[darcs-users] hunk pattern recipe

Lele Gaifax lele at metapensiero.it
Sat Dec 3 09:45:00 UTC 2011


On Sat, 03 Dec 2011 15:13:56 +0900
"Stephen J. Turnbull" <stephen at xemacs.org> wrote:

>  > Goal: prevent the creation of too many hunks
>  > caused by alternating lines within an information entity that are
>  > modified, and lines in between that are not modified.
> 
> What do you mean be "too many hunks"?  Why is "many hunks" a problem?

Indeed. Maybe Rob meant more generically "spurious hunks" (for the
particular context he is working).

That's my main complaint against other VCs (I point to bzr here, the
other VC I use these days), being "imprecise" when it comes to
"hunk-size-reduction" indentifying too little hunks, that forces me to
either temporarily "shelve" some changes or move them slightly around
to help the software recognize the "atomic" hunks I want, instead of
collecting not-exactly-adiacent changes into a single one, just
because they are closer than a threshold.

This is where darcs shines, at least when working on common
programming language sources.

I remember David used to mention different strategies, for example to
make it do-the-right-thing on TeX sources, where you sometime want to
work on smaller chunks than blank-line-separated paragraphs, for
example when the text reflows after a single word has been inserted in
a sentence.

ciao, lele.
-- 
nickname: Lele Gaifax | Quando vivrò di quello che ho pensato ieri
real: Emanuele Gaifas | comincerò ad aver paura di chi mi copia.
lele at metapensiero.it  |                 -- Fortunato Depero, 1929.


More information about the darcs-users mailing list