[XML-SIG] [ pyxml-Bugs-562276 ] xmlproc 0.7.1 reports incorrect version

[noreply@sourceforge.net]

Bugs item #562276, was opened at 2002-05-30 06:29
You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=106473&aid=562276&group_id=6473

Category: xmlproc
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Matthew Gruenke (mgruenke)
Assigned to: Lars Marius Garshol (larsga)
Summary: xmlproc 0.7.1 reports incorrect version

Initial Comment:
On line 19 of xmlproc.py, in release 0.7.1 it says:

    version="0.70"

When I assume it means to say:

    version="0.71"


BTW, I think it's a mistake to try to treat versions as
real numbers.  They are similar, especially in
notation, but really very different.  I prefer to
implement versions as a 3-element tuple, which does
make compatibility checking a little more cumbersome
than a simple numerical comparison.  The other
alternative would be to have a version_major,
version_minor, and version_patch (or version.major,
version.minor, and version.patch, if you have a version
class), if you prefer something more structured than a
tuple.

Anyhow, if your versioning semantics follow the
traditional UNIX shared-library model, a difference in
the major version of a library is usually used to
signify an incompatibility with programs written to a
previous major version.  If this convention is
followed, than the simple numeric comparison suggested
by treating version IDs as real numbers is actually
incorrect.

In my python modules (e.g.
http://sourceforge.net/projects/xml-extractor), I
actually run the .py file through sed, to expand the
version numbers, which are maintained in a separate
file that gets sourced by my release script (which
passes the values to sed).  Admittedly, it's arguable
whether such trouble to centralize version numbers in a
single place is really worthwhile, but it does solve
the problem of keeping in synch what version the
setup.py script thinks it has, and the version reported
by the actual module.



----------------------------------------------------------------------

You can respond by visiting: 
http://sourceforge.net/tracker/?func=detail&atid=106473&aid=562276&group_id=6473