Page Menu
Home
WickedGov Phorge
Search
Configure Global Search
Log In
Files
F7433834
tinyve.dm.converter.js
No One
Temporary
Actions
Download File
Edit File
Delete File
View Transforms
Subscribe
Flag For Later
Award Token
Size
4 KB
Referenced Files
None
Subscribers
None
tinyve.dm.converter.js
View Options
/**
* TinyVE converter — convert between html DOM and linear model representations
*
* This is a toy version of ve.dm.Converter which illustrates the main concepts
*/
tinyve
.
dm
.
converter
=
{};
/**
* Test whether a DOM node should be represented as a text annotation in the linear model
*
* In full VE, this is done with a model registry so extensions can add types
*
* @see {ve.dm.ModelRegistry#isAnnotation}
* @param {Node} node A dom node, like `p` or `b` or `img`
* @return {boolean} Whether the node should be represented as a text annotation in the linear model
*/
tinyve
.
dm
.
converter
.
isNodeAnnotation
=
function
isNodeAnnotation
(
node
)
{
return
node
.
tagName
&&
[
'b'
,
'i'
,
'a'
].
includes
(
node
.
tagName
.
toLowerCase
()
);
};
/**
* Test whether a DOM node's contents should be linearized when creating the linear model
*
* In full VE, this is done with a node factory so extensions can add types
*
* @see {ve.dm.NodeFactory#canNodeContainContent}
* @param {Node} node A DOM node, like p or div
* @return {boolean} Whether the node's contents should be linearized when creating the linear model
*/
tinyve
.
dm
.
converter
.
canNodeContainContent
=
function
canNodeContainContent
(
node
)
{
return
node
.
tagName
&&
[
'p'
,
'h1'
,
'h2'
,
'h3'
].
includes
(
node
.
tagName
.
toLowerCase
()
);
};
/**
* Create ContentBranchNode linearized data from a DOM node like p, h1, h2 etc
*
* This is only for nodes that should have linearized content (like p, h1 or h2). Linearization
* is NOT used for "structural" nodes like `div` or `li`. These cannot directly contain content.
*
* In full VE, this process cannot be cleanly separated from converting structural nodes,
* because we auto-generate paragraphs if we do encounter, say, a text node directly inside a
* div. The requirement to round-trip whitespace makes that code very fiddly.
*
* Here we represent annotations directly strings like '<b>' or '<a href="foo">', but real VE
* represents them with hashes like 'h8e79483a29e62bdb' that are keys in a HashValueStore
* owned by the document. The corresponding values are instances of ve.dm.Annotation, e.g.
* ve.dm.BoldAnnotation and ve.dm.LinkAnnotation.
*
* @see {ve.dm.Converter#getDataFromDomSubtree}
* @see {ve.dm.HashValueStore}
* @see {ve.dm.Annotation}
* @param {Node} contentBranchDomNode A DOM node for a p, h1, h2 etc
* @param {string[]} annotations List of open annotations, e.g. [ '<b>', '<a href="foo">' ]
* @return {Array} Linear data
*/
tinyve
.
dm
.
converter
.
linearize
=
function
linearize
(
contentBranchDomNode
,
annotations
=
[]
)
{
function
getOpenTag
(
node
)
{
return
node
.
cloneNode
().
outerHTML
.
replace
(
/(<.*?>).*/
,
'$1'
);
}
const
data
=
[];
Array
.
from
(
contentBranchDomNode
.
childNodes
).
forEach
(
(
childNode
)
=>
{
if
(
childNode
.
nodeType
===
3
)
{
// Text node
const
chars
=
Array
.
from
(
childNode
.
textContent
);
if
(
annotations
.
length
===
0
)
{
// Push each character separately, e.g. 'f', 'o', 'o'
data
.
push
(
...
chars
);
}
else
{
// Push each character separately, each annotated, e.g.
// [ 'f', [ '<b>', '<a href="foo">' ] ],
// [ 'o', [ '<b>', '<a href="foo">' ] ],
// [ 'o', [ '<b>', '<a href="foo">' ] ]
const
annotatedChars
=
chars
.
map
(
(
ch
)
=>
[
ch
,
annotations
.
slice
()
]
);
data
.
push
(
...
annotatedChars
);
}
}
else
if
(
tinyve
.
dm
.
converter
.
isNodeAnnotation
(
childNode
)
)
{
annotations
.
push
(
getOpenTag
(
childNode
)
);
data
.
push
(
...
tinyve
.
dm
.
converter
.
linearize
(
childNode
,
annotations
)
);
annotations
.
pop
();
}
else
{
throw
new
Error
(
'Unsupported node'
);
}
}
);
return
data
;
};
/**
* Create linear data for a DOM subtree
*
* @see {ve.dm.Converter#getDataFromDomSubtree}
* @param {Node} node The DOM node to process, e.g. document.body
* @return {Array} Linear data
*/
tinyve
.
dm
.
converter
.
getDataFromDomSubtree
=
function
getDataFromDomSubtree
(
node
)
{
const
data
=
[];
Array
.
from
(
node
.
childNodes
).
forEach
(
(
childNode
)
=>
{
if
(
childNode
.
nodeType
===
3
)
{
// Text node outside p/h1/h2 etc (e.g. whitespace) ... just ignore(!)
return
;
}
const
tag
=
childNode
.
tagName
.
toLowerCase
();
data
.
push
(
{
type
:
tag
}
);
if
(
tinyve
.
dm
.
converter
.
canNodeContainContent
(
childNode
)
)
{
data
.
push
(
...
tinyve
.
dm
.
converter
.
linearize
(
childNode
)
);
}
else
{
data
.
push
(
...
tinyve
.
dm
.
converter
.
getDataFromDomSubtree
(
childNode
)
);
}
data
.
push
(
{
type
:
'/'
+
tag
}
);
}
);
return
data
;
};
File Metadata
Details
Attached
Mime Type
text/html
Expires
Sun, Nov 22, 12:06 (1 h, 52 m)
Storage Engine
local-disk
Storage Format
Raw Data
Storage Handle
7d/ce/3c53080ac602acdcbe62c9183831
Default Alt Text
tinyve.dm.converter.js (4 KB)
Attached To
Mode
rMWPROD MediaWiki Production
Attached
Detach File
Event Timeline
Log In to Comment