Re: [boost] Best way to notify of regresssion test failures?

From: Beman Dawes (bdawes_at_[hidden])
Date: 2002-01-21 16:46:59

At 11:40 AM 1/21/2002, David Abrahams wrote:
 >
 >----- Original Message -----
 >From: "Beman Dawes" <bdawes_at_[hidden]>
 >
 >> As a further refinement, we might associate an email address (or
 >addresses)
 >> with each regression test row, and email a notification if (1) the test
 >> newly begins failing, (2) the test fails, period.  Presumably option 
(1)
 >> would be run before a release, while option (2) would be run less
 >> often.  The point being to notify, without flooding developers with 
email
 >> they would soon hate.
 >
 >I would like that very much. Having your email address in the system 
would
 >not be compulsory would it?

No, certainly not.

 > People could unsubscribe from this service if
 >they were overwhelmed.

Sure.

 >I would want the ability to associate multiple email addresses.

Yes.

Hum.  The whole thing might not be all that difficult to put 
together.  I've already got a script that mashes all the lines of an HTML 
table into a single line, so failures can be pulled out by grep.  I've also 
go a script that distinguishes new failures from pre-existing failures.

If the grep result were sorted on test name, then it would be easy to 
associate with a similarly sorted file that associates the test name with a 
string of email addresses.  (That file can live in CVS, so developers can 
add or remove themselves without bother.)

Then all that is left is to send an email message.  Let's see...  Python 
smtplib can handle that.  I think we're home free.

Maybe I'll try to hack something together.

--Beman