On 2/17/06, Mark D. Baushke <[EMAIL PROTECTED]> wrote:
Clarification: keywords (e.g. $RCSfile: $ $Name: $) are RCS entities ....
Hmm. Now that I look it up (http://www.badgertronics.com/writings/cvs/keywords.html), this is incorrect. $Name: $ seems to be a cvs specific keyword and thus must be expanded by CVS. It would be logical then to assume that if the sticky tag changes (i.e. cvs stat shows 'Sticky Tag: 1.1.1.1' not 'Sticky Tag:') then $Name: $ should change as well.
And from some cvs manual:
-----------------
http://www.badgertronics.com/writings/cvs/keywords.html
A little testing should show that this is true.
Hmmm again. I just tried to reproduce and some curious behaviour was observed:
- created a file and added to cvs with no keywords
- checkout file
- add $Name: $ $RCSfile$ $Author$
- commit
- tag (TAG_0)
- up (TAG_0)
- Observe: RCS keywords are properly expanded... but $Name: $ not expanded. cvs stat shows 'Sticky Tag: TAG_0'
- delete file
- cvs up
- Observe: All keywords expanded.
Maybe it should be expanded the same as RCS keywords. Any reason why not?
Regards,
--Russ
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Igor Okulist <[EMAIL PROTECTED]> writes:
> > In general, the $Name$ substitution is intended for a clean checkout of
> > your sources based on the tag.
>
> Mark,
>
> Yes, I understand the above feature.
No, from your comments, I do not believe you understand.
Clarification: keywords (e.g. $RCSfile: $ $Name: $) are RCS entities ....
Hmm. Now that I look it up (http://www.badgertronics.com/writings/cvs/keywords.html), this is incorrect. $Name: $ seems to be a cvs specific keyword and thus must be expanded by CVS. It would be logical then to assume that if the sticky tag changes (i.e. cvs stat shows 'Sticky Tag: 1.1.1.1' not 'Sticky Tag:') then $Name: $ should change as well.
And from some cvs manual:
-----------------
http://www.badgertronics.com/writings/cvs/keywords.html
$Name: $
If a sticky tag is in effect, this is the name of that tag. Otherwise
blank.
$Name: guide_release_1 $
A little testing should show that this is true.
Hmmm again. I just tried to reproduce and some curious behaviour was observed:
- created a file and added to cvs with no keywords
- checkout file
- add $Name: $ $RCSfile$ $Author$
- commit
- tag (TAG_0)
- up (TAG_0)
- Observe: RCS keywords are properly expanded... but $Name: $ not expanded. cvs stat shows 'Sticky Tag: TAG_0'
- delete file
- cvs up
- Observe: All keywords expanded.
Maybe it should be expanded the same as RCS keywords. Any reason why not?
Regards,
--Russ
The only reliable way you will get the files to substitute $Name$
for you is to do an initial checkout with of the tag you want.
CVS was not really designed to do this kind of hackery for you.
> I guess then I am looking for a re-checkout. Since many files
> are shared between different tags, build would be much faster than
> clean checkout.
Doing a 'cvs checkout' where cvs sees an existing tree will just be
converted into a 'cvs update' the local side will not see any modified
timestamps and will therefore not do the round-trip to the server to get
the replacements for $Name$.
> For "re-checkout" I should 'cd' just above the cvs controlled
> directory and use 'cvd -d'. Any other flags I should pass to cvs ?
No.
> Also does cvs store which files have/do not have kewords?
No, or at least not separately from the actual conents of the file
in the repository.
-- Mark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (FreeBSD)
iD8DBQFD9YIvCg7APGsDnFERAhdiAKCNuL4ca57OEcg0mQE84lFnLrMa9ACghHQ+
A8JiXZQ69a/tQBJZsce23R4=
=jhH1
-----END PGP SIGNATURE-----
_______________________________________________
Info-cvs mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/info-cvs
_______________________________________________ Info-cvs mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/info-cvs
