organic/src/compare/util.rs

use std::str::FromStr;

use super::diff::DiffEntry;
use super::diff::DiffStatus;
use super::elisp_fact::GetElispFact;
use super::sexp::Token;
use crate::compare::diff::compare_ast_node;
use crate::compare::sexp::unquote;
use crate::types::AstNode;
use crate::types::GetStandardProperties;
use crate::types::StandardProperties;

/// Check if the child string slice is a slice of the parent string slice.
fn is_slice_of(parent: &str, child: &str) -> bool {
    let parent_start = parent.as_ptr() as usize;
    let parent_end = parent_start + parent.len();
    let child_start = child.as_ptr() as usize;
    let child_end = child_start + child.len();
    child_start >= parent_start && child_end <= parent_end
}

/// Get the byte offset into source that the rust object exists at.
///
/// These offsets are zero-based unlike the elisp ones.
fn get_rust_byte_offsets<'b, 's, S: StandardProperties<'s> + ?Sized>(
    original_document: &'s str,
    rust_ast_node: &'b S,
) -> (usize, usize) {
    let rust_object_source = rust_ast_node.get_source();
    debug_assert!(is_slice_of(original_document, rust_object_source));
    let offset = rust_object_source.as_ptr() as usize - original_document.as_ptr() as usize;
    let end = offset + rust_object_source.len();
    (offset, end)
}

pub(crate) fn compare_standard_properties<
    'b,
    's,
    S: GetStandardProperties<'s> + GetElispFact<'s> + ?Sized,
>(
    original_document: &'s str,
    emacs: &'b Token<'s>,
    rust: &'b S,
) -> Result<(), Box<dyn std::error::Error>> {
    assert_name(emacs, rust.get_elisp_fact().get_elisp_name())?;
    assert_bounds(original_document, emacs, rust.get_standard_properties())?;
    Ok(())
}

pub(crate) fn assert_name<'b, 's, S: AsRef<str>>(
    emacs: &'b Token<'s>,
    name: S,
) -> Result<(), Box<dyn std::error::Error>> {
    let name = name.as_ref();
    let children = emacs.as_list()?;
    let first_child = children
        .first()
        .ok_or("Should have at least one child.")?
        .as_atom()?;
    if first_child != name {
        Err(format!(
            "AST node name mismatch. Expected a (rust) {expected} cell, but found a (emacs) {found} cell.",
            expected = name,
            found = first_child
        ))?;
    }
    Ok(())
}

/// Assert that the character ranges defined by upstream org-mode's :standard-properties match the slices in Organic's StandardProperties.
///
/// This does **not** handle plain text because plain text is a special case.
pub(crate) fn assert_bounds<'b, 's, S: StandardProperties<'s> + ?Sized>(
    original_document: &'s str,
    emacs: &'b Token<'s>,
    rust: &'b S,
) -> Result<(), Box<dyn std::error::Error>> {
    let standard_properties = get_emacs_standard_properties(emacs)?; // 1-based
    let (begin, end) = (
        standard_properties
            .begin
            .ok_or("Token should have a begin.")?,
        standard_properties.end.ok_or("Token should have an end.")?,
    );
    let (rust_begin, rust_end) = get_rust_byte_offsets(original_document, rust); // 0-based
    let rust_begin_char_offset = (&original_document[..rust_begin]).chars().count() + 1; // 1-based
    let rust_end_char_offset =
        rust_begin_char_offset + (&original_document[rust_begin..rust_end]).chars().count(); // 1-based
    if rust_begin_char_offset != begin || rust_end_char_offset != end {
        Err(format!("Rust bounds (in chars) ({rust_begin}, {rust_end}) do not match emacs bounds ({emacs_begin}, {emacs_end})", rust_begin = rust_begin_char_offset, rust_end = rust_end_char_offset, emacs_begin=begin, emacs_end=end))?;
    }

    Ok(())
}

struct EmacsStandardProperties {
    begin: Option<usize>,
    #[allow(dead_code)]
    post_affiliated: Option<usize>,
    #[allow(dead_code)]
    contents_begin: Option<usize>,
    #[allow(dead_code)]
    contents_end: Option<usize>,
    end: Option<usize>,
    #[allow(dead_code)]
    post_blank: Option<usize>,
}

fn get_emacs_standard_properties<'b, 's>(
    emacs: &'b Token<'s>,
) -> Result<EmacsStandardProperties, Box<dyn std::error::Error>> {
    let children = emacs.as_list()?;
    let attributes_child = children
        .iter()
        .nth(1)
        .ok_or("Should have an attributes child.")?;
    let attributes_map = attributes_child.as_map()?;
    let standard_properties = attributes_map.get(":standard-properties");
    Ok(if standard_properties.is_some() {
        let mut std_props = standard_properties
            .expect("if statement proves its Some")
            .as_vector()?
            .into_iter();
        let begin = maybe_token_to_usize(std_props.next())?;
        let post_affiliated = maybe_token_to_usize(std_props.next())?;
        let contents_begin = maybe_token_to_usize(std_props.next())?;
        let contents_end = maybe_token_to_usize(std_props.next())?;
        let end = maybe_token_to_usize(std_props.next())?;
        let post_blank = maybe_token_to_usize(std_props.next())?;
        EmacsStandardProperties {
            begin,
            post_affiliated,
            contents_begin,
            contents_end,
            end,
            post_blank,
        }
    } else {
        let begin = maybe_token_to_usize(attributes_map.get(":begin").map(|token| *token))?;
        let end = maybe_token_to_usize(attributes_map.get(":end").map(|token| *token))?;
        let contents_begin =
            maybe_token_to_usize(attributes_map.get(":contents-begin").map(|token| *token))?;
        let contents_end =
            maybe_token_to_usize(attributes_map.get(":contents-end").map(|token| *token))?;
        let post_blank =
            maybe_token_to_usize(attributes_map.get(":post-blank").map(|token| *token))?;
        let post_affiliated =
            maybe_token_to_usize(attributes_map.get(":post-affiliated").map(|token| *token))?;
        EmacsStandardProperties {
            begin,
            post_affiliated,
            contents_begin,
            contents_end,
            end,
            post_blank,
        }
    })
}

fn maybe_token_to_usize(
    token: Option<&Token<'_>>,
) -> Result<Option<usize>, Box<dyn std::error::Error>> {
    Ok(token
        .map(|token| token.as_atom())
        .map_or(Ok(None), |r| r.map(Some))?
        .map(|val| {
            if val == "nil" {
                None
            } else {
                Some(val.parse::<usize>())
            }
        })
        .flatten() // Outer option is whether or not the param exists, inner option is whether or not it is nil
        .map_or(Ok(None), |r| r.map(Some))?)
}

/// Get a named property from the emacs token.
///
/// Returns Ok(None) if value is nil or absent.
pub(crate) fn get_property<'b, 's, 'x>(
    emacs: &'b Token<'s>,
    key: &'x str,
) -> Result<Option<&'b Token<'s>>, Box<dyn std::error::Error>> {
    let children = emacs.as_list()?;
    let attributes_child = children
        .iter()
        .nth(1)
        .ok_or("Should have an attributes child.")?;
    let attributes_map = attributes_child.as_map()?;
    let prop = attributes_map.get(key).map(|token| *token);
    match prop.map(|token| token.as_atom()) {
        Some(Ok("nil")) => return Ok(None),
        _ => {}
    };
    Ok(prop)
}

/// Get a named property containing an unquoted atom from the emacs token.
///
/// Returns None if key is not found.
pub(crate) fn get_property_unquoted_atom<'b, 's, 'x>(
    emacs: &'b Token<'s>,
    key: &'x str,
) -> Result<Option<&'s str>, Box<dyn std::error::Error>> {
    Ok(get_property(emacs, key)?
        .map(Token::as_atom)
        .map_or(Ok(None), |r| r.map(Some))?)
}

/// Get a named property containing an quoted string from the emacs token.
///
/// Returns None if key is not found.
pub(crate) fn get_property_quoted_string<'b, 's, 'x>(
    emacs: &'b Token<'s>,
    key: &'x str,
) -> Result<Option<String>, Box<dyn std::error::Error>> {
    Ok(get_property(emacs, key)?
        .map(Token::as_atom)
        .map_or(Ok(None), |r| r.map(Some))?
        .map(unquote)
        .map_or(Ok(None), |r| r.map(Some))?)
}

/// Get a named property containing a boolean value.
///
/// This uses the elisp convention of nil == false, non-nil == true.
///
/// Returns false if key is not found.
pub(crate) fn get_property_boolean<'b, 's, 'x>(
    emacs: &'b Token<'s>,
    key: &'x str,
) -> Result<bool, Box<dyn std::error::Error>> {
    Ok(get_property(emacs, key)?
        .map(Token::as_atom)
        .map_or(Ok(None), |r| r.map(Some))?
        .unwrap_or("nil")
        != "nil")
}

/// Get a named property containing an unquoted numeric value.
///
/// Returns None if key is not found.
pub(crate) fn get_property_numeric<'b, 's, 'x, N: FromStr>(
    emacs: &'b Token<'s>,
    key: &'x str,
) -> Result<Option<N>, Box<dyn std::error::Error + 's>>
where
    <N as FromStr>::Err: std::error::Error,
    <N as FromStr>::Err: 's,
{
    let unparsed_string = get_property(emacs, key)?
        .map(Token::as_atom)
        .map_or(Ok(None), |r| r.map(Some))?;
    let parsed_number = unparsed_string
        .map(|val| val.parse::<N>())
        .map_or(Ok(None), |r| r.map(Some))?;
    Ok(parsed_number)
}

pub(crate) fn compare_children<'b, 's, 'x, RC>(
    source: &'s str,
    emacs: &'b Token<'s>,
    rust_children: &'x Vec<RC>,
    child_status: &mut Vec<DiffEntry<'b, 's>>,
    this_status: &mut DiffStatus,
    message: &mut Option<String>,
) -> Result<(), Box<dyn std::error::Error>>
where
    AstNode<'b, 's>: From<&'x RC>,
{
    let emacs_children = emacs.as_list()?;
    let emacs_children_length = emacs_children.len() - 2;
    if emacs_children_length != rust_children.len() {
        *this_status = DiffStatus::Bad;
        *message = Some(format!(
            "Child length mismatch (emacs != rust) {:?} != {:?}",
            emacs_children_length,
            rust_children.len()
        ));
    }
    for (emacs_child, rust_child) in emacs_children.iter().skip(2).zip(rust_children.iter()) {
        child_status.push(compare_ast_node(source, emacs_child, rust_child.into())?);
    }
    Ok(())
}

pub(crate) fn compare_children_iter<
    'b,
    's,
    'x,
    RC: 'x,
    RI: Iterator<Item = &'x RC> + ExactSizeIterator,
>(
    source: &'s str,
    emacs: &'b Token<'s>,
    rust_children: RI,
    child_status: &mut Vec<DiffEntry<'b, 's>>,
    this_status: &mut DiffStatus,
    message: &mut Option<String>,
) -> Result<(), Box<dyn std::error::Error>>
where
    AstNode<'b, 's>: From<&'x RC>,
{
    let emacs_children = emacs.as_list()?;
    let emacs_children_length = emacs_children.len() - 2;
    if emacs_children_length != rust_children.len() {
        *this_status = DiffStatus::Bad;
        *message = Some(format!(
            "Child length mismatch (emacs != rust) {:?} != {:?}",
            emacs_children_length,
            rust_children.len()
        ));
    }
    for (emacs_child, rust_child) in emacs_children.iter().skip(2).zip(rust_children) {
        child_status.push(compare_ast_node(source, emacs_child, rust_child.into())?);
    }
    Ok(())
}

pub(crate) fn assert_no_children<'b, 's>(
    emacs: &'b Token<'s>,
    this_status: &mut DiffStatus,
    message: &mut Option<String>,
) -> Result<(), Box<dyn std::error::Error>> {
    let emacs_children_length = emacs.as_list()?.len();
    // 2, one for the name of the node and one for the properties. Children would come after that.
    if emacs_children_length != 2 {
        *this_status = DiffStatus::Bad;
        *message = Some(format!(
            "Should have no children but emacs has {:?} children.",
            emacs_children_length - 2,
        ));
    }
    Ok(())
}