Re: [boost] [1.35.0] Filename lengths, path lengths, leading numerics, multiple periods?

From: John Maddock (john_at_[hidden])
Date: 2007-11-23 12:16:10

Beman Dawes wrote:
> In trying to do merges into the release branch, I'm running into
> precommit check failures due to filename and path issues.
>
> I've run the inspection report for name issues.
>
> See http://mysite.verizon.net/beman/inspect-names.html
>
> The inspect program is treating as errors:
>
> * Filenames longer than 31 characters.

I believe this is still a restriction for Mac OS filenames?

> * Filenames beginning with a non-alphabetic character.
>
> * Directory depth greater than 8.

Very hard to avoid with DocBook generated documentation?

> * Path length greater than 100 characters.

Very hard to avoid with DocBook generated documentation?

> * More than one period ('.') in a name.

Should be work-roundable.

> These requirements were put in place to ensure that Boost on a CD
> would
> be readable on systems that only supported the minimum ISO 9660:1988
> standard.
>
> Since then, ISO 9660:1999 has come out, removing many of the old
> restrictions. See
> http://www.y-adagio.com/public/standards/iso_cdromr/tocont.htm
>
> The only remaining ISO 9660:1999 restriction that affects Boost is a
> maximum path length of 207. To calculate that, we assume a root in the
> form boost_X_XX_X.
>
> I burned the current trunk onto a CD, and verified that the very deep
> and long paths could be read without problems on Windows (XP SP2),
> Linux (Ubuntu 7.10), and Mac OS X (Leopard).

Whew :-)

> I recommend we remove all restrictions except for a path length limit
> of 207 characters.

I suspect the 31-character limit may still be valid, but perhaps someone can
disabuse me of that?

John.

PS Where are we with the release process?  Do we have some kind of schedule
/ deadlines for being ready to ship: I'm a little concerned about some of
the regex library tests which haven't changed in ages, but are none the less
failing due to exceptions thrown from Boost.Thread and Boost.Test.  I'm not
sure that this is necessarily these libraries fault, but we really ought to
try and get to the cause of this...