Starting ecFlowUI

The command to start the application from ECMWF workstations is:

module load ecflow/dev
ecflow_ui

Starting ecFlowUI for the first time

When starting ecFlowUI for the first time, the application contains an empty tree view:

The first thing to do is to populate it with servers.

Go to the Servers menu; this should contain all the servers that you have configured for ecFlowview (these are copied over once, when you first start ecFlowUI; from then on, the two lists are not synchronised).

Selecting a server will add it to the tree view. For large servers, a Loading indicator may appear, but the user interface is still responsive.

Once loaded, a server appears in the tree view. Server and suite nodes show, in brackets, the total number of children under them. Any number of servers can be added to the view.

Viewing the tree

The tree nodes are expanded and collapsed using the small icons to the left of each node (these may vary in appearance depending on your workstation setup). The middle mouse button does not serve this function anymore!

At the top of the tree is a set of breadcrumbs, showing the hierarchy up to the currently-selected node. Clicking on a node in the breadcrumbs will select the node in the tree.

The configuration menu at the top-right of the panel allows access to various display options, not all of which are yet implemented.

When a node is selected, the toolbar at the top of the tree view has a button which reveals options to display various information for the selected node, such as Output and Script. Clicking one of these pops up a floating Info Panel displaying the selected information in a tab.

The panel can be detached or frozen by using the configuration menu at the top-right of the panel:

It is also possible to have a permanently docked Info Panel: from the main window, select the Add Info Panel button from the main toolbar.

The information panel appears at the bottom of the window.

This docked information panel is linked to the currently-selected node, but can also be detached and frozen using the configuration menu at the top-right of the panel.

Layout

Each panel in ecFlowUI is resizeable and moveable, following a dashboard approach. For example, drag the information panel to the right of the tree view to get an arrangement like this:

Interacting with the servers

In the tree view, the right-click context menu provides most of the standard commands. Hovering over a menu option shows, in the status bar, the command that will be sent to the server.

It is possible to select multiple nodes with the mouse and select a command which will be sent to them all. Also, using the Custom... menu entry, any command may be typed in.

Configuration and setup

The list of available servers can be managed through the Servers | Manage Servers menu.

Global server settings, plus other settings such as colour setup, can be found in the Edit | Preferences menu.

Settings per server can be found in the Information panel, in the Settings tab (when a server node is selected in the tree).

Table view

From the toolbar at the top of the window, a new Table View panel can be added. This provides an alternative, flat, way to view the nodes, and allows sorting and filtering of nodes. Work on this is incomplete, but it is in progress.

Tabs and windows

ecFlowUI supports multiple windows and multiple tabs, available from the File menu. Each tab contains its own list of active servers, and the tab control itself shows the status of each server in that tab.

Likewise, each window is separate from the other windows, each containing their own set of tabs and therefore servers.

Searching for nodes

ecFlowUI has a built-in fast, powerful search engine. To activate the search dialogue, click the Search button on the toolbar, or choose Tools | Search from the main menu. Alternatively, right-click on a node in the tree and select Search from the context menu - the search will be performed on that node and its descendants.

The search dialogue is split into three parts - the editor, the generated query and the results.

The Scope section allows you to specify which server(s) to restrict the search to. ALL means that all servers in the current tab will be searched. The root node for the search can also be set here - only that node and its descendants will be searched.

The Nodes section by default specifies no filtering, and thus will find all nodes. The node name and path can be specified using 3 possible syntaxes:

• Matches is a wildcard match, similar to UNIX file syntax, where "*" means "any number of any characters" and "?" means "one of any character", e.g. "*monitor*". If no wildcard characters are used, the result will be that only those nodes whose names/paths exactly match the string will be found. This is often the simplest way of searching. See http://doc.qt.io/qt-5/qregexp.html#wildcard-matching.
• Regexp is a regular expression match, using the Qt regexp syntax. The regexp equivalent of the above example would be ".*monitor.*". See http://doc.qt.io/qt-5/qregexp.html#introduction.
• Contains finds nodes whose names/paths contain the given regular expression as a substring; similar to Regexp, but can find more nodes since the regexp only needs to match a substring rather than the whole name.

The Type selection box by default has no entries checked. This means that the type of the node will not be checked, so nodes of all types may be found. Checking the node type boxes restricts the search to only those node types.

The Status selection box works in a similar way to the Type selection box.

The Flag box is also similar - if nothing is selected, then nodes will be found regardless of their flags; otherwise, the search will be restricted to just those nodes that contain the specified flags.

The Query section shows the query which is generated from the selections in the dialogue.

Once the options have been set, click the Search button. The results panel will be populated with the found nodes. The nodes have the normal context menu, allowing you to perform actions on them straight from the results box. Multiple node selection is allows. Clicking on a node here also selects it in the node tree.

An important thing to note about the search results is that the found nodes will be updated when ecFlowUI syncs with the server. The same nodes will be shown in the results, but their latest state will be reflected.

Current work

See ecFlowUI current priorities. This page lists the work in progress by priority and by subject.

Reporting issues

Issues can be reported directly to Sandor Kertesz or Iain Russell, or entered into JIRA as sub-issues of ECFLOW-92 - Getting issue details... STATUS .