CPAN Tester Error Reports

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

CPAN Tester Error Reports

Geoff McLane
Hi all,

Re: CPAN Tester Error Reports

@benkasminbullock has reported some CPAN Testers Matrix (http://matrix.cpantesters.org/) compile failures on certain platforms. Thanks for these.

Naturally the HTML Tidy development group can only test, and potentially fix problems, in platforms they have access to...

We certainly do want to support just about **every** platform, and invite developers with platforms that fail the CPAN Testers Matrix to join us, to give HTML Tidy the broadest reach possible.

Some present failures are -



Naturally, if a developer with one of these platform finds a fix, then would very much appreciate a Pull Request (https://github.com/htacg/tidy-html5/pulls), or a patch to here, to https://github.com/htacg/tidy-html5/issues, or direct...

Look forward to your continued support.

Regards,
Geoff.

Reply | Threaded
Open this post in threaded view
|

Re: CPAN Tester Error Reports

Geoff McLane
Transfer of comments from HTML Tidy issues: -

=========================================

gnukfreebsd compilation #302
@benkasminbullock commented 13 hours ago

Currently the default compilation fails on gnukfreebsd.

See this:


and this:


There doesn't seem to be a gnukfreebsd handler in the options in tidyplatform.h.

========================================

As you can see here:


there have been some problems with compiling on Darwin. It seems that "uint" does not get defined on this platform.

=========================================

As stated any help from developers with these platform much appreciated...

Reply | Threaded
Open this post in threaded view
|

Re: CPAN Tester Error Reports

Richard A. O'Keefe
In reply to this post by Geoff McLane
(1) Download from https://github.com/htacg/tidy-html5
(2) Unpack.
(3) Read README.
(4) WHOOPS!  README is not a plain text file, as its
    name suggests, it's a directory!
(5) Read the stuff in README/
(6) cd build/cmake
    cmake ../.. -DMAKE_INSTALL_PREFIX=$HOME/local

CMake Warning (dev):
  Policy CMP0042 is not set: MACOSX_RPATH is enabled by default.  Run "cmake
  --help-policy CMP0042" for policy details.  Use the cmake_policy command to
  set the policy and suppress this warning.

  MACOSX_RPATH is not specified for the following targets:

   tidy-share

This warning is for project developers.  Use -Wno-dev to suppress it.

m% cmake --help-policy CMP0042
CMP0042
-------

``MACOSX_RPATH`` is enabled by default.

CMake 2.8.12 and newer has support for using ``@rpath`` in a target's install
name.  This was enabled by setting the target property
``MACOSX_RPATH``.  The ``@rpath`` in an install name is a more
flexible and powerful mechanism than ``@executable_path`` or ``@loader_path``
for locating shared libraries.

CMake 3.0 and later prefer this property to be ON by default.  Projects
wanting ``@rpath`` in a target's install name may remove any setting of
the ``INSTALL_NAME_DIR`` and ``CMAKE_INSTALL_NAME_DIR``
variables.

This policy was introduced in CMake version 3.0.  CMake version
3.0.1 warns when the policy is not set and uses OLD behavior.  Use
the cmake_policy command to set it to OLD or NEW explicitly.

(7) make

[ 15%] Building C object CMakeFiles/tidy-static.dir/src/pprint.c.o
/tmp/tidy/src/pprint.c:1030:52: warning: '&&' within '||'
      [-Wlogical-op-parentheses]
  ...1) == end) || ((ix + 1) < end) && (isspace(doc->lexer->lexbuf[ix+1]))) )
                ~~ ~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/tmp/tidy/src/pprint.c:1030:52: note: place parentheses around the '&&'
      expression to silence this warning
  ...1) == end) || ((ix + 1) < end) && (isspace(doc->lexer->lexbuf[ix+1]))) )
                                    ^
                   (                                                      )
1 warning generated.

I personally regard this advice as expressing contempt
for programmers.  However, given that the line in
question already has no fewer than SEVEN pairs of
parentheses more than the compiler or human readers
need, adding an eighth pair wouldn't make it significantly
more horrible than it already is.

[ 37%] Building C object CMakeFiles/tidy-static.dir/src/streamio.c.o
/tmp/tidy/src/streamio.c:955:19: warning: unused variable 'Symbol2Unicode'
      [-Wunused-const-variable]
static const uint Symbol2Unicode[] =
                  ^
This is a table to convert characters in the "Symbol"
font to Unicode.  The function DecodeSymbolFont()
that would have used it is #if 0-ed out.

These are the only issues reported by /usr/bin/cc
which is
Apple LLVM version 5.1 (clang-503.0.38) (based on LLVM 3.4svn)
Target: x86_64-apple-darwin13.4.0
Thread model: posix
on this Mac OS X 10.9.5 box (Darwin Kernel 13.4.0).

In particular, there doesn't appear to be any problem
with 'uint' on this system.


(8) Going to src/ and doing

cc --analyze -I../include *.c

produces these reports.  Clang isn't always right,
but the last time I was certain clang was wrong about
a warning it turned out I had misunderstood what I
was being warned about and clang had been trying to
tell me about a real bug, so it's always worth
investigating when clang complains.

access.c:1605:29: warning: Value stored to 'flag' is never read
                            flag = 1;
                            ^      ~
access.c:2150:13: warning: Access to field 'content' results in a dereference of
      a null pointer (loaded from variable 'node')
        if (node->content != NULL && node->content->tag == NULL)
            ^~~~~~~~~~~~~
access.c:2756:30: warning: Access to field 'content' results in a dereference of
      a null pointer (loaded from variable 'node')
             TY_(nodeIsText)(node->content) )
                             ^~~~~~~~~~~~~
access.c:2771:22: warning: Access to field 'content' results in a dereference of
      a null pointer (loaded from variable 'node')
        for ( node = node->content; node; node = node->next )
                     ^~~~~~~~~~~~~
access.c:3237:21: warning: Access to field 'content' results in a dereference of
      a null pointer (loaded from variable 'node')
    for ( content = node->content; content != NULL; content = content->next )
                    ^~~~~~~~~~~~~
5 warnings generated.
attrs.c:1341:24: warning: Access to field 'next' results in a dereference of a
      null pointer (loaded from variable 'first')
                temp = first->next;
                       ^~~~~~~~~~~
attrs.c:1838:13: warning: Access to field 'value' results in a dereference of a
      null pointer (loaded from variable 'attval')
        if (attval->value == NULL)
            ^~~~~~~~~~~~~
2 warnings generated.
config.c:811:9: warning: Value stored to 'c' is never read
        c = FirstChar( cfg );
        ^   ~~~~~~~~~~~~~~~~
config.c:831:17: warning: Value stored to 'c' is never read
                c = AdvanceChar( cfg );
                ^   ~~~~~~~~~~~~~~~~~~
2 warnings generated.
fileio.c:81:5: warning: Dereference of null pointer
    TidyFree( fin->unget.allocator, fin );
    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./forward.h:58:37: note: expanded from macro 'TidyFree'
#define TidyFree(allocator, block) ((allocator)->vtbl->free((allocator)...
                                    ^~~~~~~~~~~
1 warning generated.
lexer.c:2113:43: warning: Access to field 'element' results in a dereference of
      a null pointer (loaded from variable 'container')
            matches = TY_(tmbstrncasecmp)(container->element, lexer->lex...
                                          ^~~~~~~~~~~~~~~~~~
1 warning generated.
localize.c:1763:65: warning: Access to field 'element' results in a dereference
      of a null pointer (loaded from variable 'element')
  ...TidyWarning, node, fmt, node->element, element->element);
                                            ^~~~~~~~~~~~~~~~
localize.c:1785:49: warning: Access to field 'element' results in a dereference
      of a null pointer (loaded from variable 'element')
        messageNode(doc, TidyWarning, rpt, fmt, element->element);
                                                ^~~~~~~~~~~~~~~~
localize.c:1789:49: warning: Access to field 'element' results in a dereference
      of a null pointer (loaded from variable 'element')
        messageNode(doc, TidyWarning, rpt, fmt, element->element, nodedesc);
                                                ^~~~~~~~~~~~~~~~
localize.c:1800:60: warning: Access to field 'element' results in a dereference
      of a null pointer (loaded from variable 'element')
        messageNode(doc, TidyWarning, node, fmt, nodedesc, element->element);
                                                           ^~~~~~~~~~~~~~~~
localize.c:1835:63: warning: Access to field 'element' results in a dereference
      of a null pointer (loaded from variable 'element')
        messageNode(doc, TidyError, node, fmt, node->element, element->element);
                                                              ^~~~~~~~~~~~~~~~
5 warnings generated.
parser.c:213:14: warning: Access to field 'parent' results in a dereference of a
      null pointer (loaded from variable 'element')
    parent = element->parent;
             ^~~~~~~~~~~~~~~
parser.c:222:9: warning: Access to field 'content' results in a dereference of a
      null pointer (loaded from variable 'parent')
    if (parent->content == element)
        ^~~~~~~~~~~~~~~
parser.c:426:27: warning: Access to field 'model' results in a dereference of a
      null pointer (loaded from field 'tag')
                    if ( (element->tag->model & CM_INLINE) &&
                          ^~~~~~~~~~~~~~~~~~~
parser.c:500:15: warning: Access to field 'model' results in a dereference of a
      null pointer (loaded from field 'tag')
        if ( (element->tag->model & CM_INLINE) &&
              ^~~~~~~~~~~~~~~~~~~
parser.c:737:11: warning: Access to field 'versions' results in a dereference of
      a null pointer (loaded from field 'tag')
         (node->tag->versions & VERS_PROPRIETARY) != 0 )
          ^~~~~~~~~~~~~~~~~~~
parser.c:794:23: warning: Access to field 'parent' results in a dereference of a
      null pointer (loaded from variable 'element')
            element = element->parent;
                      ^~~~~~~~~~~~~~~
parser.c:1424:5: warning: Access to field 'element' results in a dereference of
      a null pointer (loaded from variable 'node')
                        (node->element != NULL &&
                         ^~~~~~~~~~~~~
parser.c:2105:15: warning: Access to field 'model' results in a dereference of a
      null pointer (loaded from field 'tag')
        if (!(node->tag->model & CM_INLINE) &&
              ^~~~~~~~~~~~~~~~
parser.c:2180:11: warning: Access to field 'model' results in a dereference of a
      null pointer (loaded from field 'tag')
    if (!(element->tag->model & CM_OPT))
          ^~~~~~~~~~~~~~~~~~~
parser.c:2322:17: warning: Access to field 'last' results in a dereference of a
      null pointer (loaded from variable 'parent')
            if (parent->last == node)
                ^~~~~~~~~~~~
parser.c:2809:13: warning: Access to field 'model' results in a dereference of a
      null pointer (loaded from field 'tag')
        if (node->tag->model & CM_ROWGRP)
            ^~~~~~~~~~~~~~~~
parser.c:3398:26: warning: Access to field 'tag' results in a dereference of a
      null pointer (loaded from variable 'field')
        if (node->tag == field->tag && node->type == EndTag)
                         ^~~~~~~~~~
parser.c:3447:11: warning: Access to field 'model' results in a dereference of a
      null pointer (loaded from field 'tag')
    if (!(field->tag->model & CM_OPT))
          ^~~~~~~~~~~~~~~~~
parser.c:3447:11: warning: Access to field 'tag' results in a dereference of a
      null pointer (loaded from variable 'field')
    if (!(field->tag->model & CM_OPT))
          ^~~~~~~~~~
parser.c:4215:40: warning: Access to field 'model' results in a dereference of a
      null pointer (loaded from field 'tag')
        if (node->type == StartTag && (node->tag->model & CM_FRAMES))
                                       ^~~~~~~~~~~~~~~~
parser.c:4222:48: warning: Access to field 'model' results in a dereference of a
      null pointer (loaded from field 'tag')
        else if (node->type == StartEndTag && (node->tag->model & CM_FRAMES))
                                               ^~~~~~~~~~~~~~~~
16 warnings generated.
pprint.c:1214:11: warning: Value stored to 'indAttrs' is never read
          indAttrs = no;
          ^          ~~
pprint.c:1845:8: warning: Access to field 'parent' results in a dereference of a
      null pointer (loaded from variable 'node')
  if ( node->parent != NULL )
       ^~~~~~~~~~~~
pprint.c:1998:9: warning: Value stored to 'contentIndent' is never read
        contentIndent = 0;
        ^               ~
3 warnings generated.
streamio.c:689:13: warning: Value stored to 'c' is never read
            c = 0;
            ^   ~
streamio.c:699:17: warning: Value stored to 'c' is never read
                c = 0;
                ^   ~
2 warnings generated.
tmbstr.c:154:22: warning: Dereference of null pointer
      (loaded from variable 's1')
    while (c = (uint)(*s1), TY_(ToLower)(c) == TY_(ToLower)((uint)(*s2)))
                     ^~~~~
1 warning generated.
utf8.c:306:33: warning: Assigned value is garbage or undefined
                        theByte = buf[tempCount - 1];
                                ^ ~~~~~~~~~~~~~~~~~~
1 warning generated.

"Value stored ... is never read" warnings appear to
be correct.  "Dereference of null pointer" warnings
don't mean that the pointer in question is certain
to be NULL, only that clang cannot see anything that
ensures that it *isn't* NULL.  You can shut clang up
about these warnings by adding assert(ptr != NULL),
with the benefit that if the pointer *is* NULL you will
get an assertion failure which is somewhat more helpful
than a crash (Darwin not producing core dumps).
The wornings in streamio.c appear to be correct.

I *think* the warning in utf8.c is mistaken, but I've
been wrong about that kind of thing before.