[RFC] bzr.jrydberg.versionedfile

[John Arbash Meinel]

Johan Rydberg wrote:
> John Arbash Meinel <john at arbash-meinel.com> writes:
> 
> 
>>>Versioned files are owned by the store that they are located in, and
>>>only the store should add files to the identity map.  Also, there may
>>>be several stores using the same map, so using "weave-" + file_id as
>>>key just don't cut it.  
>>
>>What stores are sharing the identity map? Isn't there 1 map per branch?
>>So even if you have revision-store and text-store sharing the map, they
>>don't have overlapping ids.
> 
> 
> I can imagine that we might have some freaky stores in the future.
> Better to be prepared than have to rewrite things later.
> 
> But the fundamental reason I want the API to be generic is that the
> identity map is just that; a generic way to cache information
> throughout the lifetime of a transaction.  If you want to protect
> yourself against collisions; use a more specific key.  In the case of
> versioned files, only the store should know about the cached entries,
> so it puts itself in the key.

So actually the map is per transaction.

> 
> 
>>So the next word to turn to (for me) is 'text'. And I think 'lines'
>>conveys the idea of a sequence of lines quite well.
> 
> 
> When you look at a paper, do you think of the text on it as a set of
> lines, or as a stream of characters/words?  

Interesting thought. It is hard to be accurate considering this
discussion. When I see an email, I see it as a big blob. Which I then
break down probably into paragraphs, and then into words. (I'm trying to
just look at a text and think of what I see).
But I can't really say that when I see a big text that I see a bunch of
lines. I think I mostly see a bunch of paragraphs.
Which is probably why I prefer "bzr log --short" which breaks up chunks
by a newline, rather than "bzr log --long" which breaks it up using a
separator line. Because in my head, the separator line gets merged into
the surrounding texts, and I can't chunk up the revisions.

> 
> This is really a no-issue; but I like esthetics.  I can rewrite the
> code tens times if it doesn't look good. (and sometimes I just don't
> care for some reason.)  But open source is also about compromise.
> 
> I'm +1 for *_text.
> 
> ~j
> 

I'm not sure what you mean by *_text. I think you mean "get_text()",
etc. Which I'm okay with, but what are they returning?

John
=:->


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 256 bytes
Desc: OpenPGP digital signature
Url : https://lists.ubuntu.com/archives/bazaar/attachments/20051221/f52199a4/attachment.pgp