# #################################################################
# Copyright (c) 2009 Nokia Corporation and/or its subsidiary(-ies).
# All rights reserved.
# 
# Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
# 
# * Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
# * Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
# * Neither the name of Nokia Corporation nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
# 
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, 
# THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS 
# BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 
# SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 
# CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.#
#
# #################################################################

Nokia CodeScanner Command Line Tool Release Notes
=================================================

Version 2.1.4
=============
- Copyright changes done.

Version 2.1.3
=============
- Updated the function parser in CodeScanner to improve check for
leaving functions called in non-leaving functions. There should
be less false positives now.


Version 2.1.2
=============
- Added new IAD rule to check flag usage. IA must not use R&D flags 
nor feature flags via preprocessor statements in the source code. 
This means bldvariant.hrh and productvariant.hrh should not be used.
This new rule is set to low severity in the "Other" category by 
default. It can be configured by adding a <flags> script element 
in the configuration file. 
Example :
<codescannerConfig>
  <scripts>
    <flags category="other" 
           enable="true" 
           severity="low"/>
  </scripts>
</codescannerConfig>

- Added new IAD rule to check Central Repository usage. IA cannot 
assume that the Central Repository is set up fully. This means the 
return value of CRepository::get() cannot be ignored.
This new rule is set to low severity in the "Other" category by 
default. It can be configured by adding a <crepository> script 
element in the configuration file. 
Example :
<codescannerConfig>
  <scripts>
    <crepository category="other" 
                 enable="true" 
                 severity="low"/>
  </scripts>
</codescannerConfig>

- Added new IAD rule to check customizable icons. Due to device 
customization requirements, IA must not remove any customization 
done by the variant team. This means IA cannot include customizable 
icons.
This new rule is set to low severity in the "Other" category by 
default. It can be configured by adding a <customizableicons> 
script element in the configuration file. Names of icons can be 
specified in the <iconsRE> element.
Example :
<codescannerConfig>
  <scripts>
    <customizableicons category="other" 
                       enable="true" 
                       severity="low">
      <iconsRE>iconA.bmp|iconB.svg|iconC</iconsRE>
    </customizableicons>
  </scripts>
</codescannerConfig>


Version 2.1.1
=============
- Added support for user-defined rules. This can be done by
adding a <customrules> element to the configuration file.
Example :
<codescannerConfig>
  <customrules>
    <customrule>
      <name>myOwnRule</name>
      <keyword type=”class”>CMyOwnClass</keyword>
      <filetype>h</filetype>
      <severity>low</severity>
      <title>My CodeScanner rule targeting CMyOwnClass</title>
      <description>Locate definition of CMyOwnClass</description>
    </customrule>
  </customrules>
</codescannerConfig>
More information on the syntax of this new element can be found in 
the document "CodeScanner_XML_ConfigFile_Spec.doc".

- When scanning with low severity rules, CodeScanner now 
generates informative messages instead of warnings when the 
StdOut output format is selected.

- Added support to disable individual error/warning via
CodeScanner commands embedded in comments. The command has the 
following format : 
  codescanner(::<rule name>)+
Example :
void CTestCase::Foo()
{
  array->At(a); // codescanner::accessArrayElementWithoutCheck
  User::Leave(KErrNone); /* codescanner::LeaveNoError */
}
A complete list of CodeScanner rule names can be found in the 
document "CodeScanner_Rules.doc".

- Added new rule to check whether Cancel() function is called 
in active object's destructor.
Cancel() should always be called in active object's destructor 
to cancel an outstanding request if there is one. If there is 
no request pending then Cancel() just does nothing, but if we 
do not call Cancel() when having an outstanding request a panic 
will be raised.
This new rule is set to low severity in the "Coding Standards"
category by default. It can be configured by adding a 
<missingcancel> script element in the configuration
file. 
Example :
<codescannerConfig>
  <scripts>
    <missingcancel category="codingstandards" 
                   enable="true" 
                   severity="low"/>
  </scripts>
</codescannerConfig>

-Updated the function parser in CodeScanner to filter out constant 
initialization.

- Updated active object checking rule to ignore After() from 
RTimer type member variables.

- Updated to skip excluded folders specified in configuration file 
when generating component summary reports.


Version 2.1.0
=============
- Previously CodeScanner has been distributed as a single-file
executable which packs all the DLLs and library modules it needed 
into an archive attached to itself. During runtime, it creates a 
directory (_MEI<id>) in the user's temp directory, and extract 
these files there. When execution is complete, it recursively 
removes the entire directory it created. 
Unfortunately the cleanup process sometimes fail. So over time 
user's temp directory can be littered with these directories, 
which can cause CodeScanner to fail if it tries to reuse a 
directory that was created from a previous run.
To address this problem, we separated all these modules from 
CodeScanner.exe and shipping them "as is". This change should be
transparent from user's point of view, except when copying
CodeScanner to other locations on user's system.

- Updated configuration file format to allow passing arguments 
to CodeScanner. One can now add an <arguments> element to the 
configuration file with one or more of the following 
child elements :
<input> - specify additional directory/file to scan
<outputformat> (html|xml|std) - specify output format
<lxr> - url to lxr site
<lxrversion> - lxr version
<timestampedoutput> (on/off) - create time-stamped output directory 
Example:
<codescannerConfig>
  <arguments>
    <input>C:\Testing\CodeScanner_Tests\Bug_6393\Includes</input>
    <outputformat>html|xml</outputformat>
    <lxr>http://s60lxr/source/</lxr>
    <lxrversion>S60_3_2_200736</lxrversion>
    <timestampedoutput>on</timestampedoutput>
  </arguments>
</codescannerConfig>

- Added new rule to check stack-based resource objects not put 
on the cleanup stack.
If a stack-based resource object is not put on the cleanup stack 
with CleanupClosePushL() or CleanupResetAndDestroyPushL(), and 
a leaving function or ELeave is called, a memory leak occurs. 
CodeScanner occasionally gives false positives for this issue. 
Individual cases should be investigated.
This new rule is set to low severity in the "Code Review"
category by default. It can be configured by adding a 
<resourcenotoncleanupstack> script element in the configuration
file. 
Example :
<codescannerConfig>
  <scripts>
    <resourcenotoncleanupstack category="codereview" 
                               enable="true" 
                               severity="low"/>
  </scripts>
</codescannerConfig>

- Added new rule to check inheritance order of M and C classes. 
If a C class inherits first from an M class and then a C class,
a panic can occur when trying to pop a CBase pointer pointing 
to such a class from the cleanup stack when in fact a pointer 
pointing to the first predecessor, the mixin class, was popped 
instead.
This new rule is set to high severity in the "Can Panic"
category by default. It can be configured by adding a 
<inheritanceorder> script element in the configuration file. 
Example :
<codescannerConfig>
  <scripts>
    <inheritanceorder category="canpanic" 
                      enable="true" 
                      severity="high"/>
  </scripts>
</codescannerConfig>

- Updated rule for L-functions that cannot leave. It is now
possible to specify functions to be ignored by this rule in the
configuration file. This is done by adding a <LFunctionIgnoreRE> 
element to the <LFunctionCantLeave> script element.
Example:
<codescannerConfig>
  <scripts>
    <LFunctionCantLeave enable="true">
      <LFunctionIgnoreRE>RunL</LFunctionIgnoreRE>
    </LFunctionCantLeave>
  </scripts>
</codescannerConfig>

- Updated description of rule for badly-named enum members.

- Updated NULL equality check rule to ignore pointer comparisons 
with NULL inside various __ASSERT_XXX() macros.

- Removed duplicate rule for hard-coded external drive letters. 
There is now only 1 rule that check for this problem and it is 
a low severity rule in the Coding Standard category.

- Improved rule for accessing array element by [] without 
checking range.


Version 2.0.9
=============
- Updated check for Open() to allow assignment of return value on 
a different line.
Example :
  TInt error = 
    myFile.Open();


Version 2.0.8
=============
- Updated check for C class not inheriting from another C class.
It is now possible to specify C class that does not inherit from
another C class in the configuration file. This is done by adding
a <cclassIgnoreRE> element to the <missingcclass> script element.
Example :
<missingcclass enable="true">
  <cclassIgnoreRE>CLogger|CKeypadData</cclassIgnoreRE>
</missingcclass>

- Updated check for ignored Open() return value.
It is now possible to specify object or class with Open() method
that does not return any value in the configuration file. 
This is done by adding an <openIgnoreRE> element to the 
<open> script element. For a local variable, one can specify 
either the type or variable name in the <openIgnoreRE> element. 
For other types of object, one should specify the object name in 
the <openIgnoreRE> element. 
Example:
<open enable="true">
  <openIgnoreRE>CKeyStore|iKeyStore|lKeyStore</openIgnoreRE>
</open>

- Fixed a bug where some scripts cannot be disabled from the
configuration file.


Version 2.0.7
=============
- Updated standard console output to include severity level and
category information in the error/warning messages.

- Incorporated Psyco module to improve scanning performance. Please
refer to the document "MIT license.txt" for licensing info.


Version 2.0.6
=============
- Updated check for CleanupStack::Pop(AndDestroy)() to ignore
CleanupStack::Pop(TInt, TAny*) and 
CleanupStack::PopAndDestroy(TInt, TAny*)

- Updated check for Open() to allow handler of return value on a 
different line.
Example :
  User::LeaveIfError(
    myFile.Open());

- Updated check for At() to ignore timer objects.
Example :
  TTime now;
  iTimer.At(iStatus, now);

- Improved handling of scanning of code with #ifdef/#else/#endif
preprocessor commands. CodeScanner sometimes can be confused by
code surrounded by these preprocessor commands, e.g. generating
warnings on member variable deleted incorrectly, even when it 
happens inside a destructor. This has been fixed.


Version 2.0.5
=============
- Added new option to specify additional source directory/file :
"-i <source dir/file>" -- specify additional directory/file to scan
Example : codescanner -i C:\resources -i C:\includes
          C:\sources C:\results


Version 2.0.4
=============
- Updated HTML generation to use '/' instead of '\' when generating
links; this change is needed for Mozilla Firefox compatibility.

- Added new option to toggle creation of time-stamped directory for
results :
"-t on/off" -- create a time-stamped directory for results; default
               is on
Example : codescanner -t off C:\sources C:\results

- Added new option to configure output format of results:
"-o html|xml|std" -- specify output format : HTML, XML or StdOut;
                     default output format is HTML
Multiple output formats can be specified at the same time.
Example : codescanner -o xml,html C:\sources C:\results

- Updated configuration file format to allow user setting the
category and severity of scripts individually. Each <scripts> element
contains one or more Script elements, each of which corresponds to a
CodeScanner rule. Each Script element has the following attributes:
enable (boolean) – specifies whether a script is enabled by CodeScanner.
severity (string) – specifies the severity level of a script.
category (string) – specifies the category of a script.
Example :
<codescannerConfig>
  <scripts>
    <baddefines enable="true" severity="low" category="codingstandards"/>
    <friend enable="true" severity="low" category="codereview"/>
    <leave enable="true" severity="high" category="canpanic"/>
  </scripts>
</codescannerConfig>


Version 2.0.3
=============
- Updated HTML generation so that links with the '#' character can be
opened by Mozilla Firefox correctly.

- Added new options to generate links to LXR server instead of local 
file system. The new options are :
"-x" -- specify URL of LXR site
"-r" -- specify LXR version

Example : codescanner -x http://some.lxr.server/sources/ -r S60_3_2_12
C:\sources C:\results

Once scanning is completed, the results will point to the remote 
LXR server instead of the local "C:\sources" folder.
*****This new feature has been contributed by James Uther of Nokia*****


Version 2.0.2
=============
- Updated HTML generation so that source file links can be opened by
Mozilla Firefox.

- Improved handling of function parameter list spanning multiple lines.
Example :
void CMenuBar::MyMenu (const HBufC* a,
                       const TDesC& b)
{ ... }

- Improved handling of TRAP block spanning multiple lines.
Example :
  TRAPD(
    err,
    WaitDialogL(aErrorCode)
  );

- Updated handling of NULL equality check to allow comparison with NULL
in return statement. 
Example :
  return (ptr != NULL);

- Updated handling of hard-coded numbers so that '2' is no longer 
considered a 'magic number'.


Version 2.0.1b
==============
- When checking for leaving functions called in non-leaving functions,
Cases where macros are used for class declarations are now filtered out.
Example :
NONSHARABLE_CLASS(CMyClass) : public CActive 
{ ... }

- When generating individual component report, text containing HTML tags
(eg. <s>, <i>) are now cleaned up before written to file. This
prevents unintended text formatting.

- Corrected spelling mistake in usage info.


Version 2.0.0a
==============
- This is the first development release of Nokia's version of CodeScanner CLI. 
Our goals for this release are to :
1) support feature-functions available in Mobile Innovation's CodeScanner v 1.11
2) address reported issues

The following are known issues from CodeScanner v 1.11 :
1) CodeScanner output is not entirely compatible with Mozilla Firefox. 
For security reasons, Mozila Firefox and Mozilla Suite do not allow links to 
local files (and directories). This includes linking to files on the hard drive, 
on mapped network drives, and those accessible via UNC paths.
2) CodeScanner treats 2 as a ‘magic number’. In User Interface code, '2' is 
often used to calculate offsets and should not always be replaced by variable.
3) CodeScanner does not ignore CleanupStack::Pop() without parameters 
if CleanupClosePushL() or similar utility is used.
4) CodeScanner flags all use of #if in .h files irrespective of context.

Licensing Notes
===============
CodeScanner is built with Python 2.5.1 but does NOT use any of the following
files or modules :
python2.5-2.5.1/Lib/test/decimaltestdata/abs.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/add.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/base.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/clamp.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/compare.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/decimal128.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/decimal32.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/decimal64.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/divide.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/divideint.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/inexact.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/max.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/min.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/minus.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/multiply.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/normalize.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/plus.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/power.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/quantize.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/randomBound32.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/randoms.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/remainder.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/remainderNear.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/rescale.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/rounding.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/samequantum.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/squareroot.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/subtract.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/testall.decTest 
python2.5-2.5.1/Lib/test/decimaltestdata/tointegral.decTest 
/Demo/curses/tclock.py python2.5-2.5.1/Modules/expat/ascii.h 
python2.5-2.5.1/Modules/expat/asciitab.h 
python2.5-2.5.1/Modules/expat/expat.h 
python2.5-2.5.1/Modules/expat/expat_external.h 
python2.5-2.5.1/Modules/expat/iasciitab.h 
python2.5-2.5.1/Modules/expat/latin1tab.h 
python2.5-2.5.1/Modules/expat/utf8tab.h 
python2.5-2.5.1/Modules/expat/xmlparse.c 
python2.5-2.5.1/Modules/expat/xmlrole.c 
python2.5-2.5.1/Modules/expat/xmlrole.h 
python2.5-2.5.1/Modules/expat/xmltok.c 
python2.5-2.5.1/Modules/expat/xmltok.h 
python2.5-2.5.1/Modules/expat/xmltok_impl.c 
python2.5-2.5.1/Modules/expat/xmltok_impl.h 
python2.5-2.5.1/Doc/html/about.dat 
python2.5-2.5.1/Doc/html/stdabout.dat 
python2.5-2.5.1/Modules/expat/pyexpatns.h 
python2.5-2.5.1/Lib/robotparser.py 
python2.5-2.5.1/Lib/ctypes/macholib/README.ctypes 
python2.5-2.5.1/Lib/wsgiref.egg-info

Please refer to the document "python 2_5_1 licenses.txt" for licensing info.