Boost logo

Boost :

From: David Abrahams (dave_at_[hidden])
Date: 2005-05-31 20:26:46


Rob Stewart <stewart_at_[hidden]> writes:

> From: David Abrahams <dave_at_[hidden]>
>> Rob Stewart <stewart_at_[hidden]> writes:
>> > From: David Abrahams <dave_at_[hidden]>
>> >> "Hendrik Schober" <SpamTrap_at_[hidden]> writes:
>> >> > David Abrahams <dave_at_[hidden]> wrote:
>> >> >>
>> >> It's currently:
>> >>
>> >> ---------------------------------------------------------------------
>> >> *** If you don't need Boost.Python, you can ignore this section ***
>> > ^
>> > Add a semicolon.
>>
>> These are not supposed to be sentences. The semicolon adds nothing.
>> If I were trying to follow rules for writing English prose the ***s
>> would have to go too.
>
> You capitalized the first word in the first line, but not the
> first in the second line. Thus, I took them as related clauses
> needing a semicolon. You could capitalize the first word of the
> second line and it would work well enough, too.

OK.

>> >> Well, no, but telling you that we found 4471 targets and we're
>> >> updating 1123 of them has to be more cryptic than helpful!
>> >
>> > How about something along these lines:
>> >
>> > Found 4471 items to build, of which 3348 were already current.
>> > Updated 1123 items.
>>
>> Not really accurate. Most of the items are sources (headers, even),
>> so they're not "items to build." They're just nodes in the dependency
>> graph.
>
> I suggested two things in the above. One part explains why some
> weren't updated.

That part was fine.

> The other tried to find an alternative for "targets" which you
> suggested was too cryptic. Is the rest acceptable once a good
> "targets" alternative is found?

Maybe. Seems to me it would be better to give the number that were
missing or out-of-date rather than the number that were current. That
way, you could see the difference between that and the number updated
more clearly. Of course "updated" is silly at that point, I think,
because we haven't started doing any build actions yet IIRC.

> Found 4471 dependencies, of which 3348 were already current.
> Updated 1123 dependencies.

Unfortunately, "dependencies" is wrong too. The nodes in the
dependency graph includes things that are only dependents and not
dependencies. And if you were to count edges in the graph (the other
way to understand the word "dependency" you'd get different numbers --
ones that are even less useful).

>> > There's always the old standby approach of printing, in
>> > succession, the following strings:
>> >
>> > "\r-"
>> > "\r\"
>> > "\r|"
>> > "\r/"
>> > repeat
>> >
>> > That avoids spewing thousands of lines or characters of output,
>> > yet gives active feedback.
>>
>> Does that really work reliably and portably, or are some people going to see
>>
>> -\|/-\|/-\|/-\|/-\|/-\|/....
>
> I suppose that there might be a dumb terminal emulator that won't
> do the right thing, but I wouldn't expect that to be at all
> common. Even emacs's shell mode, for which $TERM is "dumb,"
> works (tested using
>
> printf "testing -"; sleep 1; printf "\rtesting \\"; sleep 1; printf "\rtesting |"; sleep 1; printf "\rtesting /"; sleep 1; printf "\rtesting...done\n"
>
> from bash).

Okay, I'm all for it.

-- 
Dave Abrahams
Boost Consulting
www.boost-consulting.com

Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk