The Wonderful World of Javascript
Tags: google, sync
Sun Jun 17 19:04:32 PDT 2001
The Wonderful World of Javascript
I started thinking about a way to take a visible page in a browser, and configure some Plucker parameters on that page, and then have it gathered for you, or have the values build a home.html file for you, so you can Pluck that page at your next frequency.
I’ve never done Javascript programming before (can I really call it that?), but here’s what I have so far after about 2 hours of tinkering (it’s a Javascript “popup” wizard walkthrough sort of thing), which will prompt for the URL (it snarfs it from the current page), an expiration time on the cookie (defaults to 1 day), and then the maxdepth of that URL, and sets two local cookies for those values. When it’s all working (or I hit the string length limit on bookmark Location fields), it will walk the user through the full
gamut of settings for that URL.
javascript:q7Hm8=prompt('This webpage has ' + document.links.length + ' links.
Below is the URL of the current webpage. We can now store this value in a
local cookie on your machine. Once your selected Plucker parameters are set,
we can then create a home.html file for you with these parameters. ',location.href);
if(q7Hm8!=null) {
// Prompt for two separate vals
// which will be stored in two
// cookies on the local machine
pVj5D=86400000*prompt('Expires in...(days)','1');
m3xD=prompt('How deep would you like to gather' +
location.href,'1');
// Prompt for the maxdepth value here
1m4g3s=prompt('How deep would you like to gather' +
location.href,'1');
dT9v=new Date;dT9v.setTime(dT9v.getTime()+pVj5D);
// Store the first cookie (URL)
void(document.cookie='PluckerURL='+escape(q7Hm8)+';
expires='+dT9v.toGMTString());
// Store the second cookie (maxdepth)
void(document.cookie='PluckerMaxdepth=
'+escape(m3xD)+';
expires='+dT9v.toGMTString());}else{void(null)
}
So far, this works, and properly sets the cookies. Why do I want to do this in a Javascript fashion? because my goal
here is to make a Plucker “bookmarklet” that most users can use to ease their pain of Plucker configuration
and sync’ing.
An example of how it can be used (for this current page) is here
Reading the cookies back is a bit simpler. I won’t bore you with the code, but it allows me to do some interesting things with it. The problem with Javascript is that I can’t launch a local client application (plucker-build in this example, which is used to actually gather the content itself), nor can I write to a local configuration file. I can, however, put the data in a popup browser window, and have the user do a SaveAs from the menu on that window. It’s not the best solution, but short of having to write several different plugins for each architecture, it will work for the moment.
I’m still trying to find out if Javascript has checkbox and <option select…> type of elements so I can make a real application out of this, in a client-side popup dialog “wizard” thingy.
Why am I not doing this fully in a webpage-style application? Because then you lose focus of the webpage you’re on that you want to Pluck, and this must run client-side. I suppose I could make a webpage that contained similar code, and then submitted the URL to my server for final parsing and gathering, but my server doesn’t have the bandwidth for that right now.
This all started from my original search
google bookmarklet. You can click on it and it will pop up an entry dialog, or you can highlight words on the webpage, and click on it (go ahead, try it) and it will send that highlighted text to google as the search criteria.
I started playing a bit, and came up with another weird one to translate
the current webpage into German (or any other language)
Mantis
The bugtracker is up. Currently supports Plucker, pilot-link, pilot-mailsync, POSE, and a few other projects. I made some cosmetic changes to the layout, and cleaned up some of the PHP code. I’m trying to learn the language, but it’s slow going. Worked for a year with rasmus, and didn’t even use the language once for anything production. EEP!
As the Task List Turns… (unordered)
- Embedded Linux Course, tightening deadlines, scope creep.
- Plucker Bookmark Assistant needs a version update and will begin handling IE as well as Netscape and Mozilla bookmarks. I don’t know what format Konqueror uses, or some of the other browsers, but I can support them too, in time. I found URI::Bookmark which may help a bit.
- The Plucker Perl Spider needs a revisit, and an update. Hasn’t been touched in 11 months. EEP! I have all new ideas for it.
- pilot-link 0.9.5 needs to be released. I wish we could get the final parts of that Ralf’s USB sync fixes pushed in.
- Update the main gnu-designs webpage. I have a new design idea here that I’m tinkering with.
- Chest x-ray.
- …buncha other things…
It’s been a busy Sunday. This week promises to be incredibly productive for me. I still have to be cranking basically 26 pages a day out of this Embedded Linux Course.
My hands hurt.