#### Is this adding or improving a _feature_ or fixing a _bug_?
Example.
#### What's the new behavior?
Adds a new example that is an input event logger, for more easily seeing which input/keyboard/selection events are firing when editing in a Slate editor.
#### Have you checked that...?
* [x] The new code matches the existing patterns and styles.
* [x] The tests pass with `yarn test`.
* [x] The linter passes with `yarn lint`. (Fix errors with `yarn prettier`.)
* [x] The relevant examples still work. (Run examples with `yarn watch`.)
#### Is this adding or improving a _feature_ or fixing a _bug_?
Bug.
#### What's the new behavior?
Fixes selection operations from being duplicated.
#### How does this change work?
Previously the selection properties were compared by reference, but paths are immutable `List` objects, which always show up as having changed, resulting in extra selection operations that without any real changes. We now use `Immutable.is` to remove those duplicates, fixing the undo history stack.
#### Have you checked that...?
<!--
Please run through this checklist for your pull request:
-->
* [x] The new code matches the existing patterns and styles.
* [x] The tests pass with `yarn test`.
* [x] The linter passes with `yarn lint`. (Fix errors with `yarn prettier`.)
* [x] The relevant examples still work. (Run examples with `yarn watch`.)
#### Does this fix any issues or need any specific reviewers?
Fixes: #2006
#### Is this adding or improving a _feature_ or fixing a _bug_?
Improvement.
#### What's the new behavior?
- Tweaking the declarative schema definition syntax to make it easier to represent more complex states, as well as enable it to validate previously impossible things.
- Rename `validateNode` to `normalizeNode` for clarity.
- Introduce `validateNode`, `checkNode`, `assertNode` helpers for more advanced use cases, like front-end API validation of "invalid" fields that need to be fixed before they are sent to the server.
#### How does this change work?
The `schema.blocks/inlines/document` entries are now a shorthand for a more powerful `schema.rules` syntax. For example, this now allows for declaratively validating by a node's data, regardless of type:
```js
{
rules: [
{
match: {
data: { id: '2kd293lry' },
},
nodes: [
{ match: { type: 'paragraph' }},
{ match: { type: 'image' }},
]
}
]
}
```
Previously you'd have to use `validateNode` for this, since the syntax wasn't flexible enough to validate nodes without hard-coding their `type`.
This also simplifies the "concatenation" of schema rules, because under the covers all of them are implemented using the `schema.rules` array, so they simply take effect in order, just like everything else in plugins.
#### Have you checked that...?
<!--
Please run through this checklist for your pull request:
-->
* [x] The new code matches the existing patterns and styles.
* [x] The tests pass with `yarn test`.
* [x] The linter passes with `yarn lint`. (Fix errors with `yarn prettier`.)
* [x] The relevant examples still work. (Run examples with `yarn watch`.)
#### Does this fix any issues or need any specific reviewers?
Fixes: #1842Fixes: #1923
#### Is this adding or improving a _feature_ or fixing a _bug_?
Feature.
#### What's the new behavior?
This pull request adds paths to `Range` objects, including the selection. The paths and keys are kept in sync automatically, so that you can use whichever is ideal for your use case.
This should allow us to use paths for lots of the internal logic, which are much quicker to work with than keys since they avoid having to lookup the key in the document and can just traverse right to the node in question.
#### How does this change work?
`Range` objects have two new properties:
```js
range.anchorPath
range.focusPath
```
(Eventually these will be `range.anchor.path` and `range.focus.path` when points are introduced.)
When operations occur and whenever ranges are created/normalized, the paths are updated and kept in sync with the keys.
#### Have you checked that...?
<!--
Please run through this checklist for your pull request:
-->
* [x] The new code matches the existing patterns and styles.
* [x] The tests pass with `yarn test`.
* [x] The linter passes with `yarn lint`. (Fix errors with `yarn prettier`.)
* [x] The relevant examples still work. (Run examples with `yarn watch`.)
#### Does this fix any issues or need any specific reviewers?
Fixes: https://github.com/ianstormtaylor/slate/issues/1408
Fixes: https://github.com/ianstormtaylor/slate/issues/1567
* initial simple decorations (mark-like), many tests added
* allow decorators to be set by focus, anchor tags - add tests
* handle one more edge case with decorations in hyperscript
* apply prettier cleanup
* apply linting rules
* update changelog
* ensure always normalize decoration ranges
* reapply prettier after latest adjustments
* all operations apply now update decorations with selection
* ranges can now be 'atomic', will invalidate if contents change
* lint, prettier cleanups
* add atomic invalidation tests, update hyperscript usage
* fix linter errors
* minor cleanup
* slight refactor for simplicity
* remove a couple superfluous lines
* update in response to review
* drop unnecessarily committed add'l file
* remove the need for explicit anchor, focus prop on decoration tags
* update hyperscript use to match latest syntax in #1777
* atomic -> isAtomic
* Fix spell check bug by add data-text:true
* Fix spell check bug by spell check add length to a leaf
* Fix tests to use data-text:true for marks
* Rename data-text to data-slate-leaf; Remove setRef; unlift attributes in leaf
* Update examples with data-*
* Add attributes to document
* Fix renderMark in all documents
* Prettier markdown
* Set up webpack configuration for building examples
* Configure react-hot-loader in development
* Improve config, set gh-pages to use examples/dist directory
* PR feedback
* Rename App.js in git
* add ability to click multiple emojis
currently, after clicking one emoji, user needs to click in the editor before being able to add another emoji
this fixes that by jumping to next text block and focusing selection to make it obvious where next insertion will take place.
* lint code
* Add Prettier, with basic config and ESLint integration
* Apply Prettier to all files using `yarn lint --fix`
* Tell Prettier to ignore an empty text in a test output.
* Run Prettier on JS files not handled by ESLint, and lint them too
* Implement first working rollup config for main slate pkg
* Convert slate-base64-serializer to rollup
* Convert slate-dev-logger to rollup
* Convert slate-html-serializer to rollup
* Convert slate-hyperscript to rollup
* Convert slate-plain-serializer to rollup
* Convert slate-prop-types to rollup
* Convert (mostly) slate-react to rollup
* Convert slate-simulator to rollup
* Misc cleanup and configuration tweaks/fixes
* Convert slate-schema-violations to rollup
* Successful rollup build for top-level examples
* Add plugin to replace process.env.NODE_ENV
* Only rebuild modules and dev examples in watch mode
* Enable sourcemaps for development builds
* Force debug to use browser version, remove builtins plugin
* Remove is-image from example
It relies on node `path` and wouldn't work well in-browser anyway
* Use browser version of react-dom/server
* Move stray require to import
* Configure examples to watch child package output
* Fix tests
* Remove unneeded preferBuiltins from resolve config
* Use more precise files array to ensure sourcemaps aren't included
* Use lodash instead of lodash.throttle
It's pulled in anyway since slate-react needs slate, so using the
minipackage actually causes code duplication
* Improve naming/fix UMD builds, update UMD doc
* Add rollup configs to linting, add a missing dep to package.json
* Use longform rollup CLI flags
* Add rollup-plugin-auto-external to reduce external module configuration
* Combine rollup config into a unioned helper
* Centralize to a single rollup configuration
* Update dist structure and package field naming for PR feedback
* Add comments and address PR feedback on rollup config
* i.e. -> e.g.
* Add some spacing to the configuration to improve readability
* Add a bit more spacing
* Remove umd from example Slate unpkg link
* added export constant enum for schema violations
* updated examples to use the schema violations enum
* use SchemaViolations enum in tests and docs
* fixed path for schema violations import
* rename state to value in slate core, as deprecation
* rename all references to state to value in slate core
* migrate slate-base64-serializer
* migrate slate-html-serializer
* migrate slate-hyperscript
* migrate slate-plain-serializer
* migrate slate-prop-types
* migrate slate-simulator
* fix change.setState compat
* deprecate references to state in slate-react
* remove all references to state in slate-react
* remove `value` and `schema` from props to all components
* fix default renderPlaceholder
* fix tests
* update examples
* update walkthroughs
* update guides
* update reference
* Include Array.prototype.includes polyfill
In https://github.com/ianstormtaylor/slate/commit/070a700
the `default` feature set was added
and the `Array.prototype.includes` feature removed
from the polyfill.io request.
I think it's great to include the `default` feature set,
but, `Array.prototype.includes` is not included in the `default` set,
so we have to explicitly request it.
IE 11 needs an `Array.prototype.includes` polyfill
in order for the examples to work.
* Update dev.html
* Update index.html
Slate uses `Array.prototype.includes`
in at least one place:
https://github.com/ianstormtaylor/slate/blob/b558872/packages/slate/src/schemas/core.js#L62
So, request an `Array.prototype.includes` polyfill
from polyfill.io
if the user's browser needs it,
which is the case with IE 11.
Also requests minified JavaScript,
since this code is used
for the slatejs.org website,
and there's no need to serve
non-minified code there.
