[petsc-dev] dates of last change on PETSc web pages

[Barry Smith]

On Mar 20, 2012, at 10:27 PM, Jed Brown wrote:

> On Tue, Mar 20, 2012 at 22:15, Barry Smith <bsmith at mcs.anl.gov> wrote:
> http://mercurial.selenic.com/wiki/KeywordExtension  seems to be what we should use. It appears that the hg makes sure the working copy always has the value (Date of last change in our case) in files one sees.
> 
> I don't really want to read diffs of all those dates. 

   You will not be reading diffs of dates if the keyword extension works properly (i.e. like the RCS stuff works)

>  
> Proving Linus thinks like Satish:
> 
> Not sure this will ever be in Git. To quote Linus: "The whole notion of keyword substitution is just totally idiotic. It's trivial to do "outside" of the actual content tracking, if you want to have it when doing release trees as tar-balls etc."
> 
> Are we that concerned with src/docs/website being unnamed before writing the tarball?

    What does unnamed mean? 

    I am concerned about any PETSc webpages anywhere not having dates on them because as you pointed out it can be problematic. So manually written webpages get their dates automatically when "accessed" from hg and automatically generated webpages get their dates when the manual page is made (nothing gets the dates when the tarball is made).


> 
> I think the date should go into the generated man pages while they are generated because that is when you know the modification date of the file.

    Correct and they will.  In factor for generated manual pages shouldn't we also put the PETSc version next to the date, easy enough for Satish to do.


   Barry