Merge pull request #15 from JuliaString/spj/fixjoin

Fix join bug, better reporting of errors in calcstats

Author: ScottPJones

Project.toml

@@ -4,7 +4,7 @@ authors  = ["ScottPJones <[email protected]>"]
 keywords = ["Strings"]
 license  = "MIT"
 uuid     = "e79e7a6a-7bb1-5a4d-9d64-da657b06f53a"
-version = "1.0.5"
+version = "1.0.6"
 
 [deps]
 Unicode = "4ec0a83e-493e-50e2-b9ac-8f72acf5a8f5"

README.md

@@ -14,9 +14,7 @@
 [gitter-img]:   https://badges.gitter.im/Join%20Chat.svg
 [gitter-url]:   https://gitter.im/JuliaString/Lobby?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge
 
-[travis-url]:   https://travis-ci.org/JuliaString/StrBase.jl
-[travis-s-img]: https://travis-ci.org/JuliaString/StrBase.jl.svg
-[travis-m-img]: https://travis-ci.org/JuliaString/StrBase.jl.svg?branch=master
+[checks]:       https://img.shields.io/github/checks-status/JuliaString/StrBase.jl/master
 
 [codecov-url]:  https://codecov.io/gh/JuliaString/StrBase.jl
 [codecov-img]:  https://codecov.io/gh/JuliaString/StrBase.jl/branch/master/graph/badge.svg
@@ -27,8 +25,8 @@
 
 | **Julia Version** | **Unit Tests** | **Coverage** |
 |:------------------:|:------------------:|:---------------------:|
-| [![][julia-release]][julia-url] | [![][travis-s-img]][travis-url] | [![][codecov-img]][codecov-url]
-| Julia Latest | [![][travis-m-img]][travis-url] | [![][codecov-img]][codecov-url]
+| [![][julia-release]][julia-url] | | [![][codecov-img]][codecov-url]
+| Julia Latest | [![][checks]][pkg-url] | [![][codecov-img]][codecov-url]
 
 See the documentation in the [Strs](https://github.com/JuliaString/Strs.jl) package,
 this package contains the basic functionality, which can be used by other string related packages.

src/io.jl

@@ -379,7 +379,7 @@ join(strings::AbstractArray{<:MaybeSub{T}}) where {C<:Quad_CSEs,T<:Str{C}} =
     _join(C, strings)
 
 join(strings::AbstractArray{<:MaybeSub{<:Str}}, delim) =
-    _joincvt(_calc_type(strings), strings, delim)
+    _joincvt(calc_type(strings), strings, delim)
 join(strings::AbstractArray{<:MaybeSub{T}},
      delim) where {C<:Union{Text1CSE, BinaryCSE, ASCIICSE, Latin_CSEs},T<:Str{C}} =
     _join(C, strings, delim)

src/stats.jl

@@ -1,7 +1,7 @@
 #=
 Statistics for Unicode character/string types
 
-Copyright 2017 Gandalf Software, Inc., Scott P. Jones
+Copyright 2017-2021 Gandalf Software, Inc., Scott P. Jones
 Licensed under MIT License, see LICENSE.md
 =#
 struct LineCounts
@@ -48,13 +48,24 @@ function calcstats(lines)
     minlen = typemax(Int)
     maxlen = 0
     maxtyp = 0
+    linnum = 0
     for l in lines
+        linnum += 1
         len = length(l)
         minlen = min(minlen, len)
         maxlen = max(maxlen, len)
         flags = 0
+        chrnum = 0
         for chr in l
-            ch = chr%UInt32
+            chrnum += 1
+            ch = UInt32(0)
+            try
+                ch = chr%UInt32
+            catch ex
+                typeof(ex) == InterruptException ||
+                    println("Bad character: $chr at line $linnum, pos $chrnum")
+                rethrow()
+            end
             t = ch <= 0x7f ? 1 :
                 ch <= 0xff ? 2 :
                 ch <= 0x7ff ? 3 :