Deprecating the use of C<< \cI<X> >> to specify a printable character.

Starting in 5.14, we deprecated the use of "\cI<X>" when this
results in a printable character. For instance, "\c:" is just
a fancy way of writing "z". Starting in 5.28, this will be a
fatal error.

This also includes certain usage in regular expressions with the
experimental (?[ ]) construct, or when "use re 'strict'" is in
effect (also experimental).

diff --git a/dquote.c b/dquote.c
index e02308e..ef03046 100644 (file)
--- a/dquote.c
+++ b/dquote.c
@@ -46,10 +46,10 @@ Perl_grok_bslash_c(pTHX_ const char source, const bool output_warning)
         clearer[i++] = result;
         clearer[i++] = '\0';
 
-        Perl_ck_warner(aTHX_ packWARN(WARN_SYNTAX),
-                        "\"\\c%c\" is more clearly written simply as \"%s\"",
-                        source,
-                        clearer);
+        Perl_ck_warner_d(aTHX_ packWARN2(WARN_SYNTAX,WARN_DEPRECATED),
+                          "\"\\c%c\" is more clearly written simply as \"%s\". "
+                          "This will be a fatal error in Perl 5.28",
+                          source, clearer);
     }
 
     return result;

diff --git a/pod/perldeprecation.pod b/pod/perldeprecation.pod
index 074ffd9..adb0737 100644 (file)
--- a/pod/perldeprecation.pod
+++ b/pod/perldeprecation.pod
@@ -218,6 +218,26 @@ C<use AutoLoader 'AUTOLOAD';>.
 This feature was deprecated in Perl 5.004, and will be fatal in Perl 5.28.
 
 
+=head3 Use of C<< \cI<X> >> to specify a printable character.
+
+In a double quoted context, Perl has the C<< \c >> construct to write
+control characters in a readable way. For instance, the tab character
+can be written as C<< \cI >> (I<< control-I >>), and the escape 
+character can be written as C<< \c[ >>.
+
+Due to implementation details, the C<< \c >> construct can be used
+to create regular, printable, characters as well. For instance,
+C<< \c >> maps a C<< , >> to C<< l >>; that is C<< \c, >> is an
+obscure way of writing C<< l >>. And not only that, it's also not
+portable between ASCII and EBCDIC platforms.
+
+Using the C<< \c >> construct with an argument which maps to a
+printable character was deprecated in Perl 5.14, and will be a fatal
+error in Perl 5.28. You're recommended to just write the intended
+character instead.
+
+
+
 =head2 Perl 5.26
 
 =head3 C<< --libpods >> in C<< Pod::Html >>

diff --git a/pod/perldiag.pod b/pod/perldiag.pod
index 7700f73..08e1c55 100644 (file)
--- a/pod/perldiag.pod
+++ b/pod/perldiag.pod
@@ -1539,7 +1539,7 @@ Perhaps you need to copy the value to a temporary, and repeat that.
 
 Note that ASCII characters that don't map to control characters are
 discouraged, and will generate the warning (when enabled)
-L</""\c%c" is more clearly written simply as "%s"">.
+L</""\c%c" is more clearly written simply as "%s". This will be a fatal error in Perl 5.28">.
 
 =item Character following \%c must be '{' or a single-character Unicode property name in regex; marked by <-- HERE in m/%s/
 
@@ -1646,13 +1646,15 @@ See L<charnames/CUSTOM ALIASES>.
 
 (W unopened) You tried chdir() on a filehandle that was never opened.
 
-=item "\c%c" is more clearly written simply as "%s"
+=item "\c%c" is more clearly written simply as "%s". This will be a fatal error in Perl 5.28
 
-(W syntax) The C<\cI<X>> construct is intended to be a way to specify
-non-printable characters.  You used it for a printable one, which
-is better written as simply itself, perhaps preceded by a backslash
-for non-word characters.  Doing it the way you did is not portable
-between ASCII and EBCDIC platforms.
+(D deprecated, syntax) The C<\cI<X>> construct is intended to be a
+way to specify non-printable characters.  You used it for a printable
+one, which is better written as simply itself, perhaps preceded by
+a backslash for non-word characters.  Doing it the way you did is
+not portable between ASCII and EBCDIC platforms.
+
+This usage is going to result in a fatal error in Perl 5.28.
 
 =item Cloning substitution context is unimplemented
 
@@ -3122,13 +3124,14 @@ code.  To avoid this a future version of perl will throw an exception when
 any of sysread(), recv(), syswrite() or send() are called on handle with the
 C<:utf8> layer.
 
-=item "%s" is more clearly written simply as "%s" in regex; marked by S<<-- HERE> in m/%s/
-
-(W regexp) (only under C<S<use re 'strict'>> or within C<(?[...])>)
+=item "%s" is more clearly written simply as "%s" in regex. This will be a fatal error in Perl 5.28; marked by S<<-- HERE> in m/%s/
 
-You specified a character that has the given plainer way of writing it,
-and which is also portable to platforms running with different character
-sets.
+(W deprecated, regexp) (only under C<S<use re 'strict'>> or within C<(?[...])>)
+The C<\cI<X>> construct is intended to be a way to specify non-printable
+characters.  You used it for a printable one, which is better written
+as simply itself, perhaps preceded by a backslash for non-word
+characters.  Doing it the way you did is not portable between ASCII
+and EBCDIC platforms.
 
 =item '%s' is not a code reference
 

diff --git a/regcomp.c b/regcomp.c
index 27e9dc2..d865c73 100644 (file)
--- a/regcomp.c
+++ b/regcomp.c
@@ -819,6 +819,13 @@ static const scan_data_t zero_scan_data =
                                        REPORT_LOCATION_ARGS(loc));      \
 } STMT_END
 
+#define        vWARN4dep(loc, m, a1, a2, a3) STMT_START {                             \
+    __ASSERT_(PASS2) Perl_warner(aTHX_ packWARN2(WARN_REGEXP,WARN_DEPRECATED), \
+                                       m REPORT_LOCATION,                      \
+                                      a1, a2, a3,                             \
+                                       REPORT_LOCATION_ARGS(loc));             \
+} STMT_END
+
 #define        ckWARN4reg(loc, m, a1, a2, a3) STMT_START {                     \
     __ASSERT_(PASS2) Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP),        \
                                           m REPORT_LOCATION,            \
@@ -16866,20 +16873,22 @@ S_regclass(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth,
                     literal[d++] = (char) value;
                     literal[d++] = '\0';
 
-                    vWARN4(RExC_parse,
-                           "\"%.*s\" is more clearly written simply as \"%s\"",
+                    vWARN4dep(RExC_parse,
+                           "\"%.*s\" is more clearly written simply as \"%s\". "
+                           "This will be a fatal error in Perl 5.28",
                            (int) (RExC_parse - rangebegin),
                            rangebegin,
                            literal
-                        );
+                    );
                 }
                 else if isMNEMONIC_CNTRL(value) {
-                    vWARN4(RExC_parse,
-                           "\"%.*s\" is more clearly written simply as \"%s\"",
+                    vWARN4dep(RExC_parse,
+                           "\"%.*s\" is more clearly written simply as \"%s\". "
+                           "This will be a fatal error in Perl 5.28",
                            (int) (RExC_parse - rangebegin),
                            rangebegin,
                            cntrl_to_mnemonic((U8) value)
-                        );
+                    );
                 }
             }
         }

diff --git a/t/lib/warnings/regcomp b/t/lib/warnings/regcomp
index 2b084c5..44ef5c3 100644 (file)
--- a/t/lib/warnings/regcomp
+++ b/t/lib/warnings/regcomp
@@ -31,12 +31,12 @@ BEGIN {
 use warnings;
 $a = qr/\c,/;
 $a = qr/[\c,]/;
-no warnings 'syntax';
+no warnings 'syntax', 'deprecated';
 $a = qr/\c,/;
 $a = qr/[\c,]/;
 EXPECT
-"\c," is more clearly written simply as "l" at - line 9.
-"\c," is more clearly written simply as "l" at - line 10.
+"\c," is more clearly written simply as "l". This will be a fatal error in Perl 5.28 at - line 9.
+"\c," is more clearly written simply as "l". This will be a fatal error in Perl 5.28 at - line 10.
 ########
 # This is because currently a different error is output under
 # use re 'strict', so can't go in reg_mesg.t

diff --git a/t/lib/warnings/toke b/t/lib/warnings/toke
index 1bbbb3c..3d5044d 100644 (file)
--- a/t/lib/warnings/toke
+++ b/t/lib/warnings/toke
@@ -1340,12 +1340,12 @@ BEGIN {
 use warnings;
 $a = "\c,";
 $a = "\c`";
-no warnings 'syntax';
+no warnings 'syntax', 'deprecated';
 $a = "\c,";
 $a = "\c`";
 EXPECT
-"\c," is more clearly written simply as "l" at - line 9.
-"\c`" is more clearly written simply as "\ " at - line 10.
+"\c," is more clearly written simply as "l". This will be a fatal error in Perl 5.28 at - line 9.
+"\c`" is more clearly written simply as "\ ". This will be a fatal error in Perl 5.28 at - line 10.
 ########
 # toke.c
 BEGIN {

diff --git a/t/re/reg_mesg.t b/t/re/reg_mesg.t
index 3ec90e0..050448b 100644 (file)
--- a/t/re/reg_mesg.t
+++ b/t/re/reg_mesg.t
@@ -532,10 +532,10 @@ my @warning = (
                   ],
     '/a{1,1}?\x{100}/' => 'Useless use of greediness modifier \'?\' {#} m/a{1,1}?{#}\x{100}/',
     "/(?[ [ % - % ] ])/" => "",
-    "/(?[ [ : - \\x$colon_hex ] ])\\x{100}/" => "\": - \\x$colon_hex \" is more clearly written simply as \":\" {#} m/(?[ [ : - \\x$colon_hex {#}] ])\\x{100}/",
-    "/(?[ [ \\x$colon_hex - : ] ])\\x{100}/" => "\"\\x$colon_hex\ - : \" is more clearly written simply as \":\" {#} m/(?[ [ \\x$colon_hex - : {#}] ])\\x{100}/",
-    "/(?[ [ \\t - \\x$tab_hex ] ])\\x{100}/" => "\"\\t - \\x$tab_hex \" is more clearly written simply as \"\\t\" {#} m/(?[ [ \\t - \\x$tab_hex {#}] ])\\x{100}/",
-    "/(?[ [ \\x$tab_hex - \\t ] ])\\x{100}/" => "\"\\x$tab_hex\ - \\t \" is more clearly written simply as \"\\t\" {#} m/(?[ [ \\x$tab_hex - \\t {#}] ])\\x{100}/",
+    "/(?[ [ : - \\x$colon_hex ] ])\\x{100}/" => "\": - \\x$colon_hex \" is more clearly written simply as \":\". This will be a fatal error in Perl 5.28 {#} m/(?[ [ : - \\x$colon_hex {#}] ])\\x{100}/",
+    "/(?[ [ \\x$colon_hex - : ] ])\\x{100}/" => "\"\\x$colon_hex\ - : \" is more clearly written simply as \":\". This will be a fatal error in Perl 5.28 {#} m/(?[ [ \\x$colon_hex - : {#}] ])\\x{100}/",
+    "/(?[ [ \\t - \\x$tab_hex ] ])\\x{100}/" => "\"\\t - \\x$tab_hex \" is more clearly written simply as \"\\t\". This will be a fatal error in Perl 5.28 {#} m/(?[ [ \\t - \\x$tab_hex {#}] ])\\x{100}/",
+    "/(?[ [ \\x$tab_hex - \\t ] ])\\x{100}/" => "\"\\x$tab_hex\ - \\t \" is more clearly written simply as \"\\t\". This will be a fatal error in Perl 5.28 {#} m/(?[ [ \\x$tab_hex - \\t {#}] ])\\x{100}/",
     "/(?[ [ $B_hex - C ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ $B_hex - C {#}] ])/",
     "/(?[ [ A - $B_hex ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ A - $B_hex {#}] ])/",
     "/(?[ [ $low_mixed_alpha - $high_mixed_alpha ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ $low_mixed_alpha - $high_mixed_alpha {#}] ])/",
@@ -611,10 +611,10 @@ my @warning_only_under_strict = (
     '/[\N{U+FF}-\x{100}]/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+FF}-\x{100}{#}]/',
     '/[\N{U+100}-\x{101}]/' => "",
     "/[%-%]/" => "",
-    "/[:-\\x$colon_hex]\\x{100}/" => "\":-\\x$colon_hex\" is more clearly written simply as \":\" {#} m/[:-\\x$colon_hex\{#}]\\x{100}/",
-    "/[\\x$colon_hex-:]\\x{100}/" => "\"\\x$colon_hex-:\" is more clearly written simply as \":\" {#} m/[\\x$colon_hex\-:{#}]\\x{100}/",
-    "/[\\t-\\x$tab_hex]\\x{100}/" => "\"\\t-\\x$tab_hex\" is more clearly written simply as \"\\t\" {#} m/[\\t-\\x$tab_hex\{#}]\\x{100}/",
-    "/[\\x$tab_hex-\\t]\\x{100}/" => "\"\\x$tab_hex-\\t\" is more clearly written simply as \"\\t\" {#} m/[\\x$tab_hex\-\\t{#}]\\x{100}/",
+    "/[:-\\x$colon_hex]\\x{100}/" => "\":-\\x$colon_hex\" is more clearly written simply as \":\". This will be a fatal error in Perl 5.28 {#} m/[:-\\x$colon_hex\{#}]\\x{100}/",
+    "/[\\x$colon_hex-:]\\x{100}/" => "\"\\x$colon_hex-:\" is more clearly written simply as \":\". This will be a fatal error in Perl 5.28 {#} m/[\\x$colon_hex\-:{#}]\\x{100}/",
+    "/[\\t-\\x$tab_hex]\\x{100}/" => "\"\\t-\\x$tab_hex\" is more clearly written simply as \"\\t\". This will be a fatal error in Perl 5.28 {#} m/[\\t-\\x$tab_hex\{#}]\\x{100}/",
+    "/[\\x$tab_hex-\\t]\\x{100}/" => "\"\\x$tab_hex-\\t\" is more clearly written simply as \"\\t\". This will be a fatal error in Perl 5.28 {#} m/[\\x$tab_hex\-\\t{#}]\\x{100}/",
     "/[$B_hex-C]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[$B_hex-C{#}]/",
     "/[A-$B_hex]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[A-$B_hex\{#}]/",
     "/[$low_mixed_alpha-$high_mixed_alpha]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[$low_mixed_alpha-$high_mixed_alpha\{#}]/",