Home >  Archive >  2010 >  July >  11


 

Previous / Next

A Remote Edit Test Post

By Marc Barrot on Sunday, July 11, 2010 at 8:45 AM.

I have created this post with Scripting2 on 7/11/10; 1:44:51 PM BST. Permanent link to this item in the archive.

Hopefully, my next modification will be from iJot. Permanent link to this item in the archive.

Well, not quite yet. It seems the latest released verion of scripting2, which I am using to write this post, is 0.28. The 2 new OPML attributes used to identify the publishing server and user name start with version 0.29. Permanent link to this item in the archive.

In the meantime, I'll work from a locally doctored OPML file. Permanent link to this item in the archive.

Well, with a little help from my friend(s), I'm now on version 0.29a :-) Permanent link to this item in the archive.

On with the experiment... Permanent link to this item in the archive.

Now that I have reposted, the top heading of this entry does include the 'server' and 'username' attributes. Permanent link to this item in the archive.

Right, I now have all the interface coded in webOutliner and iJot. All that's left is coding the actual xml-rpc transaction on iJot's back end. This should be pretty straightforward, so Murphy willing, my next modification to this post should be from iJot. Permanent link to this item in the archive.

So... Is this modification being published ? Wow it sure works as advertised. Permanent link to this item in the archive.

Now I need to release the code to the alpha server. Permanent link to this item in the archive.

Done. Permanent link to this item in the archive.

Back on the scripting2 end, everything almost work. It does work pretty well actually, although I can no longer use the version in my workspace, which is now outdated by my various modifications in iJot. Permanent link to this item in the archive.

There is still a bug though: even if see the modifications in the published page, and reload the source outline in scripting2 for further modification, the opml version which was saved by edit.scripting2.com no longer loads in iJot. More investigations are in order. Permanent link to this item in the archive.

Ok, so when the OPML is saved on edit.scripting2.com by the xml-rpc server as a result from my call to scripting2.savePost, the order of the attributes for the xml tag line is inverted again: Permanent link to this item in the archive.

The outline starts with <?xml encoding="utf-8" version="1.0"?> even though the opml text passed to the savePost function as third parameter was <?xml version="1.0" encoding="utf-8"?> Permanent link to this item in the archive.

The attributes inversion seems to cause all browser based XML parsers to reject the XML as malformed. Permanent link to this item in the archive.

But curiously, the OPML documents I generate directly from the scripting2 tool have the attributes in the proper sequence.  Permanent link to this item in the archive.

Found out the code in scripting2 was expecting an OPML document encoded in ISO-8859-1. As a matter of fact, the whole of Frontier seems to be dealing in ISO-8859-1. Permanent link to this item in the archive.

Whereas iJot's XML is all encoded in UTF-8. Should not be a problem, but it turns out the code to re-order the tag attributes in scripting2 expects an ISO-8859-1 encoding declaration. Permanent link to this item in the archive.

So now iJot re-encodes the OPML text to ISO-8859-1 before saving it back to scripting2. Let's see if this works better... Permanent link to this item in the archive.

The opml is saved properly. But this post seems to have problems rendering on scripting2... Permanent link to this item in the archive.




 
About the author

Contact me

My sites

Recent stories

Calendar

July 2010
Sun
Mon
Tue
Wed
Thu
Fri
Sat
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31

Jun   Aug

To-do

Done


A picture named xmlMini.gif



© Copyright 2010 Marc Barrot. Last update: Tuesday, July 20, 2010 at 4:34 PM Eastern. Last build: 7/29/2010; 6:11:17 PM. "It sure is worse than it appears" RSS feed for Scripting2 Reverse Engineering


Previous / Next