@knolo/cli 6.2.0 → 6.2.1

package/templates/icp-knowledge-canister/canisters/knolo-core-rust/src/kar/json.rs

@@ -0,0 +1,486 @@
+//! Research JSON canonical form.
+//! Object keys sort by UTF-16 code units. Arrays keep their order.
+//! Solidus, U+2028, and U+2029 stay literal, matching `JSON.stringify`.
+
+use std::cmp::Ordering;
+
+#[derive(Debug, Clone, PartialEq)]
+pub enum Json {
+    Null,
+    Bool(bool),
+    Int(i64),
+    String(String),
+    Array(Vec<Json>),
+    Object(Vec<(String, Json)>),
+}
+
+impl Json {
+    pub fn object(pairs: Vec<(&str, Json)>) -> Json {
+        Json::Object(
+            pairs
+                .into_iter()
+                .map(|(key, value)| (key.to_string(), value))
+                .collect(),
+        )
+    }
+
+    pub fn get(&self, key: &str) -> Option<&Json> {
+        match self {
+            Json::Object(pairs) => pairs
+                .iter()
+                .find(|(name, _)| name == key)
+                .map(|(_, value)| value),
+            _ => None,
+        }
+    }
+
+    pub fn as_object(&self) -> Option<&[(String, Json)]> {
+        match self {
+            Json::Object(pairs) => Some(pairs),
+            _ => None,
+        }
+    }
+
+    pub fn as_str(&self) -> Option<&str> {
+        match self {
+            Json::String(value) => Some(value),
+            _ => None,
+        }
+    }
+
+    pub fn as_int(&self) -> Option<i64> {
+        match self {
+            Json::Int(value) => Some(*value),
+            _ => None,
+        }
+    }
+}
+
+pub fn parse_json(input: &str) -> Result<Json, String> {
+    let mut parser = Parser { input, index: 0 };
+    parser.skip();
+    let value = parser.value()?;
+    parser.skip();
+    if parser.index != parser.input.len() {
+        return Err("trailing JSON".into());
+    }
+    Ok(value)
+}
+
+pub fn canonicalize(value: &Json) -> String {
+    let mut out = String::new();
+    write_canonical(value, &mut out);
+    out
+}
+
+pub fn stringify(value: &Json) -> String {
+    let mut out = String::new();
+    write_stringify(value, &mut out);
+    out
+}
+
+pub fn digest_json(value: &Json) -> String {
+    let raw = crate::sha256(canonicalize(value).as_bytes());
+    let mut hex = String::with_capacity(64);
+    for byte in raw {
+        hex.push_str(&format!("{byte:02x}"));
+    }
+    format!("sha256-{hex}")
+}
+
+pub fn utf16_cmp(left: &str, right: &str) -> Ordering {
+    left.encode_utf16().cmp(right.encode_utf16())
+}
+
+struct Parser<'a> {
+    input: &'a str,
+    index: usize,
+}
+
+impl<'a> Parser<'a> {
+    fn skip(&mut self) {
+        while let Some(byte) = self.input.as_bytes().get(self.index).copied() {
+            if byte == b' ' || byte == b'\n' || byte == b'\r' || byte == b'\t' {
+                self.index += 1;
+            } else {
+                break;
+            }
+        }
+    }
+
+    fn value(&mut self) -> Result<Json, String> {
+        self.skip();
+        let byte = *self
+            .input
+            .as_bytes()
+            .get(self.index)
+            .ok_or_else(|| "truncated JSON".to_string())?;
+        match byte {
+            b'n' => self.literal("null", Json::Null),
+            b't' => self.literal("true", Json::Bool(true)),
+            b'f' => self.literal("false", Json::Bool(false)),
+            b'"' => Ok(Json::String(self.string()?)),
+            b'[' => self.array(),
+            b'{' => self.object(),
+            b'-' | b'0'..=b'9' => self.number(),
+            _ => Err("invalid JSON value".into()),
+        }
+    }
+
+    fn literal(&mut self, text: &str, value: Json) -> Result<Json, String> {
+        if self.input[self.index..].starts_with(text) {
+            self.index += text.len();
+            Ok(value)
+        } else {
+            Err(format!("invalid JSON literal {text}"))
+        }
+    }
+
+    fn array(&mut self) -> Result<Json, String> {
+        self.index += 1;
+        let mut items = Vec::new();
+        self.skip();
+        if self.eat(b']') {
+            return Ok(Json::Array(items));
+        }
+        loop {
+            items.push(self.value()?);
+            self.skip();
+            if self.eat(b']') {
+                return Ok(Json::Array(items));
+            }
+            if !self.eat(b',') {
+                return Err("invalid JSON array".into());
+            }
+        }
+    }
+
+    fn object(&mut self) -> Result<Json, String> {
+        self.index += 1;
+        let mut pairs = Vec::new();
+        self.skip();
+        if self.eat(b'}') {
+            return Ok(Json::Object(pairs));
+        }
+        loop {
+            self.skip();
+            if !self.eat(b'"') {
+                return Err("JSON object key must be a string".into());
+            }
+            let key = self.string_body()?;
+            self.skip();
+            if !self.eat(b':') {
+                return Err("JSON object missing colon".into());
+            }
+            let value = self.value()?;
+            if pairs.iter().any(|(name, _): &(String, Json)| name == &key) {
+                return Err("duplicate JSON key".into());
+            }
+            pairs.push((key, value));
+            self.skip();
+            if self.eat(b'}') {
+                return Ok(Json::Object(pairs));
+            }
+            if !self.eat(b',') {
+                return Err("invalid JSON object".into());
+            }
+        }
+    }
+
+    fn number(&mut self) -> Result<Json, String> {
+        let start = self.index;
+        if self.eat(b'-') {}
+        if self.eat(b'0') {
+            if self
+                .input
+                .as_bytes()
+                .get(self.index)
+                .is_some_and(|byte| byte.is_ascii_digit())
+            {
+                return Err("invalid JSON number".into());
+            }
+        } else if self
+            .input
+            .as_bytes()
+            .get(self.index)
+            .is_some_and(|byte| byte.is_ascii_digit())
+        {
+            while self
+                .input
+                .as_bytes()
+                .get(self.index)
+                .is_some_and(|byte| byte.is_ascii_digit())
+            {
+                self.index += 1;
+            }
+        } else {
+            return Err("invalid JSON number".into());
+        }
+        let mut fractional = false;
+        if self.eat(b'.') {
+            fractional = true;
+            if !self
+                .input
+                .as_bytes()
+                .get(self.index)
+                .is_some_and(|byte| byte.is_ascii_digit())
+            {
+                return Err("invalid JSON number".into());
+            }
+            while self
+                .input
+                .as_bytes()
+                .get(self.index)
+                .is_some_and(|byte| byte.is_ascii_digit())
+            {
+                self.index += 1;
+            }
+        }
+        if self
+            .input
+            .as_bytes()
+            .get(self.index)
+            .is_some_and(|byte| *byte == b'e' || *byte == b'E')
+        {
+            fractional = true;
+            self.index += 1;
+            if self
+                .input
+                .as_bytes()
+                .get(self.index)
+                .is_some_and(|byte| *byte == b'+' || *byte == b'-')
+            {
+                self.index += 1;
+            }
+            if !self
+                .input
+                .as_bytes()
+                .get(self.index)
+                .is_some_and(|byte| byte.is_ascii_digit())
+            {
+                return Err("invalid JSON number".into());
+            }
+            while self
+                .input
+                .as_bytes()
+                .get(self.index)
+                .is_some_and(|byte| byte.is_ascii_digit())
+            {
+                self.index += 1;
+            }
+        }
+        let token = &self.input[start..self.index];
+        if !fractional {
+            let parsed = token
+                .parse::<i64>()
+                .map_err(|_| format!("JSON integer out of range: {token}"))?;
+            return Ok(Json::Int(parsed));
+        }
+        let number: f64 = token
+            .parse()
+            .map_err(|_| format!("invalid JSON number: {token}"))?;
+        if number.fract() == 0.0 && number >= i64::MIN as f64 && number <= i64::MAX as f64 {
+            let whole = number as i64;
+            if whole as f64 == number && (whole as f64).abs() <= 9007199254740991.0 {
+                return Ok(Json::Int(whole));
+            }
+        }
+        Err("KAR JSON numbers must be safe integers".into())
+    }
+
+    fn string(&mut self) -> Result<String, String> {
+        if !self.eat(b'"') {
+            return Err("JSON string expected".into());
+        }
+        self.string_body()
+    }
+
+    fn string_body(&mut self) -> Result<String, String> {
+        let mut out = String::new();
+        loop {
+            let byte = *self
+                .input
+                .as_bytes()
+                .get(self.index)
+                .ok_or_else(|| "unterminated JSON string".to_string())?;
+            if byte == b'"' {
+                self.index += 1;
+                return Ok(out);
+            }
+            if byte == b'\\' {
+                self.index += 1;
+                let escape = *self
+                    .input
+                    .as_bytes()
+                    .get(self.index)
+                    .ok_or_else(|| "truncated JSON escape".to_string())?;
+                self.index += 1;
+                match escape {
+                    b'"' => out.push('"'),
+                    b'\\' => out.push('\\'),
+                    b'/' => out.push('/'),
+                    b'b' => out.push('\u{0008}'),
+                    b'f' => out.push('\u{000c}'),
+                    b'n' => out.push('\n'),
+                    b'r' => out.push('\r'),
+                    b't' => out.push('\t'),
+                    b'u' => out.push(self.unicode_escape()?),
+                    _ => return Err("invalid JSON string escape".into()),
+                }
+                continue;
+            }
+            if byte < 0x20 {
+                return Err("raw control character in JSON string".into());
+            }
+            let ch = self.input[self.index..]
+                .chars()
+                .next()
+                .ok_or_else(|| "invalid UTF-8 in JSON".to_string())?;
+            self.index += ch.len_utf8();
+            out.push(ch);
+        }
+    }
+
+    fn unicode_escape(&mut self) -> Result<char, String> {
+        let unit = self.hex4()?;
+        if (0xD800..=0xDBFF).contains(&unit) {
+            if self.input[self.index..].starts_with("\\u") {
+                self.index += 2;
+                let second = self.hex4()?;
+                if (0xDC00..=0xDFFF).contains(&second) {
+                    let point =
+                        0x10000 + (((unit as u32 - 0xD800) << 10) | (second as u32 - 0xDC00));
+                    return char::from_u32(point)
+                        .ok_or_else(|| "invalid JSON surrogate pair".to_string());
+                }
+            }
+            return Err("lone JSON surrogate".into());
+        }
+        if (0xDC00..=0xDFFF).contains(&unit) {
+            return Err("lone JSON surrogate".into());
+        }
+        char::from_u32(unit as u32).ok_or_else(|| "invalid JSON unicode escape".to_string())
+    }
+
+    fn hex4(&mut self) -> Result<u16, String> {
+        let end = self.index + 4;
+        let text = self
+            .input
+            .get(self.index..end)
+            .ok_or_else(|| "truncated JSON unicode escape".to_string())?;
+        let value =
+            u16::from_str_radix(text, 16).map_err(|_| "invalid JSON unicode escape".to_string())?;
+        self.index = end;
+        Ok(value)
+    }
+
+    fn eat(&mut self, byte: u8) -> bool {
+        if self.input.as_bytes().get(self.index) == Some(&byte) {
+            self.index += 1;
+            true
+        } else {
+            false
+        }
+    }
+}
+
+fn write_canonical(value: &Json, out: &mut String) {
+    match value {
+        Json::Array(items) => {
+            out.push('[');
+            for (index, item) in items.iter().enumerate() {
+                if index > 0 {
+                    out.push(',');
+                }
+                write_canonical(item, out);
+            }
+            out.push(']');
+        }
+        Json::Object(pairs) => {
+            let mut keys: Vec<&(String, Json)> = pairs.iter().collect();
+            keys.sort_by(|left, right| utf16_cmp(&left.0, &right.0));
+            out.push('{');
+            for (index, (key, child)) in keys.iter().enumerate() {
+                if index > 0 {
+                    out.push(',');
+                }
+                write_string(key, out);
+                out.push(':');
+                write_canonical(child, out);
+            }
+            out.push('}');
+        }
+        other => write_scalar(other, out),
+    }
+}
+
+fn write_stringify(value: &Json, out: &mut String) {
+    match value {
+        Json::Array(items) => {
+            out.push('[');
+            for (index, item) in items.iter().enumerate() {
+                if index > 0 {
+                    out.push(',');
+                }
+                write_stringify(item, out);
+            }
+            out.push(']');
+        }
+        Json::Object(pairs) => {
+            out.push('{');
+            for (index, (key, child)) in pairs.iter().enumerate() {
+                if index > 0 {
+                    out.push(',');
+                }
+                write_string(key, out);
+                out.push(':');
+                write_stringify(child, out);
+            }
+            out.push('}');
+        }
+        other => write_scalar(other, out),
+    }
+}
+
+fn write_scalar(value: &Json, out: &mut String) {
+    match value {
+        Json::Null => out.push_str("null"),
+        Json::Bool(true) => out.push_str("true"),
+        Json::Bool(false) => out.push_str("false"),
+        Json::Int(value) => out.push_str(&value.to_string()),
+        Json::String(value) => write_string(value, out),
+        Json::Array(_) | Json::Object(_) => write_canonical(value, out),
+    }
+}
+
+fn write_string(value: &str, out: &mut String) {
+    out.push('"');
+    for ch in value.chars() {
+        match ch {
+            '"' => out.push_str("\\\""),
+            '\\' => out.push_str("\\\\"),
+            '\u{0008}' => out.push_str("\\b"),
+            '\u{000c}' => out.push_str("\\f"),
+            '\n' => out.push_str("\\n"),
+            '\r' => out.push_str("\\r"),
+            '\t' => out.push_str("\\t"),
+            other if (other as u32) < 0x20 => {
+                out.push_str(&format!("\\u{:04x}", other as u32));
+            }
+            other => out.push(other),
+        }
+    }
+    out.push('"');
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn line_separators_stay_literal() {
+        let value = Json::String("\u{2028}\u{2029}".into());
+        assert_eq!(canonicalize(&value), "\"\u{2028}\u{2029}\"");
+        assert_eq!(stringify(&value), "\"\u{2028}\u{2029}\"");
+    }
+}

package/templates/icp-knowledge-canister/canisters/knolo-core-rust/src/kar/mod.rs

@@ -0,0 +1,14 @@
+//! Native evaluate and verify for public KAR V5 conformance vectors.
+//! Authoring, coverage reports, Cortex, and composition stay in TypeScript.
+
+mod closure;
+mod cover;
+mod evaluate;
+mod json;
+mod model;
+mod session;
+
+pub use session::{
+    create_kar_dossier_v5, evaluate_kar_v5, render_kar_dossier, verify_kar_dossier_v5,
+    verify_kar_v5, KarDossier, KarEvaluation, KarRunError, KarVerification,
+};