[Convert markdown] Script to convert Markdown to Plain text, using an Espanso trigger. #espanso #powershell #python

conv_markdown.yml

## Espanso trigger to accompany ConvertMarkdown.ps1

  - trigger: :conv
    replace: '{{output}}'
    vars:
      - name: clipb
        type: clipboard
      - name: output
        type: script
        params:
          args: 
            - pwsh
            - /home/stephen/.config/espanso/scripts/ConvertMarkdown.ps1 
            - -markdownText 
            - "{{clipb}}"


## Neater inline solution using Python libraries instead, 
## to convert Markdown→HTML→Plain text:

  - trigger: :conv
    replace: '{{output}}'
    vars:
      - name: clipb
        type: clipboard
      - name: output
        type: script
        params:
          args:
            - python
            - -c
            - |
              from bs4 import BeautifulSoup
              from markdown import markdown
              import re
              args = """{{clipb}}"""
              html = markdown(args)
              html = re.sub(r'<pre>(.*?)</pre>', ' ', html)
              html = re.sub(r'<code>(.*?)</code>', ' ', html)
              text = ''.join(BeautifulSoup(html, "html.parser").findAll(text=True))
              print(text)

ConvertMarkdown.ps1

## Script to convert Markdown to Plain text, using an Espanso trigger:

#!/usr/bin/env pwsh

param (
    [string]$markdownText
)

# Function to convert Markdown to plain text
function Convert-MarkdownToText {
    param (
        [string]$markdown
    )

    # Split the input into lines to process each line individually
    $lines = $markdown -split "`n"

    $inCodeBlock = $false
    $plainTextLines = $lines | ForEach-Object {
        # Toggle code block flag when encountering ``` lines
        if ($_ -match '^\s*```') {
            $inCodeBlock = -not $inCodeBlock
            # Skip the ``` lines themselves
            return
        }

        # If inside a code block, capture content directly without modification
        if ($inCodeBlock) {
            return $_
        }

        # Process each line for Markdown syntax if not in a code block
        $_ -replace '^\s*#{1,6}\s*(.+)$', '$1' |                    # Remove headers
        ForEach-Object { $_ -replace '\*\*([^\*]+)\*\*', '$1' } |    # Bold
        ForEach-Object { $_ -replace '\*([^\*]+)\*', '$1' } |        # Italic
        ForEach-Object { $_ -replace '__(.+?)__', '$1' } |           # Bold underline
        ForEach-Object { $_ -replace '_(.+?)_', '$1' } |             # Italic underline
        ForEach-Object { $_ -replace '\[([^\]]+)\]\([^\)]+\)', '$1' } | # Links
        ForEach-Object { $_ -replace '!\[([^\]]*)\]\([^\)]+\)', '$1' } | # Images
        ForEach-Object { $_ -replace '``(.+?)``', '$1' } |           # Inline code
        ForEach-Object { $_ -replace '`(.+?)`', '$1' } |             # Inline code
        ForEach-Object { $_ -replace '^\s*[-\*\+]\s+', '' } |        # Unordered list markers
        ForEach-Object { $_ -replace '^\s*\d+\.\s+', '' } |          # Ordered list markers
        ForEach-Object { $_ -replace '^\s*>\s+', '' } |              # Blockquotes
        ForEach-Object { $_ -replace '\[\[([^\]]+)\]\]', '$1' }      # Convert [[text]] to text
    }

    # Join the processed lines back into a single string
    $plainText = ($plainTextLines -join "`n").Trim()

    return $plainText
}

# Convert markdownText from input
$plainText = Convert-MarkdownToText -markdown $markdownText
Write-Output $plainText