirString discarded, since nothing after XIR should be coupled with it. Then, when we later construct a XIR stream for writting, XirString will no longer be available and our previously known escape is lost, so the writer will have to re-escape. Further, if we look at XirString's generic for the XirStringEscaper---it uses phantom, which hints that maybe it's not in the best place. Indeed, I've already acknowledged that only a reader unescapes and only a writer escapes, and that the rest of the system works with normal (unescaped) values, so only readers and writers should be part of this process. I also already acknowledged that XirString would be lost and only the unescaped SymbolId would be used. So what's the point of XirString, then, if it won't be a useful optimization beyond the temporary echo writer? Instead, we can take the XirStringWriter and implement two caches on that: mapping SymbolId from escaped->unescaped and vice-versa. These can be simple vectors, since SymbolId is a 32-bit value we will not have much wasted space for symbols that never get read or written. We could even optimize for preinterned symbols using markers, though I'll probably not do so, and I'll explain why later. If we do _that_, we get even _better_ optimizations through caching that _will_ apply in the general case (so, not just for echo), and we're able to ditch XirString entirely and simply use a SymbolId. This makes for a much more friendly API that isn't leaking implementation details, though it _does_ put an onus on the caller to pass the encoder to both the reader and the writer, _if_ it wants to take advantage of a cache. But that burden is not significant (and is, again, optional if we don't want it). So, that'll be the next step.
2021-11-10 09:42:18 -05:00
// The attribute value,
// having just been read from XML,
// must have been escaped to be parsed properly.
// If it parsed but it's not technically escaped according to
// the spec,
// that's okay as long as we can read it again,
// but we probably should still throw an error if we
// encounter such a situation.
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
let value = escaper
.unescape(
attr.value
.as_ref()
.intern_utf8()
.map_err(Error::from_with_span(span_value))?,
)
.map_err(Error::from_with_span(span_value))?
.into();
tamer: xir::reader: Initial introduction of spans This is a large change, and was a bit of a tedious one, given the comprehensive tests. This introduces proper offsets and lengths for spans, with the exception of some quick-xml errors that still need proper mapping. Further, this still uses `UNKNOWN_CONTEXT`, which will be resolved shortly. This also introduces `SpanlessError`, which `Error` explicitly _does not_ implement `From<SpanlessError>` for---this forces the caller to provide a span before the error is compatable with the return value, ensuring that spans will actually be available rather than forgotten for errors. This is important, given that errors are generally less tested than the happy path, and errors are when users need us the most (so, need span information). Further, I had to use pointer arithmetic in order to calculate many of the spans, because quick-xml does not provide enough information. There's no safety considerations here, and the comprehensive unit test will ensure correct behavior if the implementation changes in the future. I would like to introduce typed spans at some point---I made some opinionated choices when it comes to what the spans ought to represent. Specifically, whether to include the `<` or `>` with the open span (depends), whether to include quotes with attribute values (no), and some other details highlighted in the test cases. If we provide typed spans, then we could, knowing the type of span, calculate other spans on request, e.g. to include or omit quotes for attributes. Different such spans may be useful in different situations when presenting information to the user. This also highlights gaps in the tokens emitted by XIR, such as whitespace between attributes, the `=` between name and value, and so on. These are important when it comes to code formatting, so that we can reliably reconstruct the XML tree, but it's not important right now. I anticipate future changes would allow the XIR reader to be configured (perhaps via generics, like a strategy-type pattern) to optionally omit these tokens if desired. Anyway, more to come. DEV-10934
2022-04-08 11:03:46 -04:00
tokbuf.push_front(Token::AttrName(name, span_name));
tokbuf.push_front(Token::AttrValue(value, span_value));
}
Ok(found)
}
}
tamer: Xirf::Text refinement This teaches XIRF to optionally refine Text into RefinedText, which determines whether the given SymbolId represents entirely whitespace. This is something I've been putting off for some time, but now that I'm parsing source language for NIR, it is necessary, in that we can only permit whitespace Text nodes in certain contexts. The idea is to capture the most common whitespace as preinterned symbols. Note that this heuristic ought to be determined from scanning a codebase, which I haven't done yet; this is just an initial list. The fallback is to look up the string associated with the SymbolId and perform a linear scan, aborting on the first non-whitespace character. This combination of checks should be sufficiently performant for now considering that this is only being run on source files, which really are not all that large. (They become large when template-expanded.) I'll optimize further if I notice it show up during profiling. This also frees XIR itself from being concerned by Whitespace. Initially I had used quick-xml's whitespace trimming, but it messed up my span calculations, and those were a pain in the ass to implement to begin with, since I had to resort to pointer arithmetic. I'd rather avoid tweaking it. tameld will not check for whitespace, since it's not important---xmlo files, if malformed, are the fault of the compiler; we can ignore text nodes except in the context of code fragments, where they are never whitespace (unless that's also a compiler bug). Onward and yonward. DEV-7145
2022-07-27 15:49:38 -04:00
/// Whether the byte represents XML whitespace.
///
/// This is quick-xml's whitespace predicate,
/// and corresponds to the
/// [nonterminal `S` in the XML specification][xmlspec-s].
///
/// [xmlspec-s]: https://www.w3.org/TR/xml/#NT-S
pub fn is_xml_whitespace_u8(b: u8) -> bool {
match b {
b' ' | b'\r' | b'\n' | b'\t' => true,
_ => false,
}
}
/// Whether the character represents XML whitespace.
///
/// See [`is_xml_whitespace_u8`].
pub fn is_xml_whitespace_char(c: char) -> bool {
match c {
' ' | '\r' | '\n' | '\t' => true,
_ => false,
}
}
impl<'s, B, S> Iterator for XmlXirReader<'s, B, S>
where
B: BufRead,
S: Escaper,
{
type Item = ParsedResult<ParsedObject<Token, Error>>;
/// Produce the next XIR [`Token`] from the input.
///
/// For more information on how this reader operates,
/// see [`XmlXirReader`].
fn next(&mut self) -> Option<Self::Item> {
self.tokbuf
.pop_back()
.map(|tok| Ok(Parsed::Object(tok)))
.or_else(|| {
self.refill_buf().map(|result| {
result.map(Parsed::Object).map_err(ParseError::StateError)
})
})
}
}
#[cfg(test)]
mod test;