The configtx code uses an 'append(...)' against a slice, and records the
new slice address when computing the config map.  Through sheer dumb
luck, this code works for config groups which are only nested 2 levels
deep, because the append call triggers a true new memory allocation.
However, for config groups 3 levels deep (such as consortium groups),
the append call actually re-uses the underlying memory for the slice.

This causes the path to be corrupted internally for the consortium group
items, and cause the wrong policy to be resolved when checking for a
policy which has been specified relatively.

This fix manually allocates new memory, copies it, and then appends the
element.

Change-Id: I0f4df619e006cdfebba60173156bda597d42a544
Signed-off-by: Jason Yellick <jyellick@us.ibm.com>

The current channel ID checks are to ensure that the IDs comply with the
Kafka topic naming restrictions.

However, it is possible that two different channel IDs such as TestChain
and testchain would both map to the same couchdb name, causing an
intersection and failure.

This CR changes the logic to test for the intersection of the
restrictions on CouchDB and Kafka.  In particular, the logic now
requires that Channel IDs:

1. Contain only lower case ASCII alphanumerics, dots '.' and dashes '-'
2. Are shorter than 250 characters.
3. Start with a letter

Note that this is not a true intersection, because the ledger must still
map '.' to '_' for CouchDB, but this mapping is bijective so is safe and
free from collisions.

Because the configuration key name checking was leveraging this same
code path, the existing function was duplicated, to leave the same logic
checks in place for the config.

Although the code diff is relatively high, this is largely due to the
copying and renaming.

Change-Id: Ie957cf9b8a075233bfcc5d3748e1a91e795ff067
Signed-off-by: Jason Yellick <jyellick@us.ibm.com>

note that bddtests/regression/go/ote/ote.go is just
whitespace alignment affected by linter.
add check_spelling.sh
Add check_spelling to Makefile

Change-Id: I7029e1223abede0d05758a8b3d435be061824083
Signed-off-by: Christopher Ferris <chrisfer@us.ibm.com>

The configtxgen tool currently only prints the organizations in the
channel creation configtx.  This is not enough information for an admin
to decide whether to sign or not.

This CR makes the configtx printing much verbose, including the read
set, write set, and computes the delta set for easy identification of
non-overlap between the read and write sets.

Change-Id: I3f9bfc2839ec595b1f2fc9a0cb149797ed4f862b
Signed-off-by: Jason Yellick <jyellick@us.ibm.com>
Signed-off-by: Kostas Christidis <kostas@christidis.io>

Existing `make linter` will miss the accesscontrol and common
source code directory. This patchset fix the missing and the
linter checking problem in the missed paths.

This patchset fixes FAB-2902.

https://jira.hyperledger.org/browse/FAB-2902



Change-Id: I504b856441642a0e12ed3fc7d1bb20b90cd1358c
Signed-off-by: Baohua Yang <baohyang@cn.ibm.com>

https://jira.hyperledger.org/browse/FAB-2511



The previous non-MVCC configuration system required that the modified
version numbers of the configuration elements be equal to the global
configuration sequence lock.

This makes the global sequence explicit, while allowing the individual
elements to move (or not move) their version numbers locally.  This
finally enables the partial update functionality advertised in FAB-1880.

Change-Id: Ic2f554b864e91435a8812d56eb8247cc1fa33720
Signed-off-by: Jason Yellick <jyellick@us.ibm.com>

https://jira.hyperledger.org/browse/FAB-2201



The configtx code currently shortcuts by assuming that the writeset
contains exactly the entire config. In order to support partial updates,
this writeset should be overlayed on top of the existing config, then
transformed back into a new config.

This CR still depends on the writeset containing the entire config, but,
readies for the conversion to a partial writeset.

Change-Id: I35a71a02437865abe4178c47adfb7c5479b86a22
Signed-off-by: Jason Yellick <jyellick@us.ibm.com>

https://jira.hyperledger.org/browse/FAB-2197



The configtx processing currently performs validation while applying
configuration.  In order to split this process, the config map
construction will need to be used in two different code paths, so this
CR factors it out.

Change-Id: Ib230c3f363d490fb89b83fcd64bf2c9ec4fc67a4
Signed-off-by: Jason Yellick <jyellick@us.ibm.com>