(And of course, whever my stupid OE sees an HTML
attachement, it goes into HTML mode)
I was about to vote -1, actually.
Basically, at the very least, I would have liked to release 1.0
before.
Besides that change would force an application to use log4j. Something like
an HTTP client library should be as lightweight and as unobtrusive as possible
IMO.
I also don't like log4j a lot anymore (but that's my personal opinion, so
you can ignore that).
There is a setDebug call on HTTPClient.
Not true. Slide doesn't use log4j. As many projects probably have, it has a logger interface, and an implementation of the logger interface which can use log4j (by using it is 100% optional). Actually, what I would do eventually in the client lib is get rid of all the logging calls, and replace them by using hooks like the ConnectionInterceptor. That way the application gets all the significant events about what's going on, and can decide what (or what not) to log. This is much more flexible, and allows for embedding the client cleanly.
There was no vote or anything like that.
Remy |
Title: [httpclient] logging and testing changes
- [httpclient] logging and testing changes Waldhoff, Rodney
- Re: [httpclient] logging and testing changes Morgan Delagrange
- Re: [httpclient] logging and testing changes Jason van Zyl
- RE: [httpclient] logging and testing changes Remy Maucherat
- RE: [httpclient] logging and testing changes Waldhoff, Rodney
- Re: [httpclient] logging and testing changes Remy Maucherat
- Re: [httpclient] logging and testing change... Jason van Zyl
- RE: [httpclient] logging and testing ch... Paulo Gaspar
- Re: [httpclient] logging and testi... Jason van Zyl
- RE: [httpclient] logging and t... Paulo Gaspar
- RE: [httpclient] logging and testing changes Kevin Seguin
- Re: [httpclient] logging and testing changes Scott Sanders
- Re: [httpclient] logging and testing change... Remy Maucherat
- [httpclient] [VOTE] HTTP client 1.0 rel... Remy Maucherat