Markdown formatting plugin for dprint.
This uses the pulldown-cmark parser for markdown.
dprint / dprint-plugin-markdown Goto Github PK
View Code? Open in Web Editor NEWMarkdown code formatting plugin for dprint.
License: MIT License
dprint-plugin-markdown's Introduction
dprint-plugin-markdown's People
Contributors
Stargazers
Watchers
dprint-plugin-markdown's Issues
Backticks not correctly maintained in this scenario
Given the following text, it should format as is:
````text
# std/io
## readLines
```ts
import * as path from "https://deno.land/std/path/mod.ts";
## Rest of the file
````
Space between list items
Would be good to make this configurable and probably the default.
Image tag is incorrectly assumed to be a link reference
Version 0.9.5:
Example:
1. Test You can try it here: https://dprint.dev/playground/#code/FDCMDoAIBUFMGcAukCEBtAlgWwIYHNZIN4BdACnAHoB3AJwHsA7PAGwE9Kd54BXLWACYB9RPSEAjWEJYZGAayG1YAM1hLGAY1jgADswCUQA/language/markdown
it yields:
Line 3, column 63: Unexpected token `)` while parsing link reference definition.
e_link_reference.png)
additional fenced code block formatting
Hello,
Apologies if this is the wrong place to ask this question. Is there anyway to extend the code formatting? I noticed dprint can format fenced js and css. I'm more specifically trying to format fenced svelte, vue and react.
🐛 Parsing bug for combination of blockquote containing a bullet-list and following text
@dsherret , Hello! 👋🏻
I just ran into what appears to be a parsing bug and wanted to let you know about it.
Describe the bug
C:>dprint --version
dprint 0.18.2
C:>cat .dprint.json
{
"$schema": "https://dprint.dev/schemas/v0.json",
"projectType": "openSource",
"incremental": true,
"indentWidth": 2,
"lineWidth": 100,
"useTabs": true,
"typescript": {
// ref: <https://dprint.dev/plugins/typescript/config>
"deno": true,
// * customize preferences (overrides of some Deno formatting choices)
// ref: <https://github.com/dprint/dprint-plugin-typescript/blob/44b6cf562e511a308f4a7183dc98fb19cdf21d07/src/configuration/builder.rs#L51>
// ref: <https://github.com/denoland/deno/blob/f46e39c5c5/cli/tools/fmt.rs#L311>
"lineWidth": 100,
"preferSingleLine": true,
"quoteStyle": "preferSingle",
//
"ignoreNodeCommentText": "dprint-ignore", // from Deno's "deno-fmt-ignore"
"ignoreFileCommentText": "dprint-ignore-file", // from Deno's "deno-fmt-ignore-file"
"memberExpression.linePerExpression": true,
"memberExpression.preferSingleLine": true,
"module.sortImportDeclarations": "caseInsensitive",
"module.sortExportDeclarations": "caseInsensitive"
},
"json": { "preferSingleLine": true },
"markdown": { "textWrap": "always", "lineWidth": 99999 },
"prettier": { "printWidth": 100, "singleQuote": true, "tabWidth": 2 },
"rustfmt": {},
"includes": ["**/*.{ts,tsx,js,jsx,cjs,mjs,json,md,mkd,rs,yaml,yml}"],
"excludes": [
".history",
".changelog/*.tpl.*",
"CHANGELOG{,.}*",
"**/node_modules",
"**/*-lock.json",
"**/target",
"**/vendor"
],
"plugins": [
// ref: <https://plugins.dprint.dev>
// "https://plugins.dprint.dev/typescript-0.51.0.wasm",
// "https://plugins.dprint.dev/json-0.12.3.wasm",
// "https://plugins.dprint.dev/markdown-0.9.5.wasm",
// "https://plugins.dprint.dev/toml-0.4.1.wasm",
// // ref: <https://github.com/dprint/dprint-plugin-rustfmt/releases>
// "https://plugins.dprint.dev/rustfmt-0.4.0.exe-plugin@c6bb223ef6e5e87580177f6461a0ab0554ac9ea6b54f78ea7ae8bf63b14f5bc2",
// // ref: <https://github.com/dprint/dprint-plugin-prettier/releases>
// "https://plugins.dprint.dev/prettier-0.2.2.exe-plugin@63b06beba3acd51e6d23379c47db342a1a24f3d88f3c52a775e2db3426124582"
"https://plugins.dprint.dev/typescript-0.59.0.wasm",
"https://plugins.dprint.dev/json-0.13.1.wasm",
"https://plugins.dprint.dev/markdown-0.11.1.wasm",
"https://plugins.dprint.dev/toml-0.5.2.wasm",
// ref: <https://github.com/dprint/dprint-plugin-rustfmt/releases>
"https://plugins.dprint.dev/rustfmt-0.4.0.exe-plugin@c6bb223ef6e5e87580177f6461a0ab0554ac9ea6b54f78ea7ae8bf63b14f5bc2",
// ref: <https://github.com/dprint/dprint-plugin-prettier/releases>
"https://plugins.dprint.dev/prettier-0.3.0.exe-plugin@c6c227493e655717b5f3d9c811ba576c82f2b060317bb233c6ec014958fbee19"
]
}Input Code
> - Officia exercitation voluptate ea esse do culpa consequat amet consectetur mollit dolor tempor.
>
> ...Actual Output
An error occurs during reformatting...
[ERROR] Error formatting text. Line 4, column 1: Unexpected token `>` while parsing link reference definition.
>
~
Let me know if I can be of help tracking down the source of the issue.
Centered columns do not always have the correct width in a cell
Example:
Column Title|Test|Asd
--:|--|:---:
Testing|asdff|tesd
This is a test|asd|tesOutput:
| Column Title | Test | Asd |
| -------------: | ----- | :--: |
| Testing | asdff | tesd |
| This is a test | asd | tes |Maintain blank line for list inside list
1. Get [VS Community 2019](https://www.visualstudio.com/downloads/) with
"Desktop development with C++" toolkit and make sure to select the following
required tools listed below along with all C++ tools.
- Visual C++ tools for CMake
- Windows 10 SDK (10.0.17763.0)
- Testing tools core features - Build Tools
- Visual C++ ATL for x86 and x64
- Visual C++ MFC for x86 and x64
- C++/CLI support
- VC++ 2015.3 v14.00 (v140) toolset for desktopAdd configuration options that allow (partial) enforcement of the semantic line breaks specification
I've recently come across https://sembr.org/ and I am a big fan so far.
In our markdown documents, we are already trying to follow the "one sentence per line" rule.
This makes our Git diffs more readable because natural language tends to be edited on a sentence basis (i.e. rewrite a single sentence, split it in two, merge two into one, etc).
This behaviour is captured in requirement 4 of sembr.
I haven't looked into the parser and formatter yet but it would be amazing if dprint would make it easy to have markdown documents follow sembr.
Handle incorrectly indented list with sub code blocks more gracefully
Would be ok if the list was indented correctly:
Input Code
## Section
- List item 1
Paragraph content at same indentation as list item 1
```
code block at same indentation as list item 1
```
- Sublist item A
Paragraph content at same indentation as sublist item A
```
code block at same indentation as sublist item A
```
More paragraph content at same indentation as list item 1
```
code block at same indentation as list item 1
```
- List item 2
Paragraph content at same indentation as list item 2
```
code block at same indentation as list item 2
```
Output code
## Section
- List item 1
Paragraph content at same indentation as list item 1
```
code block at same indentation as list item 1
```
- Sublist item A
Paragraph content at same indentation as sublist item A
```
code block at same indentation as sublist item A
```
More paragraph content at same indentation as list item 1
```
code block at same indentation as list item 1
```
- List item 2
Paragraph content at same indentation as list item 2
```
code block at same indentation as list item 2
```
Desired Output
## Section
- List item 1
Paragraph content at same indentation as list item 1
```
code block at same indentation as list item 1
```
- Sublist item A
Paragraph content at same indentation as sublist item A
```
code block at same indentation as sublist item A
```
More paragraph content at same indentation as list item 1
```
code block at same indentation as list item 1
```
- List item 2
Paragraph content at same indentation as list item 2
```
code block at same indentation as list item 2
```
Should add blank line between html and code block
<!-- prettier-ignore-start -->
```jsShould keep double backticks when enclosing text uses backticks
Look at the source for this: `test`. It's formatted without the spaces which causes it to lose the backticks.
`test`test<-- goes to
Also: testing `this` out
Space erroneously added after first item in list and before sub item when last item has a code block
## Comparison to Node.js
- Deno does not use `npm`
- It uses modules referenced as URLs or file paths
- Uses "ES Modules" and does not support `require()`. Third party modules are
imported via URLs:
\`\`\`javascript
import * as log from "https://deno.land/std/log/mod.ts";
\`\`\`Emdash followed by code block causes panic
test—`cargo install gvm`
Should not remove `>` on blank line in markdown block quote
For example, the following:
> Testing this out
> Testing
>
> Testing
> Testing asdf
Should format as-is.
Markdown config for emphasis and strong text kind
There's multiple ways to represent strong and emphasis text decorations:
https://github.com/adam-p/markdown-here/wiki/Markdown-Cheatsheet#emphasis
Need newline after table
Try this:
## Verbs
The following verbs are supported:
| Verb | Meaning |
| ----- | -------------------------------------------------------------- |
| `%` | print a literal percent |
| `t` | evaluate arg as boolean, print `true` or `false` |
| `b` | eval as number, print binary |
| `c` | eval as number, print character corresponding to the codePoint |
| `o` | eval as number, print octal |
| `x X` | print as hex (ff FF), treat string as list of bytes |
| `e E` | print number in scientific/exponent format 1.123123e+01 |
| `f F` | print number as float with decimal point and no exponent |
| `g G` | use %e %E or %f %F depending on size of argument |
| `s` | interpolate string |
| `T` | type of arg, as returned by `typeof` |
| `v` | value of argument in 'default' format (see below) |
| `j` | argument as formatted by `JSON.stringify` |
## Width and PrecisionConfiguration for changing ignore comment text
Right now ignore comments have the text "dprint-ignore", "dprint-ignore-start" and "dprint-ignore-end", but this should be configurable.
It is being really lazy now and using regex, but that should be removed and replaced with character comparisons.
Excess cells in a table row disappear in output
Input
| name | description |
| ------------------------ | ------------------------------ |
| `get(key: name): string | null` | Get the value of the string. |
Expected output
| name | description |
| ------------------------------- | ---------------------------- |
| `get(key: name): string | null` | Get the value of the string. |
Actual output
| name | description |
| ----------------------- | ----------- |
| `get(key: name): string | null` |
Blank lines in code blocks are removed when using carriage return line feeds
Kind of strange. Look into it.
Double space between strikethrough and emphasis
~~v8~~ _can't implement_
Support formatting code blocks within markdown
Waiting on dprint/dprint#237
Newline or blank line should be respected in lists because it affects rendered output
The following should format as-is:
1. test
2. test
3. test
```shell
test
```additional indentation for task lists
Thanks for dprint! It reduces the time to format my knowledge base consisting of ~1600 Markdown files from 87 seconds (Prettier) to 0.2 seconds! 💥
I noticed different indentation behavior for task lists when always wrapping text. Prettier formats them with 6 spaces of indentation. This separates the check boxes visually from the text (playground example):
- [x] an issue with a very very long title that flows into subsequent lines in
order to demonstrate how Prettier uses six spaces of indentation for task
lists
- [ ] another issueDprint indents task lists with only two spaces. This somewhat hides the checkboxes in the text (playground example):
- [x] an issue with a very very long title that flows into subsequent lines in
order to demonstrate how dprint uses only two spaces of indentation for task
lists
- [ ] another issueBoth styles work but I think the Prettier version looks nicer and is a bit more ergonomic. How do you feel about matching Prettier's formatting of task lists in dprint? If you agree and provide some guidance, I'm happy to take a stab at implementing this.
Escape open and close angle brackets in text
- **--allow-net=\<allow-net\>** Allow network access. You can specify an
optional, comma separated list of domains to provide a allow-list of allowed
domains.Support dprint-ignore-file directive
From denoland/deno#10185
Two lists with task lists causes strange formatting
Why?
## Supported Builtins
- [ ] assert
- [x] node globals _partly_Formats as:
## Supported Builtins
- [ ]
assert
- [x]
node globals _partly_Support numbered markdown lists with repeated numbers
1. Item
1. Item
1. ItemIs currently reformatted as:
1. Item
2. Item
3. ItemHowever, both render the same (see the source of this):
- Item
- Item
- Item
The form that repeats the same number can be more preferable, as it's easier to see what changed in diffs and add/remove items without modifying every other list item after it.
Could a new setting be added to prefer this form, or maintain the numbering (if it's consistent)?
`proseWrap: always` should not wrap link title
Thanks for dprint! I love it and hope it takes over the formatting world and we never have to wait for Prettier again! I would like to propose to never wrap links even if textWrap is set to always.
Details
Prettier never wraps links even if their title is longer than 80 characters. In this example in the Prettier playground the link is only on line 2. Dprint currently wraps links: in the equivalent example in the dprint playground the link ends up on lines 2-3 in the formatted output.
Rationale
Keeping links on a single line makes processing markdown text line by line so much easier, for example grepping with regexes, showing only matching lines in search results, or when syntax highlighting. I'd also argue that keeping the link on one line improves readability of the formatted output since links could be considered a logical unit in the content. It is easier to read Markdown when links stay "together". This change wouldn't affect HTML rendered from the Markdown.
Implementation
If you agree and provide some guidance around how to implement this, I'm happy to take a stab at it.
quadruple backtick sections in markdown are outdented
Even though marked ignored, this markdown fails to stay in format:
* some list
<!-- dprint-ignore-start -->
````md
:rocket:[censored]
```ini
DEBUG=true
```
````
<!-- dprint-ignore-end -->it seems to receive a diff on fmt:
text code blocks are reformatted
Markdown 0.10.0
expected:
```text
│
HTTP ┌───────────────┐
┌─────────────────────────────────────────┼─────────────────────┤ component ├─┐
│ │ └───────────────┘ │
│ ▼ │
│ ┌───────────────────┐ │
│ │ │ │
│ │ Blablabla │ │
│ │ │ │
│ └───────────────────┘ │
│ │ │
│ │ │
│ helloWorld() │
│ │ │
│ ▼ │
│ ┌───────────────────┐ │
│ │ │ │
│ │ Bla │ │
│ │ │ │
│ └───────────────────┘ │
│ │ │
└─────────────────────────────────────────┼───────────────────────────────────────┘
▼
```to stay as-is. Instead the first pipe is shifted (spaces collapsed):
```text
│
HTTP ┌───────────────┐
┌─────────────────────────────────────────┼─────────────────────┤ component ├─┐
│ │ └───────────────┘ │
│ ▼ │
│ ┌───────────────────┐ │
│ │ │ │
│ │ Blablabla │ │
│ │ │ │
│ └───────────────────┘ │
│ │ │
│ │ │
│ helloWorld() │
│ │ │
│ ▼ │
│ ┌───────────────────┐ │
│ │ │ │
│ │ Bla │ │
│ │ │ │
│ └───────────────────┘ │
│ │ │
└─────────────────────────────────────────┼───────────────────────────────────────┘
▼
```(first line)
cc @shanloid
Use newline group on links
So that they prefer to stay on the same line
Need newline before codeblock after text in list
- `options['--']` - when true, populate `parsedArgs._` with everything before
the `--` and `parsedArgs['--']` with everything after the `--`. Here's an
example:
\`\`\`ts
// $ deno run example.ts -- a arg1
import { parse } from "https://deno.land/std/flags/mod.ts";
console.dir(parse(Deno.args, { "--": false }));
// output: { _: [ "a", "arg1" ] }
console.dir(parse(Deno.args, { "--": true }));
// output: { _: [], --: [ "a", "arg1" ] }
\`\`\`Support formatting code blocks with markdown code
Might as well 😀
Ability to turn off text wrapping
"textWrapping": "maintain" | "force" | "never"
Tables not formatted properly when header has a column count less than the body
From #33. I think perhaps the parser is not parsing this as a table for some reason.
Handle non-breaking spaces
For example:
testing testing testing testing testing `code` testing testing testing testing `code` testingLink reference after list when indented will disappear
Example:
* Testing
[test]: asdfOutputs:
* TestingNeed to handle multiple lists after one another
The following should format as-is:
- Testing
- Other
* Testing
* OtherThis is important because changing the deliminator starts a new list. See https://spec.commonmark.org/0.29/#loose
Incorrectly trims whitespace
Source:
`` Edit post `slug-here` ``Expected:
`` Edit post `slug-here` ``Which gets rendered as:
<p><code>Edit post `slug-here`</code></p>Actual:
``Edit post `slug-here` ``Which gets rendered as:
<p><code>Edit post `slug-here` </code></p>Which contains a trailing whitespace in a code block, which then looks incorrect when displayed.
Ref: denoland/deno#10313
Ignore file directive works in the whole file
When working on a blog post @lucacasonato discovered that ignore directive for the file (in our case deno-fmt-ignore-file) makes dprint to ignore file if it's placed anywhere in the file.
Ideally this directive should only have effect if it's placed on the first line.
Need newline after block quotes
> This is an unstable feature. Learn more about
> [unstable features](./stability.md).
By default the namespace is not available in worker scope.Table dash count subtraction with overflow error
https://github.com/littletof/prettyBenching/blob/696e3541fc5a52e35776f93c704ddae31a1a3d44/README.md
thread 'test_specs' panicked at 'attempt to subtract with overflow', src\parsing\parse.rs:638:32
Problem line:
let dashes_count = column_width - (if has_left_colon { 1 } else { 0 }) - (if has_right_colon { 1 } else { 0 });
format_code_block_text should avoid string copy when tag doesn't match
If the tag doesn't match anything, then the user should be able to say to format with the original string instead. This will avoid a needless copy.
Formatting Breaks Valid Markdown
Input/Output:
Expected:
A list should not be deleted
Support ignore comments
As title.
meta information not retained in shape without new line
Input
---
title: File System APIs
category: Runtime
weight: 2
---
^-- no new line at the end.
Output
---
## title: File System APIs category: Runtime weight: 2
Expected Behaviour
---
title: File System APIs
category: Runtime
weight: 2
---
Line breaks with double spaces and/or backslash not supported
backlash
Input:
foo<space><space>\
bar<space><space>\
baz
Output:
foo
bar
baz
Expected output:
foo<space>\
bar<space>\
baz
spaces
Input:
foo<space><space>
bar<space><space>
baz
Output:
foo<space>
bar<space>
baz
Expected output:
foo<space><space>
bar<space><space>
baz
ref: denoland/deno#9727
Keep indents of code blocks and do not convert code blocks to fenced automatically
I thought it would be nice to convert code blocks to be fenced automatically, but in practice it can be annoying since some markdown renders have different indentation requirements.
Extra blank line added after dprint-ignore comment when file is CRLF
Weird:
Testing:
<!-- dprint-ignore -->
```json
{
"extends": "https://dprint.dev/path/to/config/file.v1.json",
// ...omitted...
}
```Formats as:
Testing:
<!-- dprint-ignore -->
```json
{
"extends": "https://dprint.dev/path/to/config/file.v1.json",
// ...omitted...
}
```Reference links after a table are swallowed
Input
# Test
| Company |
| -------------------------- |
| [GitHub][github] |
| [Amazon Web Services][aws] |
[github]: https://github.com
[aws]: https://aws.amazon.com
A paragraph.
Expected Output
# Test
| Company |
| -------------------------- |
| [GitHub][github] |
| [Amazon Web Services][aws] |
[github]: https://github.com
[aws]: https://aws.amazon.com
A paragraph.
Actual Output
# Test
| Company |
| -------------------------- |
| [GitHub][github] |
| [Amazon Web Services][aws] |
A paragraph.
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.