-
Notifications
You must be signed in to change notification settings - Fork 120
Docs
- Folder Structure
- Commands/Shortcuts
-
Understanding the Data Model
3.1 Data Types
3.2 Working with the Data Types
3.3 Views - Data Storage/Persistence
- Metaprogramming
- Cursor and Caret
- Drag and Drop
- Layout Rendering Algorithm
The main directory structure is organized as follows. Tests are located in a subdirectory named __tests__
in in each directory.
-
/src/App.css
- Legacy styles. New components should use inline styles. -
/src/actions
- Redux reducers and action-creators are co-located. Prefer reducers when possible, as they are pure functions that are more easily testable and composable. Only define an action creator if it requires a side effect. Use util/reducerFlow to compose reducers. -
/src/constants.ts
- Constant values. For constants that are only used in a single module, start by defining them in the module itself. They can be moved toconstants.js
if they need to be used in multiple modules or if there is a strong case to define them separately, e.g. an app-wide configuration that may need to be changed or tweaked. -
/src/components
- React components -
/src/hooks
- React hooks -
/src/redux-enhancers
- Redux enhancers -
/src/redux-middleware
- Redux middleware -
/src/selectors
- Select, compute, and possibly memoize slices from the Redux state. -
/src/shortcuts
- Keyboard and gesture shortcuts -
/src/util
- Miscellaneous
There are more than 50 commands within the app that are available to the user for editing, navigating, and other activities. Each can be activated with a toolbar button, keyboard shortcut, and/or gesture. The command palette brings up a list of all commands on desktop (Command/Ctrl + P).
- Commands: https://github.com/cybersemics/em/wiki/Commands
- Source: https://github.com/cybersemics/em/tree/main/src/shortcuts
We describe em to users in terms of creating, editing, and organizing their thoughts.
Thought
is the data type for a specific thought in the tree. Thoughts are stored in the Redux state as state.thoughts.thoughtIndex
. Individual thoughts can be retrieved with the selector getThoughtById
.
The rank
of a Thought
is a number
used to determine the thought's sort order among its siblings.
Ranks are unique within a single context. There is no relationship between ranks across contexts.
Ranks are relative; the absolute value does not matter. What matters is only if a rank is greater than or less than other ranks in the same context.
A new thought will be assigned a rank depending on where it is inserted:
- at the end of a context → rank of last thought + 1
- at the beginning of a context → rank of first thought - 1 (may be negative!)
- in the middle of a context → rank halfway between surrounding siblings (may be fractional!)
Negative ranks allow new thoughts to be efficiently inserted at the beginning of a context without having to modify the ranks of all other siblings. e.g. If a thought is placed before a thought with ranks 0
, it will be assigned a rank of -1
.
Fractional ranks allow new thoughts to efficiently be inserted between any two siblings without having to modify the ranks of other siblings. e.g. If a thought is placed between thoughts with ranks 5
and 6
, it will be assigned a rank of 5.5
.
importJSON
autoincrements the ranks of imported thoughts across contexts for efficiency and may result in different ranks that would be produced by manually adding the thoughts, but the sibling-relative ordering will be the same.
The word context refers to the ancestor path of a thought. e.g. c
is in the context a/b
:
- a
- b
- c
/** A sequence of thoughts from contiguous contexts. */
type Path = ThoughtId[]
e.g. ['kv9a-vzva-ac4n', '2mv0-atk3-tjlw', 'vkwt-ftz1-094z']
may represent the Path
for a/b/c
:
- a
- b
- c
- d
A Path
always starts at the ROOT thought. The ROOT thought is not part of the Path
itself, but is implied as the starting point. i.e. The first id of a Path
(e.g. kv9a-vzva-ac4n
) is a child of the ROOT thought. The Path
representing the ROOT thought itself is a special case defined as [HOME_TOKEN]
.
An important Path
in em is the thought that is being edited: state.cursor
. When the user clicks on thought A
, state.cursor
will be set to [idOfA]
. Navigating to a subthought will append the child's id to the cursor
. So hitting ArrowDown
on A
will set the cursor to [idOfA, idOfB]
, etc.
Circular Paths
are allowed. This is possible because of the Context View, described below, which allows jumping across the hierarchy.
/** A contiguous Path with no cycles. */
export type SimplePath = Path & Brand<'SimplePath'>
A SimplePath
is a Path
that has not crossed any Context Views, and thus has no cycles. Typescript is not expressive enough to capture this property in a type, but we can use brand types to require explicit casting, thus minimizing the chance of using a Path
with cycles when a SimplePath
is required. A Brand type is a nominal type that disallows implicit conversion. See: https://spin.atomicobject.com/2018/01/15/typescript-flexible-nominal-typing/.
/** An object that contains a list of contexts where a lexeme appears in
different word forms (plural, different cases, emojis, etc). */
export interface Lexeme {
contexts: ThoughtId[],
created: Timestamp,
lastUpdated: Timestamp,
updatedBy: string
}
A Lexeme
stores all the contexts where a thought appears in identical and near-identical word forms (ignoring case, plurality, emojis, etc). Lexemes are stored in state.thoughts.lexemeIndex
and keyed by a hash of the thought value. Think of these as the inbound links to a thought.
e.g. The Lexeme
for cat contains two contexts, Animals
and Socrates
:
- Animals
- Cats
- Dogs
- My Pets
- Socrates
- cat
Usage Tip: Use the getLexeme
selector to get the Lexeme
for a thought value.
There are several common selectors that are used to access a parent, child, or sibling of a thought. Consider the following thought structure:
- a
- b
- c
- x
- y
- z
- To get the parent
Thought
ofThought
c
:thoughtB ≅ getThoughtById(state, thoughtC.parentId)
. See: getThoughtById. - To get the parent
Path
ofPath
a/b/c
:pathAB ≅ rootedParentOf(state, pathABC)
. See: rootedParentOf.- Note:
rootedParentOf
always return a validPath
. If passed a child of the ROOT context,rootedParentOf
returns[HOME_TOKEN]
. If parentOf is passed a child of the ROOT context, it returns an empty array, which is not a validPath
.parentOf
should only be used if one or more additional thoughts are immediately appended, e.g.appendToPath(parentOf(path), thoughtId)
. This is not currently reflected in the return type ofparentOf
.
- Note:
- To get the children of
Thought
c
:childrenOfC ≅ getAllChildrenAsThoughtsById(state, thoughtC.id)
. If you have theContext
ofc
rather than its ID, you can use:childrenOfC ≅ getAllChildren(state, contextABC)
. See: getChildren.ts. - To get the next sibling of
Thought
y:thoughtZ ≅ nextSibling(state, 'y', contextABC)
. See: nextSibling. - To get the previous sibling of
Thought
y:thoughtX ≅ prevSibling(state, 'y', contextABC)
. See: prevSibling.
Those are just the most basic. There are many selectors and util functions which can be used to traverse, navigate, and convert between Thought
, ThoughtId
, Context
, and Path
:
- basic traversal
- parentOfThought - Returns the parent Thought of a given ThoughtId.
- nextSibling/prevSibling - Gets the next/previous sibling of a thought, according to its parent's sort preference.
-
rootedParentOf - Gets the parent Context/Path of a given Context/Path. If passed a child of the root thought, returns
[HOME_TOKEN]
or[ABSOLUTE_TOKEN]
as appropriate. - parentOf - Gets the parent of a Context or Path. Use rootedParentOf instead unless you are immediately appending an additional thoughtnn
-
appendToPath - Appends one or more child nodes to a
Path
or SimplePath. Ensures the root thought is removed.
- children
- getChildren - Gets all visible children of a Context, unordered.
- getAllChildren - Returns the subthoughts (as ThoughtIds) of the given context unordered. If the subthoughts have not changed, returns the same object reference.
- getAllChildrenAsThoughts - Returns the subthoughts (as Thoughts) of the given context unordered.
- getAllChildrenAsThoughtsById - Returns the subthoughts (as Thoughts) of the given ThoughtId unordered.
- getAllChildrenSorted - Gets all children of a Context sorted by rank or sort preference.
- getChildrenRanked - Gets all children of a Context sorted by their ranking. Returns a new object reference even if the children have not changed.
- getChildrenRankedById - Gets all children of a ThoughtId sorted by their ranking. Returns a new object reference even if the children have not changed.
- getChildrenSorted - Gets all visible children of a Context sorted by rank or sort preference.
- lookup/conversion
- getThoughtById - Gets a Thought by its ThoughtId.
- pathToThought - Gets the head Thought of a path.
- pathToContext - Converts a Path to a Context.
-
contextToPath - DEPRECATED: Converts a Context to a Path. This is a lossy function! If there is a duplicate thought in the same context, it takes the first. It should be removed. Build up the
Path
from information that is already in scope, or use thoughtToPath instead. - thoughtToPath - Generates the Path for a Thought by traversing upwards to the root thought.
- head - Gets the last ThoughtId or value in a Path or Context.
- childIdsToThoughts - Converts a list of ThoughtIds to a list of Thoughts. If any one of the thoughts are not found, returns null.
- thoughtToContext - Generates the Context for a Thought by traversing upwards to the ROOT thought.
- contextToThought - Gets the head Thought of a context.
- contextToThoughtId - Recursively finds the thought represented by the context and returns the id. This is the part of the independent migration strategy. Will likely be changed to some other name later.
- unroot - Removes the root token from the beginning of a Context or Path.
- ancestors/descendants
- getAncestorBy - Traverses the thought tree upwards from the given thought and returns the first ancestor that passes the check function.
- getDescendantContexts - Generates a flat list of all descendant Contexts. If a filterFunction is provided, descendants of thoughts that are filtered out are not traversed.
- getDescendantThoughtIds - Generates a flat list of all descendant Paths. If a filterFunction is provided, descendants of thoughts that are filtered out are not traversed.
- ancestors - Returns a subpath of ancestor children up to the given thought (inclusive).
- isDescendant
- isDescendantPath
- predicates
- isHome - Returns true if the Thoughts or Path is the home context.
- isRoot - Returns true if the Thoughts or Path is the one of the root contexts.
- hasChild
- hasLexeme
- equalPath
- equalThoughtRanked
- equalThoughtSorted
- equalThoughtValue
- attributes
- lexemes
- getLexeme - Gets the Lexeme of a given value.
- getLexemeById - Gets the Lexeme at the given lexemeIndex key.
- getContexts - Returns all of the Thoughts of a thought's Lexeme.
- context view
-
appendChildToPath - Appends the head of a child
SimplePath
to a parentPath
. In case of a parent with an active context view, it appends the head of the parent of the childPath. - getChildPath
- simplifyPath - Infers the path from a path that may cross one or more context views.
-
appendChildToPath - Appends the head of a child
In normal view (default), a thought's children are rendered in a collapsible tree.
- a
- m [cursor]
- x
- y
- b
- m
- y
- z
Enter Option + Shift + S
, or click the button in the toolbar to activate the Context View on a given thought. This will show all the contexts that a thought appears in.
e.g. Given a normal view...
- a
- m
- x
- y
- b
- m
- y
- z
Activating the context view on m
(indicated by ~
) renders:
- a
- m~ [cursor]
- a
- b
- y
- z
- b
- m
- y
- z
a
and b
are listed under a/m~
because they are the contexts that m
appears in. They are the inbound links to m
, as opposed to the outbound links that are rendered from a context to a child.
Note: The ranks of the contexts are autogenerated and do not correspond with the rank of the thought within its context, but rather the sorted order of the contexts in the context view.
Usage Tip: Use getContexts()
or getThought(...).contexts
to get the contexts for a thought value.
Descendants of contexts within a context view are rendered recursively. The Child
thoughts that are generated from the list of contexts mentioned above can render their own Child
thoughts (defaulting to Normal View). But what Context
to use? When the parent is in Normal View, a Path
is converted to a Context
. When the parent, is in Context View, e.g. ['a', 'm']
, we have direct access to the Context
not from a Path
but from getContexts('m')
: [{ context: ['a'], rank: 0 }, { context: ['b'], rank: 1 }]
. We then combine the desired Context
with the head thought to render the expected Child
thoughts. See the following example.
Note: The cursor
here is circular. The underlying data structure allows for cycles. This is possible because only a fixed number of levels of depth are shown at a time.
(~
indicates context view)
- a
- m~
- a [cursor]
- x
- y
- b
- b
- m
- y
- z
The ThoughtContexts
for m
are [{ context: ['a'], rank: 0 }, { context: ['b'], rank: 1 }]
. Where do x
and y
come from? They are the children of ['a', 'm']
. When the Context View of m
is activated, and the context ['a']
is selected, it renders the children of ['a', 'm']
.
When working with Context Views, it is necessary to switch between the full Path
that crosses multiple Context Views, and the contiguous SimplePath
segments that make it up. This is the only way to get from a Path
that crosses multiple Context Views to a single Context
, which does not allow cycles.
This more verbose and explicit representation of a transhierarchical Path
and its different Context View boundaries is called a contextChain
. contextChain
is not stored in state
, but derived from the cursor
via splitChain(state, cursor)
. Consider the following:
- a
- m
- x
- m
- b
- m
- y
- m
When cursor
is a/m~/b/y
, then contextChain
is (ranks omitted for readability):
[
['a', 'm'],
['b', 'y']
]
That is, the cursor
consists of the initial segment a/m
, then we enter the Context View of m
, then b/y
.
This allows the cursor
to move across multiple Context Views. A more complicated example (copy and paste into em to test):
- Books
- Read
- C. S. Peirce
- Philosophical Writings
- Three Categories
- Philosophy of Math
- Philosophy Logic
- Semiotics
- Personal
- Influences
- Gregory Bateson
- Michael Polanyi
- C. S. Peirce
- Philosophy
- Philosophy of Math
- Statistical Inference
- Probability as Potential
- Philosophy of Science
- Metaphysics
- Sri Aurobindo
- Forrest Landry
- Potentiality
- Probability as Potential
- Potentiality vs Actuality
The cursor /Books/Read/C.S. Peirce/Philosophical Writings/Philosophy of Math~/Philosophy/Probability as Potential~/Potentiality/Potentiality vs Actuality
spans two Context Views (Philosophy of Math
and Probability as Potential
), thus there are three segments in the contextChain
:
[
['Books', 'Read', 'C. S. Peirce', 'Philosophical Writings', 'Philosophy of Math'],
['Philosophy', 'Probability as Potential'],
['Potentiality', 'Potentiality vs Actuality'],
]
Other functions related to contextChain
are:
Thoughts are stored in the underlying objects lexemeIndex
and thoughtIndex
, which map hashed values to Lexemes
and ids to Thoughts
, respectively. Only visible thoughts are loaded into state. The pullQueue
is responsible for loading additional thoughts into state that need to be rendered.
- state (Redux)
- local (IndexedDB via Dexie)
- remote (Firebase) [optional; if user is logged in]
The syncing and reconciliation logic is done by pull
, push
, reconcile
, and updateThoughts
.
Metaprogramming provides the ability to alter em's behavior from within em itself through hidden subthoughts called metaprogramming attributes. Metaprogramming attributes begin with =
and are hidden unless showHiddenThoughts
is toggled on from the toolbar. Generally an attribute will affect only its parent context.
Note: User settings are stored as metaprogramming thoughts within [EM, 'Settings']
. See INITIAL_SETTINGS for defaults.
List of possible metaprogramming attributes:
-
=bullets
Hide the bullets of a context. Options:Bullets
,None
. -
=children
Apply attributes to all children. Currently only works with=style
and=bullets
. e.g. This would makeb
andc
the colortomato
:- a - =children - =style - color - tomato - b - c
-
=focus
When the cursor is on this thought, hide parent and siblings for additional focus. Options:Normal
,Zoom
. -
=hidden
The thought is only displayed whenshowHiddenThoughts === true
. -
=immovable
The thought cannot be moved. -
=label
Display alternative text, but continue using the real text when linking contexts. Hide the real text unless editing. -
=note
Display a note in smaller text underneath the thought. -
=options
Specify a list of allowable subthoughts. -
=pin
Keep a thought expanded. Options:true
,false
. -
=pinChildren
Keep all thoughts within a context expanded. Options:true
,false
. -
=readonly
The thought cannot be edited, moved, or extended. -
=style
Set CSS styles on the thought. May also use=children/=style
or=grandchildren/=style
. -
=uneditable
The thought cannot be edited. -
=unextendable
New subthoughts may not be added to the thought. -
=view
Controls how the thought and its subthoughts are displayed. Options:List
,Table
,Prose
.
More information: https://github.com/cybersemics/em/wiki/Browser-Selection
[TODO: Consolidate browser selection documentation]
The cursor is a dark gray circle surrounding the bullet of the active thought. It is stored as a Path
in state.cursor
. Only one thought can have the cursor at a time. All shortcuts operate on the cursor thought or its children, so it serves as the main point of interaction for the user while editing. The cursor is not the browser selection (see below), however the cursor thought contains the browser selection (caret) while editing.
You can call setCursor
to set state.cursor
. setCursor
does not set the browser selection, although it does maintain some state for the position of the caret in caretOffset
.
The caret is the native browser selection, i.e. window.getSelection()
. We use the name "caret" because it is shorter, and is distinguishable from "cursor". Unless otherwise specified, the caret refers to a browser selection that is collapsed, i.e. no text is selected.
Caret position is set by selection.set(...)
. This is typically handled automatically by the Editable
component. Each Editable
instance checks if the caret should be active on that thought when it is rendered. That is, it maintains the browser selection even when thoughts are re-rendered during navigation. There are other checks related to edit mode on mobile, drag-and-drop, etc.
On desktop, the caret is always on the cursor thought.
On mobile, the caret is only set when in edit mode. Otherwise the cursor
changes without any browser selection. This allows the user to navigate thoughts without opening the virtual keyboard. Edit mode is stored in state.editing
. When the user closes their mobile keyboard, state.editing
is set to false.
- To enter
editing
mode, the user taps on the cursor thought or activates a shortcut that modifies a visible thought, such asnewThought
,clearText
,subcategorizeOne
, etc. - Tapping on a non-cursor thought while not in edit mode will not activate edit mode.
- To close
editing
mode, the user closes the virtual keyboard or navigates to the root.
em uses the react-dnd library for drag-and-drop functionality.
There there a variety of components that utilize drag-and-drop.
DragAndDropToolbarButton - Defines canDrag
, beginDrag
, endDrag
, canDrop
and drop
for dragging or dropping a toolbar button when customizing the toolbar.
Screen.Recording.2024-05-15.at.3.58.06.PM.mov
DragAndDropThought - Defines canDrag
, beginDrag
, endDrag
, canDrop
and drop
for dragging or dropping a thought. Used by the Thought
component so that every thought can be dragged, and every thought can serve as a drop target (inserted before, i.e. as the previous sibling).
Screen.Recording.2024-05-15.at.3.22.30.PM.mov
DropChild - Allows a thought to be dropped as a subthought of a thought that is empty or collapsed.
Screen.Recording.2024-05-15.at.3.02.28.PM.mov
DropEnd - Allows a thought to be dropped at the end of a list of subthoughts.
Screen.Recording.2024-05-15.at.3.08.43.PM.mov
Many DropEnd components can be rendered consecutively when there is a cliff, i.e. the depth decreases by more than 1:
Screen.Recording.2024-05-15.at.3.29.05.PM.mov
DropUncle - Allows a thought to be dropped before the next hidden uncle. For example, in the video below, the thought c
is dragged onto a DropUncle
drop target that allows the user to drop the thought before the next hidden uncle e
(after the hidden parent a
).
Screen.Recording.2024-06-22.at.6.40.05.PM.copy.mov
Favorites (also contains a drop target) - Defines canDrag
, beginDrag
, endDrag
, canDrop
and drop
for dragging or dropping a thought within the favorites panel.
Screen.Recording.2024-05-15.at.3.55.22.PM.mov
These drop targets slide out from the right side of the screen as soon as a drag begins, providing an alternative means of executing a command on a thought.
Screen.Recording.2024-05-15.at.3.31.11.PM.mov
DragAndDropContext - Provide drag-and-drop context to the entire component hierarchy.
DragOnly - A container fragment that only renders its children when state.dragInProgress
is true. Strictly performance related. Useful for short circuiting child components with expensive selectors.
DropHover - Renders a blue bar at the insertion point when a valid drop target is being hovered over.
DragAndDropSubthoughts - Defines canDrop
and drop
for dropping a thought as a subthought (i.e. child). Used by DropChild
and DropEnd
since they both involve dropping a thought as a subthought.
Rather than rely on the hierarchy of the DOM, thoughts are rendered as a flat list of siblings, and are absolutely positioned to create a visual hierarchy. This allows smooth animation of thoughts across levels between any arbitrary states.
The rendering algorithm is contained in the TreeLayout component.
Each rendering of the tree generates two lists of nodes:
-
treeThoughts
- TheLayoutTree
does an in-order traversal of the thought tree from the Redux state, generating a linear sequence of all visible thoughts from top to bottom. EachTreeThought
includes metadata such as depth, whether it has any children, if it's in a table view, etc, that are needed to render the thought. It is recalculated whenever the Redux state changes. -
treeThoughtsPositioned
- Next, the x,y coordinate are calculated for each visible thought. The accumulated y value is incremented by the height of each thought so that the next thought is rendered in the correct position. When the depth increases, the x value is increased (indent), and when the depth decreases, the x value is decreased (outdent). This list is recalculated whenever the width or height of a thought changes.
Thoughts are rendered in two passes:
- The first render positions all thoughts with an estimated height based on the current font size.
- Next, the VirtualThought component measures each thought's height in a layout effect and passes it up to the
LayoutTree
viaonResize
. If any of the heights differ from their original estimate, a re-render is triggered. - The second render recalculates
treeThoughtsPositioned
with each thought's measured height. Any updates to thought y coordinates are smoothly animated into place via a CSS transition.
[In progress]
linearizeTree
useSizeTracking ->
useSingleLineHeight
useTreeThoughtsPositioned
useAutofocusViewport <- useNavAndFooterHeight
div - vertical (instant)
div - horizontal (slow)
for each node
div - absolutely positioned for animating across levels
VirtualThought
DropCliff
The table view is rendered within the same algorithm, with some differences to how the x,y coordinates are calculated.
Take the following table:
- a
- =view
- Table
- b
- b1
- b2
- c
- d
- e
In the first step (treeThoughts
), properties are added for each node that is part of a table.
e.g.
-
isTable
: a -
isTableCol1
: b, c -
isTableCol2
: b1, b2, d -
isTableCol2Child
: e
In the second step (treeThoughtsPositioned
), the x,y coordinates are generated for each thought.
- After col1 is rendered (e.g.
b
), the x coordinate is incremented by the thought's width, and the y coordinate is held at the same value, so that col2 is rendered to the right of col1 (e.g.b1
). - Additional thoughts in col2 (e.g.
b2
) are rendered one after another just like a normal list of thoughts. - The next thought in col1 (e.g.
c
) is rendered at the x coordinate of its previous sibling and a y coordinate that clears the height of both columns in the previous row (e.g.max(b, b1 + b2)
).
To support nested thoughts, the width of col1 thoughts are accumulated in tableCol1Widths
so that deeper thoughts can be rendered at a deeper x coordinate. In combination with ycol1Ancestors
, which stores the depth and y coordinate of col1 thoughts, this allows any level of nesting, including nested tables.
Lastly, the the entire tree is shifted left as the cursor moves deeper. Combined with the autofocus functionality which fades out ancestors the deeper the user moves, this allows the user to navigate throughout their thoughtspace while keeping the cursor and nearby thoughts front and center. The variables indentDepth
and indentCursorAncestorTables
are used to calculate how much the tree is shifted left.