[Sugar-devel] Technique to extract all records from xapian DB?

[Tomeu Vizoso]

On Wed, Dec 16, 2009 at 20:21, Martin Langhoff
<martin.langhoff at gmail.com> wrote:
> On Wed, Dec 16, 2009 at 9:46 PM, Tomeu Vizoso <tomeu at sugarlabs.org> wrote:
>> On Wed, Dec 16, 2009 at 17:48, Martin Langhoff
>> <martin.langhoff at gmail.com> wrote:
>>> On Fri, Dec 11, 2009 at 2:35 PM, Tomeu Vizoso <tomeu at sugarlabs.org> wrote:
>>> Yeah. It's a very complex beast that, after following the callstack
>>> hither and yon, only makes a v simple mapping. I've written a 20 line
>>> mini-reader class using xapian directly. Instead of following the DS
>>> code, I used the CLI tools that xapian includes, and it is a trivial
>>> thing. Needs a bit more time but it'll work.
>>
>> Sounds good, if you need later to move to python, we can do something
>> easily if we know the prefixes for each field.
>
> The code is all python using the xapian bindings. I used the CLI tools
> to figure out the data layout...  because I hadn't spotted the mapping
> array you've just given me :-)
>
>> I think it should be stable for the fields we care about, which is a fixed set:
>>
>> http://git.sugarlabs.org/projects/sugar-datastore/repos/mainline/blobs/sucrose-0.82/src/olpc/datastore/model.py#line379
>
> Will use that and stop guessing. They still have odd prefixes which I
> have to strip out, some do, some don't... the whole thing is rather
> strange. I don't know if I want to know the hysterical raisins behind
> it...

In case you change your mind... ;)

http://xapian.org/docs/quartzdesign.html

Regards,

Tomeu

-- 
«Sugar Labs is anyone who participates in improving and using Sugar.
What Sugar Labs does is determined by the participants.» - David
Farning