Job Views: Cards, Tree, and Table
Last updated
Last updated
Job cards in CryoSPARC are displayed by default in a masonry grid referred to as the “Cards View”. Cards are laid out from left to right and then in sequential rows organized by a sort attribute and sort order. The default sort attribute is the “Date Created”, and the default sort order is ascending (this means the most recent jobs will appear at the bottom of the view, while the oldest will appear at the top).
The cards view is a scrollable page that progressively loads job information as you navigate. This means that you can use the scroll bar or your mouse wheel to navigate to any point in the view, whether you are working on 100 jobs or 1000 jobs, without needing to change pages.
The target button on the filter bar allows you to quickly navigate the view to the currently selected job. This can also be done by pressing the T key on your keyboard. The adjoined arrow button will open a dropdown menu with options for the “Last Running Job”, “Last Completed Job”, or “Last Created Job”. Selecting an option from this menu will select the corresponding job and then navigate you to it.
The job search menu can be opened by clicking on the job count button in the footer. This menu contains a list of jobs in the current view (workspace, project, or instance). The menu items include a colour indicator for the job’s current status, as well as its job ID and job type. The list can be filtered by typing into the input at the bottom of the menu. Clicking on a job in this list will select the job and then immediately navigate you to it.
The job search menu is limited to the first 1000 jobs in the view, sorted by the attribute and order that you have selected.
A variety of filters are available to help you quickly and easily find sets of jobs that match a specific criteria (eg. jobs that were created between two dates, or have a specific job type). All jobs that match the filter will continue to be displayed, and those that do not will be removed from the view. Filters can be applied additively to create a more granular criteria for matching jobs (eg. all Ab-Initio jobs that have a status of completed or failed). More information on the filter system can be found in the "Filters and Sorting” section of the guide.
The tree view is a unique view available for jobs inside of a workspace or project. This view lays out all jobs as an interconnected network branching out from the first job(s) in the processing pipeline. Each job is linked to previous and subsequent jobs with coloured lines representing the flow of inputs and outputs between them.
The tree view is in many ways a “superset” of the cards view, which is to say that the job cards themselves look and behave for all intents and purposes identically to the card view, and jobs also retain all of the same information and actions. All systems available in the card view for navigation (ie. targeting and searching) are also available in the tree view.
When initially navigating into the tree view, the view will load with all jobs visible and centred horizontally and vertically within the viewing area. The scroll wheel is set to zoom in and out of the view by default, and you can click and drag anywhere in the viewing area to navigate.
The zoom or pan modes are shown in the bottom right of the view, as the magnifying glass and four-arrows icons respectively. These control the scroll behaviour. In zoom mode (default), scrolling will zoom the view in or out, while in pan mode, scrolling will pan the view horizontally or vertically. Holding the command/ ctrl key while scrolling will invert the action. For example, by holding down the command key while in zoom mode, scrolling would cause the view to pan instead. Both options are also mapped to their own keyboard shortcuts. Zoom mode is mapped to the Z key while pan mode is mapped to the X key beside it.
The two options for the click mode, drag and select, are shown in the bottom right of the view as the hand and pointer icons respectively. By default, the click mode is set to drag. This mode allows you to click, hold, and drag the tree view to navigate around it. By switching to select mode you can click, hold, and drag the cursor to select multiple jobs in a rectangular selection box (much like drag selection of files on the desktop of most modern operating systems). Drag select will, by default, start a new selection each time you begin dragging the selection box. You can retain your current selection and select additional jobs by holding the command / ctrl key while making your selection. Drag mode is mapped to the C key while select mode is mapped to the V key.
The scale switcher is located on the bottom left of the viewing area and allows you to reset the view to show all of the available jobs, select a custom zoom level, and cycle through multiple preset zoom levels. The R key will reset the view, while holding the shift key and then pressing the R key will cycle through the preset zoom levels.
Filters can be applied in the tree view in the same way as in the cards view, but in order to retain the spatial context of job trees, filtered jobs will still appear in the tree view in the same position but as dotted outlines, with no input or output connections.
Jobs that are not filtered out appear as usual and will have their input/output connections rendered if they connect to other jobs that have also not been filtered out.
In order to correctly render the layout of the tree view, it is sometimes necessary to include job cards for jobs that are either not in the current workspace or have been deleted.
Jobs that are not in the current workspace but must be shown in the tree are referred to as “dependent” jobs. These jobs will appear with a solid orange border and an unlinked icon in their footer. They include two buttons that are visible on the card when it is hovered. These buttons allow you to link or move the job into the current workspace.
Dependent jobs cannot be selected or interacted with at all unless they have been linked or moved into the current workspace.
Note: command key represents command on Mac and ctrl on Windows or Linux
Z key will set the scroll mode to zoom.
X key will set the scroll mode to pan.
C key will set the click mode to drag.
V key will set the click mode to select.
T will reset the view on the currently selected job or set of jobs.
R will reset the view to show all jobs.
shift + R will cycle through zoom levels (0.25x, 0.5x, 1x).
Holding the command key while scrolling will invert scroll mode to pan or zoom (depending on which is currently selected).
The table view presents jobs with their information laid out in a highly structured and predictable manner. The rows represent jobs, and the columns represent the different types of information that the jobs contain.
The table view is particularly useful for accounting purposes, as it allows you to view a condensed representation of you job data, much like a spreadsheet.
All of the filters and sorting options that are available in the card and tree view are available in the table view as well. Jobs can be quickly sorted by clicking the column header (eg. project, job, or status), which will set the sort attribute and the sort order, clicking the header again will invert the sort order.
Clicking a row will select the associated job, and clicking it again will deselect it. The checkbox at the far left of the row shows whether the job is selected or not (and can also be clicked to toggle the selection on or off).
The table view (as with the other views) can be saved as a CSV by clicking the download button at the bottom right of the jobs footer. This will export the entire data selection with the same filter, sorting, and granularity options as set in the table view.
Job groups are an organizational feature that allow you to collapse multiple jobs into a single card to maintain the legibility of a workspace. They are designed to compact branches of jobs that are not being used for active processing but are still required for reference or continued processing in the future.
Job groups are simple constructs that are meant to be created and deleted without much overhead. Options for a title and description allow you to clarify a group’s purpose, and can help to document the thought process behind branches of processing when coming back later.
To create a group you must select all of the jobs you would like to have included in it. From here you can select the “Group Jobs” option from the job sidebar actions menu, or right click on any of the selected jobs to open the quick actions menu and select the option from there. Alternatively you can use the keyboard shortcut command + G . All of these options will open the job group dialog with fields available for a title, description, and colour. All of these fields are optional, and if you would prefer to leave them blank a default title and colour will be applied. Clicking the “Create Group” button at the bottom of the dialog will create the group (This button is focused by default when the dialog opens, allowing the group to be created with all default fields by pressing the enter key).
Groups can be selected by clicking on the group card or by clicking the group menu button on the footer to open the group search menu and choose the group you would like to select. This will also navigate the view to the selected group.
Groups are meant to be lightweight, and because of this, they do not have any specific metadata associated with them. When selecting a group you are essentially just selecting the jobs that are in that group. This is reflected in the sidebar, which is identical to the multi-selection sidebar that is shown when selecting multiple jobs whether they are grouped or not.
All groups, and jobs that belong to a group, will appear with a group widget on their card. This widget includes the group ID and a +/- button.
Clicking the +/- button will expand or collapse the group. Expanding the group will show all of the jobs contained in the group directly in the view, while collapsing the group will show only the group card. Keyboard shortcuts are also available for expanding and collapsing groups. When a group or group job is selected or hovered, you can press the E key to either expand or collapse it.
Clicking the group ID will select all of the jobs in the group, this allows you to select all grouped jobs whether they are collapsed into a group card or expanded as individual jobs.
The view footer also includes +/- toggle buttons beside the group count that allow you to expand or collapse all of the groups in the workspace.
You can add a single job or multiple jobs to an existing group by selecting the jobs and then right clicking any of them to open the quick actions menu. From here you can navigate to the “Add to Group” or “Add {x} Jobs to Group” option and select the group you wish to add these jobs to from the submenu.
You can remove any number of jobs from an existing group by selecting the jobs and then opening the quick actions menu by right clicking on any of them. From here you can select the “Remove From Group” or “Remove Jobs from Group” option. Alternatively you can use the command + shift + G keyboard shortcut to quickly remove jobs from a group. If you remove all of the jobs from a group, the group itself will automatically be removed.
A group can be removed by hovering the group card to show its action buttons on the bottom right hand corner of the card, and then clicking the ungroup button. This will not delete the jobs within the group, only remove the group itself.
Groups have all of the same navigation options as job cards. This means that a selected group or a selection of group jobs, can be automatically panned and zoomed to by clicking the target button on the filter bar (or by pressing the T key).
Groups can also be searched and found in the group menu to select and navigate to the group card or the expanded group job cards. This menu can be accessed by clicking the group menu button in the footer.
A Job Group is simply a set of jobs. However, not every set of jobs can be made into a Job Group. For a set of jobs to be a valid group, the following rule must be true: For any two jobs in the set, all the jobs on the chain connecting those two jobs must also be inside the set.
This rule is a necessary condition in order for the Job Group to be rendered as a single card in the tree view and is therefore a requirement.
As the simplest example of this rule, consider a job chain A → B → C.
The set A,C is not a valid group, because B is on the chain connecting A and C. However, A,B, B,C, or A,B,C are valid groups.
When creating a group, adding jobs to a group, or removing jobs from a group, if the rule above is violated, a Missing Connections error will occur. Sometimes, with a large group or a complex tree, it is not easy to tell which jobs are causing the error.
When you encounter this error, we advise that you carefully inspect the selection of jobs that you have made, and make sure that you have not missed any jobs that are on chains between jobs that are inside the group. It is important to make sure to check for, and link or move, any “dependent” jobs from other workspaces that are needed in order to make a valid group into the current workspace.
Note: command key represents command on Mac and ctrl on Windows or Linux
O will toggle the outputs view on all cards.
E will toggle the expanded/collapsed state of a group when selected or hovered (hover takes precedence).
T will navigate the view to the currently selected job or set of jobs.
command + G will group selected jobs.
command + shift + G will ungroup selected jobs.
