Mailinglist Archive: zypp-devel (68 mails)

[Jan Kupec <jkupec@xxxxxxx>]

I will just write a "diff" against the last status:

Jan Kupec wrote:
> Done:

- I separated things that several metadata files have in common from
PrimaryFileReader into FileReaderBase class.

> - YUMParser class using RepomdFileReader to get a list of files to parse
> and calling corresponding *FileReaders to do the job. YUMParser also
> defines callbacks which recieve data from parsers and use CacheStore to
> put it into database.
> - primary.xml.gz parser almost complete, the rest are just stubs.
> 
> 
> Pending:
> 
> - patch parser
Made some progress here: need to parse <atoms> yet.

> - suse-primary extensions (are there any docs about these?)
> - filelists (both from filelists.xml.gz and primary.xml.gz) [easy]
> - other.xml [easy]
> - groups
> - patterns, products
> 
> 
> CacheStore API needed for:
> 
> - storing resolvable details (attributes other than NVRAD)

Duncan provided one.

> - inserting filelist entries
> - changelog entries
> 
> 
> My previous estimate was too optimistic :O( I'd say i need five more
> days to have all YUMParser jobs done.

This time the estimate is (hopefully) right, but so far i spent only
something like two and a half day on the parser (i have been busy
sorting yast2maintainers bugs and fixing some important bugs of mine).

So, optimistically, the parser will be ready by the end of the week.

Jano

PS: Question for Michael:

I see a lot of classes in libzypp are designed following a certain
pattern with public API separated from implementation. Is it the Bridge
pattern? I have not found time to learn much about it so far. Should i
design YUMParser's classes that way?

-- 
To unsubscribe, e-mail: zypp-devel+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: zypp-devel+help@xxxxxxxxxxxx