Merge branch 'cleaning_2' of github.com:Wandalen/wTools into cleaning_2

Author: Wandalen

module/move/unilang_instruction_parser/changelog.md

@@ -0,0 +1,6 @@
+# Changelog
+
+* [Increment 1 | 2025-07-05 10:34 UTC] Added failing test for incorrect command path parsing.
+* [Increment 2 | 2025-07-05 10:58 UTC] Correctly parse command paths instead of treating them as arguments.
+*   Investigated and documented the correct usage of `strs_tools::string::split::SplitOptionsFormer` with dynamic delimiters to resolve lifetime issues.
+* [Increment 1 | 2025-07-06 06:42 UTC] Investigated `strs_tools` API issues and proposed switching to `regex` for string splitting.

module/move/unilang_instruction_parser/src/config.rs

@@ -1,120 +1,51 @@
-//! Defines configuration options for the unilang parser.
-use strs_tools::string::split::SplitOptionsFormer;
-use strs_tools::string::parse_request::OpType;
+//! Configuration options for the unilang instruction parser.
+//!
+//! This module defines the `UnilangParserOptions` struct, which allows
+//! customization of parsing behavior, including delimiters, operators,
+//! and error handling.
 
-/// High-level options for configuring the `unilang` parser.
-///
-/// These options control various aspects of the parsing process, such as how quotes and delimiters
-/// are handled, and rules for argument parsing. These options are then translated into
-/// lower-level settings for the `strs_tools::string::split::SplitOptionsFormer` which performs
-/// the initial tokenization of the input string.
-#[derive(Debug, Clone, PartialEq, Eq)]
-#[allow(clippy::struct_excessive_bools)]
+// Removed SplitOptionsFormer import as it's no longer used here.
+
+/// Configuration options for the unilang instruction parser.
+#[ derive( Debug, Clone ) ]
 pub struct UnilangParserOptions
 {
-  /// Defines pairs of characters or strings that denote the start and end of a quoted value.
-  ///
-  /// For example, `vec![("\"", "\""), ("'", "'")]` would recognize both double-quoted
-  /// and single-quoted strings. The parser will extract the inner content of these quotes.
-  /// Escape sequences within these quoted values are handled by the parser.
-  pub quote_pairs : Vec<( &'static str, &'static str )>,
-  /// A list of strings that act as primary delimiters or operators in the unilang syntax.
-  ///
-  /// This typically includes:
-  /// - `"::"` for separating named argument names from their values.
-  /// - `";;"` for separating multiple instructions within a single input string.
-  /// - `"?"` for requesting help on a command.
-  /// These delimiters are preserved during tokenization and used by the parser to
-  /// determine the structure of commands and arguments.
-  #[allow(clippy::doc_lazy_continuation)]
-  /// These delimiters are preserved during tokenization and used by the parser to
-  /// determine the structure of commands and arguments.
-  pub main_delimiters : Vec<&'static str>,
-  /// If `true`, leading and trailing whitespace will be stripped from each token produced
-  /// by the underlying `strs_tools` splitter before classification.
-  /// Defaults to `true`.
-  pub strip_whitespace : bool,
-  /// If `true`, the parser will return an error if a named argument is duplicated within a single instruction.
-  ///
-  /// For example, `cmd name::val1 name::val2` would cause an error.
-  /// If `false` (the default), the last occurrence of a duplicated named argument "wins", effectively
-  /// overwriting previous values for that argument name.
-  pub error_on_duplicate_named_arguments : bool,
-  /// If `true` (the default), the parser will return an error if a positional argument
-  /// is encountered after any named argument has already been parsed for that instruction.
-  ///
-  /// For example, `cmd name::val pos_arg` would cause an error.
-  /// If `false`, positional arguments can be interleaved with or follow named arguments,
-  /// e.g., `cmd name1::val1 pos1 name2::val2 pos2`.
+  /// If true, a positional argument after a named argument will result in a parse error.
   pub error_on_positional_after_named : bool,
-  /// If `true` (the default), whitespace characters (space, tab, newline, carriage return)
-  /// will also act as separators between tokens, in addition to `main_delimiters`.
-  /// If `false`, only `main_delimiters` will separate tokens, and whitespace might become
-  /// part of unquoted values.
+  /// If true, duplicate named arguments will result in a parse error.
+  pub error_on_duplicate_named_arguments : bool,
+  /// Pairs of quote characters (e.g., `("\"", "\"")`, `("'", "'")`).
+  pub quote_pairs : Vec< ( String, String ) >,
+  /// Main delimiters used for splitting the input string.
+  pub main_delimiters : Vec< String >,
+  /// If true, whitespace is considered a separator.
   pub whitespace_is_separator : bool,
 }
 
 impl Default for UnilangParserOptions
 {
-  /// Creates a default set of parser options.
-  ///
-  /// Default values are:
-  /// - `quote_pairs`: `vec![("\"", "\""), ("'", "'")]`
-  /// - `main_delimiters`: `vec![ "::", ";;", "?" ]`
-  /// - `strip_whitespace`: `true`
-  /// - `error_on_duplicate_named_arguments`: `false` (last one wins)
-  /// - `error_on_positional_after_named`: `true` (strict order)
-  /// - `whitespace_is_separator`: `true`
   fn default() -> Self
   {
     Self
     {
-      quote_pairs : vec![ ( "\"", "\"" ), ( "'", "'" ) ],
-      main_delimiters : vec![ "::", ";;", "?" ], // Corrected: removed duplicate line
-      strip_whitespace : true,
-      error_on_duplicate_named_arguments : false,
       error_on_positional_after_named : true,
-      whitespace_is_separator : true,
+      error_on_duplicate_named_arguments : true,
+      quote_pairs : vec!
+      [
+        ( "\"".to_string(), "\"".to_string() ),
+        ( "'".to_string(), "'".to_string() ),
+      ],
+      main_delimiters : vec!
+      [
+        "::".to_string(),
+        ";;".to_string(),
+        ".".to_string(),
+        "?".to_string(),
+        // Removed spaces and tabs from here, as strs_tools should handle whitespace as separator
+      ],
+      whitespace_is_separator : true, // Reverted to true
     }
   }
 }
 
-impl UnilangParserOptions
-{
-  /// Translates these high-level `UnilangParserOptions` into a `SplitOptionsFormer`
-  /// instance, which is used by the `strs_tools::string::split` module for initial
-  /// tokenization of the input string.
-  ///
-  /// This method configures the splitter based on the defined quote pairs, delimiters,
-  /// and whitespace handling rules.
-  #[allow(clippy::must_use_candidate)]
-  pub fn to_split_options_former<'s>( &'s self, src : &'s str ) -> SplitOptionsFormer<'s>
-  {
-    let mut prefixes = Vec::with_capacity( self.quote_pairs.len() );
-    let mut postfixes = Vec::with_capacity( self.quote_pairs.len() );
-    for (prefix, postfix) in &self.quote_pairs
-    {
-      prefixes.push( *prefix );
-      postfixes.push( *postfix );
-    }
-
-    let mut effective_delimiters = self.main_delimiters.clone();
-    if self.whitespace_is_separator
-    {
-      effective_delimiters.extend( vec![ " ", "\t", "\n", "\r" ] );
-    }
-
-    let mut former = SplitOptionsFormer::new( OpType::Vector( Vec::new() ) );
-    former.src( src );
-    former.delimeter( OpType::Vector( effective_delimiters ) );
-    former.preserving_empty( false );
-    former.preserving_delimeters( true );
-    former.stripping( self.strip_whitespace );
-    former.quoting( !self.quote_pairs.is_empty() );
-    former.quoting_prefixes( prefixes );
-    former.quoting_postfixes( postfixes );
-    former.preserving_quoting( true );
-
-    former
-  }
-}
+// Removed the to_split_options_former method.