| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |\ |
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Unfortuantely, #1437 brought in a build bug that caused dist tarballs
to be created always using the last tag that could be found on the
tree. This lead to `master` building tarballs labelled `2.1.0`.
The new approach includes extensive comments to explain the approach,
fixes the bug, and for an encore adds -dirty if you're building a
CouchDB with local changes that aren't committed to git.
|
| | |
| |
| |
| | |
Closes #1309
|
| | | |
|
| | | |
|
| |/ |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
Fixes #927
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
| |
the released tarball should have -RCx in the name, but not in the
extracted file, otherwise we can't simply rename the final artefact for
our public release.
|
| |
|
|
|
|
|
|
|
|
|
| |
git-describe may have changed output formats. The previous regex doesn't
allow any trailing content, and git-describe always appends -g<SHA> unless
abbrev=0 is added.
This approach pulls out only the matching tag, leaving behind any
trailing garbage from git-describe, and as a bonus allows tagging a
commit from within a branch, and running a release directly from the
branch without needing to re-check out the tag itself.
|
| |
|
|
|
|
| |
- send a readable error response from failed config set
- trust but verify admin-supplied content in separate function
- return specific error conditions for logging
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
Previously on view compaction, the new index state would not carry
over the associated database update and purge sequences. As a
result, views who were compacted at least once could no longer
use the ?update_seq query option, or purge properly.
Closes #984
|
| | |
|
| |
|
|
|
| |
I couldn’t find a cleaner way to add _bulk_docs with
`new_edits: false`.
|
| |
|
|
|
|
| |
This allows for using Mango queries for finding docs with conflicts.
Closes #1101
|
| | |
|
| | |
|
| |
|
|
| |
Complete deletion tests with not found
|
| |
|
|
|
|
|
|
|
|
|
|
| |
After the aebdbc452573f70f4e50d88af5814d0fbe936333 stemming is done separately
from merge so stem interactive option didn't take effect. That is mostly ok as
speed improvements should reduce the need for that option, but it still might
be nice to keep the option (just in case).
Also, a nice side effect is it removes an extra external function from
couch_key_tree module and simplifies the tests a bit.
Related PR: #958
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
When we delete files with context option set to `delete`,
`couch_file` respects configuration flag "enable_database_recovery"
and just renames the files in case it's set to true.
This change removes context for compaction files deleted
during database deletion to make sure we are actually
erasing them and not just renaming and leaving behind.
|
| |\
| |
| | |
Fix for a function_clause crash in the couch_native_process for Erlang views
|
| | | |
|
| |/
|
|
| |
Crash was caused by a missing implementation of ddoc function for <<"views">> FunPath, implementation is based on FilterFun but matches return values of the erlang:put() which is called in the native Emit function and also expects ok and false when docs were not emitted.
|
| |
|
|
| |
Thanks to Just Some Enterprises for donating the compute power!
|
| | |
|
| |
|
|
| |
Issue #1396
|
| |
|
|
|
|
|
|
| |
This moves the sort check into the is_usable function for all indexes.
For map/reduce indexes it can add any constant fields e.g {a: {"$eq": 4}
to the prefix of the sort because it won't affect the actual sort but
will increase the chance that an index is selected. This is a user
experience fix to help a user if they don't add all the columns for an
index to the sort fields.
|
| |
|
|
|
|
|
|
|
|
|
| |
Passing closures around is fragile and prevents smooth upgrading. Instead pass
a tuple with a data from the receiver closure explicitly and convert to back to
a local fun locally on each node.
This is a preparatory commit before the switch. To ensure attachment uploading
requests are successful, would need to first install this change on all the
nodes. Then in a separate upgrade step, switch fabric.erl to call
fabric_doc_atts:receiver instead fabric_doc_attatchments:recevier.
|
| |
|
|
| |
Previously, when we fail authorization but pass authentication, we use
the old HttpReq which drops user_ctx. This change leaves user_ctx.
|
| |
|
|
|
|
| |
This uses the new `couch_util:set_mqd_off_heap/0` function to set
message queues to off_heap for some of our critical processes that
receive a significant amount of load in terms of message volume.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
In Erlang VMs starting with version 19.0 have a new process_flag to
store messages off the process heap. This is extremely useful for
processes that can have huge numbers of messages in their mailbox. For
CouchDB this is most usually observed when couch_server backs up with a
large message queue which wedges the entire node.
This utility function will set a process's message_queue_data flag to
off_heap in a way that doesn't break builds of CouchDB on older Erlang
VMs while automatically enabling the flag on VMs that do support it.
|
| |
|
|
|
|
|
|
|
|
| |
The `should_merge_tree_to_itself` and `should_merge_tree_of_odd_length`
tests were both invalid as merging does not support merging of anything
other than a linear path. This failure was covered up by the fact that
the stem operation will detect and cover up any errors from a failed
merge.
Co-Authored-By: Nick Vatamaniuc <vatamane@apache.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is two related optimizations for stemming revisions. The first
optimization re-writes the stemming algorithm to drop it from an O(N^2)
to O(N) operation by using a depth first search through the tree and
tracking which revisions are within `revs_limit` revs from a leaf
dropping any revision that exceeds that limit.
The second optimization is just that we avoid calling stemming more
often than necessary by switching away from using `merge/3` to `merge/2`
and then calling `stem/2` only when necessary.
Co-Authored-By: Nick Vatamaniuc <vatamane@apache.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It was brought to my attention that the active size looked a bit funny
occasionally as it could be larger than the file size. Given that active
size is defined to be the number of bytes used in a view file it must be
strictly greater than or equal to zero as well as less than or equal to
file size. Thus the great hunt had begun!
While I'd love more than anything to regail you, Dear Reader, with the
tales, the joys, the highs, and yes, even the lows of this search, it
turns out that I cannot. I must not. For there were none!
Turns out this was a trivial bug where we were re-using the ExternalSize
calculation instead of applying `couch_btree:size/1` to all of the
btrees in the `#mrview` records. Simple bug comes with a correspondingly
simple fix.
I also noticed that the info tests were broken and not being run so I
spent a few minutes cleaning those up to make the various assumptions.
|
| |\
| |
| | |
Add compile's command line options, introduce `ERL_OPTS` and make `bin_opt_info` optional.
|
