Migrate from &str to &[u8] in ixdtf (#4918)

Author: nekevss

Cargo.lock

@@ -28,6 +28,7 @@ duration = []
 
 [dependencies]
 displaydoc = { workspace = true }
+utf8_iter.workspace = true
 
 [dev-dependencies]
 serde-json-core = { workspace = true, features = ["std"] }

utils/ixdtf/Cargo.toml

@@ -25,7 +25,7 @@
 //!
 //! let ixdtf_str = "2024-03-02T08:48:00-05:00[America/New_York]";
 //!
-//! let result = IxdtfParser::new(ixdtf_str).parse().unwrap();
+//! let result = IxdtfParser::from_str(ixdtf_str).parse().unwrap();
 //!
 //! let date = result.date.unwrap();
 //! let time = result.time.unwrap();
@@ -41,7 +41,7 @@
 //! assert_eq!(offset.hour, 5);
 //! assert_eq!(offset.minute, 0);
 //! assert!(!tz_annotation.critical);
-//! assert_eq!(tz_annotation.tz, TimeZoneRecord::Name("America/New_York"));
+//! assert_eq!(tz_annotation.tz, TimeZoneRecord::Name("America/New_York".as_bytes()));
 //! ```
 //!
 //! ## Date/Time Strings
@@ -142,9 +142,7 @@
 //! let example_one =
 //!     "2024-03-02T08:48:00-05:00[u-ca=iso8601][America/New_York]";
 //!
-//! let mut ixdtf = IxdtfParser::new(example_one);
-//!
-//! let result = ixdtf.parse();
+//! let result = IxdtfParser::from_str(example_one).parse();
 //!
 //! assert_eq!(result, Err(ParserError::AnnotationKeyLeadingChar));
 //! ```
@@ -158,9 +156,9 @@
 //! ```rust
 //! use ixdtf::{parsers::IxdtfParser, ParserError};
 //!
-//! let example_one = "2024-03-02T08:48:00-05:00[u-ca=iso8601][!u-ca=japanese]";
+//! let example_two = "2024-03-02T08:48:00-05:00[u-ca=iso8601][!u-ca=japanese]";
 //!
-//! let result = IxdtfParser::new(example_one).parse();
+//! let result = IxdtfParser::from_str(example_two).parse();
 //!
 //! assert_eq!(result, Err(ParserError::CriticalDuplicateCalendar));
 //! ```
@@ -173,10 +171,10 @@
 //! ```rust
 //! use ixdtf::{parsers::IxdtfParser, ParserError};
 //!
-//! let example_one =
+//! let example_three =
 //!     "2024-03-02T08:48:00-05:00[u-ca=iso8601][!answer-to-universe=fortytwo]";
 //!
-//! let result = IxdtfParser::new(example_one).parse();
+//! let result = IxdtfParser::from_str(example_three).parse();
 //!
 //! assert_eq!(result, Err(ParserError::UnrecognizedCritical));
 //! ```
@@ -207,19 +205,17 @@
 //! ```rust
 //! use ixdtf::parsers::{IxdtfParser, records::TimeZoneRecord};
 //!
-//! let example_one = "2024-03-02T08:48:00+01:00[!America/New_York]";
-//!
-//! let mut ixdtf = IxdtfParser::new(example_one);
+//! let example_two = "2024-03-02T08:48:00+01:00[!America/New_York]";
 //!
-//! let result = ixdtf.parse().unwrap();
+//! let result = IxdtfParser::from_str(example_two).parse().unwrap();
 //!
 //! let tz_annotation = result.tz.unwrap();
 //! let offset = result.offset.unwrap();
 //!
 //! // The time zone annotation and offset conflict with each other, and must therefore be
 //! // resolved by the user.
 //! assert!(tz_annotation.critical);
-//! assert_eq!(tz_annotation.tz, TimeZoneRecord::Name("America/New_York"));
+//! assert_eq!(tz_annotation.tz, TimeZoneRecord::Name("America/New_York".as_bytes()));
 //! assert_eq!(offset.hour, 1);
 //! ```
 //!
@@ -254,8 +250,8 @@
 //!
 //! let mut answer = None;
 //!
-//! let _ = IxdtfParser::new(example_with_custom_key).parse_with_annotation_handler(|annotation| {
-//!     if annotation.key == "answer-to-universe" {
+//! let _ = IxdtfParser::from_str(example_with_custom_key).parse_with_annotation_handler(|annotation| {
+//!     if annotation.key == "answer-to-universe".as_bytes() {
 //!         answer.get_or_insert(annotation);
 //!         // Found our value! We don't need `ixdtf` to handle this annotation.
 //!         return None
@@ -268,7 +264,7 @@
 //! let answer = answer.unwrap();
 //!
 //! assert!(answer.critical);
-//! assert_eq!(answer.value, "fortytwo");
+//! assert_eq!(answer.value, "fortytwo".as_bytes());
 //! ```
 //!
 //! It is worth noting that in the above example the annotation above found is a critically flagged

utils/ixdtf/README.md

@@ -21,7 +21,7 @@ use crate::{
 /// Strictly a parsing intermediary for the checking the common annotation backing.
 pub(crate) struct AnnotationSet<'a> {
     pub(crate) tz: Option<TimeZoneAnnotation<'a>>,
-    pub(crate) calendar: Option<&'a str>,
+    pub(crate) calendar: Option<&'a [u8]>,
 }
 
 /// Parse a `TimeZoneAnnotation` `Annotations` set
@@ -53,15 +53,15 @@ pub(crate) fn parse_annotation_set<'a>(
 pub(crate) fn parse_annotations<'a>(
     cursor: &mut Cursor<'a>,
     mut handler: impl FnMut(Annotation<'a>) -> Option<Annotation<'a>>,
-) -> ParserResult<Option<&'a str>> {
+) -> ParserResult<Option<&'a [u8]>> {
     let mut calendar: Option<Annotation<'a>> = None;
 
     while cursor.check_or(false, is_annotation_open) {
         let annotation = handler(parse_kv_annotation(cursor)?);
 
         match annotation {
             // Check if the key is the registered key "u-ca".
-            Some(kv) if kv.key == "u-ca" => {
+            Some(kv) if kv.key == "u-ca".as_bytes() => {
                 // Check the calendar
                 match calendar {
                     Some(calendar)
@@ -124,7 +124,7 @@ fn parse_kv_annotation<'a>(cursor: &mut Cursor<'a>) -> ParserResult<Annotation<'
 }
 
 /// Parse an `AnnotationKey`.
-fn parse_annotation_key<'a>(cursor: &mut Cursor<'a>) -> ParserResult<&'a str> {
+fn parse_annotation_key<'a>(cursor: &mut Cursor<'a>) -> ParserResult<&'a [u8]> {
     let key_start = cursor.pos();
     assert_syntax!(
         is_a_key_leading_char(cursor.next_or(ParserError::AnnotationKeyLeadingChar)?),
@@ -147,7 +147,7 @@ fn parse_annotation_key<'a>(cursor: &mut Cursor<'a>) -> ParserResult<&'a str> {
 }
 
 /// Parse an `AnnotationValue`.
-fn parse_annotation_value<'a>(cursor: &mut Cursor<'a>) -> ParserResult<&'a str> {
+fn parse_annotation_value<'a>(cursor: &mut Cursor<'a>) -> ParserResult<&'a [u8]> {
     let value_start = cursor.pos();
     cursor.advance();
     while let Some(potential_value_char) = cursor.next() {

utils/ixdtf/src/lib.rs

@@ -66,7 +66,7 @@ pub(crate) fn parse_date_duration(cursor: &mut Cursor) -> ParserResult<DateDurat
 
     while cursor.check_or(false, |ch| ch.is_ascii_digit()) {
         let mut value: u32 = 0;
-        while cursor.check_or(false, |c| c.is_ascii_digit()) {
+        while cursor.check_or(false, |ch| ch.is_ascii_digit()) {
             let digit = cursor
                 .next_digit()?
                 .ok_or_else(|| ParserError::abrupt_end("DateDuration"))?;
@@ -127,13 +127,13 @@ pub(crate) fn parse_time_duration(cursor: &mut Cursor) -> ParserResult<Option<Ti
 
     cursor.advance();
     assert_syntax!(
-        cursor.check_or(false, |ch| ch.is_ascii_digit()),
+        cursor.check_or(false, |c| c.is_ascii_digit()),
         TimeDurationDesignator,
     );
 
     let mut time: (u32, u32, u32, Option<u32>) = (0, 0, 0, None);
     let mut previous_unit = TimeUnit::None;
-    while cursor.check_or(false, |ch| ch.is_ascii_digit()) {
+    while cursor.check_or(false, |c| c.is_ascii_digit()) {
         let mut value: u32 = 0;
         while cursor.check_or(false, |c| c.is_ascii_digit()) {
             let digit = cursor

utils/ixdtf/src/parsers/annotations.rs

@@ -66,7 +66,7 @@ pub(crate) const fn is_time_designator(ch: char) -> bool {
 #[inline]
 /// Checks if char is a space.
 pub(crate) const fn is_space(ch: char) -> bool {
-    ch == '\u{0020}'
+    ch == ' '
 }
 
 /// Checks if char is a `DateTimeSeparator`.