tryfabric / martian Goto Github PK

_{Discuss in #public-issue-34--add-a-way-to-stringify-non-paragraph-elements}

It could be useful if someone is using parseRichText and they don't have just a paragraph, but other elements too

_{Discuss in #public-issue-10--bug-markdowntorichtext-not-working-if-there-are-multiple-new-li}

import { markdownToRichText } from "@instantish/martian";

const md = 'hello  \n  \n[sadsa](http://google.com)  \n  \n  \n  \n  \n**Bye**'
const rawRichTextArr = markdownToRichText(md);

error:

'Unsupported markdown element: {"type":"root","children":[{"type":"paragraph","children":[{"type":"text","value":"hello","position":{"start":{"line":1,"column":1,"offset":0},"end":{"line":1,"column":6,"offset":5}}}],"position":{"start":{"line":1,"column":1,"offset":0},"end":{"line":1,"column":8,"offset":7}}},{"type":"paragraph","children":[{"type":"link","title":null,"url":"http://google.com","children":[{"type":"text","value":"sadsa","position":{"start":{"line":3,"column":2,"offset":12},"end":{"line":3,"column":7,"offset":17}}}],"position":{"start":{"line":3,"column":1,"offset":11},"end":{"line":3,"column":27,"offset":37}}}],"position":{"start":{"line":3,"column":1,"offset":11},"end":{"line":3,"column":29,"offset":39}}},{"type":"paragraph","children":[{"type":"strong","children":[{"type":"text","value":"Bye","position":{"start":{"line":8,"column":3,"offset":54},"end":{"line":8,"column":6,"offset":57}}}],"position":{"start":{"line":8,"column":1,"offset":52},"end":{"line":8,"column":8,"offset":59}}}],"position":{"start":{"line":8,"column":1,"offset":52},"end":{"line":8,"column":8,"offset":59}}}],"position":{"start":{"line":1,"column":1,"offset":0},"end":{"line":8,"column":8,"offset":59}}}'

Issue happens if there are more than one new lines.
another example where it would break:

const md = `
hello

bye

go
`;

expected output: it should convert to richText object preserving new lines

_{Discuss in #public-issue-26--deploy-new-nelease}

Hello,

Would it be possible to release the new version of the library to NPM?

Thank you very much for the library. It is really useful!

Hi, thanks for this package!

I noted in the documentation that it should automatically try and spread data that would be too long into multiple blocks, however after using markdownToBlocks() with some data that contained many lines and line breaks, the result was a paragraph block with a rich_text array that had over 100 items, which is over Notion's API limit for rich_text items.

When reading the docs, I thought it might try and spread the excess rich_text items into other blocks, but looking at the source code it will either truncate the array or leave it as it is - I think we should create X-amount of blocks until all excess rich_text items can fit?

_{Discuss in #public-issue-16--simple-table-support}

https://developers.notion.com/changelog/simple-table-support

This means that the unsupported change can be fixed! I'll see if I can get it done some time this week unless it is already being worked on.

Hello, thanks for maintaining this tool. 🙏

I am trying to create a basic table, but am unable to do so. My table looks like this:

| Date       | Exercise      |
|------------|---------------|
| 27.08.2022 | exercise 0    |
| 30.08.2022 | exercise 1    |
| 01.09.2022 | exercise 2    |
| 02.09.2022 | exercise 3    |

However the request fails with:

@notionhq/client warn: request fail {
  code: 'validation_error',
  message: 'body failed validation: body.children[21].table.table-width should be not present, instead was `5`.'
}
/**Path**/node_modules/@notionhq/client/build/src/errors.js:162
        return new APIResponseError({
               ^

APIResponseError: body failed validation: body.children[21].table.table-width should be not present, instead was `5`.

I am using markdownToBlocks function with package versions:

    "@notionhq/client": "2.2.0",
    "@tryfabric/martian": "1.2.4",

Anyone else facing similar issue?

_{Discuss in #public-issue-38--feature-add-parse-yaml-front-matter}

maybe add some new feature

like https://www.npmjs.com/package/yaml-front-matter this package

now i write a obsidian plugin sync obsidian md to notion ,

but my obsidian md have yaml fromt matter

i write markdown with yaml front matter , use this package i can parses yaml head.

maybe you can add this new feature.

_{Discuss in #public-issue-29--numberbullet-list-within-blockquote-will-be-ignored}

The markdown string blow will produce a empty Blockquote notion block using markdownToBlocks.

> - list item 1
> - list item 2

[
  {
    "object": "block",
    "type": "quote",
    "quote": {
      "text": []
    }
  }
]

I think it should be keep as rich text just as other block

https://github.com/instantish/martian/blob/master/src/parser/internal.ts#L147-L159

* test
	* test2
		* test3

it can work

but when bulleted list more than four it don't work to upload to notion

* test
	* test2
		* test3
			* test4

_{Discuss in #public-issue-8--automatically-bump-version}

We could build our own action or I'm sure there are ones available in the marketplace that we could use to automatically bump the version in package.json when we cut releases

EasyChris/obsidian-to-notion#33

I did try to use browserify to make it work in my react project but couldn't, does this only work in node env?

_{Discuss in #issue-1--images-should-render-as-image-block-instead-of-paragraph-block}

First of all thanks for making this library public. 💜

I am trying to convert markdown to Notion blocks. Everything works fine except images.

The images are currently rendered as a paragraph block

{
   "object":"block",
   "type":"paragraph",
   "paragraph":{
      "text":[
         {
            "type":"text",
            "annotations":{
               "bold":false,
               "strikethrough":false,
               "underline":false,
               "italic":false,
               "code":false,
               "color":"default"
            },
            "text":{
               "content":"https://i0.wp.com/css-tricks.com/wp-content/uploads/2021/09/s_C6DD37D8ED67AB804309E3524813327121D8EA01ECCF02AF2243019807FF6CCA_1631271964338_notion-front-page-hero.png?resize=1000,716&ssl=1"
            }
         }
      ]
   }
},

They should be rendered like

{
   "object":"block",
   "type":"image",
   "image":{
      "type":"external",
      "external":{
         "url":"https://i0.wp.com/css-tricks.com/wp-content/uploads/2021/09/s_C6DD37D8ED67AB804309E3524813327121D8EA01ECCF02AF2243019807FF6CCA_1631271964338_notion-front-page-hero.png?resize=1000,716&ssl=1"
      }
   }
},

Can you tell me how to do so?

Thanks

_{Discuss in #issue-2--undefined-error-on-parsing-empty-list}

First of all. thanks for making this open-source love the way it's helping us at @meetshepherd. 🙌🏻

Issue

If the markdown has an empty NumberedListItem or BulletedListItem parsing list to notion blocks fails.

This is what I mean by having an empty NumberedListItem or BulletedListItem 👇🏻
(they have the formatting of a Bulleted/Number list with no text)

Empty BulletedListItem	Empty NumberedListItem

Error Screenshot

_{Discuss in #public-issue-32--library-sets-the-wrong-table-header}

Markdown:

| First Header  | Second Header |
| ------------- | ------------- |
| Content Cell  | Content Cell  |
| Content Cell  | Content Cell  |

Current:

Expected:

_{Discuss in #public-issue-13--latest-version-of-martian-isnt-published-to-npm}

The NPM published version is v1.0.3, but the latest GH version is v1.1.1, so it should be updated on NPM too. :)

Awesome to see this package becoming relatively popular!!

We noticed recently that some text when converting through markdownToBlocks gets a bad URL value, with blocks that looks like:

[
  {
    "object": "block",
    "type": "paragraph",
    "paragraph": {
      "rich_text": [
        {
          "type": "text",
          "annotations": {
            "bold": false,
            "strikethrough": false,
            "underline": false,
            "italic": false,
            "code": false,
            "color": "default"
          },
          "text": {
            "content": "https://60f06c1c5a1772003b824d76-gclpzqfgft.chromatic.com/?path=/story/poi-map--create-point-poi",
            "link": {
              "type": "url",
              "url": "url" // <- this fails Notion client
            }
          }
        }
      ]
    }
  },
  // ...<other blocks>
]

Where the source is simply the link text https://60f06c1c5a1772003b824d76-gclpzqfgft.chromatic.com/?path=/story/poi-map--create-point-poi

This would trigger an error when creating the blocks on Notion:

@notionhq/client warn: request fail {
  code: 'validation_error',
  message: 'Content creation Failed. Fix the following: \n' +
    'Invalid URL for link.\n' +
    'Invalid URL for link.\n' +
    'Invalid URL for link.'
}

At the moment, we employ a patch mutation script to address this as the following:

const seekFixFalseURLs = (blocks) => { // mutates blocks
  blocks.forEach((block) => {
    ((block[block.type] || {}).rich_text || []).forEach((rt) => {
      if ((rt[rt.type].link || {}).url === 'url') {
        rt[rt.type].link.url = rt[rt.type].content
      }
    })
  })
}

_{Discuss in #public-issue-20--feat-use-generics-in-block-interface}

Using a generic in the definition of the Blocks interface would allow us to better enforce supported block types and get more accurate typings

I'm facing this issue when I use the function markdownToBlocks.

@notionhq/client warn: request fail {
  code: 'validation_error',
  message: 'body failed validation: body.image.type should be not present, instead was `"external"`.'
}

_{Discuss in #public-issue-35--errors-from-parserichtext-are-inconsistent}

We're only checking if the first Markdown element is a paragraph while ignoring other children of the root.
The result is that the package throws an error when the first element is unsupported, but not when it's the second or third: it would be better to either ignore all of them or throw an error for each.

_{Discuss in #public-issue-22--equation-block-and-inline-math-support}

Thanks for the great package!

Martian seems not support math equations in markdown. For example:

Lift($L$) can be determined by Lift Coefficient ($C_L$) like the following
equation.

$$
L = \frac{1}{2} \rho v^2 S C_L
$$

would be converted into plain text paragraph, keeping all $ and $$ symbols.

I found unified plugin remark-math may be useful. It would be amazing if Martian add support for math equations.

_{Discuss in #public-issue-11--release-v110v111-}

@instantish/martian latest version is 1.0.3.
So I use @tryfabric/martian(1.1.0) instead of that.

_{Discuss in #public-issue-30--notion-api-now-supports-tables}

https://developers.notion.com/reference/block#table-blocks

Notion API now supports tables. Could you add support of table into martian?(Or you just forgot to edit the README?)

_{Discuss in #public-issue-6--add-checks-for-all-property-value-limits}

Warn or truncate values that exceed the allowed limits to avoid unexpected failures https://developers.notion.com/reference/errors#limits-for-property-values

Problem

Paragraph text with multiple dollar amounts incorrectly creates an equation block.

Working Example

markdownToBlock('A painting by John Doe sold for $9M at auction. A previous piece of his sold for $5M last year.');

Actual Result

Generated JSON block

[
  {
    "object": "block",
    "type": "paragraph",
    "paragraph": {
      "rich_text": [
        {
          "type": "text",
          "annotations": {
            "bold": false,
            "strikethrough": false,
            "underline": false,
            "italic": false,
            "code": false,
            "color": "default"
          },
          "text": {
            "content": "A painting by John Doe sold for "
          }
        },
        {
          "type": "equation",
          "annotations": {
            "bold": false,
            "strikethrough": false,
            "underline": false,
            "italic": false,
            "code": false,
            "color": "default"
          },
          "equation": {
            "expression": "9M at auction. A previous piece of his sold for "
          }
        },
        {
          "type": "text",
          "annotations": {
            "bold": false,
            "strikethrough": false,
            "underline": false,
            "italic": false,
            "code": false,
            "color": "default"
          },
          "text": {
            "content": "5M last year."
          }
        }
      ]
    }
  }
]

Screenshot when saved to Notion API

Expected Result

[
  {
    "object": "block",
    "type": "paragraph",
    "paragraph": {
      "rich_text": [
        {
          "type": "text",
          "annotations": {
            "bold": false,
            "strikethrough": false,
            "underline": false,
            "italic": false,
            "code": false,
            "color": "default"
          },
          "text": {
            "content": "A painting by John Doe sold for $9M at auction. A previous piece of his sold for $5M last year."
          }
        }
      ]
    }
  }
]

I've noticed that markdownToRichText completely ignores new lines break both \n(both at the end and the beginning of a line) and double space(at the end of the line).
Is there a way to make these things work? I would love to have child text block be created for the newlines break because with without it, there is no indentation in the whole callout block

_{Discuss in #public-issue-15--nested-numberbullet-list-combination-ends-up-with-error}

I've a deeply nested list items in my .md file, like this:

1. **Something important**

    1. Sub step one
        - something to remember
        - parameters: 
          - param 1
          - param 2
          - param 3

This formats correctly in GH:

1. Something important

   1. Sub step one
      • something to remember
      • parameters:
        • param 1
        • param 2
        • param 3

but when converted to blocks and pushed as children element in the Notion API call I get error:
'{"object":"error","status":400,"code":"validation_error","message":"body failed validation: body.children[1].numbered_list_item.children[0].numbered_list_item.children[0].numbered_list_item.children should be not present, instead was `[{\\"object\\":\\"block\\",\\"type\\":\\"bulleted_list_item\\",\\"bullete...`."}'.
I tried recreating that styling manually in Notion and it seems to work.