AnnotatedStrings, and a string styling stdlib (#49586)

Author: LilithHafner

NEWS.md

@@ -8,6 +8,11 @@ New language features
   difference between `public` and `export` is that `public` names do not become
   available when `using` a package/module. ([#50105])
 * `ScopedValue` implement dynamic scope with inheritance across tasks ([#50958]).
+* A new `AbstractString` type, `AnnotatedString`, is introduced that allows for
+  regional annotations to be attached to an underlying string. This type is
+  particularly useful for holding styling information, and is used extensively
+  in the new `StyledStrings` standard library. There is also a new `AnnotatedChar`
+  type, that is the equivalent new `AbstractChar` type.
 
 Language changes
 ----------------
@@ -51,6 +56,17 @@ New library features
 Standard library changes
 ------------------------
 
+#### StyledStrings
+
+* A new standard library for handling styling in a more comprehensive and structured way.
+* The new `Faces` struct serves as a container for text styling information
+  (think typeface, as well as color and decoration), and comes with a framework
+  to provide a convenient, extensible (via `addface!`), and customisable (with a
+  user's `Faces.toml` and `loadfaces!`) approach to
+  styled content.
+* The new `@styled_str` string macro provides a convenient way of creating a
+  `AnnotatedString` with various faces or other attributes applied.
+
 #### Package Manager
 
 #### LinearAlgebra

base/client.jl

@@ -271,6 +271,10 @@ function exec_options(opts)
     interactiveinput = (repl || is_interactive::Bool) && isa(stdin, TTY)
     is_interactive::Bool |= interactiveinput
 
+    # load terminfo in for styled printing
+    term_env = get(ENV, "TERM", @static Sys.iswindows() ? "" : "dumb")
+    global current_terminfo = load_terminfo(term_env)
+
     # load ~/.julia/config/startup.jl file
     if startup
         try
@@ -416,11 +420,9 @@ function run_main_repl(interactive::Bool, quiet::Bool, banner::Symbol, history_f
         end
     end
     # TODO cleanup REPL_MODULE_REF
-
     if !fallback_repl && interactive && isassigned(REPL_MODULE_REF)
         invokelatest(REPL_MODULE_REF[]) do REPL
             term_env = get(ENV, "TERM", @static Sys.iswindows() ? "" : "dumb")
-            global current_terminfo = load_terminfo(term_env)
             term = REPL.Terminals.TTYTerminal(term_env, stdin, stdout, stderr)
             banner == :no || Base.banner(term, short=banner==:short)
             if term.term_type == "dumb"

base/exports.jl

@@ -1089,8 +1089,15 @@ public
     Generator,
     ImmutableDict,
     OneTo,
+    AnnotatedString,
+    AnnotatedChar,
     UUID,
 
+# Annotated strings
+    annotatedstring,
+    annotate!,
+    annotations,
+
 # Semaphores
     Semaphore,
     acquire,

base/regex.jl

@@ -212,14 +212,18 @@ julia> hr
 "11"
 ```
 """
-struct RegexMatch <: AbstractMatch
-    match::SubString{String}
-    captures::Vector{Union{Nothing,SubString{String}}}
+struct RegexMatch{S<:AbstractString} <: AbstractMatch
+    match::SubString{S}
+    captures::Vector{Union{Nothing,SubString{S}}}
     offset::Int
     offsets::Vector{Int}
     regex::Regex
 end
 
+RegexMatch(match::SubString{S}, captures::Vector{Union{Nothing,SubString{S}}},
+           offset::Union{Int, UInt}, offsets::Vector{Int}, regex::Regex) where {S<:AbstractString} =
+    RegexMatch{S}(match, captures, offset, offsets, regex)
+
 """
     keys(m::RegexMatch) -> Vector
 
@@ -423,9 +427,35 @@ function match(re::Regex, str::Union{SubString{String}, String}, idx::Integer,
     return result
 end
 
+function _annotatedmatch(m::RegexMatch{S}, str::AnnotatedString{S}) where {S<:AbstractString}
+    RegexMatch{AnnotatedString{S}}(
+        (@inbounds SubString{AnnotatedString{S}}(
+            str, m.match.offset, m.match.ncodeunits, Val(:noshift))),
+        Union{Nothing,SubString{AnnotatedString{S}}}[
+            if !isnothing(cap)
+                (@inbounds SubString{AnnotatedString{S}}(
+                    str, cap.offset, cap.ncodeunits, Val(:noshift)))
+            end for cap in m.captures],
+        m.offset, m.offsets, m.regex)
+end
+
+function match(re::Regex, str::AnnotatedString)
+    m = match(re, str.string)
+    if !isnothing(m)
+        _annotatedmatch(m, str)
+    end
+end
+
+function match(re::Regex, str::AnnotatedString, idx::Integer, add_opts::UInt32=UInt32(0))
+    m = match(re, str.string, idx, add_opts)
+    if !isnothing(m)
+        _annotatedmatch(m, str)
+    end
+end
+
 match(r::Regex, s::AbstractString) = match(r, s, firstindex(s))
 match(r::Regex, s::AbstractString, i::Integer) = throw(ArgumentError(
-    "regex matching is only available for the String type; use String(s) to convert"
+    "regex matching is only available for the String and AnnotatedString types; use String(s) to convert"
 ))
 
 findnext(re::Regex, str::Union{String,SubString}, idx::Integer) = _findnext_re(re, str, idx, C_NULL)
@@ -671,18 +701,19 @@ function _replace(io, repl_s::SubstitutionString, str, r, re)
     end
 end
 
-struct RegexMatchIterator
+struct RegexMatchIterator{S <: AbstractString}
     regex::Regex
-    string::String
+    string::S
     overlap::Bool
 
-    function RegexMatchIterator(regex::Regex, string::AbstractString, ovr::Bool=false)
-        new(regex, string, ovr)
-    end
+    RegexMatchIterator(regex::Regex, string::AbstractString, ovr::Bool=false) =
+        new{String}(regex, String(string), ovr)
+    RegexMatchIterator(regex::Regex, string::AnnotatedString, ovr::Bool=false) =
+        new{AnnotatedString{String}}(regex, AnnotatedString(String(string.string), string.annotations), ovr)
 end
 compile(itr::RegexMatchIterator) = (compile(itr.regex); itr)
-eltype(::Type{RegexMatchIterator}) = RegexMatch
-IteratorSize(::Type{RegexMatchIterator}) = SizeUnknown()
+eltype(::Type{<:RegexMatchIterator}) = RegexMatch
+IteratorSize(::Type{<:RegexMatchIterator}) = SizeUnknown()
 
 function iterate(itr::RegexMatchIterator, (offset,prevempty)=(1,false))
     opts_nonempty = UInt32(PCRE.ANCHORED | PCRE.NOTEMPTY_ATSTART)
@@ -727,7 +758,7 @@ julia> rx = r"a.a"
 r"a.a"
 
 julia> m = eachmatch(rx, "a1a2a3a")
-Base.RegexMatchIterator(r"a.a", "a1a2a3a", false)
+Base.RegexMatchIterator{String}(r"a.a", "a1a2a3a", false)
 
 julia> collect(m)
 2-element Vector{RegexMatch}: