Epte

This is the page where Erich has his notes he needs in order to operate in the Peder's wiki environment, and to accomplish projects having to do with this environment. Any discussions with Peder will be in the Discussion area. Erich needs this page to be his notes to himself for reference.

Peder, please, don't touch this page. If you do not respect my wishes in this, I will be forced to keep my notes elsewhere. I'm doing this for you, because you asked me to and because I trust you, and it might help you see why I find this environment so difficult. I am tracking my understanding as I learn this system for your benefit.

Displaying your inner thought processes to clients is being very vulnerable and usually unwise.

Quickbooks integration

Sites

• Stamp Invoice
• Operationscan
• Residentialscan
• Commercialscan
• Ha! Here's the site I was thinking this was: homepage

Understanding Transclusion

Let's analyze the Demoscan page. Here's the text of it:

The code for Demoscan

{{refresh}}                           (Reload the page)
{{stamp_invoice_message}}             (isn't working.  intentionally so I don't commit?)

[[Demoscan/instructions]]
<br>
'''Note: Use only Firefox''' When you are done, click on Article to reload file list
<br>

[[*Picker.html|nowiki]]     
[[*Gen_validator2.html|nowiki.php]]   (nothing visible.  probably validates the data)
[[*popupdiv.html|nowiki.php]]         (displays "show" and "hide")
[[*demoscan.html|nowiki.php]]         (main form)
[[*requirefile.html|nowiki.php]]      (nothing visible)

{{change}}                            (the change dial for change requests)

And some links:

• Demoscan
  • Refresh Template
  • Stamp Invoice Message Template
  • Alert.html: Stamp Invoice is just Alert piped through nowiki
  • nowiki.php
  • Gen_validator2.html
  • Picker.html
  • popupdiv.html
  • demoscan.html
  • requirefile.html
  • Change Request Template
  • Change Request Target Page (gets there somehow)

Learning

nowiki.php

Ahh!! Finally understanding!

The name "nowiki.php" was misleading. I had thought it was suppressing output that would otherwise be wikified, like sending to /dev/null. Rather, nowiki.php allows the contents of the wiki page it is passed to be rendered as if there were no wiki engine. So passing a foo.html file to nowiki.php is equivalent to having an extra-wiki foo.html file and browsing straight to it.

Transclusion syntax

I can't use applies on div0 articles from the wikiexpert context. All experimentation will be done on the div0 level.

• You "transclude" another article with the double brackets, and can "pipe" them through "transforms"

• Use + in trsnslucsion link to embed and pipe a heading
• Use * in transclusion link to pipe thru transforms and embed the result
• See talk:document.php and apply examples

• Double braces is for a "template" transclusion. I'm still not clear on templates from having read the suggested reading.

URL Syntax Findings

So, why doesn't the stamp_invoice_message template include correctly into demoscan? The same thing happens in residential scan. The apply isn't getting applied, and the link is interpreted as the target *alert.html and named nowiki.php, as if done in double brackets without the * apply operator.

So... stamp_invoice_message never worked? Or is it just because of my permissions? There don't seem to be any error messages generated from the wiki engine that would suggest one way or the other.

Ideas for Peder

Now that I'm beginning to understand how this works, I have some ideas for how Peder's framework could be improved:

• Keep pages modular. For example, it is fortunate that Picker.html serves your needs as html head block and style sheet universally. But what if it didn't? What if you needed more flexibility that that? It would have been better from the start to have the DHTML lib in its own page, the stylesheet be in its own page, and then one html-head transform or page to tie them together.

Frustrations:

• can't be logged into div0 and wikiexpert simultaneously
• At least 12 theoretical URL permutations out of which 2 work. And I know I've seen ADEFT.com URLS... I don't feel I can trust my intuition.
• I can't easily use vim as my native editor
• Open fresh browser. Try to edit Demoscan. Login. Try to edit Demoscan. Have to login again. This has happened multiple times. I'm not that inable to type my password. Though maybe my b key is sticking... Hmm.. no. It just happened again.
• If http://www.div0.com/wiki/index.php/Fcre is a "subdirectory", how can I do an "ls" at the index.php level to see all such "subdirectories"?
• If http://www.div0.com/wiki/index.php/Fcre is a "subdirectory" how can I "grep -r" within just that subdirectory? I understand I can search the whole filesystem using the wiki search, but what if that isn't specific enough?
• nowiki tags don't work -- I get xmcode tags and html comments. As a workaround, to render applies in the article, I know of two syntaxes, one is to use XwLink templates per apply and the other is to break the brackets with nowiki tags per apply. I want a code block sort of thing. That is, there should be a <noxmwiki></noxmwiki> pair.
• Either there's a login timeout that is set very low, or logging into div0 logs me out of wikiexpert and vice versa.
• MediaWiki template syntax allows for parameters and variables. However, doing [[Epte/UrlExperiment | foo=bar ]] just sets the title, inappropriately, IMHO