Hi, First of all, I'm so much confortable with your modification, and I understand that you keep the original behaviour. I'm proposeing to rewrite http-client. Sorry for the confusion.
--daishi At Thu, 14 Feb 2008 15:36:42 -0800 (PST), Elf wrote: > > > On Fri, 15 Feb 2008, Daishi Kato wrote: > > > Hi, > > > > Why on earth do you want to add "Connection: close" in HTTP/1.0? > > I think it's meaning-less, I'd propose just > > to remove adding this header in http:POST, http:GET and http:send-request. > > > > Although I'm not sure why http:GET and http:send-request has been > > adding "Connection: close". > > why did i add it? i didnt add it. it was there originally when i started > modifying the one function. the goal was not to do a complete rewrite of the > http egg, it was to make one particular procedure work correctly over its > inputs. (the previous state had erroneous assumptions on content.) i left it > there because a) it doesnt hurt anything to make it explicit b) it was put > there initially by someone who knew what they were doing and c) i had neither > the time nor inclination to try tracing and refactoring an entire egg, > especially one that is widely used and is a core component of a lot of > interesting very active things. the one method changed will act exactly the > way it did before i changed it for the various eggs that use it. if i > randomly > remove things, i cannot guarantee that things wont break. (not that i can > guarantee this regardless, but i can demonstrate that for all eggs currently > in the repo that use http-client will get exactly the same request objects > sent out as they did before i changed it.) > > > -elf > > > ****************************** > XREA.COM -Free Web Hosting- > http://www.xrea.com/ > ****************************** _______________________________________________ Chicken-users mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/chicken-users
