Skip to content

Conversation

@sandipnd
Copy link

No description provided.

@mnunberg
Copy link

This will be merged into master when it has undergone proper testing and when the feature is available on the server side.

This fixes a build issue in the library on MSVCs which support C99

Change-Id: I21b1916642084321fc24fb7570e2b4be5b8cc5d9
This allows polling the configuration at a specified interval (by
default, 10s) to see if new nodes have been added. This is needed to
take advantage of new non-KV nodes, for which the traditional "reactive"
not-my-vbucket reply is not applicable.

Change-Id: I4a8541eef08f09a680d08718b41625b87a324b52
This is required to allow the row parser to be in C++. This newer
version of jsonsl supports C++ drop-in mode

Change-Id: I0c58b65ae51583a13c336b42973df01dff204894
Change-Id: I5eda81b6f1d23ce88e99ae1d3e0fbcbf33a6113b
We should *always* log when a given error is unrecognized by the library
or is being squashed into a network error. This avoids the trouble of
forcing the user to re-run with detailed_errcodes or use strace in the
event of unrecognized/suppressed errors.

Change-Id: I197352868188edfeb2b98be1fec4d60fe0aaf3ad
Change-Id: If2adfedcd497a07d126ff4e95c1f6d989de95138
Change-Id: Iecc1250d81c3716696dc99fd3b3edb601bd9770f
This is helpful in situations where values are already in the format of
multiple fragments, and allows the user to avoid making a new contiguous
buffer just for this purpose.

Change-Id: I5554a5d1eddda481f2977bfecab5b0fa835d5192
Change-Id: I7d8bd6dda89dd4ffabc3488bcd1a45c78026f3c6
Change-Id: I83dad8346d2ae200d07b7f09b4d41772e3db0d97
Change-Id: I37968f133d26c6150f72d4cc231a0a307afd4042
This adds a load of subdoc tests - mainly ensuring we conform at the
protocol layer. Additional multipath and input validation tests are
still required.

Change-Id: I58303e134c6199ea727c414ef306b7dc431813ef
Change-Id: I758f5a1a50f33485a2388733bef221e3c3dc4299
The subdoc examples were previously local to my machine.

Change-Id: I04656641d2bb0b3d23da7a57c8ebc8f0ca5e08ac
Also ensure that empty keys are not sent on commands which do not allow
them (return LCB_EINVAL instead).

Change-Id: I23b65a27d6608413cae8967981c94d05aabd8a5a
Change-Id: I40e57a4e9fb1ee701c54ce66d7af28490c73994c
Change-Id: I62edb92b19e3e1c70fac8ee54f7e05e6e406313c
Change-Id: Icd680dc95150aa5b504f49fdd1a0dff0e118a96b
These commands are very similar as far as inputs are concerned

Change-Id: I6dad834171e231546817962e394b10474f48ce07
These are works in progress wrt naming. Also, durability for subdoc
isn't implemented yet (perhaps at the lcb-level, DURSTORE makes a bit
more sense?)

Change-Id: Ia930b302dc2d6379568890e9d3cdad987a7efc97
This uses traits to determine if a given command supports a given
feature.

Change-Id: If681b34626377659fca213ff7872f7c415d86dd1
Change-Id: I703c52689bed644638c0940686ba99d43cd7b902
Centralize more things inside traits system so logic looks more
streamlined

Change-Id: I1d01c863004bcd601d66b39e7901e911e15f0adb
@mnunberg
Copy link

This has already been merged.

@mnunberg mnunberg closed this Apr 11, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants