Asynchronous (Lazy) Results Fetching

When results are big, you can prevent the user browser having to fetch them at every page load but only on demand.

When the fmt parameter puts the results into a tab which is not open at page load, this will significantly improve page load times and transfer volume, when users do not open the tabs.

Mechanics

Simply add the fetch parameter into the header.

Example:

 ```bash lp fetch=md_async_example
 <heavy results causing evaluation>
 ```

Result:

1. At evaluation, the results will then not be put into the markdown directly but something like that instead (see src/lcdoc/lp.py ).

    <xterm />

         remote_content

    ![](./media/mypage.md_async_example.ansi)

1. The results will be written into a .ansi file, next to the original markdown page (into a subdirectory "media")
2. fn_frm will be the link to that file.
3. The javascript shipped with lp will pick up the xterm tags and render them. When it sees 'remote_content`, it will send an XHR to the server, downloading the results. That way the server can still remain static.

Note

The results are fetched only when the xterm tag is to be displayed - i.e. not while the tab containing it is closed.