You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@couchdb.apache.org by faust 1111 <fa...@gmail.com> on 2010/04/29 19:37:44 UTC

Two kinds of Attachments in doc

Hi guy's

I need modeling this:

Audio
  has_attachment
    cover_img
    trac_mp3

{
  _attachments: {
     'super_cover.png': {content_type: ''},
     'my_track.mp3': {}
  }
}

how later i can identify, which file is cover and which is track?
Now i have only this idea:
  store file name in additional field cover_attach_name, trac_attach_name.

But may be another way?

Re: Two kinds of Attachments in doc

Posted by Anh <7z...@gmail.com>.
Your first approach would work, or you could also create an additional
document for one attachment, which has a "type" property.

For example:

{
   "_id": "mytrack.mp3"
   "_attachments": {
      "file": {content_type: ''}
   }
}

{
   "_id": "super_cover.png",
   "song": "mytrack.mp3"
   "type": "cover",
   "_attachments": {
      "file": {content_type: ''}
   }
}

"Splitting" it like this may offer some additional flexibility for views, etc.
Hard to say without knowing the use-cases better.

Hope this helps.


On Thu, Apr 29, 2010 at 11:11 AM, faust 1111 <fa...@gmail.com> wrote:
> It's not simple to me,
> to get url for attach i must parse _attachments.each.content_type == /^image/
>
>
>
> 2010/4/29 Randall Leeds <ra...@gmail.com>:
>> content_type starts with image/ it's the cover, if it starts with
>> audio/ it's the track?
>>
>> On Thu, Apr 29, 2010 at 10:37, faust 1111 <fa...@gmail.com> wrote:
>>> Hi guy's
>>>
>>> I need modeling this:
>>>
>>> Audio
>>>  has_attachment
>>>    cover_img
>>>    trac_mp3
>>>
>>> {
>>>  _attachments: {
>>>     'super_cover.png': {content_type: ''},
>>>     'my_track.mp3': {}
>>>  }
>>> }
>>>
>>> how later i can identify, which file is cover and which is track?
>>> Now i have only this idea:
>>>  store file name in additional field cover_attach_name, trac_attach_name.
>>>
>>> But may be another way?
>>>
>>
>

Re: Two kinds of Attachments in doc

Posted by faust 1111 <fa...@gmail.com>.
It's not simple to me,
to get url for attach i must parse _attachments.each.content_type == /^image/



2010/4/29 Randall Leeds <ra...@gmail.com>:
> content_type starts with image/ it's the cover, if it starts with
> audio/ it's the track?
>
> On Thu, Apr 29, 2010 at 10:37, faust 1111 <fa...@gmail.com> wrote:
>> Hi guy's
>>
>> I need modeling this:
>>
>> Audio
>>  has_attachment
>>    cover_img
>>    trac_mp3
>>
>> {
>>  _attachments: {
>>     'super_cover.png': {content_type: ''},
>>     'my_track.mp3': {}
>>  }
>> }
>>
>> how later i can identify, which file is cover and which is track?
>> Now i have only this idea:
>>  store file name in additional field cover_attach_name, trac_attach_name.
>>
>> But may be another way?
>>
>

Re: Two kinds of Attachments in doc

Posted by Randall Leeds <ra...@gmail.com>.
content_type starts with image/ it's the cover, if it starts with
audio/ it's the track?

On Thu, Apr 29, 2010 at 10:37, faust 1111 <fa...@gmail.com> wrote:
> Hi guy's
>
> I need modeling this:
>
> Audio
>  has_attachment
>    cover_img
>    trac_mp3
>
> {
>  _attachments: {
>     'super_cover.png': {content_type: ''},
>     'my_track.mp3': {}
>  }
> }
>
> how later i can identify, which file is cover and which is track?
> Now i have only this idea:
>  store file name in additional field cover_attach_name, trac_attach_name.
>
> But may be another way?
>