WWW Data: a World Wide Web Data Base
The installation procedure of WWW Data is divided into a set of steps. All of them have to be executed properly before the system can be run.
At the moment there are four different ports of WWW Data. WWW data requires a C/C++ compiler. The user generated schema is converted by WDCOMP into a C module that needs to be compiled and linked to the WWW Data distribution files.
The following list shows WWW Data requirements on the compiler per port. When no compiler version is shown, the one offered for the listed version of the operating system will be the right one.
The WWW Data Download page takes to a second page containing various compressed files. These are the compressed files:
You should download the set files corresponding to your system.
The distribution files are "pkzip" archives.
Supposing that the files set has been downloaded in the directory
c:\tmp
these are the commands that have to be executed:
c: (the actual target disk must be put here, so instead of c: we could have d:, e: and so on) cd \ (to go to the root) pkunzip -d c:\tmp\htm_w32.zip (to decompress the HTML files) pkunzip -d c:\tmp\www_w32.zip (to decompress WWW Data files) pkunzip -d c:\tmp\uti_w32.zip (to decompress the registration utilities)
When using a Windows version on "pkunzip" care must be taken so that the original directories/files names are kept.
The distribution files are "tar" archives, compressed with "gzip".
Supposing that the files set has been downloaded in the directory
~/tmp
these are the commands that have to be executed:
cd ~ (to go to the home directory) gzip -dc ~/tmp/htm_xxx.tgz | tar xvf - (to decompress the HTML files) gzip -dc ~/tmp/www_xxx.tgz | tar xvf - (to decompress WWW Data files) gzip -dc ~/tmp/uti_xxx.tgz | tar xvf - (to decompress the registration utilities)
The World Wide Web Server has to be set up in such a way that the WWW Data CGI can access all the relevant HTML files. Supposing that the server is similar to the NCSA one, hereafter are listed the required changes.
In the file "access.conf" an entry similar to the following has to be inserted. This change will make the SWWINGS directory visible to the WWW Server. The actual syntax of the SWWINGS directory name depends on the operating system. In this example the Windows syntax has been used.
<Directory C:/SWWINGS/> </Directory>
In the file "srm.conf", and namely in the "alias" section, an entry similar to the following has to be inserted. This change will allow the hyper-links in the WWW Data HTML pages to work properly. The actual syntax of the SWWINGS directory name depends on the operating system. In this example the Windows syntax has been used.
Alias /swwings/ C:/SWWINGS/
After all the steps described above have been performed, it will be possible to create the first schema as described in the Detailed Description.
When the schema is ready, it will only be necessary to put its name in the distributed "makefile". This is accomplished by changing its first line, as follows:
# Insert in the CSCHEMA variable the actual schema name CSCHEMA = library
After this operation, the command
make all
will generate all the WWW DATA executable files.
NOTE
It is recommended to put these applications, as well as the registration utilities, in a different directory than the original ones. The batch files "go.bat" (or "go" in UNIX) in WWWDATA and UTILS directories will put these executable files in the directory /SWWINGS/BIN. It is also recommended to add this directory to the environment variable PATH; this way all the WWW Data executable files and the registration utilities will be directly available.