From: Rush Manbert (rush_at_[hidden])
Date: 2007-12-03 12:57:50
Rene Rivera wrote:
> Vladimir Prus wrote:
>>It's worth to generalize this, and allow a project to specify custom
>>function used to compute target paths. The question, then, is what
>>do to if the rule is changed by the user in a way that path XXX
>>is possible with previous version of the rule and with new version of
>>the rule, but with different properties.
>>Maybe I should actually implement MD5 build signatures.
> Maybe people should read the various suggestions I posted about shorter
> paths over the past few *years* ;-) The last one of which is about user
> designated feature paths
> <http://article.gmane.org/gmane.comp.lib.boost.build/14852>. It could
> certainly be extended to specify a user rule (with the @rule syntax) to,
> for example, calculate a hash.
Not really a reply to Rene, just appending to the latest post on the thread.
From a user's perspective, I would much prefer a clearly defined set of
abbreviations used to create paths over a MD5 hash. I think it's good to
be able to look at a path and figure out what it represents, even if I
need to parse the abbreviations carefully. I have found from my own
projects that you can be very aggressive with abbreviations and still
get useful names that people can understand. A hash loses all such meaning.
Just my 2 cents.
Boost-Build list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk