d73cd14dfb
* docs: add mermaid docs * fix link * add development.mdx * add codebase * add parseEdge * add subgraph * convertors * tweaks * structure the codebase * add docs for new diagram type * decouple docs for flowchart parser * fix broken links * use diagram instead of chart * revert package.json * tweaks * point root folder of mermaid-to-excalidraw to installation * remove installation from sidebar children * fix link * update flowchart parser diagram * update parser overview diagram * update more diagrams * remove note about change in element stroke * update diagram * increase size limit for excalidraw production bundle * tweaks * add link * move codebase after introduction * dummy --------- Co-authored-by: dwelle <luzar.david@gmail.com>
177 lines
5.0 KiB
Plaintext
177 lines
5.0 KiB
Plaintext
# Flowchart Parser
|
|
|
|
As mentioned in the previous section, we use [getDiagramFromText](https://github.com/mermaid-js/mermaid/blob/00d06c7282a701849793680c1e97da1cfdfcce62/packages/mermaid/src/Diagram.ts#L80) to parse the full defination and get the [Diagram](https://github.com/mermaid-js/mermaid/blob/00d06c7282a701849793680c1e97da1cfdfcce62/packages/mermaid/src/Diagram.ts#L15) json from it.
|
|
|
|
In this section we will be diving into how the [flowchart parser](https://github.com/excalidraw/mermaid-to-excalidraw/blob/master/src/parser/flowchart.ts#L256) works behind the scenes.
|
|
|
|
|
|
|
|
We use `diagram.parser.yy` attribute to parse the data. If you want to know more about how the `diagram.parse.yy` attribute looks like, you can check it [here](https://github.com/mermaid-js/mermaid/blob/00d06c7282a701849793680c1e97da1cfdfcce62/packages/mermaid/src/diagrams/flowchart/flowDb.js#L768), however for scope of flowchart we are using **3** APIs from this parser to compute `vertices`, `edges` and `clusters` as we need these data to transform to [ExcalidrawElementSkeleton](https://github.com/excalidraw/excalidraw/blob/master/src/data/transform.ts#L133C13-L133C38).
|
|
|
|
|
|
For computing `vertices` and `edge`s lets consider the below svg generated by mermaid
|
|
|
|
|
|
|
|
|
|
## Computing the vertices
|
|
|
|
We use `getVertices` API from `diagram.parse.yy` to get the vertices for a given flowchart.
|
|
|
|
Considering the same example this is the response from the API
|
|
|
|
```js
|
|
{
|
|
"start": {
|
|
"id": "start",
|
|
"labelType": "text",
|
|
"domId": "flowchart-start-1414",
|
|
"styles": [],
|
|
"classes": [],
|
|
"text": "start",
|
|
"props": {}
|
|
},
|
|
"stop": {
|
|
"id": "stop",
|
|
"labelType": "text",
|
|
"domId": "flowchart-stop-1415",
|
|
"styles": [],
|
|
"classes": [],
|
|
"text": "stop",
|
|
"props": {}
|
|
}
|
|
}
|
|
```
|
|
The dimensions and position is missing in this response and we need that to transform to [ExcalidrawElementSkeleton](https://github.com/excalidraw/excalidraw/blob/master/src/data/transform.ts#L133C13-L133C38), for this we have our own parser [`parseVertex`](https://github.com/excalidraw/mermaid-to-excalidraw/blob/master/src/parseMermaid.ts#L178) which takes the above response and uses the `svg` together to compute position, dimensions and cleans up the response.
|
|
|
|
The final output from `parseVertex` looks like :point_down:
|
|
|
|
```js
|
|
{
|
|
"start": {
|
|
"id": "start",
|
|
"labelType": "text",
|
|
"text": "start",
|
|
"x": 0,
|
|
"y": 0,
|
|
"width": 67.796875,
|
|
"height": 44,
|
|
"containerStyle": {},
|
|
"labelStyle": {}
|
|
},
|
|
"stop": {
|
|
"id": "stop",
|
|
"labelType": "text",
|
|
"text": "stop",
|
|
"x": 117.796875,
|
|
"y": 0,
|
|
"width": 62.3828125,
|
|
"height": 44,
|
|
"containerStyle": {},
|
|
"labelStyle": {}
|
|
}
|
|
}
|
|
```
|
|
|
|
|
|
## Computing the edges
|
|
|
|
The lines and arrows are considered as `edges` in mermaid as shown in the above diagram.
|
|
We use `getEdges` API from `diagram.parse.yy` to get the edges for a given flowchart.
|
|
Considering the same example this is the response from the API
|
|
|
|
```js
|
|
[
|
|
{
|
|
"start": "start",
|
|
"end": "stop",
|
|
"type": "arrow_point",
|
|
"text": "",
|
|
"labelType": "text",
|
|
"stroke": "normal",
|
|
"length": 1
|
|
}
|
|
]
|
|
```
|
|
|
|
Similarly here the dimensions and position is missing and we compute that from the svg. The [`parseEdge`](https://github.com/excalidraw/mermaid-to-excalidraw/blob/master/src/parseMermaid.ts#L245) takes the above response along with `svg` and computes the position, dimensions and cleans up the response.
|
|
|
|
The final output from `parseEdge` looks like :point_down:
|
|
|
|
```js
|
|
[
|
|
{
|
|
"start": "start",
|
|
"end": "stop",
|
|
"type": "arrow_point",
|
|
"text": "",
|
|
"labelType": "text",
|
|
"stroke": "normal",
|
|
"startX": 67.797,
|
|
"startY": 22,
|
|
"endX": 117.797,
|
|
"endY": 22,
|
|
"reflectionPoints": [
|
|
{
|
|
"x": 67.797,
|
|
"y": 22
|
|
},
|
|
{
|
|
"x": 117.797,
|
|
"y": 22
|
|
}
|
|
]
|
|
}
|
|
]
|
|
```
|
|
## Computing the Subgraphs
|
|
|
|
`Subgraphs` is collection of elements grouped together. The Subgraphs map to `grouping` elements in Excalidraw.
|
|
|
|
Lets consider the below example :point_down:
|
|
|
|
|
|
|
|
We use `getSubgraphs` API to get the subgraph data for a given flowchart.
|
|
Considering the same example this is the response from the API
|
|
|
|
```js
|
|
[
|
|
{
|
|
"id": "one",
|
|
"nodes": [
|
|
"flowchart-a2-1399",
|
|
"flowchart-a1-1400"
|
|
],
|
|
"title": "one",
|
|
"classes": [],
|
|
"labelType": "text"
|
|
}
|
|
]
|
|
```
|
|
|
|
For position and dimensions we use the svg to compute. The [`parseSubgraph`](https://github.com/excalidraw/mermaid-to-excalidraw/blob/master/src/parseMermaid.ts#L139) takes the above response along with `svg` and computes the position, dimensions and cleans up the response.
|
|
|
|
|
|
```js
|
|
[
|
|
{
|
|
"id": "one",
|
|
"nodes": [
|
|
"flowchart-a2-1399",
|
|
"flowchart-a1-1400"
|
|
],
|
|
"title": "one",
|
|
"labelType": "text",
|
|
"nodeIds": [
|
|
"a2",
|
|
"a1"
|
|
],
|
|
"x": 75.4921875,
|
|
"y": 0,
|
|
"width": 121.25,
|
|
"height": 188,
|
|
"text": "one"
|
|
}
|
|
]
|
|
``` |