improved README

README.md

@@ -6,7 +6,7 @@ A module on top of [chjj/marked](https://github.com/chjj/marked) to extract a pa
 Run the following commands to download and install the application:
 
 ```sh
-$ npm i markdown-extract --save
+npm i markdown-extract --save
 ```
 
 ## Usage
@@ -24,10 +24,12 @@ nd it s [markdown version](https://github.com/maboiteaspam/markdown-extract/blob
 
 It exposes an unique function, with three signatures
 
+```js
     mdExtract (what[, markedOpt[, text]])
     mdExtract (what, markedOpt)
     mdExtract (what, text)
-
+```
+
 * __what__: The lookup object, regexp, or string
 * __markedOpt__: The options passed to `marked`, default to `{gfm:true}`
 * __text__: The markdown text to parse, it not provided it will fallback to `README.md` file on `cwd`
@@ -40,16 +42,16 @@ It will return only their text, as an array of string, in order of appearance in
 * if it s a `string` or a `regexp`, it selects nodes with matching type or text property
 * if it s an `array of string`, it selects nodes of those types
 * if it s a `number`, it selects nodes with given depth
-* if it s an `object`, then it can have type/text/depth properties with values that matches above logic.
+* if it s an `object`, then it can have `type`/`text`/`depth` properties with values that matches above logic.
 * if it s an `object`, and it s have a property (`grepNextParagraph` || `gnp`)==`true`, 
 then, the search is made based on `heading` nodes, 
-the results is al the following text of each matched `heading`.
+the results is all the following `text` node of each matched `heading`.
 
 The last option is the most interesting to me as you can match a `heading` such `## Usage`, 
 but gets only it s following `text`.
 
-```
-mdExtract ({type: /heading/, text: /Usage/, gnp: true}).join('\n')
+```js
+    mdExtract ({type: /heading/, text: /Usage/, gnp: true}).join('\n')
 ```
 
 ## How to contribute