diff options
Diffstat (limited to '2017/ttcn-netdev2.2/paper')
| -rw-r--r-- | 2017/ttcn-netdev2.2/paper/isea.bbl | 20 | ||||
| -rw-r--r-- | 2017/ttcn-netdev2.2/paper/isea.sty | 301 | ||||
| -rw-r--r-- | 2017/ttcn-netdev2.2/paper/isea.tex | 445 |
3 files changed, 766 insertions, 0 deletions
diff --git a/2017/ttcn-netdev2.2/paper/isea.bbl b/2017/ttcn-netdev2.2/paper/isea.bbl new file mode 100644 index 0000000..5233a5a --- /dev/null +++ b/2017/ttcn-netdev2.2/paper/isea.bbl @@ -0,0 +1,20 @@ +\begin{thebibliography}{} + +\bibitem[\protect\citeauthoryear{Unknown}{2013}]{ETSI} +\newblock TTCN-3 Tutorial +\newblock ETSI Centre for Testing and Interoperability +\newblock \url{http://www.ttcn-3.org/files/ETSI\_TTCN3\_Tutorial.pdf} + +\bibitem[\protect\citeauthoryear{Unknown}{2002-2015}]{Ericsson} +\newblock TTCN-3 Course Presentation Material +\newblock Ericsson Test Solutions and Competence Center +\newblock \url{http://www.ttcn-3.org/files/TTCN3\_P.pdf} + +\bibitem[\protect\citeauthoryear{Unknown}{2002-2015}]{ETSI} +\newblock ETSI svn repository +\newblock \url{http://oldforge.etsi.org/websvn/} + + + + +\end{thebibliography} diff --git a/2017/ttcn-netdev2.2/paper/isea.sty b/2017/ttcn-netdev2.2/paper/isea.sty new file mode 100644 index 0000000..4658cf4 --- /dev/null +++ b/2017/ttcn-netdev2.2/paper/isea.sty @@ -0,0 +1,301 @@ +%Filename: isea.sty +% +\typeout{Conference Style for ISEA 2015 for LaTeX 2e -- version of 2nd, December 2014} +% Modified by B. Bogart and based on the ICCC.sty style file. The original ICCC notice follows +% +% NOTICE: DO NOT MODIFY THIS FILE WITHOUT CHANGING ITS NAME. This style +% file is called iccc.sty. It is lifted from the aaai.sty file used by the AAAI +% and consists of minimal modification to allow its use for ICCC. These modifications +% were done by Dan Ventura in September 2010. The original AAAI notice follows. +% +% NOTICE: DO NOT MODIFY THIS FILE WITHOUT CHANGING ITS NAME. This style +% file is called aaai.sty. Modifications to this file are permitted, +% provided that your modified version does not include the acronym "aaai" +% in its name, that credit to the authors and supporting agencies is +% retained, and that further modification or reuse is not restricted. This +% file was originally prepared by Peter F. Patel-Schneider, liberally +% using the ideas of other style hackers, including Barbara Beeton. It was +% modified in April 1999 by J. Scott Penberthy and George Ferguson. It was +% modified in 2007 by AAAI. It was modified in February 2009 +% and in November 2009 by Hans W. Guesgen and Giuseppe De Giacomo. It +% was further modified in March 2010 by AAAI. +% The original preparation of this file was supported by +% Schlumberger Palo Alto Research, AT\&T Bell Laboratories, AAAI, and +% Morgan Kaufmann Publishers. +% +% WARNING: This style is NOT guaranteed to work. It is provided in the +% hope that it might make the preparation of papers easier, but this style +% file is provided "as is" without warranty of any kind, either express or +% implied, including but not limited to the implied warranties of +% merchantability, fitness for a particular purpose, or noninfringement. +% You use this style file at your own risk. Standard disclaimers apply. +% +% Do not use this file unless you are an experienced LaTeX user. To +% satisfy AAAI's requirements, you must change your paper's configuration +% to use Times fonts. AAAI will not accept your paper if it is formatted +% using obsolete type 3 Computer Modern bitmapped fonts. Please ensure +% that your version of dvips maps to type 1 fonts. Place this document in +% a file called aaai.sty in the TeX search path. (Placing it in the same +% directory as the paper should also work.) +% +% You must also format your paper for US letter-sized paper. +% +% There are undoubtably bugs in this style. If you would like to submit +% bug fixes, improvements, etc. please let us know. Please use the contact form +% at www.aaai.org. +% +% \documentclass[letterpaper]{article} +% \usepackage{aaai} +% \usepackage{times} +% \usepackage{helvet} +% \usepackage{courier} +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% +% PDFMARK for TeX and GhostScript +% Uncomment and complete the following for metadata if +% your paper is typeset using TeX and GhostScript (e.g +% if you use .ps or .eps files in your paper): +% \special{! /pdfmark where +% {pop} {userdict /pdfmark /cleartomark load put} ifelse +% [ /Author (John Doe, Jane Doe) +% /Title (Input Your Paper Title Here) +% /Subject (Input the Proceedings Title Here) +% /Keywords (AAAI, artificial intelligence) +% /DOCINFO pdfmark} +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% +% PDFINFO for PDFTeX +% Uncomment and complete the following for metadata if +% your paper is typeset using PDFTeX +% \pdfinfo{ +% /Title (Input Your Paper Title Here) +% /Subject (Input the Proceedings Title Here) +% /Author (John Doe, Jane Doe) +% } +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% +% Uncomment if you want to use section numbers +% and change the 0 to a 1 or 2 +% \setcounter{secnumdepth}{0} +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% +% \title{Title} +% \author{Author 1 \and Author 2 \\ Address line \\ Address line \And +% Author 3 \\ Address line \\ Address line} +% \begin{document} +% \maketitle +% ... +% \bibliography{Bibliography-File} +% \bibliographystyle{aaai} +% \end{document} +% \pubnote{\em To appear, AAAI-10} % optional, remove for submission +% +% \pubnote is for printing the paper yourself, and should not be used in +% submitted versions!!!! +% Author information can be set in various styles: +% For several authors from the same institution: +% \author{Author 1 \and ... \and Author n \\ +% Address line \\ ... \\ Address line} +% if the names do not fit well on one line use +% \author{Author 1 \\ {\bf Author 2} \\ ... \\ {\bf Author n} \\ +% Address line \\ ... \\ Address line} +% For authors from different institutions: +% \author{Author 1 \\ Address line \\ ... \\ Address line +% \And ... \And +% Author n \\ Address line \\ ... \\ Address line} +% To start a separate ``row'' of authors use \AND, as in +% \author{Author 1 \\ Address line \\ ... \\ Address line +% \AND +% Author 2 \\ Address line \\ ... \\ Address line \And +% Author 3 \\ Address line \\ ... \\ Address line} +% If the title and author information does not fit in the area allocated, +% place \setlength\titlebox{height} +% after the \documentstyle line +% where {height} is something like 2.5in +% PHYSICAL PAGE LAYOUT +\setlength\topmargin{-0.25in} \setlength\oddsidemargin{-0.25in} +\setlength\textheight{9.0in} \setlength\textwidth{7.0in} +\setlength\columnsep{0.25in} \newlength\titlebox \setlength\titlebox{2.25in} +\setlength\headheight{0pt} \setlength\headsep{0pt} +%\setlength\footheight{0pt} \setlength\footskip{0pt} +\thispagestyle{empty} \pagestyle{empty} +\flushbottom \twocolumn \sloppy +% We're never going to need a table of contents, so just flush it to +% save space --- suggested by drstrip@sandia-2 +\def\addcontentsline#1#2#3{} +% gf: PRINT COPYRIGHT NOTICE +%\def\copyright@year{\number\year} +%\def\copyright@text{Copyright \copyright\space \copyright@year, +%Association for the Advancement of Artificial Intelligence (www.aaai.org). +%All rights reserved.} +%\def\copyright@on{T} +%\def\nocopyright{\gdef\copyright@on{}} +%\def\copyrighttext#1{\gdef\copyright@on{T}\gdef\copyright@text{#1}} +%\def\copyrightyear#1{\gdef\copyright@on{T}\gdef\copyright@year{#1}} +% gf: End changes for copyright notice (used in \maketitle, below) +% Title stuff, taken from deproc. +\def\maketitle{\par +\begingroup % to make the footnote style local to the title +\def\thefootnote{\fnsymbol{footnote}} +% gf: Don't see why we'd want the footnotemark to be 0pt wide +%\def\@makefnmark{\hbox to 0pt{$^{\@thefnmark}$\hss}} +\twocolumn[\@maketitle] \@thanks +\endgroup +% gf: Insert copyright slug unless turned off +%\if T\copyright@on\insert\footins{\noindent\footnotesize\copyright@text}\fi +% gf: And now back to your regular programming +\setcounter{footnote}{0} +\let\maketitle\relax \let\@maketitle\relax +\gdef\@thanks{}\gdef\@author{}\gdef\@title{}\let\thanks\relax} +\def\@maketitle{\vbox to \titlebox{\hsize\textwidth +%%% AAAI changed: 03/05/2010 +%%\linewidth\hsize \vskip 0.625in minus 0.125in \centering +\linewidth\hsize \vskip 0.625in minus 0.125in \centering +%%% END changed +{\LARGE\bf \@title \par} \vskip 0.2in plus 1fil minus 0.1in +{\def\and{\unskip\enspace{\rm and}\enspace}% +\def\And{\end{tabular}\hss \egroup \hskip 1in plus 2fil + \hbox to 0pt\bgroup\hss \begin{tabular}[t]{c}\Large\bf}% +\def\AND{\end{tabular}\hss\egroup \hfil\hfil\egroup +\vskip 0.25in plus 1fil minus 0.125in +% hg: Changed Large to normalsize on next line + \hbox to \linewidth\bgroup\normalsize \hfil\hfil +\hbox to 0pt\bgroup\hss \begin{tabular}[t]{c}\Large\bf} +\hbox to \linewidth\bgroup\normalsize \hfil\hfil +\hbox to 0pt\bgroup\hss \begin{tabular}[t]{c}\Large\bf\@author +\end{tabular}\hss\egroup +\hfil\hfil\egroup} +\vskip 0.3in plus 2fil minus 0.1in +}} +\renewenvironment{abstract}{\centerline{\bf +Abstract}\vspace{0.5ex}\begin{quote}\small}{\par\end{quote}\vskip 1ex} +% jsp added: +\def\pubnote#1{\thispagestyle{myheadings} +\pagestyle{myheadings} +\markboth{#1}{#1} +\setlength\headheight{10pt} \setlength\headsep{10pt} +} +% SECTIONS with less space +\def\section{\@startsection {section}{1}{\z@}{-2.0ex plus +-0.5ex minus -.2ex}{3pt plus 2pt minus 1pt}{\Large\bf\centering}} +\def\subsection{\@startsection{subsection}{2}{\z@}{-2.0ex plus +-0.5ex minus -.2ex}{3pt plus 2pt minus 1pt}{\large\bf\raggedright}} +\def\subsubsection{\@startsection{subparagraph}{3}{\z@}{-6pt plus +%%% DIEGO changed: 29/11/2009 +%% 2pt minus 1pt}{-1em}{\normalsize\bf}} +-2pt minus -1pt}{-1em}{\normalsize\bf}} +%%% END changed +\setcounter{secnumdepth}{0} +% add period to section (but not subsection) numbers, reduce space after +%\renewcommand{\thesection} +% {\arabic{section}.\hskip-0.6em} +%\renewcommand{\thesubsection} +% {\arabic{section}.\arabic{subsection}\hskip-0.6em} +% FOOTNOTES +\footnotesep 6.65pt % +\skip\footins 9pt plus 4pt minus 2pt +\def\footnoterule{\kern-3pt \hrule width 5pc \kern 2.6pt } +\setcounter{footnote}{0} +% LISTS AND PARAGRAPHS +\parindent 10pt +\topsep 4pt plus 1pt minus 2pt +\partopsep 1pt plus 0.5pt minus 0.5pt +\itemsep 2pt plus 1pt minus 0.5pt +\parsep 2pt plus 1pt minus 0.5pt +\leftmargin 10pt \leftmargini\leftmargin \leftmarginii 10pt +\leftmarginiii 5pt \leftmarginiv 5pt \leftmarginv 5pt \leftmarginvi 5pt +\labelwidth\leftmargini\advance\labelwidth-\labelsep \labelsep 5pt +\def\@listi{\leftmargin\leftmargini} +\def\@listii{\leftmargin\leftmarginii +\labelwidth\leftmarginii\advance\labelwidth-\labelsep +\topsep 2pt plus 1pt minus 0.5pt +\parsep 1pt plus 0.5pt minus 0.5pt +\itemsep \parsep} +\def\@listiii{\leftmargin\leftmarginiii +\labelwidth\leftmarginiii\advance\labelwidth-\labelsep +\topsep 1pt plus 0.5pt minus 0.5pt +\parsep \z@ \partopsep 0.5pt plus 0pt minus 0.5pt +\itemsep \topsep} +\def\@listiv{\leftmargin\leftmarginiv +\labelwidth\leftmarginiv\advance\labelwidth-\labelsep} +\def\@listv{\leftmargin\leftmarginv +\labelwidth\leftmarginv\advance\labelwidth-\labelsep} +\def\@listvi{\leftmargin\leftmarginvi +\labelwidth\leftmarginvi\advance\labelwidth-\labelsep} +\abovedisplayskip 7pt plus2pt minus5pt% +\belowdisplayskip \abovedisplayskip +\abovedisplayshortskip 0pt plus3pt% +\belowdisplayshortskip 4pt plus3pt minus3pt% +% Less leading in most fonts (due to the narrow columns) +% The choices were between 1-pt and 1.5-pt leading +\def\normalsize{\@setfontsize\normalsize\@xpt{11}} % 10 point on 11 +\def\small{\@setfontsize\small\@ixpt{10}} % 9 point on 10 +\def\footnotesize{\@setfontsize\footnotesize\@ixpt{10}} % 9 point on 10 +\def\scriptsize{\@setfontsize\scriptsize\@viipt{10}} % 7 point on 8 +\def\tiny{\@setfontsize\tiny\@vipt{7}} % 6 point on 7 +\def\large{\@setfontsize\large\@xipt{12}} % 11 point on 12 +\def\Large{\@setfontsize\Large\@xiipt{14}} % 12 point on 14 +\def\LARGE{\@setfontsize\LARGE\@xivpt{16}} % 14 point on 16 +\def\huge{\@setfontsize\huge\@xviipt{20}} % 17 point on 20 +\def\Huge{\@setfontsize\Huge\@xxpt{23}} % 20 point on 23 +%%%% named style for aaai, included here for ease of use +% This section implements citations for the ``named'' bibliography style, +% modified for AAAI use. +% This file can be modified and used in other conferences as long +% as credit to the authors and supporting agencies is retained, this notice +% is not changed, and further modification or reuse is not restricted. +% The ``named'' bibliography style creates citations with labels like +% \citeauthoryear{author-info}{year} +% these labels are processed by the following commands: +% \cite{keylist} +% which produces citations with both author and year, +% enclosed in square brackets +% \shortcite{keylist} +% which produces citations with year only, +% enclosed in square brackets +% \citeauthor{key} +% which produces the author information only +% \citeyear{key} +% which produces the year information only +\def\leftcite{(}\def\rightcite{)} +\def\cite{\def\citeauthoryear##1##2{\def\@thisauthor{##1}% +\ifx \@lastauthor \@thisauthor \relax \else##1 \fi ##2}\@icite} +\def\shortcite{\def\citeauthoryear##1##2{##2}\@icite} +\def\citeauthor{\def\citeauthoryear##1##2{##1}\@nbcite} +\def\citeyear{\def\citeauthoryear##1##2{##2}\@nbcite} +% internal macro for citations with () and with breaks between citations +% used in \cite and \shortcite +\def\@icite{\leavevmode\def\@citeseppen{-1000}% +\def\@cite##1##2{\leftcite\nobreak\hskip 0in{##1\if@tempswa , ##2\fi}\rightcite}% +\@ifnextchar [{\@tempswatrue\@citex}{\@tempswafalse\@citex[]}} +% internal macro for citations without [] and with no breaks +% used in \citeauthor and \citeyear +\def\@nbcite{\leavevmode\def\@citeseppen{1000}% +\def\@cite##1##2{{##1\if@tempswa , ##2\fi}}% +\@ifnextchar [{\@tempswatrue\@citex}{\@tempswafalse\@citex[]}} +% don't box citations, separate with ; and a space +% also, make the penalty between citations a parameter, +% it may be a good place to break +\def\@citex[#1]#2{% +\def\@lastauthor{}\def\@citea{}% +\@cite{\@for\@citeb:=#2\do +{\@citea\def\@citea{;\penalty\@citeseppen\ }% +\if@filesw\immediate\write\@auxout{\string\citation{\@citeb}}\fi +\@ifundefined{b@\@citeb}{\def\@thisauthor{}{\bf ?}\@warning +{Citation `\@citeb' on page \thepage \space undefined}}% +{\csname b@\@citeb\endcsname}\let\@lastauthor\@thisauthor}}{#1}} +%Ignore the key when generating the Reference section. +\def\@lbibitem[#1]#2{\item\if@filesw +{ \def\protect##1{\string ##1\space}\immediate +\write\@auxout{\string\bibcite{#2}{#1}}}\fi\ignorespaces} +\def\thebibliography#1{\section*{References\@mkboth +{REFERENCES}{REFERENCES}}\list +{}{\labelwidth 0in\leftmargin\labelwidth +%%% DIEGO removed +%%\advance\leftmargin\labelsep +%%% END removed +%%% DIEGO changed +\itemsep .01in % original +%%\itemsep -.0125in % reduced space between bib entries +%%% END changed +} +\def\newblock{\hskip .11em plus .33em minus .07em} +\sloppy\clubpenalty4000\widowpenalty4000 +\sfcode`\.=1000\relax} +\let\endthebibliography=\endlist diff --git a/2017/ttcn-netdev2.2/paper/isea.tex b/2017/ttcn-netdev2.2/paper/isea.tex new file mode 100644 index 0000000..d329f49 --- /dev/null +++ b/2017/ttcn-netdev2.2/paper/isea.tex @@ -0,0 +1,445 @@ +% This file is isea.tex. It contains the formatting instructions for and acts as a template for submissions to ISEA 2015. It is based on the ICCC formats and instructions. It uses the files isea.sty, isea.bst and isea.bib, the first two of which also borrow from AAAI IJCAI formats and instructions. +% Modified from ICCC.tex by B. Bogart + +\documentclass[letterpaper]{article} +\usepackage{isea} +\usepackage[pdftex]{graphicx} +\usepackage{times} +\usepackage{helvet} +\usepackage{courier} +\usepackage[numbers]{natbib} +\usepackage[colorlinks=true,linkcolor=blue,urlcolor=blue]{hyperref} +\usepackage{listings} +\lstset{ + basicstyle=\footnotesize, + frame=top, + frame=bottom +} +\pdfinfo{ +/Title (TTCN-3 and Eclipse TITAN for testing protocol stacks) +/Author (Harald Welte)} +% The file isea.sty is the style file for ISEA 2015 proceedings. +% +\title{TTCN-3 and Eclipse TITAN for testing protocol stacks [draft]} +\author{Harald Welte\\ +sysmocom - systems for mobile communications GmbH\\ +Berlin, Germany\\ +laforge@gnumonks.org\\ +\newline +\newline +} +\setcounter{secnumdepth}{0} + +\begin{document} +\maketitle +\begin{abstract} +Implementations of networking protocol stacks are in need of thorough testing in order to ensure not only +their security, but also their interoperability and compliance to relevant standards and +specifications. Implementing test suites for verification of implementations of TCP/IP communications +protocols in the IETF world has traditionally been done in any number of ways, by using general-purpose +programming languages such as C, C++, Java, Python and others. In the ITU/ETSI world, TTCN-3 has been +develpoed as a domain-specific language for the specific use case of writing protocol conformance tests. +This paper acts as an introduction into both TTCN-3 as well as an open source TTCN-3 compiler, Eclipse TITAN. +\end{abstract} + +\section{Keywords} + +TTCN-3, conformance testing, protocol testing, validation, TITAN + +\section{Introduction} + +\subsection{Protocol Testing} + +Testing network protocols is important for a variety of reasons, such as +\begin{itemize} +\item conformance to a specification +\item ensuring interoperability, including quirks for known-broken other implementations +\item network security +\end{itemize} + +The focus of this paper (as well as the focus of TTCN-3) is functional testing. It is not primarily +performance testing, though some people use TTCN-3 even in those scenarios. The fact that it's not a scripted +language or a VM but a compiled language executing native code helps with that. + +In the implementation of a functional test of a given communication protocol, there are typically repeating +patterns of certain building blocks, such as +\begin{itemize} +\item encoding and decoding messages between their wire format and some higher-level representation +\item matching received messages against templates to validate their contents or act depending on various + parts of their contents +\item waiting for any number of different events/conditions guarded by one or multiple timers +\end{itemize} + +TTCN-3 offers unique language capabilities to address related problems in a very productive and expressive +way. + +\subsection{The TTCN-3 Language} + +TTCN was originally called the {\em Tree and Tabular combined Notation}, but has since been +renamed to the {\em Test and Testing Control Notation}. It is a purpose-built language with +the sole purpose of implementing testing. + +\subsubsection{History} + +TTCN has been designed as an internationally standardized language purely for testing. Its origins go back to +1883, when ISO was working on Open Systems Interconnection (OSI) conformance testing methodology and +framework. It was first standardized as ISO 9646-3 in 1992. Within the ITU / OSI / ISO world, TTCN was +well-established and widely used in conformance tests for protocols like those used in ISDN. + +In 1997, STF 133 was formed by ETSI MTS to produce TTCN version 3 in co-operation with ITU-T SG10. Main +contributions from Nortel, Ericsson, Telelogic, Nokia. More than 200 members participated in the STF133 +discussion group, and the language was finally standardized in October 2000. + +Ever since, TTCN-3 has gained widespread adoption within the telecom industry. Several major telecom +equipment vendors such as Nokia and Ericsson have publicly disclosed some of the extensive in-house testing +they do using TTCN-3. A market of TTCN-3 compilers and tools has been well-established, but the proprietary +nature of those compilers and toolchains has put TTCN-3 out of reach from the Free Software community. + +Meanwhile, as the classic telecom sector became more and more involved with IP and Internet technologies, +conformance testing specifications / test suites for Internet centric protocols have been developed by telecom +standardization bodies. + +For example, ETSI has meanwhile developed and published test suites for IPv6, SIP, DIAMETER, electronic +passports, DMR (Digital Mobile Radio) and 6LoWPAN in TTCN-3. Those test suites are available via the ETSI +Subversion repository that can be browsed at \url{http://oldforge.etsi.org/websvn/} + +Also, TTCN-3 test suites have been specified by other bodies for technologies and protocols such as CoAP, +MQTT in the IoT area, or MOST and AUTOSAR in the automotove domain. + +\subsection{Eclipse TITAN} + +Around 2000, Ericsson internally started development of a TTCN-3 toolchain, which developed into a complete, +proven product that was adopted and is used extensively both inside Ericsson, as well as licensed to third +parties. Over the years, it has developed into approximately 300,000 lines of Java and 1,6 million lines of +C++ code, including extensive self-testing code for TITAN. + +TITAN includes not only a TTCN-3 compiler for translating TTCN-3 to C++ and the corresponding runtime +libraries, but also a number of other tools, such as a parallel executor for executing test components and +result reporting, a Makefile generator, log filtering, report generator, coverage analysis and much more. + +Being a commercially developed and supported tool, TITAN also includes a set of extensive user and developer +manuals. + +In 2015, Ericsson decided to transform TITAN into an open source project under the umbrella of the Eclipse +foundation. It is subsequently licensed under Eclipse Public License. Unlike other large software projects +developed as proprietary software and dumped to the community, this did not mark the end of Ericsson +involvement. To the contrary, the are very active in maintenance of the software ever since, with daily +commits to the public git repositories by the Ericsson team, regular releases and participation on forums and +mailing lists. + +While Eclipse Titan has some optional GUI components within the Eclipse IDE framework, such as the Titan +Designer and Titan LogViewer, neither the compiler nor the runtime libraries for executing the actual test +suites require those UI components. The entire toolchain can be used from the command line, driven from +classic Makefiles. + +Together with the TTCN-3 compiler and utilities, Ericsson has also been releasing an ever-growing list of +TTCN-3 source code implementations for a variety of protocols. This includes native TTCN-3 implementations of +IP, ICMP, IPv6, L2TP, GRE, HTTP, ICMP, RTP, SCTP, SDP, TCP and UDP. It also includes test ports for using +the underlaying operating system protocol stack(s), such as the IPL4asp for using the regular socket API +as provided by Linux+libc. + + +\section{Key TTCN-3 Language Features} + +TTCN-3 is a high-level, abstract language. The code itself is platform independent, as well as test +environment independent. In TTCN-3, you define only the abstract messages/signals as they are exchanged +between the test system and the tested entity. The transport layers and connections are provided and handled +by the tools Message encoding (serialization) and decoding (deserialization) is part of the tool/environment, +and not part of the test definition itself. + +The strong points of TTCN-3 for use in protocol tests are: +\begin{itemize} +\item A rich data/type system +\item Parametric templating and powerful template matching +\item Behavior specification using the {\em alt} and {\em default} behaviors +\item Separation of tests, test adapter and codec +\end{itemize} + + +\subsection{TTCN-3 data/type system} + +TTCN-3 is a strongly typed language. This provides the advantage that many type incompatibilities can already +be caught at the compile time, as opposed to weakly typed languages, where type errors are often only +discovered at runtime and are hence hard to catch, particularly in rarely used code such as error paths. + +\subsubsection{Basic types} + +The simple basic types of TTCN-3 include {\tt integer}, {\tt float}, {\tt boolean}. + +There {\tt verdicttype} is a special type for storing the preliminary and final verdicts of test execution. +It has five distinct values: {\tt none}, {\tt pass}, {\tt inconc}, {\tt fail}, {\tt error}. The useful +property is that a verdicttype varialbe can only get {\em worse}, but never better. So if any part of a test +case has ever set the verdict to {\tt fail} or {\tt error}, no follow-up assignments of the variable can ever +turn it into {\tt pass} anymore. This greatly simplifies control flow handling in erroneous conditions while +writing test cases. + +Basic strign types include {\tt bitstring}, {\tt hexstring}, {\tt octetstring}, {\tt charstring} (IA5) as well +as {\tt universal charstring} (UCS-4). + +\subsubsection{Structured Types} + +Using the {\tt record}, {\tt set}, {\tt union}, {\tt record of} and {\tt set of} TTCN-3 structured types, +programmers can create abstract container types. + +FIXME: example + +TTCN-3 also knows a {\tt enumberated} type, like many other programming languages. + +\subsubsection{Not-used and omit} + +Until any variable or field of a structured type has been assigned an explicit value, it is {\em +unbound}. Whenever a value is expected, and that value is unbound, the TTCN-3 runtime will create an error. +It's therefore not possible to e.g. accidentially encode unspecified data! + +If the programmer wishes to explicitly state that a structured type's optional field is not present, the +special value {\tt omit} may be used. + +\subsubsection{Sub-Typing} + +You can derive a new child type from an existing parent type by restricting the new type's domain to a subset +of the parent type's value domain. You can use various sub-typing constructs such as value range, value list, +length restriction and patterns, see Listing \ref{subtyping}. + +\begin{lstlisting}[caption={Example of TTCN-3 sub-typing},label=subtyping] +type integer MyIntRange (1 .. 100); +type integer MyIntRange8 (0 .. infinity); +type charstring MyCharRange ("k" .. "w"); + +type charstring SideType ("left", "right"); +type integer MyIntListRange (1..5, 7, 9); + +type record length(0..10) of integer RecOfInt; + +type charstring CrLfTermString (pattern "*\r\n"); +\end{lstlisting} + + +\subsection{Parametric templates} + +When sending messages of a given protocol, templating helps in abstraction, readability and productivity. +TTCN-3 has quite extensive templating capabilities. + +Templates can be parametric, i.e. they can take arguments just like you would have function arguments in an +encoding function in a C program. + +\begin{lstlisting}[caption={Example of parametric templates},label=paramtric] +type record MyMessageType { + integer field1 optional, + charstring field2, + boolean field3 +}; + +template MyMessageType tr_MyTemplate + (boolean pl_param) := { + field1 := ?, // present, but any value + field2 := ("B", "O", "Q"), + field3 := pl_param +}; +\end{lstlisting} + +Templates can be hierarchical, so from the most generic to the most specific case, you can specify ever more +concrete templates, depending on the need. + +FIXME: Example + +Templates can further be specified as {\tt subset}, {\tt superset}, {\tt permutation} (any order of elements). + +Templates can also be used for the receiving side. Here, an incoming, already-decoded message is matched +against one or multiple receive templates. Matching against a fixed template wouldn't give much improvement +over e.g. comparing with a 'const struct' in a C-language test case. However, TTCN-3 templates can have +wild-cards and pattern matching. + +FIXME: Example + + +The built-in {\tt match()} function can be used to determine if a received +message (or actually any value) matches the template. Even more so, the fundamental {\tt receive()} function, +which is used to receive any inbound message from a test port, has built-in matching capability, so explicit +calls of {\tt match()} are rarely required. See FIXME later. + + +\subsection{TTCN-3 + TITAN encoders/decoders} + +The data/type system of TTCN-3, extended by non-standard capabilities of TITAN allows describing the messages +of practically all possible protocols. The programmer can focus on {\em expressing} the structure/syntax of +the protocol, rather than having to write an encoder/decoder by hand. + +TTCN-3 specifies ways of importing other data types or schema definitions, for example ASN.1, IDL, XSD (XML) +and JSON. TITAN allows adding encoding instructions as annotations to the TTCN-3 type definitions to +automatically encode/decode them into binary or textual forms, XML or JSON or ASN.1 (BER/CER/DER). + +Import of the formal definition of a protocol works of course only for such protocols that have a formal +definition of their encoding. While this is more often the case in modern telecom protocols, it is not so +often the case in the TCP/IP/IETF world. This is where the TITAN binary and text encoder/decoder come into +play. + +Listing \ref{lengthto} uses the TTCN-3 type language with TITAN binary codec extensions to describe a UDP heaer. +We first define a {\tt LIN2\_BO\_LAST} type as an unsigned 16bit integer with little-endian byte order, then use +this to define the UDP header as a record of four such integers to finally define a UDP packet as a record +consisting of the header and a variable-length octetstring as payload. Note the {\tt LENGTHTO} and {\tt +LENGTHINDEX} notation to express that the "len" field of the header is set to the combined length of the +header and the payload of the packet. + + +\begin{lstlisting}[caption={Example of TITAN binary codec LENGTHTO},label=lengthto] +type integer LIN2_BO_LAST (0..65535) with + { variant "FIELDLENGTH(16), + COMP(nosign), + BYTEORDER(last)" + }; + +type record UDP_header { + LIN2_BO_LAST srcport, + LIN2_BO_LAST dstport, + LIN2_BO_LAST len, + LIN2_BO_LAST cksum +} with { variant "FIELDORDER(msb)" }; + +type record UDP_packet { + UDP_heasder header + octetstring payload +} with { + variant (header) "LENGTHTO(header, payload), + LENGTHINDEX(len)" +}; +\end{lstlisting} + +The next example in Listing \ref{presence} contains a partial definition of the GRE header, where individual +flag bits at the beginning of the header determine if certain optional fields at some later part in the header +are present or not. The {\tt PRESENCE()} attribute of the TITAN binary codec provides an elegant solution to +express this: + +\begin{lstlisting}[caption={Example of TITAN Binary codec PRESENCE},label=presence] +type record GRE_Header { + BIT1 csum_present, + BIT1 rt_present, + BIT1 key_present, + ... + OCT2 protocol_type, + OCT2 checksum optional, + OCT2 offset optional, + OCT4 key otional, + ... +} with { + variant (checksum) "PRESENCE(csum_present='1', + rt_present='1'B)" + variant (offset) "PRESENCE(csum_present='1'B, + rt_present='1'B)" + variant (key) "PRESENCE(key_present='1'B)" +\end{lstlisting} + +Much more complex constructs are possible in the TTIAN binary codec, e.g. the extension octet concept found in +many telecom protocols. + +FIXME: examples for a text based protocol + + +\subsection{Abstract Communications Operations} + +Abstract communications happens on the {\em test ports} which connect the test case of the abstract test suite +(ATS) with the implementation under test (IUT). TTCN-3 supports abstract communications operations for both +asynchronous and synchronous communication. + +Asynchronous communication is what is typically used to send and receive messages with the IUT. The {\tt +send()} function is non-blocking, while {\tt receive()} is blocking\footnote{See the {\tt alt} behavioral +statement to achieve non-blocking semantics}. Arriving messages stay in the incoming queue of the destination +part. MEssages are sent in order. The receive operation examines the first message of the port's queue, but +extracts it {\em only} if the message matches the receive operations template. + +\begin{lstlisting}[caption={Example receive() operation},label=receive] +template MsgType MsgTemplate := { /* valid */ }; +var MsgType MsgVar; +PortRef.receive(MsgTemplate) -> value MsgVar; +\end{lstlisting} + + +\subsection{Program Control and Behavior specification} + +TTCN-3 offers the usual program control statements that C programmers are familiar with: {\tt if}-clauses, +{\tt for}-, {\tt while}- and {\tt do-while} loops, including {\tt break} and {\tt continue}. It also offers +somehing like the {\tt switch} statement of C, but it is called {\tt select} in TTCN-3. Furthermore, {\tt +goto} and associated labels are supported, although again with slightly different syntax than in C. + +Beyond those, TTCN-3 offers a couple of unique so-called {\em behavioral control statements} which are +further illustrated below. + +\subsubsection{The alt statement} + +In testing, quite often one is waiting for one or multiple received messages, guarded by one or multiple +timeouts. The blocking semantics of {\tt receive()} means we need some kind of a non-blocking alternative. +In TTCN-3, this is achieved by the {\tt alt} statement, which declares a set of alternatives which can happen, +but must not happen. + +The below example Listing \ref{alt} shows a code example that first sends a message through port P, and then +waits for either +\begin{itemize} +\item a response from port P matching the template {\tt resp}, at which point it will set the verdict of the + test to pass and leave the alt. +\item any other message on any other port, which it will receive but ignore, and repeat the alt. +\item a timeout of the timer T, which will set the verdict to fail and leave the alt. +\end{itemize} + +\begin{lstlisting}[caption={Example alt statement},label=alt] +P.send(req); +T.start; +alt { + [] P.receive(resp) { setverdict(pass); } + [] any port.receive { repeat; } + [] T.timeout { setverdict(fail); } +} +\end{lstlisting} + +The {\tt []} at the beginning of each line in Listing \ref{alt} is the {\em guard expression} which can be +used to restrict whether a given alternative is eligible or not. This can e.g. be used by state machines to +allow certain processing only in certain states. + +Experience shows that in more complex tests suites, there will be many alt with partially repeating content, +such as e.g. the case when the main execution timer times out, which always leads to fail. Rather than +open-coding those alternatives again and again over the code, they can be abstracted out as so-called {\em +altstep}. Those altsteps can then be activated/deactivated and then become active without any explicit code +inside each and every {\tt alt}. + +\subsubsection{The interleave statement} + +In {\tt alt}, one of the stated events must happen in order for the control flow to continue after the +statement. The {\tt interleave} statement offers an different behavior in which {\em all events must happen +exactly once, in any order}. + +\begin{lstlisting}[caption={Example interleave statement},label=interleave] +interleave { + [] P.receive(1) { ... } + [] Q.receive(4) { ... } + [] R.receive(6) { ... } +} +\end{lstlisting} + + +FIXME: synchronous API calls example + +FIXME: diagram about netfilter conntrack test suite + +FIXME: test execution + +FIXME: configuration file / moduleparam + +FIXME: logfile, including match error + +FIXME: Junit-XML output plugin + + +FIXME: fuzzing extensions + +\section{Acknowledgments} +The author acknowledges that like everyone in Free Software, he is merely standing on the shoulders of +giants\footnote{In this case, maybe rather {\em on the shoulders of titans?}}. +Without the work of the ITU and ETSI on the TTCN-3 Language and without the release of Eclipse TITAN as Free +Software by Ericsson, he would not have been able to discover TTCN-3. + +Special thanks go to Elemer Lelik for his responsiveness and help in getting started with TITAN, as well as the +excellent support of the TITAN team for fixing any bugs I reported in virtually no time. + +\bibliographystyle{isea} +\bibliography{isea} + + +\end{document} |