monotone-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Monotone-devel] Changes in nvm.basic_io.inventory


From: Thomas Keller
Subject: Re: [Monotone-devel] Changes in nvm.basic_io.inventory
Date: Mon, 23 Jul 2007 00:59:21 +0200
User-agent: Thunderbird 2.0.0.5 (Macintosh/20070716)

Stephen Leake schrieb:
> And I've finished that; all tests in tests/automate_inventory and
> tests/automate_inventory_path now pass.
> 
> There are a couple FIXMEs about adding more tests; I guess I should
> work on those while I'm on a roll :).

I'm currently skimming over the bug reports wrt automate inventory of
the last months (this [0] is a good starting point) and stumbled across
another:

https://savannah.nongnu.org/bugs/?17361

Since monotone will never - and any other cross-platform software
neither - get proper support for any special file type (and I'm not
convinced anyways that looking at these kind of files is worth in
general, maybe mtn should just deny them _altogether_ without even
listing them explicitely as ignored files...?), I think a good solution
for the above bug would be the your idea of having a "--hide-ignored"
and/or "--hide-unknown-and-ignored" flag - whats the status of this anyways?

I've also added a few more tests to tests/automate_inventory which do
currently fail, just because they do require a bit more brain work to be
figured out properly - and I currently don't have the time for that.
These are:

a) newly added nodes are missing the changes "attrs" line (since we
track attribute changes with this format, we should be able to tell if a
new node was plainly added or optionally also got some attribute attached)

b) Somehow the status screws up for one certain cornercase: drop an
existing file and add a dir with the same name or vice versa. Instead of
showing "dropped, added, known", it shows "rename_source, added, known".
This is because we tell apart drops from renames by looking at the
new_node part of an entry and specifically there at the path - if this
has a non-zero-length my ill logic assumes it has to be a rename, which
is not the case. I guess we have to do something more complex with node
IDs here...


Thomas.


[0] https://savannah.nongnu.org/bugs/?16939

-- 
only dead fish swim with the stream: http://thomaskeller.biz/blog
Am Anfang war das Wort: http://www.schäuble-muss-weg.de

Attachment: signature.asc
Description: OpenPGP digital signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]