disallow unrecognized string/char escapes (closes #21284)

NEWS.md

@@ -13,6 +13,9 @@ Language changes
  * The syntax `1.+2` is deprecated, since it is ambiguous: it could mean either
  `1 .+ 2` (the current meaning) or `1. + 2` ([#19089]).
 
+ * In string and character literals, backslash `\` may no longer
+ precede unrecognized escape characters ([#22800]).
+
 Breaking changes
 ----------------
 

base/markdown/render/html.jl

@@ -26,7 +26,7 @@ const _htmlescape_chars = Dict('<'=>"&lt;", '>'=>"&gt;",
  '"'=>"&quot;", '&'=>"&amp;",
  # ' '=>"&nbsp;",
  )
-for ch in "'`!\$\%()=+{}[]"
+for ch in "'`!\$%()=+{}[]"
  _htmlescape_chars[ch] = "&#$(Int(ch));"
 end
 

base/sparse/spqr.jl

@@ -390,9 +390,9 @@ end
 
 (\)(F::QRSparse{T}, B::StridedVecOrMat{T}) where {T} = _ldiv_basic(F, B)
 """
- (\)(F::QRSparse, B::StridedVecOrMat)
+ (\\)(F::QRSparse, B::StridedVecOrMat)
 
-Solve the least squares problem ``\min\|Ax - b\|^2`` or the linear system of equations
+Solve the least squares problem ``\\min\\|Ax - b\\|^2`` or the linear system of equations
 ``Ax=b`` when `F` is the sparse QR factorization of ``A``. A basic solution is returned
 when the problem is underdetermined.
 
@@ -404,7 +404,7 @@ julia> A = sparse([1,2,4], [1,1,1], ones(3), 4, 2)
  [2, 1] = 1.0
  [4, 1] = 1.0
 
-julia> qrfact(A)\ones(4)
+julia> qrfact(A)\\ones(4)
 2-element Array{Float64,1}:
  1.0
  0.0

base/statistics.jl

@@ -361,7 +361,7 @@ cov(X::AbstractMatrix, vardim::Int=1; corrected::Bool=true) =
 Compute the covariance between the vectors `x` and `y`. If `corrected` is `true` (the
 default), computes ``\\frac{1}{n-1}\\sum_{i=1}^n (x_i-\\bar x) (y_i-\\bar y)^*`` where
 ``*`` denotes the complex conjugate and `n = length(x) = length(y)`. If `corrected` is
-`false`, computes ``\frac{1}{n}\sum_{i=1}^n (x_i-\\bar x) (y_i-\\bar y)^*``.
+`false`, computes ``\\frac{1}{n}\\sum_{i=1}^n (x_i-\\bar x) (y_i-\\bar y)^*``.
 """
 cov(x::AbstractVector, y::AbstractVector; corrected::Bool=true) =
  covm(x, Base.mean(x), y, Base.mean(y); corrected=corrected)

src/flisp/read.c

@@ -516,7 +516,12 @@ static value_t read_string(fl_context_t *fl_ctx)
  i += u8_wc_toutf8(&buf[i], wc);
  }
  else {
- buf[i++] = read_escape_control_char((char)c);
+ char esc = read_escape_control_char((char)c);
+ if (esc == (char)c && !strchr("\\'\"$`", esc)) {
+ free(buf);
+ lerror(fl_ctx, fl_ctx->ParseError, "read: invalid escape sequence");
+ }
+ buf[i++] = esc;
  }
  }
  else {

test/libdl.jl

@@ -13,7 +13,7 @@ if !Sys.iswindows() || Sys.windows_version() >= Sys.WINDOWS_VISTA_VER
  end
 end
 @test length(filter(dlls) do dl
- return ismatch(Regex("^libjulia(?:.*)\.$(Libdl.dlext)(?:\..+)?\$"), basename(dl))
+ return ismatch(Regex("^libjulia(?:.*)\\.$(Libdl.dlext)(?:\\..+)?\$"), basename(dl))
  end) == 1 # look for something libjulia-like (but only one)
 
 # library handle pointer must not be NULL

test/strings/basic.jl

@@ -491,3 +491,7 @@ Base.endof(x::CharStr) = endof(x.chars)
 @test 'a' * 'b' == "ab"
 @test 'a' * "b" * 'c' == "abc"
 @test "a" * 'b' * 'c' == "abc"
+
+# unrecognized escapes in string/char literals
+@test_throws ParseError parse("\"\\.\"")
+@test_throws ParseError parse("\'\\.\'")

test/strings/io.jl

@@ -82,10 +82,6 @@ for i = 0:0x7f, p = ["","\0","x","xxx","\x7f","\uFF","\uFFF",
  @test string(unescape_string(string("\\x",hex(i,3),p))) == hp
 end
 
-@test "\z" == unescape_string("\z") == "z"
-@test "\X" == unescape_string("\X") == "X"
-@test "\AbC" == unescape_string("\AbC") == "AbC"
-
 @test "\0" == unescape_string("\\0")
 @test "\1" == unescape_string("\\1")
 @test "\7" == unescape_string("\\7")