Practical instructions for IRF's website

Johan Arvelius

This instruction in PDF: http://www.irf.se/~johan/webman_en.pdf

This document is a part of the full instructions in Swedish: http://spider.irf.se/intra/doc/webmanual.html

4 Practical treatment

Except scripts that automise the website there are from technical point of view three categories of documents: documents that shall be listed by an automatic function (seminars, minutes etc.), usual documents stored in a database and documents stored as files on the webserver. The two first categories is handled through the webinterface http://spider.irf.se/webinterface/intra/.

For documents stored as files an important thing to remember is that all relative links and pathes is relative to the basic file index.php. To be able to give relative pathes in the code that fits with the files place in the filestructure a symbolic link to http://spider.irf.se/index.php can be done by the command
ln -s /var/www/index.php index.php.
You must not copy index.php to the local directory, it shall be one file for the whole site in which changes shall give global effect. It's strongly recomended to make such links whenever a new directory is created.

4.1 Edit documents in the database

To edit a document stored in the database the most simple way is to go the webinterface http://spider.irf.se/webinterface/intra/ and log in. Choose the link ``Admin. Website'' in the left column and go on to the document you want to edit. If this part of the site is setup according to recommendations there should be a link ``Modify:'' at the bottom of your page leading to the database entry. There you could edit directly in the window on the screen or cut and paste the source to your favourite editor and reload it using the ``file'' button.

4.2 To add minutes, seminars and news

webfigures/protokoll.png

Figure 1: Interacton interface to add minutes

Minutes, seminars and news can easily be added to the database and will be shown by routines in lists of different kinds on the website without further treatment. Step by step the process goes as:

enter http://spider.irf.se/webinterface/intra/ and click LOGIN below the IRF-logo.

Yoy will now se a list of links to the groups you have authority to add documents to. Follow the link of your choice.

You will now enter an administration interface where it is possible to chose Admin. Protocols, Admin. Seminars or Admin. News. (Indicated in Green in Figure 1.) Follow the link of your choice.

You'll now enter al list of present documents in the chosen category according to Figure 1. There might be sub categories. Follow the link Add Protocol/Seminar/News in the sub category of your choice.

Chose and fill the fields of the form.

Title

The title is probably used in the list where the document is shown on the website.

New category

Think thoroughly before you add a new category and make shure to make a new script to get it displayed on the website.

Read permissions

Sets the reading permissions to your document. Group means the database group in which you are entering a document , Intranet is all users that has a user account at the IRF computersystem i.e. all employees as well as all temporary guests. Public means it can be read by anyone. This does not automatically mean that it in practice can be found by anyone if there is no list linking to it that is avalilable to public, but take care, there might be that someone writes such a list later!

Write permissons

Sets the permissions to edit and delete the document. Creator means your self , the administrator of the group and the main administrator (probably the PR-manager and/or a computer manager). Group means everyone in the group.

Add the document either by choosing a file from your own file system by entering a filename or use the browse button. As an alternative the text can be entered directly in the big field.

End by UPDATE DATABASE.

4.3 The link database

To use the links in the link database you will have to do an ordinary link in your document to http://spider.irf.se/link/linkname. Through the web interface to the database http://spider.irf.se/webinterface/intra/ it is possible to see what links is avalible. This is done by following the link database in the upper left corner. There are also the full link names available to cut and past in your code.

To make new entries to the link database:

Go to the webinterface http://spider.irf.se/webinterface/intra/, log in and follow the link Link database in the top left corner.

To avoid double entries, search the database for your target by following the SEARCH link and write the adress in the field.

Follow the ADD link at the top of the page.

Enter the new links name and the target adress in the fields.

Click UPDATE DATABASE.

View your new link appear in the list.

To change links in the linkdatabase:

Go to the webinterface http://spider.irf.se/webinterface/intra/, log in and follow the link Link database in the top left corner.

Follow the EDIT link for the link of your choice.

Enter the new target adress.

Click UPDATE DATABASE.

NB only the adress is possible to change. The name of the link is decided once for all and changing it should be in straight contradiction to the purpose of the database. It is possible to remove links from the database. Think thoroughly before you do so. This must only be done if you are really sure that the contents of the target files are permanently removed from the web. There might be other links than your own ot the database entry. Try to find all links to the entry if you really removes it from the link database.

4.4 Other documents, ``normal web pages'' and pictures

In this section the procedure to add web pages to the IRF website that shall share the plue field on the left. Shortly this is done by adding the file on the web server or in the database and then make liks to http://spider.irf.se/index.php with the parameter ?page=filename for files on the server or the parameters ?dbfile= databaseentry&dbsec=group for entries in the database.

4.4.1 Restrictions

There are a few technical specialities that might interfere with the script that shall read the file in a bad manner and must be avoided. For the web site to give a uniform impression there are recommendations on the design by Jenny Brännström (http://www.irf.se/intra/webbredaktion/profilmall_030606.pdf). Added files in simple HTML code with no extra manipulations on the graphical form will follow these recomendations.

Absolute coordinates: As the entered code will not form an own HTML document but in the end be added as a field in a larger document including the blue part on the left and probably a header and a footer parts placed using absolute coordinates in the code will probably end up in an unexpected and unwanted placement compared to the rest.
CSS: IRF's web site is using CSS in the design of the pages. These will even be applicated to the pages you will add. CSS's parsed by the browser will overwrite each other. This means that added pages with CSS definitions can overrule the common CSS and mess up the rest of the page. Normally CSS is defined in the head of the HTML file (between the tags <head> and </head>). This part of the entered file will be stripped off by the script and will not affect the page. If you for some reason really needs other CSS definitions than the default IRF CSS There is a possibility to write an own CSS file and link it in by another parameter to the skript http://spider.irf.se/index.php. That is explained in section .

4.4.2 Add documents to the database

To add the document to the database follow the instructions in section 4.2 with the only difference to choose Add Document.

To link to the page link http://spider.irf.se/index.php with the parameter ?dbfile=databaseentry&dbsec= group where group is the user group the document belongs to usually 'Administration' or a research program ('P1' etc.). If you link to this document through a menu of a wrapper and want to link to the document included in that wrapper you could of course also link to the adress that gets generated by the wrapper when you click your way there. This adress can be rather long and messy but it helps the visitor to understand the context by showing the right sub menu at the same time.

Example:: If I have added an entry to the database in group P1 titled ``example'' I shall write the link <a href="index.php?dbsec=P1&dbfile=example">example</a> from referring pages in the directory /var/www/ on spider. From other pages you can chose to write the partial path to test.php or the full: <a href="http://spider.irf.se/index.php?dbsec=P1&dbfile=example">example</a>

In the examples in section 4.5 we will continue to add this document in the wrapper and link to it that way.

4.4.3 Add images

To add the document to the database follow the instructions in section 4.2 with the only difference to choose Add Image.

To add the picture to a website the PHP function http://spider.irf.se/webinterface/intra/code/include_db_image.php shall be used.

4.4.4 Add documents as files

Put the file you want to add in the tree structure in the directory /var/www/ on spider. The directory where the file is placed will decide in which environment the page will be chown. I.e. a file in the directory /var/www/P1/ will get a row of links for navigation on the atmospheric physics pages at the top of the page.

To link that page a link to http://spider.irf.se/ subdirectory/wrapper.php shall be made with the parameter ?page=filename. If this is not working it is probably due to that there are no symbolic link to http://spider.irf.se/index.php from the sub directory where the file is added. Such a link can easily be added by the command
ln -s /var/www/index.php index.php.
You must not copy index.php to the local directory, it shall be one file for the whole site in which changes shall give global effect. As an alternative a link to http://spider.irf.se/index.php?page= subdirectory/filename, can be made. This will result in a page with a header etc that is common with the IRF main page. The drawback with this method is that you must make sure all links on the page has links relative to /var/www/. Even for files it is better to link to the adress generated by the wrapper as discussed in section 4.4.2.

Example:: If I have added a page /var/www/ P1/example.html I shall make the link <a href="P1/index.php?file=example">example</a> from pages in the directory /var/www/ on spider. From other pages either the relative searchpath to index.php or the whole: <a href="http://spider.irf.se/P1/index.php?file=example">example</a>.

In the examples in section 4.5 we will continue to add this in the wrapper and link to it that way.

4.5 Pages with page head

To add files to generate the dynamically generated page heads there is a template. This template also contains code in a way that makes it work as a wrapper-file. In order to add a set of web pages with a common page head e.g. a new project or instrument according to this template do as following. The wrapper file to the page shown in Figure is prepared with this template. In situations where the template is not usable there are instructions to write a new wrapper from scratch in section .

Copy the file /var/www/skeleton/wrapper.php.

Set the headline for the page. This can be done either by a image logotype, a text headline or both. If both are used the image logotype will be placed in top and the text headline under. This can be used e.g. to add a new project with the research program logo at the top of the page and then a text headline of the project. To the image logotype a link can be added by the variable $imglogourl="URL";. To the image logotype a text alternative with the same text should be entered by $logoalt="alt text";. The path to the image logo is set by $imglogo=" relativ/path/to.file"; while the text headline is set by $textlogo="Headline";. In Figure an example with image logo is shown.

Insert the names of the main links that shall be shown under the headline. They shall be entered in order by $linkname[x][0]="linkname";

Add names of any sub-links that shall fall out in menues to the main links while these are activated. These are added in the same way; $linkname[x][y]=" linkname";. Sublinks with the first coordinate x appears in order after y under the main link x when it is activated.

Links leading ``out'', id est pages that shall not be shown within the same wrapper is added by writing $linkurl[x][y]="url"; directly after the link name. For wrappers that are not top level, eg a reseach project within a project that shall have an own wrapper recomended to add such as $linkname[0][0] leading back to the previous level.

Add whatever shall be shown when the links are activated either by adding HTML-code directly or include other files in the corresponding else if clause. To include files or documents in the database that by them selves are complete HTML-documents there are two ready scripts that can be used to strip off parts of the HTML code that must not appear in the middle of the document.
- http://spider.irf.se/webinterface/intra/code/include_local.php is a PHP script that takes one argumend file=file and returnes the contents of the file file from the local filesystem stripped according to section .
- http://spider.irf.se/webinterface/intra/code/include_db.php is a PHP script that takes the arguments dbfile=file and dbsec=group, where group is the user group the document belongs to usually 'Administration' or a research program ('P1' etc.), and returns the contents of the document file in the database in the same way.
Both these script shall be called with 'include' to make sure that any PHP code in the files shall be interpreted, the scripts by themselves only returns exactly the contents in the file left after the stripping.
These scripts are hailed in the end of the file, thereby it is enough to get one file from the filesystem or the database it's enough to set the variable $file=file or $dbfile= file respectively, the admin variable is already set.

Add the file to the database according to section 4.4.2 or directly on the webserver according to section 4.4.4. It is important that it has the ending .php in the file name.

As http://spider.irf.se/index.php starts to look for a wrapper in the current directory the file shall be called wrapper.php if it shall be the defalult wrapper to the current directory. Then there must also be an symbolic link to http://spider.irf.se/index.php.

Linking to the documents that shall be included with the new wrapper explicitly is made as usual. The adresses gets a little peculiar. NB the &-signs that appears in the adresses must be written & in the links that shall be interpreted by the browser as & has a special meaning in HTML as explained in , on the other hand in the include commands there must not be any & as this is a command in PHP and not HTML.

Example 1:: To add the example document added to the database in section 4.4.2 to the sub menu to the second menu in the page head I add $linkname[1][1]="new file"; in the wrapper and then include "http://spider.irf.se/webinterface/intra/code/include_db.php?dbfile=exempel&dbsec=P1" in the else if ($link[1][1]) statement.
Example 2:: To link to the document from somewhere else I'll write a link <a href="http://spider.irf.se/P1/index.php?link[1][1]=1&chosen=program">example</a>.
Example 3:: To add the example file from section 4.4.4 next to it in the sub menu I add $linkname[1][2]="newer file"; in the wrapper and then include "http://spider.irf.se/webinterface/intra/code/include_local.php?file=exempel" in the else if ($link[1][2]) statement.
Example 4:: To link this file from somewhere else I'll write a link <a href="http://spider.irf.se/P1/index.php?link[1][2]=1&chosen=program">example</a>.

File translated from T_EX by T_TH, version 3.08.
On 14 Oct 2004, 13:44.