reopen 279168 severity 279168 normal reassign 279168 reportbug, debbugs retitle 279168: reportbug needs a direct (HTTP?) submission interface to the BTS thanks
On 8/12/06, Albert Cahalan <[EMAIL PROTECTED]> wrote:
On 8/12/06, Chris Lawrence <[EMAIL PROTECTED]> wrote: > More to the point, this is a general issue with all MUAs that depend > on an MTA to accomplish delivery for them, and (IMHO) loading down > reportbug with functionality that is only necessary to diagnose > people's broken mail systems wouldn't be a very good tradeoff. So an MTA is not allowed to have a bug? :-)
Sure... it's just not reportbug's job to figure out if your MTA has a bug, any more than it's mutt's or Thunderbird's or Evolution's or mailx's.
Spam filters seem to eat half the legit email these days. There might be several on the path to bugs.debian.org.
True enough.
Supposing that the MTA is indeed misconfigured by the user's own action (can not blame Debian config), so what? It's still important to deliver the bug report. There is no need for reportbug to diagnose broken mail systems; there is only a need to deliver the bug report if at all possible and ensure that silent failure can never happen. Feel free to tell the user "It looks like the email was eaten by a grue." before offering to try some other method.
The problem is that there's no deterministic way for reportbug (or any MUA) to find out whether a mail was ever delivered to its ultimate endpoint - SMTP just doesn't work that way. And the BTS works asynchronously, so there's no way to tell the user definitively "your email was eaten by a grue" - bugs.d.o could just be having a bad day.
> You may reopen the report, but it will probably be downgraded and > marked wontfix. Please do reopen it.
I sense a pronoun shift here. I don't consider this a bug, at least not one in reportbug per se. That said, some sort of way to send some nicely formatted XML describing a bug report to bugs.debian.org via a HTTP POST request would be nice. Considering that I've never convinced the BTS developers to implement any sort of XML output (which would radically simplify the BTS parsing code and would curtail weird breakage every time they fiddle with the web page generation), however, I doubt they're in any hurry to do input either. Chris -- Chris Lawrence - http://blog.lordsutch.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
