md-export
v1.0.5
Published
Customizable parse + convert text
Downloads
8
Maintainers
Readme
md-export
md-export can convert content between formats. It comes with parsers and template for converting Wordpress XML to Markdown but you can easily specify your own.
Quick start
# Generates frontmatted .md files of wordpress posts
npx md-export wp-export.xml --download-images
Installation
yarn global add md-export
Usage
# Convert Wordpress xml to markdown
md-export export.xml
# Download images references in text
md-export export.xml --download-images
# Using your own output template
md-export export.xml --template=my-template.md
# ...or your own parser
md-export anything.json --parser=my-json-parser.js
Generated files, and accompanied images scraped from the post are found in the output folder (default: output
).
Instructions for exporting your information from WordPress can be found here.
Options
md-export <input-file> [args]
Options:
--version Show version number [boolean]
-d, --download-images Downloads images references to post folder.
[boolean] [default: false]
--debug Log for debug purposes [boolean] [default: false]
-f, --folder-format Format of individual post folder name.
[default: "yyyy-mm-dd-"slug""]
-o, --output-dir Folder in which to put posts [default: "output"]
-i, --filter-images Regex filter for which linked images to download and
replace urls.
[default: "(?:src="(http[^"]*?)")|(?:href="(http[^"]*?(?:\.(?:apng|bmp|gif|cur
|ico|jpg|jpeg|jfif|pjpeg|pjp|png|svg|webp))))""]
-p, --parser Which parser to use for parsing input file.
[default: "./parsers/wordpress-xml"]
-s, --filter-slug Specify post slug if wish to convert a single post
-t, --template Which template to use for generating files.
[default: "./templates/gatsby.md"]
-h, --help Show help [boolean]
Examples:
md-export wordpress.xml Generates markdown files based on wordpress xml export
md-export wordpress.xml -d Downloads linked images (hosted on same domain)
Post output folder
Each post is put in an own individual folder.
/2018-11-30-how-to-markdown/index.md
Its folder name can be specified with --folder-format=YOUR-FORMAT
Default: yyyy-mm-dd-"slug"
Note that quotes are required to surround text that should not be formatted as date.
Replaced values are:
author
: The author that created the postslug
: The url slug name of the post
The rest is formatted as dates, using dateformat.
Images
All linked images in the original post from the same domain are downloaded
and put in the folder belonging to the related markdown file, when -d
is used
/2018-11-30-how-to-markdown/index.md
/2018-11-30-how-to-markdown/image-for-the-post.jpg
/2018-11-30-how-to-markdown/another-image.jpg
You can download all images to a shared folder by specifying -g=./public/images
.
/2018-11-30-how-to-markdown/index.md
/public/images/image-for-the-post.jpg
/public/images/another-image.jpg
By default, we download all links from img src=
, and all linked images
a href="path to image"
that resides within the same domain as the post.
This can be changed with --filter-images=YOUR-REGEX
These URLs are also changed in the content of the original data.
Templates
By default, we use a template of this format
---
title: "{{ title }}"
date: {{ date }}
image: {{ image }}
tags: {{ tags }}
author: {{ author }}
status: {{ status }}
---
{{ content }}
Placeholders, e.g. {{ title }}
, are replaced with the data parsed
from the input file.
The template used can be changed with --template=my-file.md
Available variables are:
author
: The author of the postcontent
: The markdown generated body of the posthtml
: The body of the post, in HTML.date
: The post date formatted asyyyy-mm-dd
slug
: The url slug of the original posttitle
: The title of the postimage
: The featured image of the article
Parsing other inputs
Parsers can be found in in the source code. These contain logic for parsing
a file into a structured format. You can add override the parser
and specify your own with --parser=YOUR-PARSER
.
If you create your own parser, it should default export a function that accepts path of file, and returns a list of an objects. It must return an array of items, where each item should have the following keys:
slug
: A slug of the itemdate
: date (optional)content
: Content as HTML (optional)
Note: You can also add more keys. These will be passed on as is to be reused in the template.