release 5.25.2c

see pod/perlcdelta.pod or http://perl11.org/cperl/perl5252cdelta.html

Author: Reini Urban

Porting/release_cperl.pod

@@ -13,7 +13,11 @@ There's no C<make release> target yet.
 
     less Porting/release_manager_guide.pod
 
-    ../do-conf-cperl-release
+    Porting/do-conf-cperl-release
+    Porting/do-make-cperl-release
+
+or:
+
     make -s -j4 test && make -s install DESTDIR=`pwd`/../cperl-5.25.2-inst
 
     mkdir ../cperl-5.25.2
@@ -28,7 +32,7 @@ There's no C<make release> target yet.
         --version 5.25.2 \
         -m rurban@cpanel.net \
         --url http://perl11.org/cperl/$t/ \
-        --deb-config Porting/debian --deb-changelog pod/perl5222cdelta.pod
+        --deb-config Porting/debian --deb-changelog pod/perl5252cdelta.pod
     done
 
 So far we have no upstream accepted C<cperl> packages yet, so we
@@ -65,7 +69,8 @@ ccflags is architecture specific.
                 -Dcf_email='rurban@cpanel.net' \
                 -Uusethreads -Doptimize='-O3 -g'
 
-    sed -i -e's,-fstack-protector-strong,-fstack-protector,' config.sh Makefile cflags
+    sed -i -e's,-fstack-protector-strong,-fstack-protector,' config.sh \
+      Makefile cflags
     touch config.h
     make -s git_version.h ECHO=true
     make -s -j4 ECHO=true
@@ -174,10 +179,24 @@ Copy the created rpms into the rpm dest dirs.
 
 perl11.org/cperl/STATUS.html, perl11.org/cperl/Changes.html and the pods.
 
-(from markdown and pod) TODO...
+From markdown and pod:
+
+    ROOT=../perl11.github.com/cperl
+    
+    #Markdown.pl Changes > $ROOT/Changes.html
+    pod2html --podpath=pod pod/perlcperl.pod > $ROOT/index.html
+    cdelta=`ls pod/perl5*cdelta.pod|sed 's,pod/perl,,; s,.pod$,,' `
+    for p in "" data cperl sub types obj cdelta delta call util guts hack $cdelta sec hist; do
+        pod2html --podpath=pod pod/perl$p.pod > $ROOT/perl$p.html
+    done
+    Markdown.pl STATUS.md > $ROOT/STATUS.html
+    
+    sed -i -e's,href="/pod/,href="/cperl/,' $ROOT/*.html
 
 =head2 publish the website
 
+    cd ../perl11.github.com/cperl
+    git ci -am'update cperl release'
     git push
 
 Check it and announce it on twitter and your blog.

STATUS.md

@@ -31,8 +31,8 @@ Windows is smoked with MSVC 10 and 12 for 32 and 64bit.
 
 The current stable release is
   [5.24.1c](https://github.com/perl11/cperl/releases/tag/cperl-5.24.1) - [perl5241cdelta](perl5241cdelta.html),
-the latest development release [5.25.0c](https://github.com/perl11/cperl/releases/tag/cperl-5.25.0) - [perl5250cdelta](perl5250cdelta.html).
-See the current changelog [cperl-5.25.2 delta](perlcdelta.html).
+the latest development release [5.25.2c](https://github.com/perl11/cperl/releases/tag/cperl-5.25.2) - [perl5250cdelta](perl5252cdelta.html).
+See the current changelog [cperlcdelta](perlcdelta.html).
 We also have [5.22.4c](https://github.com/perl11/cperl/releases/tag/cperl-5.22.4), [perl5224cdelta](perl5224cdelta.html).
 
 All tests pass. CPAN works.
@@ -99,7 +99,7 @@ For all versions see [bench-all/](bench-all/index.html)
   versions with a 'c' suffix, and 10x faster JSON and YAML usage. (esp. with cpan).
 * many typed and modernized core modules, where signatures and types make
   sense and cause not much trouble.
-* some security fixes for Unicode confusables, but more are needed (use strict 'names').
+* many security fixes for Unicode symbols. no mixed scripts, normalized, no \0.
 * handle method calls on protected stashes.
 * disallow silent overflows of hash and array indices or string/name lengths.
   New "Too many elements" error and many new "overlarge" or "too large" panics.
@@ -313,6 +313,17 @@ are limited. So they are based on master.
 
   compile/dump to native code via emacs unexec, on most platforms.
 
+* [feature/gh141-smallstring](https://github.com/perl11/cperl/issues/141)
+
+  [code](http://github.com/perl11/cperl/commits/feature/gh141-smallstring)
+
+  optimize space for small strings.
+
+and various [hash tables refactorings]((https://github.com/perl11/cperl/issues/24):
+
+feature/gh24-base-hash feature/gh24-he-array feature/gh24-oldnew-hash-table
+featurex/gh24-array_he featurex/gh24-hash-loop featurex/gh24-hash-loop+utf8
+featurex/gh24-hash-utf8
 
 ## A bit more work is needed for
 
@@ -350,6 +361,10 @@ They also revert some wrong decisions p5p already made.
   class, method and multi keywords but no dispatch, subtyping and type checks yet.
   in work.
 
+* various more hash tables:
+
+featurex/gh24-one-word-ahe featurex/gh24-open-hash
+
 ## Soon
 
 * user facing classes, multiple dispatch (fast for binary, slow for mega)

t/porting/known_pod_issues.dat

@@ -375,6 +375,7 @@ pod/perlwin32.pod	Verbatim line length including indents exceeds 79 by	7
 porting/epigraphs.pod	Verbatim line length including indents exceeds 79 by	16
 porting/release_cperl.pod	? Should you be using F<...> or maybe L<...> instead of	1
 porting/release_cperl.pod	Verbatim line length including indents exceeds 79 by	4
+porting/release_cperl.pod	line containing nothing but whitespace in paragraph	2
 porting/release_managers_guide.pod	Verbatim line length including indents exceeds 79 by	1
 porting/todo.pod	? Should you be using F<...> or maybe L<...> instead of	1
 lib/benchmark.pm	Verbatim line length including indents exceeds 79 by	2

t/uni/normalize.t

@@ -41,7 +41,7 @@ for my $c (1 .. 0x10FFFF) {
     # => 12076 confusables
 }
 
-plan (tests => 22 + (scalar(@nfc)/2));
+plan (tests => 24 + (scalar(@nfc)/2));
 my $i = 1;
 
 # first check if _is_utf8_decomposed() catches all characters for pv_uni_normalize.
@@ -76,35 +76,23 @@ while (@nfc) {
     is( $orig, $i, "orig global var \${$qfrom}" );
     is( $norm, $i, "norm global var \${$qto}" );
     my $b = B::svref_2object(\$gv);
-    if (ref $b eq 'B::GV') {
-        is( $b->NAME, $to, "normalize E+COMBINING ACUTE ACCENT => E WITH ACUTE" );
-    } else {
-        is( ref $b, "B::GV", " => $qto (".ord($to).")");
-    }
+    is( $b->NAME, $to, "normalize E+COMBINING ACUTE ACCENT => E WITH ACUTE" );
     $i++;
 
     $É = $i; ($orig, $norm, $gv) = ($É, $É, *É);
     is( $orig, $i, "orig global var \$$qfrom" );
     is( $norm, $i, "norm global var \$$qto" );
     my $b = B::svref_2object(\$gv);
-    if (ref $b eq 'B::GV') {
-        is( $b->NAME, $to, "normalize GV E+COMBINING ACUTE ACCENT => E WITH ACUTE" );
-    } else {
-        is( ref $b, "B::GV", " => $qto");
-    }
+    is( $b->NAME, $to, "normalize GV E+COMBINING ACUTE ACCENT => E WITH ACUTE" );
     $i++;
 
     sub É {$i}; $orig = É(); $norm = É(); $cv = \&É;
     is( $orig, $i, "orig sub $qfrom" );
     is( $norm, $i, "norm sub $qto" );
   TODO: {
-      local $TODO = "B::CV->NAME_HEK";
+      local $TODO = "B::CV->NAME_HEK of \\&";
       $b = B::svref_2object($cv);
-      if (ref $b eq 'B::CV') {
-          is( $b->NAME_HEK, $to, "normalize CV NAME to E WITH ACUTE" );
-      } else {
-          is( ref $b, "B::CV", " => $qto");
-      }
+      is( $b->NAME_HEK, $to, "normalize CV NAME to E WITH ACUTE" );
     }
     $i++;
 
@@ -116,15 +104,8 @@ while (@nfc) {
     sub bÉ () {$i}; ($orig, $norm, $cv) = (bÉ(), bÉ(), \&bÉ);
     is( $orig, $i, "orig const sub $qfrom" );
     is( $norm, $i, "norm const sub $qto" );
-  TODO: {
-      local $TODO = "B::CV->NAME_HEK";
-      $b = B::svref_2object(\$cv);
-      if (ref $b eq 'B::CV') {
-          is( $b->NAME_HEK, $to, "normalize CV NAME to E WITH ACUTE" );
-      } else {
-          is( ref $b, "B::CV", " => $qto");
-      }
-    }
+    $b = B::svref_2object($cv);
+    is( $b->NAME_HEK, "b".$to, "normalize CV NAME to E WITH ACUTE" );
     $i++;
 
     sub PKG_É::É {$i};
@@ -149,3 +130,11 @@ while (@nfc) {
     is( $orig, $i, "dynamic string ref \${\"$qfrom\"}");
     is( $norm, 0,  "not normalized");
 }
+
+# almost illegal unicode: double combiners
+my $qfrom = join("",map{sprintf"\\x{%x}",$_}unpack"U*","É́");
+my $qto   = join("",map{sprintf"\\x{%x}",$_}unpack"U*","É́");
+# correctly normalized, but shouldn't this warn?
+${É́} = $i; ($orig, $norm) = (${É́}, ${É́});
+is( $orig, $i, "orig global var \${$qfrom}" );
+is( $norm, $i, "norm global var \${$qto}" );