Update README.md

README.md

@@ -2,13 +2,8 @@ A collection of 38,176 emoji images from Facebook, Google, Apple, WhatsApp, Sams
 
 * Counts: Facebook=3664, Google=3664, Apple=3961, WhatsApp=3519, Samsung=3752, JoyPixels=3538, Twitter=3544, emojidex=2040, LG=3051, OpenMoji=3512, Microsoft=3931.
 * Sizes: Facebook=144x144, Google=144x144, Apple=144x144, WhatsApp=144x144, Samsung=108x108, JoyPixels=144x144, Twitter=144x144, emojidex=144x144, LG=136x128, OpenMoji=144x144, Microsoft=144x144.
-* The emoji code points are at the end of the filename. For example, the filename "astronaut-dark-skin-tone_1f9d1-1f3ff-200d-1f680.png" can be converted to an emoji in several different ways (in JavaScript):
-  * `"\u{1f9d1}\u{1f3ff}\u{200d}\u{1f680}"`
-  * `String.fromCodePoint(0x1f9d1, 0x1f3ff, 0x200d, 0x1f680)`
-  * More generally: `String.fromCodePoint(...filename.split("_")[1].split("-").map(n => parseInt(n, 16)))`
 * The tar files directly contain the image files (they're not inside a parent folder).
-
-Here's some JavaScript code to convert the file name of an emoji image into the actual Unicode emoji character:
+* The emoji code points are at the end of the filename, but there are some adjustments needed to parse them into the Unicode character consistently across all sets of emojis in this dataset. Here's some JavaScript code to convert the file name of an emoji image into the actual Unicode emoji character:
 ```js
 let filename = ...;
 let fixedFilename = filename.replace(/(no|light|medium|medium-light|medium-dark|dark)-skin-tone/, "").replace(/__/, "_").replace(/--/, "-");