#base64-table {

white-space: nowrap;

font-size: 0.6em;

column-width: 6em;

column-count: 5;

column-gap: 1em;

-moz-column-width: 6em;

-moz-column-count: 5;

-moz-column-gap: 1em;

-webkit-column-width: 6em;

-webkit-column-count: 5;

-webkit-column-gap: 1em;

}

#base64-table thead { display: none; }

#base64-table * { border: none; }

#base64-table tbody td:first-child:after { content: ':'; }

#base64-table tbody td:last-child { text-align: right; }

DOMString <span data-x="dom-atob">atob</span>(DOMString data);

user agent must convert <var>data</var> to a sequence of octets whose <var>n</var>th octet is the

eight-bit representation of the code point of the <var>n</var>th character of <var>data</var>, and

then must apply the base64 algorithm to that sequence of octets, and return the result. <ref

spec=RFC4648><!--base64--></p>

<!-- Aryeh says: This seems to be what all browsers do as of January 2011 (except IE, which

doesn't support these functions at all). -->

<!-- Aryeh says: Copies Firefox behavior as of January 2011 (4.0b8). WebKit is somewhat laxer,

and Opera throws no exceptions at all. gsnedders reports Opera's behavior causes site-compat

problems, and I figure most sites depend on Firefox if on anything, so go with that. -->

<!-- Since updated to drop whitespace, based on the arguments here:

https://lists.w3.org/Archives/Public/public-whatwg-archive/2011May/0207.html

-->

<li><p>Let <var>position</var> be a pointer into <var>data</var>, initially

pointing at the start of the string.</p></li>

<li><p>Remove all <span>ASCII whitespace</span> from <var>data</var>.</p></li>

<li><p>If the length of <var>data</var> divides by 4 leaving no remainder, then: if

<var>data</var> ends with one or two U+003D EQUALS SIGN (=) characters, remove them from

<var>data</var>.</p></li>

<li><p>If the length of <var>data</var> divides by 4 leaving a remainder of 1, throw an

<span>"<code>InvalidCharacterError</code>"</span> <code>DOMException</code> and abort these

steps.</p>

<li>

<p>If <var>data</var> contains a character that is not in the following list of

characters and character ranges, throw an <span>"<code>InvalidCharacterError</code>"</span>

<code>DOMException</code> and abort these steps:</p>

<ul class="brief">

<li>U+002B PLUS SIGN (+)

<li>U+002F SOLIDUS (/)

<li><span>ASCII alphanumeric</span>

</ul>

</li>

<li><p>Let <var>output</var> be a string, initially empty.</p></li>

<li><p>Let <var>buffer</var> be a buffer that can have bits appended to it, initially

empty.</p></li>

<li>

<p>While <var>position</var> does not point past the end of <var>data</var>, run these

substeps:</p>

<ol>

<li>

<p>Find the character pointed to by <var>position</var> in the first column of the

following table. Let <var>n</var> be the number given in the second cell of the same

row.</p>

<div id="base64-table">

<table>

<thead>

<tr>

<th>Character

<th>Number

<tbody>

<tr><td>A<td>0

<tr><td>B<td>1

<tr><td>C<td>2

<tr><td>D<td>3

<tr><td>E<td>4

<tr><td>F<td>5

<tr><td>G<td>6

<tr><td>H<td>7

<tr><td>I<td>8

<tr><td>J<td>9

<tr><td>K<td>10

<tr><td>L<td>11

<tr><td>M<td>12

<tr><td>N<td>13

<tr><td>O<td>14

<tr><td>P<td>15

<tr><td>Q<td>16

<tr><td>R<td>17

<tr><td>S<td>18

<tr><td>T<td>19

<tr><td>U<td>20

<tr><td>V<td>21

<tr><td>W<td>22

<tr><td>X<td>23

<tr><td>Y<td>24

<tr><td>Z<td>25

<tr><td>a<td>26

<tr><td>b<td>27

<tr><td>c<td>28

<tr><td>d<td>29

<tr><td>e<td>30

<tr><td>f<td>31

<tr><td>g<td>32

<tr><td>h<td>33

<tr><td>i<td>34

<tr><td>j<td>35

<tr><td>k<td>36

<tr><td>l<td>37

<tr><td>m<td>38

<tr><td>n<td>39

<tr><td>o<td>40

<tr><td>p<td>41

<tr><td>q<td>42

<tr><td>r<td>43

<tr><td>s<td>44

<tr><td>t<td>45

<tr><td>u<td>46

<tr><td>v<td>47

<tr><td>w<td>48

<tr><td>x<td>49

<tr><td>y<td>50

<tr><td>z<td>51

<tr><td>0<td>52

<tr><td>1<td>53

<tr><td>2<td>54

<tr><td>3<td>55

<tr><td>4<td>56

<tr><td>5<td>57

<tr><td>6<td>58

<tr><td>7<td>59

<tr><td>8<td>60

<tr><td>9<td>61

<tr><td>+<td>62

<tr><td>/<td>63

</table>

</div>

</li>

<li><p>Append to <var>buffer</var> the six bits corresponding to <var>number</var>, most significant bit first.</p></li>

<li><p>If <var>buffer</var> has accumulated 24 bits, interpret them as three 8-bit

big-endian numbers. Append the three characters with code points equal to those numbers to <var>output</var>, in the same order, and then empty <var>buffer</var>.</p></li>

<li><p>Advance <var>position</var> by one character.</p></li>

</ol>

</li>

<li>

<p>If <var>buffer</var> is not empty, it contains either 12 or 18 bits. If it contains

12 bits, discard the last four and interpret the remaining eight as an 8-bit big-endian number.

If it contains 18 bits, discard the last two and interpret the remaining 16 as two 8-bit

big-endian numbers. Append the one or two characters with code points equal to those one or two

numbers to <var>output</var>, in the same order.</p>

<p class="note">The discarded bits mean that, for instance, <code data-x="">atob("YQ")</code> and

<code data-x="">atob("YR")</code> both return "<code data-x="">a</code>".</p>

</li>

<li><p>Return <var>output</var>.</p></li>

<!-- Note: this function is defined explicitly here because RFC4648 does not specify how to handle

erroneous input, and no preexisting browser implementation simply throws an exception on all

erroneous input. -->

<dt id="refsRFC4648">[RFC4648]</dt>

<dd><cite><a href="https://tools.ietf.org/html/rfc4648">The Base16, Base32, and Base64 Data Encodings</a></cite>, S. Josefsson. IETF.</dd>