Mark Phippard wrote:
> I guess I did not realize that you only planned on receiving a specific
> XML file. Since you were talking about this in the context of RegEx, I
> thought you were planning on receiving the same HTML file a browser
> receives and use RegEx to extract the issues. That is why I kept
> referring to it as HTTP-scraping. I agree that if we are going to define
> a new XML format and expect bug trackers to support it, then it is no
> different.
Parsing a html file to find the issues is something that's not possible
in a way it would work with all issue trackers out there. And it would
be much more work and be error prone.
The well defined XML file is much better here.
> I still think the solution I proposed is better. In my solution, TSVN is
> just putting a framework in place that the bug tracker has to fill on the
> client. TSVN could provide a default implementation that would work with
> unauthenticated XML, but anything else leaves it to the bug tracker.
That might work. If we provide at least the use of an XML file which can
be fetched unauthenticated then it would work for most people, and all
others still could implement their own solutions.
> While it is probably more difficult to build an integration on the client
> side, I think it is better since one person can write a smallish client
> program that anyone using Bugzilla or Tigris can use. Where as a
> server-side solution requires Bugzilla, Tigris etc. to write it, and make
> it available on your project. For example, how long before SourceForge or
> the other bug hosting sites provide this to projects they host, even
> assuming that their underlying issue trackers are enhanced for it. But
> one motivate programmer can solve it for all SourceForge projects on the
> client.
Since providing an XML file isn't difficult and can be done in a day
with e.g. PHP, I don't think that issue trackers won't implement it if
users ask for this feature.
Also, parsing the html would stop working every time the issue tracker
changes something in the UI.
> Also, having a client-side program pushes the issue of things like
> authentication on to someone else, and also opens the door for supporting
> non-HTTP solutions.
That's a good argument. I'll think of a way to implement calling such
client-side scripts.
Stefan
--
___
oo // \\ "De Chelonian Mobile"
(_,\/ \_/ \ TortoiseSVN
\ \_/_\_/> The coolest Interface to (Sub)Version Control
/_/ \_\ http://tortoisesvn.tigris.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tortoisesvn.tigris.org
For additional commands, e-mail: dev-help@tortoisesvn.tigris.org
Received on Tue Feb 22 22:10:30 2005