text:{<sh>...or how to create index pages easily</sh>
The 'pullout' command in Zonk is one of its most powerful features. It scans an already created html page looking for <ref cpulli>Zonk pull family</r> commands and can assign variables depending on what it finds. It will also re-write the existing html based on the value of specified Zonk variables.
<b>IMPORTANT NOTE: because Zonk rewrites a pullout file as it reads it, and especially while Zonk is in Beta, you should always have a backup of any such file in case of problems.</b> Normally this html file will itself have been created from a Zonk datafile and template so recreating it in the event of an abnormal termination is straightforward.
The command which starts a pullout is placed in a datafile as detailed in <ref dlay>datafile layout</r>.
There is a subtle processing point that you need to be aware of here. $pullout commands are <i>not</i> executed the moment they are found in a datafile, but instead added to a first-in-first-out list of $pullout files to process. This list is only processed when, in the datafile, either $endpage is encountered or the end-of-file. The reasoning behind this is so that, when a pullout file is being processed, Zonk knows what the preceeding and next pullout files are (if any), so that links can be made to these files inside the pullout if needed. As a side effect all your variable definitions, will have been read.
Used in combination with <ref dlay>$readlog</r>, a wide range of index type pages can be created quickly and easily, whilst also maintaining the correct links between subpages automatically.
This all comes across as more complicated than it actually is in practise. If you want to use this feature I suggest you now take a look at the details of the <ref pulli>Zonk pull family</r> commands which live inside the html and study the examples given.
