Markdown Package #2937
Markdown Package #2937
Conversation
juliaroldi
requested review from
JiuqingSong,
flyingbee2012,
BryanValverdeU,
ianeli1 and
vinguyen12
| text: string, | ||
| blockquote?: ContentModelFormatContainer | ||
| ): ContentModelFormatContainer { | ||
| text = text.replace('>', ''); |
There was a problem hiding this comment.
replace function only replace the first element found, to replace all we need to use />/g
| * @internal | ||
| */ | ||
| export function applyLink(textSegment: ContentModelText): ContentModelText { | ||
| const linkRegex = /\[([^\[]+)\]\((https?:\/\/[^\s\)]+)\)/g; |
There was a problem hiding this comment.
There is already regex matches in splitLinkAndImages, so I think we just need to match once there, then create segments at once, no need to split then match again.
| text: string, | ||
| splitLinesPattern?: string | ||
| ) { | ||
| editor.formatContentModel( |
There was a problem hiding this comment.
Suggest this API does not have dependency to editor. It should just take string as input, and ContentModelDocument as output
There was a problem hiding this comment.
This whole package should not have dependency to editor
| * @internal | ||
| */ | ||
| export function createImageSegment(textSegment: ContentModelText): ContentModelImage | undefined { | ||
| const imageRegex = /\!\[([^\[]+)\]\((https?:\/\/[^\)]+)\)/g; |
| /** | ||
| * @internal | ||
| */ | ||
| export function createListFromMarkdown(text: string, patternName: string): ContentModelListItem { |
There was a problem hiding this comment.
better make the patternName strong-typed
| */ | ||
| export function createListFromMarkdown(text: string, patternName: string): ContentModelListItem { | ||
| const listType = patternName === 'ordered_list' ? 'OL' : 'UL'; | ||
| const itemText = text.trim().substring(listType == 'OL' ? 2 : 1); |
There was a problem hiding this comment.
why 2? what if list item "100"
| */ | ||
| export function createTableFromMarkdown(tableLines: string[]): ContentModelTable { | ||
| const tableDivider = tableLines[1].split('|').filter(content => content.trim() !== ''); | ||
| const tableModel = tableLines.filter(line => tableLines.indexOf(line) !== 1); |
There was a problem hiding this comment.
this looks weird. If you want to remove lines[1], just use tableLines.splice(1,1)
This change introduces the
rooosterjs-content-model-markdown package, which is responsible for converting plain text written in Markdown into a content model structure. All the basic syntax except for alternative heading, HTML and code are supported. The package provides a single public function,convertMarkdownToContentModel, which handles the conversion of plain text and inserts it into the editor.The conversion process involves dividing the plain text by line breaks and identifying the appropriate content model blocks based on the Markdown syntax. The text is divided by the regex
/\r\n|\r|\\n|\n/, but the splitLinesPattern parameter allows to replace this regex with another one.All the created content model blocks are inserted into a new model document in the order they appear in the plain text, and after all the plain text is processed, the new content model document is merged in the editor.