PPUG_copy_html

Copying diagnostics plots and web pages

Synopsis

$PP_CASE_PATH/copy_html

Copies all the diagnostics plots and html files generation after successfully running the [comp]_diagnostics scripts. Uses the env_postprocess.xml settings to define the remote web server, login, and path. It is recommended that the user setup ssh-keys to the remote web server so copy_html is not interrupted waiting for login credentials. This script does not run in parallel and so can be submitted from the login node.

./copy_html

Details

Each successful completion of the [comp]_diagnostics script will create a unique text file in the $PP_CASE_PATH/web_dirs of the form [comp]DIAG_WEBDIR*.[date]. This file contains a path string to a set of diagnostics plots and html files that need to be copied to a remote web server. The copy_html tool reads the files in $PP_CASE_PATH/web_dirs, copies the files specified in the path to the remote web server, and then updates the web_dirs file by appending a ":copied" string to the end of the path. Subsequent running of the copy_html tool will read the file and only copy files without the ":copied" string at the end of the path string.

Note: for NCAR users, you can use this helper script to setup you ssh keys on yellowstone:

mk_user_ssh_keys.ssh

Dependencies

The copy_html does not depend on any other scripts having successfully completed as it will exit if there are no files to be copied based on the content of the files in the $PP_CASE_PATH/web_dirs and the existence of the ":copied" string. The script will try to create the directories on the remote web server if permissions allow. Users copying files to the NCAR CGD web servers (http://webext.cgd.ucar.edu and http://webint.cgd.ucar.edu) should adhere to the directory and file naming conventions described here.