Xathrya Sabertooth

Installing GraphicsMagick on Linux
Installing libHaru from Source
Installing Boost Library for Linux
Installation of CMake
Wt C++ Web Toolkit Introduction

Posted: 29 Jun 2013 02:34 AM PDT

GraphicsMagick is self-claimed as swiss army knife of image processing. It provides a robust and efficient collection of tools and libraries which support reading, writing, and manipulating an image in over 88 major formats including DPX, GIF, JPEG, JPEG-2000, PNG, PDF, PNM, and TIFF.

The image processing is multithreaded using OpenMP.

In this article we will discuss about installing ImageMagick in Linux. For this purpose I use:

Slackware64 14.0
gcc

Obtain the Materials

GraphicsMagick can be obtain using two ways: download from sourceforge or clone using Mercurial.

To download the tarball, go to this link. Then, extract the content and change to its directory.

If you want to clone the repository, do this command:

  hg clone http://hg.code.sf.net/p/graphicsmagick/code GM  cd GM

At this point, for any method you choose, we are now at GraphicsMagick’s root directory.

Installation

Building ImagesMagick can be done by using following commands (assuming we are at GraphicsMagick’s root directory):

  ./configure  make  make install

Note that to do installation, you should have enough privilege.

At this point, we have installed GraphicsMagick standard installation.

Installing libHaru from Source

Posted: 29 Jun 2013 01:32 AM PDT

LibHaru is a free, cross platform, open source library for generating PDF files. libHaru can be used to produce PDF files, however it still lack at support for reading and editing existing PDF files.

The feature supported by libHaru:

Generating PDF files with lines, text, images.
Outline, text annotation, link annotation.
Compressing document with deflate-decode.
Embedding PNG, Jpeg images.
Embedding Type1 font and TrueType font.
Creating encrypted PDF files.
Using various character sets (ISO8859-1~16, MSCP1250~8, KOI8-R).
Supporting CJK fonts and encodings. You can add the feature of PDF creation by using HARU without understanding complicated internal structure of PDF. libHaru is written in ANSI C, so theoretically it supports most of the modern OSes.

In this article we will discuss about installing libHaru in Linux. For this purpose I use:

Slackware64 14.0
gcc

Obtain the Materials

libHaru use github to host the codes. To download source codes there are two methods: download as tar ball, or clone the repository.

To download the tarball, go to this link. Then, extract the content and change to its directory.

If you want to clone the repository, do this command:

  git clone https://github.com/libharu/libharu.git  cd libharu

At this point, for any method you choose, we are now at libharu root directory.

Installation

libHaru can use cmake for building. To build libHaru, do following

  cmake .  make  make install

Note that to do installation, you should have enough privilege.

Installing Boost Library for Linux

Posted: 28 Jun 2013 11:50 PM PDT

Boost is portable C++ source libraries and work well with the C++ Standard Library. It is intended to be widely useful and usable across a broad spectrum of applications.

In this article, we will discuss about how to install Boost Library within Linux. The method we use here is generic way, which means, it can be applied to any Linux distribution. However, for demonstration I will use following:

Slackware64 14.0 Linux
GCC (Compiler)

Preparation

At least you have privileges for installing the files later.

Obtain the Materials

Boost Library has official site here.

The latest version of Boost library is version 1.53.0 which can be downloaded from sourceforge (here). While there is also Boost 1.54.0 beta available, we will use the stable version at this time.

Compile & Install

Assuming the archive is boost_1_53_0.tar.bz2 and on path ~/Downloads or /home/user/Downloads. First we have to extract the library from the archive.

  cd ~/Downloads  tar -xjf boost_1_53_0.bz2  cd boost_1_53_0.bz2

Next, build bjam binary and also set some options on what to compile and where to.

Boost is a collection of libraries / modules. The whole Boost libraries will cost some time and computation for compilation. We can skip some libraries we don’t want or won’t use. To see the list of libraries need compiling, run following:

  ./bootstrap.sh --show-libraries

For example, following is the result / the list of libraries on Boost 1.53.0:

atomic
chrono
context
date_time
exception
filesystem
graph
graph_parallel
iostreams
locale
math
mpi
program_options
python
random
regex
serialization
signals
system
test
thread
timer
wave

Pick any modules you want to include in Boost, for example: filesystem,program_options, and system. In this case, we can use argument –with-library=filesystem,program_options,system to bootstrap.sh. However you can also pick all / compile all the libraries, just invoke ./bootstrap.sh without –with-library switch.

Next, choose where the library will be installed. By default, the header files will be installed to /usr/local/include and the libraries (.so, .a) will be installed to /usr/local/lib. If other path is preferred, we can use –libdir switch for specifying where library will be stored and –includedir switch for specifying where header will be stored.

Now bring all of the info and invoke like this:

  ./bootstrap.sh --libdir=/usr/local/lib64 --includedir=/usr/local/include

And you could see I will build all the libraries and store the library to /usr/local/include.

Now let’s build the library

  ./bjam

Then install by (make sure you have privileges to do)

  ./bjam install

At this point we have successfully installed Boost library

Installation of CMake

Posted: 28 Jun 2013 10:45 PM PDT

CMake is a cross platform, open-source build system consists of tools designed to build, test, and packaging software. It used to control the software compilation process using simple platform and compiler independent configuration files.

In general, CMake generates native makefiles and workspaces that can be used in the chosen compiler environment.

In this article, we will discuss about how to install CMake, either binary or from sources. The method we use here is generic way, which means, it can be applied to any Operating System. However, for demonstration I will use following machine:

Slackware64 14.0 Linux
GCC (Compiler)

Preparation

At least you have privileges for installing the files later.

Obtain the Materials

A binary and source code can be obtained here. The latest version per June 29th, 2013 is 2.8.11. Both binaries and source codes can be downloaded from here.

As I use Linux I would download source code (http://www.cmake.org/files/v2.8/cmake-2.8.11.1.tar.gz) and unfortunately no binary for Linux x64 so we download this binary (http://www.cmake.org/files/v2.8/cmake-2.8.11.1-Linux-i386.sh). If you want to download for Windows version, go grab source code (http://www.cmake.org/files/v2.8/cmake-2.8.11.1.zip) and binary (http://www.cmake.org/files/v2.8/cmake-2.8.11.1-win32-x86.exe)

Binary Installation

Assuming we download shell archive cmake-2.8.11.1-Linux-i386.sh.

Make sure the file is executable (having +x permission). Do following if you want to set the permission:

  chmod +x cmake-2.8.11.1-Linux-i386.sh

Then execute it from terminal

  ./cmake-2.8.11.1-Linux-i386.sh

Make sure you have privileges to do installation.

Binary installation for Windows system is similar. Just execute the application to initiated installation.

Source Code Compilation & Installation

There are two case for compilation, whether you have CMake or not installed in your system. If you have CMake, we can use CMake to build CMake. However, we can also build it from scratch. Either way we should extract the sourcecode from. A simple way is using command:

  tar -xf cmake-2.8.11.1-Linux-i386.tar.gz  cd cmake-2.8.11.1

From Scratch

This case will describe compilation from scratch and suitable for us who either don’t have CMake installed before or just want to compile it from scratch.

We need a compiler and a make utility and running UNIX/Mac OSX/MinGW/MSYS/Cygwin.

Run the bootstrap script in the source directory of CMake. If CMake is going to be installed on particular path, use –prefix. For this example I use /usr/local as my base path.

  ./bootstrap --prefix=/usr/local  make  make install

Unfortunately for Windows user who didn’t use Cygwin, you can’t do this method. Instead, you should download the binary version of CMake.

Build With CMake

Building CMake with CMake is like building any CMake-based project. Run CMake to the root of cmake directory (e.x: ~/cmake-2.8.11.1).

  cd ~/cmake-2.8.11.1  cmake -DCMAKE_INSTALL_PREFIX:PATH=/usr/local .  make  make install

Notice the “.” (dot) in the trail, that’s part of command.

At this point, we have successfully install CMake

Wt C++ Web Toolkit Introduction

Posted: 28 Jun 2013 07:16 PM PDT

Wt (pronounces as witty) is a C++ library for developing web application.

Wt offers abstraction of web-specific implementation details including client-server protocols, event handling, graphics support, graceful degradation (or progressive enhancement), and URL handling. The API is widget-centric, like Qt and other toolkits.

Unlike many page-base frameworks, Wt was designed for creating stateful application that are at the same time highly interactive (leveraging techniques such as WebSockets, and Ajax) and accessible (supporting plain HTML browsers), using automatic graceful degradation or progressive enhancement. Things that are natural and simple with Wt would require an impractical amount of effort otherwise: switching widgets using animations, while being perfectly indexed by search robots with clean URLs, or having a persistent chat widget open throughout, that even works in legacy browsers like Microsoft Internet Explorer 6.

Wt can acts as a stand alone Http(s)/WebSocket server or integrates through FastCGI with other web servers.

Why Wt

Page-based web frameworks (Django, Ruby on Rails, PHP, etc …) do not attempt to abstract underlying technologies (HTML/XHTML, JavaScript, CSS, Ajax, WebSockets, Comet, Forms, DHTML, SVG/VML/Canvas). As a consequence, a web developer needs to be familiar with all of these evolving technologies and is also responsible for graceful degradation when browser support is lacking. The structure of many web applications still follows mostly the page-centric paradigm of early day HTML. This means that not only will you need to implement a controller to indicate how a user moves from page to page, but when using advanced Ajax or WebSockets, you will need to design and maintain your client-server communication.

Pure Ajax frameworks on the other hand require tedious JavaScript programming to deal with browser quirks, and client-server programming to interact securely with server resources. These applications usually are not compliant with accessibility guidelines and cannot be indexed by a search robot.

Generating HTML code or filling HTML templates is prone to security problems such as XSS (Cross-Site-Scripting) by unwillingly allowing JavaScript to be inserted in the page, and CSRF (Cross-Site Request Forgery) by trusting cookies for authentication. These security problems are hard to avoid in traditional frameworks when as a developer you need to implement JavaScript functionality and thus the framework cannot filter it out.

In contrast, a web application developed with Wt is developed against a C++ API, and the library provides the necessary HTML, CSS, Javascript, CGI, SVG/VML/Canvas and Ajax code. The responsibility of writing secure and browser-portable web applications is carried by Wt. For example, if available, Wt will maximally use JavaScript, Ajax and even WebSockets, but applications developed using Wt will also function correctly when JavaScript is not available. Wt will start with a plain HTML/CGI application and progressively enhance to a rich Ajax application if possible. With Wt, security is built-in and by default.

Typical Use Scenario

High performance, complex web applications which are fully personalized (and thus cannot benefit from caching), fully Ajax enabled and at the same time entirely accessible and Search Engine Optimized.
Web-based GUIs for embedded systems benefit from the low footprint of a C++ web application server.
Web-based GUIs that require integration with (existing) C++ libraries, for example for scientific or engineering applications, or existing C++ desktop applications.

Other Benefits

Develop web applications using familiar desktop GUI patterns.
Provides an extensive set of widgets, which work regardless of JavaScript availability (but benefit from JavaScript availability).
A single specification for both client- and server-side validation and event handling.
Optionally, use XHTML and CSS for layout and decoration.
Generates standards compliant HTML or XHTML code.
Portable, anti-aliased graphics optimized for web usage (using inline VML, inline SVG, HTML5 canvas or PNG images), which can also be used to render to PDF.
Avoid common security problems since Wt has complete control over the presentation layer and proactively filters out active tags and attributes, does not expose business logic, and simplifies authentication using a stateful design.
Ultra-fast load time and low bandwidth usage, which are affected only by screen complexity, not application size. Wt implements all the common tips and tricks for optimizing application responsiveness and even optimizes per browser.
A simple API with a robust cross-browser implementation for server-initiated events aka server push (using comet or WebSockets).
Use the built-in httpd for easy development and deployment, or use the FastCGI/ISAPI connectors to deploy in existing web servers.

Features

Core Library

Supports major browsers (Firefox/Gecko, Internet Explorer, Safari, Chrome, Konqueror, and Opera) but also plain HTML browsers (Links, Lynx).
Develop and deploy on Unix/GNU Linux or Microsoft Windows (Visual Studio) environments.
Equal behavior with or without support for JavaScript or Ajax, as far as possible, by using graceful degradation or progressive enhancement.
Integrated Unicode support and pervasive localization.
Efficient rendering and (very) low latency.
Support for browser history navigation (back/forward buttons and bookmarks), pretty URLs with HTML5 History if available, and search engine optimization with a unified behavior for plain HTML or Ajax sessions.
Configurable session tracking options that include URL rewriting and cookies.
High performance, allowing deployment on low-end embedded devices, or energy-, space- and budget-friendly deployment of Internet or extranet web sites.
Completely based on event-driven async I/O: sessions are not tied to threads, and neither do open connections block threads. Instead, threads are needed only to improve concurrent request handling or for reentrant event loops.

Event Handling

Uses a modern typesafe signal/slot API for responding to events.
Listen for keyboard, mouse and focus events, and get event details (such as mouse position, modifier buttons, or keyboard key).
Automatic synchronization of form field data between browser and server.
Dynamic C++-to-JavaScript translation, by specifying stateless slot implementations. A single C++ slot implementation provides both client-side and server-side event handling: visual changes at client-side and application state at server side.
Possibility to hook in custom JavaScript (e.g. for client-side only event handling), and emit C++ signals from this custom JavaScript.
Drag&Drop API.
Timed events and server-initiated updates (“server push”)
Uses plain HTML CGI, Ajax or WebSockets

Native Painting Support

Unified painting API which uses the browsers native (vector) graphics support (inline VML, inline SVG, or HTML5 canvas), or renders to common image formats (PNG, GIF, …) or vector formats (SVG, PDF).
Supports arbitrary painter paths, clipping, text, images, transformations, drop shadow.

GUI Component

GUI component is composed of various widgets. For comprehensive examples, you can visit this link.

Built-in Security

Kernel-level memory protection protects against privacy issues arising from programming bugs, since sessions can be completely isolated from each other (in dedicated-process mode).
Supports encryption and server authentication using Secure Sockets Layer (SSL) or Transport Layer Security (TLS) through HTTPS.
Enables continuous use of HTTPS through low bandwidth requirements (fine-grained Ajax).
Built-in Cross-Site Scripting (XSS) prevention. Rendered text is always filtered against potentially malicious code, making XSS attacks against Wt applications (close to) impossible.
Not vulnerable to Cross-site Request Forgery (CSRF) because cookies for session tracking are optional, and even when used, they are never solely relied on for requests that trigger event handling code.
Not vulnerable to breaking the application logic by skipping to a particular URL, since only those events exposed in the interface can be triggered.
Session hijacking mitigation and risk prevention
DoS mitigation
A built-in authentication module implements best practices for authentication, and supports third party identity providers using OAuth 2.0, and (later) OpenID Connect

Object Relational Mapping Library

Wt comes with Wt::Dbo, a self-contained library which implements Object-Relational mapping, and thus a convenient way to interact with SQL databases from C++. Although features like optimistic concurrency control make this an ideal technology for a database driven web application (and it provides good integration with Wt’s MVC classes), the library can also be used for other applications, and does not depend on Wt. The ORM library has the following features:

No code generation, no macro hacks, no XML configuration, just modern C++!
Uses a templated visitor pattern which requires a single template method to provide the mapping: DRY and as efficient as conceivable!
You can indicate surrogate auto-incremental keys or map natural keys of any C++ type, which may also be composite (i.e. require more than one database field).
Supports optimistic concurrency control using a version field.
Maps Many-to-One and Many-to-Many relations to STL-compatible collections.
Provides schema generation (aka DDL: data definition language) and CRUD operations (aka DML: data manipulation language).
Prepared statements throughout.
Each session tracks dirty objects and provides a first-level cache.
Flexible querying which can query individual fields, objects, or tuples of any these (using Boost.Tuple).
Use a single connection or share connection pools between multiple sessions from which connections are used only during an active transaction.
Comes with Sqlite3, Firebird, MariaDB/MySQL and PostgreSQL backends, and an Oracle backend is also available on request.

Proyek Teknologi

Sabtu, 29 Juni 2013