From Atom to Sublime Text
2017-07-09 00:56:58 Author: parsiya.net(查看原文) 阅读量:79 收藏

I have moved from Atom to Sublime. Atom is a nice editor with a lot of features but it has a lot of performance issues for what I want to do.

Below is my setup for reference. When I want to do it again in a year (or a new machine) I can just use everything here or just use the config files. You can find the config files in my clone:

Install package control using: https://packagecontrol.io/installation.

Now you can install packages.

  1. Press Ctrl+Shift+P to open the command palette.
  2. Type Install and then select Package Control: Install Package.
  3. Type the name of the package you are looking for to search for it.
  4. Select the package and press Enter.
  1. Markdown Extended: Syntax highlighting.
  2. Markdown Preview.
  3. Markdown Editing: Shortcut keys (e.g. ctrl+1 means heading 1).
  4. Monokai Extended: Needed for highlighting.
  5. LiveReload: For live markdown preview.
  6. MarkdownTOC: Automatically generate clickable table of contents to markdown documents.
  7. TOML: TOML highlighting.

After installing packages, just copy the config files to the user package settings directory. On Windows it will be "%Appdata%\Sublime Text 3\Packages\User\" (don't forget the double quotes if you want to just paste it into the run prompt).

Generally sublime and each package have two types of settings, default and user. User settings are used to override default ones. Both are in JSON.

Settings can be opened via Preferences > Package Settings > Settings Default/User. I usually copy from default file to user, remove the unneeded settings and override the rest.

Package settings on Windows are at %Appdata%\Sublime Text 3\Packages\User\package-name.sublime-settings.

Main settings can be overridden in Preferences.sublime-settings or accessed via Preferences > Settings. This will open two files in one window. Left are the defaults settings and right is the user settings file. Copy from left pane to the right one and override.

Markdown Editing has its own color scheme. I don't like it. Instead I use Monokai Extended. It can be selected from Preferences > Color Scheme > Monokai Extended > Monokai Extended.

In order for it to kick in, the document type need to be set to Markdown Extended. This can be set by clicking on document type (bottom right).

Next you want to set all markdown files to be opened as Markdown Extended for syntax highlighting (including code blocks). This can be done by:

  • View (menu) > Syntax > Open all with current extension as > Markdown Extended.

This method only sets it for the current extension (e.g. md).

Go to Preferences > Settings - Syntax Specific or edit package settings Markdown Extended.sublime-settings and add the following:

{
  "extensions":
  [
    "md",
    "markdown",
    "moreextensions"
  ],
  "spell_check": true	// enable spell check
}

Note that Markdown Editing has added itself for .mdown files. You can just delete the file Markdown.sublime-settings and add the extensions in the previous file.

Enable LiveReload via Settings

Add the following to user settings for the LiveReload package LiveReload.sublime-settings:

{
  "enabled_plugins": [
   "SimpleReloadPlugin",
   "SimpleRefreshDelay"	// use SimpleRefresh for reload without delay
  ]
}

This way it does not need to be re-enabled after every start.

Enable LiveReload Manually (need to do after every Sublime launch)

  1. Open command palette and type livereload.
  2. Select LiveReload: Enable/disable plug-ins.
  3. Select Enable - Simple Reload with delay(400ms).
  4. You should see a console message saying it was enabled. Note that the menu will say Enable whether it's enabled or not, if it's already enabled choosing the menu will disable it so make sure to look at the console messages.

Configure Markdown Preview

Supposedly these two plugins work out of the box. It was not in my case. I had to add autoreload in markdown preview. I removed the github parser because it sends the markdown file to Github to be processed.

Add the following to MarkdownPreview.sublime-settings:

{
  "enable_autoreload": true,
  "enabled_parsers": ["markdown"],
  "enable_highlight": true
}

Preview Files

  1. Open command palette and select Markdown Preview: Preview in Browser.
  2. A new browser window will open to display the rendered document.
  3. Browser should update after every save.

Preview Keybind

Add the following to Default (Windows).sublime-keymap or open it via Preferences > Key Bindings.

[
  {
  "keys": [
    "alt+p"
  ],
  "command": "markdown_preview",
  "args": {
    "target": "browser",
    "parser": "markdown"
  }
  }
]
  1. Install package Markdown TOC.
  2. Add the following user settings to MarkdownTOC.sublime-settings:

    {
    "default_autoanchor": true,  // adds anchor to headings in not present
    "default_autolink": true,	 // make the ToC linkable
    "default_bracket": "round",  // use parentheses for link (e.g. [linktext](linkreference)) otherwise both will be brackets (square)
    "default_depth": 0  // 0 = unlimited depth
    }
  3. In any markdown document, put the cursor where you want the T0C. Then Tools > MarkdownTOC > Insert TOC.

  4. All settings can also be controlled in the T0C by adding them to the top comment. For example:

    <!-- MarkdownTOC style="round" autolink="true" depth=0 -->
    
    - [Heading 1]
    - [Heading 2]
    
    <!-- /MarkdownTOC -->

TOC Keybind

Add the following to the keymap Default (Windows).sublime-keymap:

{
  "keys": [
    "alt+t"
  ],
  "command": "markdowntoc_insert"  // update is markdowntoc_update
}

By default in markdown editing, bold marker is __ and italic is _. I am more used to ** for bold and * for italic.

Simply add a file named Bold and Italic Markers.tmPreferences and put it in the usual User directory.

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple Computer//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
  <key>name</key>
  <string>Bold and Italic Markers</string>
  <key>scope</key>
  <string>text.html.markdown</string>
  <key>settings</key>
  <dict>
    <key>shellVariables</key>
    <array>
      <dict>
        <key>name</key>
        <string>MD_BOLD_MARKER</string>
        <key>value</key>
        <string>**</string>
      </dict>
      <dict>
        <key>name</key>
        <string>MD_ITALIC_MARKER</string>
        <key>value</key>
        <string>*</string>
      </dict>
    </array>
  </dict>
  <key>uuid</key>
  <string>E3F0F1B0-53C8-11E3-8F96-0800200C9A66</string>
</dict>
</plist>

Essentially both do the same.

  1. Snippets are XML while Completions are JSON.
  2. Snippets are easier to read because you can have new lines in CDATA tags while you have to escape doublequotes and use special characters for new lines (e.g. \n).
  3. Only one snippet is allowed per file while you can have multiple completions in one file.

Mainly as a result of 3, I went with completions because it's just one file and easier to manage (although harder to read).

Snippet

Tools > Developer > New Snippet will create and open a template. Files are stored in the User directory similar to config files (although packages can have their own snippets). Extension for snippets is sublime-snippet.

Unofficial documentation page: http://docs.sublimetext.info/en/latest/extensibility/snippets.html.

Actual snippet is in the content tag which supports new lines inside CDATA.

For example the Snippet for my Hugo shortcode codecaption (link). Remove the space in {{[space]< to get the actual snippet or look at the repo (otherwise my theme's shortcode would break the codeblock).

<snippet>
  <content><![CDATA[{{ < codecaption title="$1" lang="$2" >}}
${3:default text}
{{ < /codecaption >}}
    ]]></content>
  <!-- Optional: Set a tabTrigger to define how to trigger the snippet -->
  <tabTrigger>codecap</tabTrigger>
  <!-- Optional: Set a scope to limit where the snippet will trigger -->
  <scope>text.html.markdown</scope>
  <!-- Optional: Description to show in the menu -->
  <description>Codecaption Hugo Shortcode</description>
</snippet>

$1 means the cursor will be there after the snippet is activated, after typing and pressing tab we will jump to $2. $3 has a default text which will be highlighted after cursor jumps to it and can be overwritten.

scopeis where the snippet will be active. Without a scope, it's active in all documents. To get the current scope press ctrl+alt+shift+p or through Tools > Developer > Show Scope Name.

Same thing can be done for imgcaption (link). Remove the space in {{[space]< to get the actual snippet.

<snippet>
	<content><![CDATA[{{ < imgcap title="$1" src="/images/2017/${2:imagepath}" >}}]]></content>
	<!-- Optional: Set a tabTrigger to define how to trigger the snippet -->
	<tabTrigger>imgcap</tabTrigger>
	<!-- Optional: Set a scope to limit where the snippet will trigger -->
	<scope>text.html.markdown</scope>
	<!-- Optional: Description to show in the menu -->
  <description>Imagecaption Hugo Shortcode</description>
</snippet>

Completions

I prefer completions because all can be in one file. For most purposes we can treat them like snippets. They are JSON files so escape " with \" and new line is \n.

Link to unofficial wiki: http://docs.sublimetext.info/en/latest/reference/completions.html.

Completions are stored in the User directory with extension .sublime-completions.

Sample completion file for markdown for the same shortcodes. Note the triggers are the same as the ones in snippet examples above. Snippets always have priority over completions. Remove the space in {{[space]<.

{
   "scope": "text.html.markdown - source - meta.tag, punctuation.definition.tag.begin",

   "completions":
   [
      { "trigger": "codecap\tCodecaption Hugo Shortcode", "contents": "{{ < codecaption title=\"$1\" lang=\"$2\" >}}\n${3:default text}\n{{ < /codecaption >}}" },
      { "trigger": "imgcap\tImagecaption Hugo Shortcode", "contents": "{{ < imgcap title=\"$1\" src=\"/images/2017/${2:1.png}\" >}}" }
   ]
}

Note the trigger, the first part is the actual trigger and everything after \t is the hint that appears in the autocompletion list similar to snippet descriptions.

There are tons of completion files for different languages and frameworks online and in package control.

Snippet and Completion Triggers

In short type the trigger (or parts of it if it's unique) and press tab. If there are snippets and completions with the same triggers, snippets always have priority.

  • Typing codecap and pressing tab will activate the snippet/completion.
  • Typing code and ctrl+space will show it in the auto-complete menu.
  • Typing snippet in the command palette will show all snippets for the current scope along with their triggers.

文章来源: https://parsiya.net/blog/2017-07-08-from-atom-to-sublime-text/
如有侵权请联系:admin#unsafe.sh