Github user kxepal commented on the pull request:

    https://github.com/apache/couchdb-chttpd/pull/33#issuecomment-98387637
  
    @nolanlawson 
    
    1) Yes. Couchbase spec assumes and actually there is no point of having 
atts_since without returning attachments back.
    
    2) `revs=true` on other hand need to specify explicitly and this option 
applied on all the docs specified in payload.
    
    3) No, `open_revs` [explicitly 
banned](https://github.com/apache/couchdb-chttpd/pull/33/files#diff-eb6d44751e9360ac262be41e9f69fab2R1532)
 to prevent confusion. According Couchbase implementation there is the 
following logic: if no `rev` field specified for doc in payload, the server 
acts as if `open_revs=all` was specified. Otherwise, it returns a specific 
revision which you'd put in this field. That's a design flaw, imho, since to 
emulate `open_revs` for multiple revisions, you have to put the same document 
into payload multiple times per each conflict revision you want to get. So your 
example is correct, unfortunately.
    
    For the last one:  
    - git clone couchdb
    - cd couchdb
    - ./configure
    - cd src/chttpd
    - curl -O https://github.com/apache/couchdb-chttpd/pull/33.patch
    - git apply 33.patch
    - cd ../../
    - make
    - dev/run
    
    alternatively, instead of patching, you can edit rebar.config.script in 
order to point chttpd to my fork and PR feature branch, but it required more 
typing and I may not guess your favourite editor (: 


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---

Reply via email to