Support sequential tasks

[dgreen]

Many projects have a sequence of tasks to be done and the tasks need to be done in a specific order.
For example,

1. Call vendors to get quotes
2. Choose a vendor and schedule work
3. Meet vendor for work

It would be nice to have a way that only the doable task (first one not completed) shows in a list. Probably lots of ways to do this but perhaps

• #project
  • #task Call vendors to get quotes
  • #task Choose a vendor and schedule work
  • #task Meet vendor for work

and some sort of keywords for the filter like "only next" which means when presented a list of tasks only show the first one not done.

[claremacrae]

Hi @dgreen, thanks for the suggestion.

Searching for the word 'sequential', there are some related requests already:

• Sequential tasks #376 - which looks to be identical, albeit without your useful example
• Introduce the concept of task dependencies #463

So I will link those to this one, and then mark this one as a duplicate.

[dgreen]

Sorry not to have noticed that you wish feature requests go to discussions. I saw that GitHub has Issue processing for feature requests, searched there for hits on sequential (both open and closed), did not find any so posted an issue.

Thanks for creating the Task Plug-in and sharing it with us. I will follow the discussion.

[claremacrae]

@dgreen Workarounds are discussed in #463.

[Silv-1]

This is a dataviewjs script which generates a tasks query.

It returns first tasks of all projects.

```dataviewjs
// Create a list which contains path and text of the first task of each project
// Tasks must have tag #task, projects must have tag #project
let projectsFirstTask = dv.pages("#project")
    .flatMap(page => page.file.tasks
        .where(task => !task.checked)
        .where(task => task.tags.contains("#task"))
        .slice(0,1))
    .map(task => ({
        path: task.path,
        text: task.text,
    }))
    .values;

// Create a tasks query string which filters tasks to match any path and text of the list
const query = `
filter by function JSON.parse(${JSON.stringify(JSON.stringify(projectsFirstTask))}).some(entry => entry.path === task.file.path && task.originalMarkdown.endsWith(entry.text))
`;

// With dv.paragraph print the string which then gets interpreted as a tasks query
dv.paragraph('```tasks\n' + query + '\n```');
```

[Silv-1]

Credits to this idea how to combine dataviewjs and task queries:

https://github.com/obsidian-tasks-group/obsidian-tasks/blob/983c93a9e0890973887759a79b5be0e93b6ca62a/resources/sample_vaults/Tasks-Demo/How%20To/Find%20tasks%20in%20notes%20with%20particular%20tag.md

[claremacrae]

Wow, I created some test data and ran this.

I added the explain instruction to the query, to see how it worked...

filter by function JSON.parse("[{\"path\":\"Obsidian/tasks/tasks sponsorship/sdfgsdfgsdfg.md\",\"text\":\"P1 Task 1 #task\"},{\"path\":\"Obsidian/tasks/tasks sponsorship/asdgsdfgsfgs.md\",\"text\":\"P2 Task 1 #task\"}]").some(entry => entry.path === task.file.path && entry.text === task.description)

This is powerful stuff - thank you for suggesting it.

[Silv-1]

Thank you. Like this you can essentially include any dataviewjs into a task query.

Just one problem I see with this query is when you have more than one task with the exact same text in one file.

[Silv-1]

Personally I use an expanded version of the query.

• This lets one mark projects as parallel or sequential by setting the parallel boolean
  • For sequential projects only the first task gets listed
  • For parallel projects all tasks get listed
• It filters all projects by the area property
  • This query shows all projects with area === Work
  • Put this query in a separate file and use dv.view(path, input) to set the area as an argument
• It does not show tasks with tag #backlog
• It groups by tasks tags.

```dataviewjs
let desiredArea = "Work";

let sequentialProjectsTasks = dv.pages("#project")
	.where(page => !page.parallel)
	.flatMap(page => page.file.tasks
		.where(task => !task.checked)
		.where(task => task.tags.contains("#task"))
		.slice(0,1))
	.map(task => ({
		path: task.path,
		text: task.text,
	}))
	.values;
let areaPagePaths = dv.pages()
	.where(page => page.area == desiredArea)
	.map(page => page.file.path)
	.values;

const query = `
not done
tags do not include #backlog

# Only first tasks of non parallel projects
# Filters out all tasks which match a path in the list but don't match the text of the first task
filter by function !JSON.parse(${JSON.stringify(JSON.stringify(sequentialProjectsTasks))}).find(entry => entry.path === task.file.path && entry.text !== task.description)

# Only tasks of projects in desired area
filter by function JSON.parse(${JSON.stringify(JSON.stringify(areaPagePaths))}).contains(task.file.path)

group by function task.tags
`;

dv.paragraph('```tasks\n' + query + '\n```');
```

[Silv-1]

I found you have to consider aspects of the description field:

For precise searches, it may help to know that description:

• first removes all each task's signifier emojis and their values,
• then removes any global filter,
• then removes an trailing spaces
• and then searches the remaining text

Source: https://publish.obsidian.md/tasks/Queries/Filters#Description

Dataview tasks text field neither removes signifier emojis with their values nor trailing spaces.

Therefore I changed entry.text === task.description to task.originalMarkdown.trim().endsWith(entry.text)