- parent.innerHTML =
- "<p>By using the <i>connectSeparated</i> attribute, it's possible to display a chart of several time series with completely independent x-values.</p> \
-\
- <p>The trick is to specify values for the series at the union of the x-values of all series. For one series' x values, specify <code>null</code> for each of the other series.</p> \
-\
- <div id='graph' style='float: right; margin-right: 50px; width: 400px; height: 300px;'></div> \
- <p>For example, say you had two series:</p> \
- <table><tr> \
- <td valign=top> \
- <table> \
- <table class='thinborder'> \
- <tr><th>x</th><th>A</th></tr> \
- <tr><td>2</td><td>2</td></tr> \
- <tr><td>4</td><td>6</td></tr> \
- <tr><td>6</td><td>4</td></tr> \
- </table> \
- </td> \
- <td valign=top style='padding-left:25px;'> \
- <table class='thinborder'> \
- <tr><th>x</th><th>B</th></tr> \
- <tr><td>1</td><td>3</td></tr> \
- <tr><td>3</td><td>7</td></tr> \
- <tr><td>5</td><td>5</td></tr> \
- </table> \
- </td> \
- </tr></table> \
-\
- <p>Then you would specify the following CSV or native data:</p> \
- <table><tr> \
- <td valign=top> \
- (CSV) \
- <pre id='csv1'></pre> \
- </td> \
- <td valign=top style='padding-left: 25px;'>\
- (native) \
- <pre id='native1'></pre> \
- </td> \
- </tr></table> \
-\
- <h3>Encoding a gap</h3>\
- <p>There's one extra wrinkle. What if one of the series has a missing \
- value, i.e. what if your series are something like </p> \
-\
- <table><tr> \
- <td valign=top> \
- <table> \
- <table class='thinborder'> \
- <tr><th>x</th><th>A</th></tr> \
- <tr><td>2</td><td>2</td></tr> \
- <tr><td>4</td><td>4</td></tr> \
- <tr><td>6</td><td>(gap)</td></tr> \
- <tr><td>8</td><td>8</td></tr> \
- <tr><td>10</td><td>10</td></tr> \
- </table> \
- </td> \
- <td valign=top style='padding-left:25px;'> \
- <table class='thinborder'> \
- <tr><th>x</th><th>B</th></tr> \
- <tr><td>1</td><td>3</td></tr> \
- <tr><td>3</td><td>5</td></tr> \
- <tr><td>5</td><td>7</td></tr> \
- </table> \
- </td> \
- </tr></table> \
-\
- <div id='graph2' style='float: right; margin-right: 50px; width: 400px; height: 300px;'></div> \
- <p>The gap would normally be encoded as a null, or missing value. But when you use <code>connectSeparatedPoints</code>, that has a special meaning. Instead, you have to use <code>NaN</code>. This is a bit of a hack, but it gets the job done.</p> \
-\
- <table><tr> \
- <td valign=top> \
- (CSV) \
- <pre id='csv2'></pre> \
- </td> \
- <td valign=top style='padding-left: 25px;'> \
- (native) \
- <pre id='native2'></pre> \
- </td> \
- </tr></table>"
+ parent.innerHTML = [
+ "<p>By using the <i>connectSeparated</i> attribute, it's possible to display a chart of several time series with completely independent x-values.</p> ",
+ "",
+ "<p>The trick is to specify values for the series at the union of the x-values of all series. For one series' x values, specify <code>null</code> for each of the other series.</p> ",
+ "",
+ "<div id='graph' style='float: right; margin-right: 50px; width: 400px; height: 300px;'></div> ",
+ "<p>For example, say you had two series:</p> ",
+ "<table><tr> ",
+ "<td valign=top> ",
+ "<table> ",
+ " <table class='thinborder'> ",
+ " <tr><th>x</th><th>A</th></tr> ",
+ " <tr><td>2</td><td>2</td></tr> ",
+ " <tr><td>4</td><td>6</td></tr> ",
+ " <tr><td>6</td><td>4</td></tr> ",
+ " </table> ",
+ "</td> ",
+ "<td valign=top style='padding-left:25px;'> ",
+ " <table class='thinborder'> ",
+ " <tr><th>x</th><th>B</th></tr> ",
+ " <tr><td>1</td><td>3</td></tr> ",
+ " <tr><td>3</td><td>7</td></tr> ",
+ " <tr><td>5</td><td>5</td></tr> ",
+ " </table> ",
+ "</td> ",
+ "</tr></table> ",
+ "",
+ "<p>Then you would specify the following CSV or native data:</p> ",
+ "<table><tr> ",
+ "<td valign=top> ",
+ "(CSV) ",
+ "<pre id='csv1'></pre> ",
+ "</td> ",
+ "<td valign=top style='padding-left: 25px;'>",
+ "(native) ",
+ "<pre id='native1'></pre> ",
+ "</td> ",
+ "</tr></table> ",
+ "",
+ "<h3>Encoding a gap</h3>",
+ "<p>There's one extra wrinkle. What if one of the series has a missing ",
+ "value, i.e. what if your series are something like </p> ",
+ "",
+ "<table><tr> ",
+ "<td valign=top> ",
+ "<table> ",
+ " <table class='thinborder'> ",
+ " <tr><th>x</th><th>A</th></tr> ",
+ " <tr><td>2</td><td>2</td></tr> ",
+ " <tr><td>4</td><td>4</td></tr> ",
+ " <tr><td>6</td><td>(gap)</td></tr> ",
+ " <tr><td>8</td><td>8</td></tr> ",
+ " <tr><td>10</td><td>10</td></tr> ",
+ " </table> ",
+ "</td> ",
+ "<td valign=top style='padding-left:25px;'> ",
+ " <table class='thinborder'> ",
+ " <tr><th>x</th><th>B</th></tr> ",
+ " <tr><td>1</td><td>3</td></tr> ",
+ " <tr><td>3</td><td>5</td></tr> ",
+ " <tr><td>5</td><td>7</td></tr> ",
+ " </table> ",
+ "</td> ",
+ "</tr></table> ",
+ "",
+ "<div id='graph2' style='float: right; margin-right: 50px; width: 400px; height: 300px;'></div> ",
+ "<p>The gap would normally be encoded as a null, or missing value. But when you use <code>connectSeparatedPoints</code>, that has a special meaning. Instead, you have to use <code>NaN</code>. This is a bit of a hack, but it gets the job done.</p> ",
+ "",
+ "<table><tr> ",
+ "<td valign=top> ",
+ "(CSV) ",
+ "<pre id='csv2'></pre> ",
+ "</td> ",
+ "<td valign=top style='padding-left: 25px;'> ",
+ "(native) ",
+ "<pre id='native2'></pre> ",
+ "</td> ",
+ "</tr></table>"].join("\n");