Improve test coverage for errors #1068

betasve · 2020-04-12T19:57:11Z

Signed-off-by: Svetoslav Blyahov svetlio.blyahoff@gmail.com

Improve configtxgen test coverage (for error messages) as @wlahti advised in my previous PR.

Type of change

Test update

Description

Start expecting exact error messages.

wlahti

Thanks for taking the initiative to improve the tests here. Much appreciated!

One important requested change (the regexp for the path doesn't work on all operating systems) and a few nits.

cmd/configtxgen/main_test.go

betasve · 2020-04-14T18:53:14Z

Pushed some changes I think (as far as I understood your points) are addressing the concerns. Please, let me know if I missed something or something else needs improvement.

wlahti

Sorry, maybe should've been clearer in the OS-specific nature of the test that fails locally. The temp directory is different depending on the developer's OS, so we need to generalize the regular expression to not care about the beginning of the path. For reference, here's the output when running that test on my mac:

=== RUN   TestUnsuccessfulChannelTxFileCreation
2020-04-14 16:30:43.549 EDT [common.tools.configtxgen.localconfig] Load -> INFO 01b Loaded configuration: /Users/wlahti/gopath/src/github.com/hyperledger/fabric/sampleconfig/configtx.yaml
2020-04-14 16:30:43.549 EDT [common.tools.configtxgen] doOutputChannelCreateTx -> INFO 01c Generating new channel configtx
2020-04-14 16:30:43.551 EDT [common.tools.configtxgen] doOutputChannelCreateTx -> INFO 01d Writing new channel tx
2020-04-14 16:30:43.551 EDT [common.tools.configtxgen] doOutputChannelCreateTx -> INFO 01e Generating new channel configtx
2020-04-14 16:30:43.553 EDT [common.tools.configtxgen] doOutputChannelCreateTx -> INFO 01f Writing new channel tx
    TestUnsuccessfulChannelTxFileCreation: main_test.go:92: 
        	Error Trace:	main_test.go:92
        	Error:      	Expect "Error writing channel create tx: open /var/folders/gy/jb454xb90513cqzgvv7111hr0000gn/T/configtxgen143477252/configtx: permission denied" to match "Error writing channel create tx: open /tmp/configtxgen.+/configtx: permission denied"
        	Test:       	TestUnsuccessfulChannelTxFileCreation
--- FAIL: TestUnsuccessfulChannelTxFileCreation (0.03s)

betasve · 2020-04-15T20:05:07Z

Yes, sorry, I didn't really understand your point but it totally makes sense. I should have thought of that. Pushed an update.

wlahti · 2020-04-17T19:01:36Z

cmd/configtxgen/main_test.go

+	err := doOutputChannelCreateTx(config, nil, "foo", configTxDest).Error()
+	assert.Error(t, doOutputChannelCreateTx(config, nil, "foo", configTxDest), "Not able to write to file")
+	assert.Regexp(t, "Error writing channel create tx: open .+/configtxgen.+/configtx: permission denied", err)


Suggested change

err := doOutputChannelCreateTx(config, nil, "foo", configTxDest).Error()

assert.Error(t, doOutputChannelCreateTx(config, nil, "foo", configTxDest), "Not able to write to file")

assert.Regexp(t, "Error writing channel create tx: open .+/configtxgen.+/configtx: permission denied", err)

assert.EqualError(t, doOutputChannelCreateTx(config, nil, "foo", configTxDest), fmt.Sprintf("Error writing channel create tx: open %s: permission denied", configTxDest))

Not trying to stall on this one, but considering I'm not a maintainer and can't merge this in anyways, can't we simplify this down to just one line and avoid the regular expression by formatting in the path since we have the variable with the exact path. Tried this out locally and it works for me. Seems more flexible to me in case somebody down the line decided to change the name of the tmpDir.

Sure, sure, I totally got your point. Pushing a change now. Let me know about the new version.

wlahti

Looks good to me now, thanks!

caod123

Some general comments on error string conventions

caod123 · 2020-04-20T19:49:55Z

cmd/configtxgen/main_test.go

@@ -34,7 +35,7 @@ func TestMain(m *testing.M) {
 }

 func TestInspectMissing(t *testing.T) {
-	assert.Error(t, doInspectBlock("NonSenseBlockFileThatDoesn'tActuallyExist"), "Missing block")
+	assert.EqualError(t, doInspectBlock("NonSenseBlockFileThatDoesn'tActuallyExist"), "Could not read block NonSenseBlockFileThatDoesn'tActuallyExist")


I know you didn't introduce these error messages, but per the general Go programming conventions, error strings should not start capitalized https://github.com/golang/go/wiki/CodeReviewComments#error-strings would you be able to go through the production code as well as the tests in this path and update the error strings accordingly? There's a lot of places in Fabric where we don't properly follow this convention but might as well start here since we're touching this path.

caod123 · 2020-04-20T19:50:14Z

cmd/configtxgen/main_test.go

@@ -49,8 +50,8 @@ func TestInspectBlock(t *testing.T) {
 func TestInspectBlockErr(t *testing.T) {
 	config := genesisconfig.Load(genesisconfig.SampleInsecureSoloProfile, configtest.GetDevConfigDir())

-	assert.Error(t, doOutputBlock(config, "foo", ""), "Error writing genesis block:")
-	assert.Error(t, doInspectBlock(""), "Could not read block")
+	assert.EqualError(t, doOutputBlock(config, "foo", ""), "Error writing genesis block: open : no such file or directory")


caod123 · 2020-04-20T19:50:41Z

cmd/configtxgen/main_test.go

@@ -59,7 +60,7 @@ func TestMissingOrdererSection(t *testing.T) {
 	config := genesisconfig.Load(genesisconfig.SampleInsecureSoloProfile, configtest.GetDevConfigDir())
 	config.Orderer = nil

-	assert.Error(t, doOutputBlock(config, "foo", blockDest), "Missing orderer section")
+	assert.EqualError(t, doOutputBlock(config, "foo", blockDest), "refusing to generate block which is missing orderer section")


Interestingly enough we follow the convention here, further demonstrating how inconsistent error handling is in Fabric

caod123 · 2020-04-20T19:51:17Z

cmd/configtxgen/main_test.go

-	assert.Error(t, doOutputBlock(config, "foo", ""), "Error writing genesis block:")
-	assert.Error(t, doInspectBlock(""), "Could not read block")
+	assert.EqualError(t, doOutputBlock(config, "foo", ""), "Error writing genesis block: open : no such file or directory")
+	assert.EqualError(t, doInspectBlock(""), "Could not read block ")


caod123 · 2020-04-20T19:51:32Z

cmd/configtxgen/main_test.go

+	assert.NoError(t, ioutil.WriteFile(configTxDest, []byte{}, 0440))
+	defer os.Remove(configTxDest)
+
+	assert.EqualError(t, doOutputChannelCreateTx(config, nil, "foo", configTxDest), fmt.Sprintf("Error writing channel create tx: open %s: permission denied", configTxDest))


caod123 · 2020-04-20T19:51:56Z

cmd/configtxgen/main_test.go

@@ -115,15 +126,15 @@ func TestBadAnchorPeersUpdates(t *testing.T) {

 	config := genesisconfig.Load(genesisconfig.SampleSingleMSPChannelProfile, configtest.GetDevConfigDir())

-	assert.Error(t, doOutputAnchorPeersUpdate(config, "foo", configTxDest, ""), "Bad anchorPeerUpdate request - asOrg empty")
+	assert.EqualError(t, doOutputAnchorPeersUpdate(config, "foo", configTxDest, ""), "Must specify an organization to update the anchor peer for")


caod123 · 2020-04-20T19:52:08Z

cmd/configtxgen/main_test.go


 	backupApplication := config.Application
 	config.Application = nil
-	assert.Error(t, doOutputAnchorPeersUpdate(config, "foo", configTxDest, genesisconfig.SampleOrgName), "Bad anchorPeerUpdate request")
+	assert.EqualError(t, doOutputAnchorPeersUpdate(config, "foo", configTxDest, genesisconfig.SampleOrgName), "Cannot update anchor peers without an application section")


lindluni · 2020-04-22T21:19:27Z

/azp run

azure-pipelines · 2020-04-22T21:19:37Z

Azure Pipelines successfully started running 1 pipeline(s).

Signed-off-by: Svetoslav Blyahov <svetlio.blyahoff@gmail.com>

wlahti · 2020-05-12T14:27:12Z

Can you rebase this on master and we'll see if we can get a maintainer to merge as-is? I think the capitalization of the error messages can be done in a follow-up PR considering they've had that capitalization for years.

caod123 · 2020-05-12T15:25:02Z

Looks like there aren't any conflicts with master, merging as is

* tests for commit hashes with rollback and reset (hyperledger#1221) The COMMIT_HASH is being added to each block both in v1.1 and >= v1.4 As this hash has to be the same on all peers, it is necessary to start adding this hash from the same block on all peers. Hence, it is necessary to reset the existing channels to the genesis block inorder to add COMMIT_HASH. Otherwise, the peer would not add COMMIT_HASH to blocks. There are two new sets of tests: (1) ensure that the COMMIT_HASH is being added to the block once a peer has been reset to the genesis block. Otherwise, COMMIT_HASH would not be added. (2) assuming that COMMIT_HASH is already present, doing a rollback/reset and reprocessing the block should not result in different COMMIT_HASH (as we assume no changes in the validation results). This test also helps to ensure that any changes to the code that computes the COMMIT_HASH do not break the backward compatibility. FAB-17832 #done Signed-off-by: manish <manish.sethi@gmail.com> Signed-off-by: senthil <cendhu@gmail.com> Co-authored-by: manish <manish.sethi@gmail.com> * [FAB-17862] Allow upgrade-dbs to drop all dbs in multiple runs (hyperledger#1236) Administrators may need to rerun upgrade-dbs in some cases, e.g., if they forgot to pass the CouchDB variable CORE_LEDGER_STATE_STATEDATABASE, it will not drop the CouchDB databases as part of upgrade-dbs. Therefore, upgrade-dbs should be able to drop all dbs when running multiple times. Signed-off-by: Wenjian Qiao <wenjianq@gmail.com> * Improve test coverege for errors (hyperledger#1068) Signed-off-by: Svetoslav Blyahov <svetlio.blyahoff@gmail.com> * test for commit hash computation's backward compatibility (hyperledger#1240) We might change the code that computes the block commit hash during refactoring or to improve the performance from time to time. When changing the hash computation code, we need to ensure that we are not changing the way commit hash has been computed in previous releases. Hence, we add a test to ensure that commit hash computation is always compatible with the first code added in v1.1 FAB-17878 Signed-off-by: senthil <cendhu@gmail.com> * Add snapshot file creator and reader functions This commit adds reusable functions for creating and reading the snapshot files. These functions are expected to be used by various components that would generate or consume the snapshot files, such as blockstorage and statedb Signed-off-by: manish <manish.sethi@gmail.com> * Use snapshot file creator in blockstorage This commit leverages the snapshot file creator to generate snapshot files and computing hashes for TXIDs. FAB-17870 Signed-off-by: manish <manish.sethi@gmail.com> * [FAB-17433] Add _lifecycle function for querying approved chaincode (hyperledger#1220) Currently, Fabric peer does not provide a function to query the details of approved chaincode definitions. On the other hand, Fabric admins will need to confirm/use the information after approval for their operations. This patch adds a function for querying the details of the chaincode definition approved by the organization itself to `_lifecycle`, so that Fabric peer can provide a function to query them. Signed-off-by: Tatsuya Sato <Tatsuya.Sato@hal.hitachi.com> * [FAB-17846] Add clear statements indicating that external chaincode builder will likely require custom packaging of peer image Signed-off-by: Tatsuya Sato <Tatsuya.Sato@hal.hitachi.com> * Externally built and launched chaincodes cleanup on signal When a peer is terminated via SIGTERM or SIGINT, it should ensure any externally built and launched chaincodes have a chance to cleanup. FAB-17155 Signed-off-by: Will Lahti <wtlahti@us.ibm.com> * Remove launchDir from session and cleanup test The implementation of the e2e test wasn't properly verifying that run directories were getting cleaned up. These changes address that. Also, instead of delegating cleanup of launchDir to the session as a special case, the session was modified to allow arbitrary callbacks to be called at exit prior to returning from wait. This is more generic than wiring through the launch directory and is simple enough. Signed-off-by: Matthew Sykes <sykesmat@us.ibm.com> * Correct HSM environment variables Signed-off-by: Brett Logan <brett.t.logan@ibm.com> * [FAB-17438] Add links to CA deployment Guide from Fabric Deployment Guide Signed-off-by: pama-ibm <pama@ibm.com> * Fix misleading doc statement The Kafka to Raft migration doc incorrectly implies that mutual TLS is required when using Raft. More correctly, Raft nodes will authenticate with eachother by presenting a client certificate, as is done in mutual TLS. However, other clients may continue to authenticate without mutual TLS. Signed-off-by: Jason Yellick <jyellick@us.ibm.com> * FAB-17874 serialize message delivery in raft UTs This commit introduces a channel to buffer ingress messages for a chain, so they are consumed with the same order. Signed-off-by: Jay Guo <guojiannan1101@gmail.com> * Remove references to libtool The package we use for pkcs11 no longer uses libtool to interact with shared libraries so we no longer need the prereq for development. Signed-off-by: Matthew Sykes <sykesmat@us.ibm.com> * Cleanup macOS prerequisites and instructions Signed-off-by: Matthew Sykes <sykesmat@us.ibm.com> * [FAB-17886] Fix default value of `sequence` parameter to empty (hyperledger#1262) Currently, default value of `sequence` parameter for `peer chaincode lifecycle` commands is set to `1`. However, the default value should be modified to `0` that means empty for the following several reasons. - According to `Validate()` in `approveformyorg.go`, this function is implemented assuming that `sequence` is `0`. This is inconsistent with the default value `1`. - The current default value `1` enables Fabric users `approveformyorg` without specifying the sequence number. Not forcing the specification of a sequence number may cause operation mistakes (e.g, updating the package-id for sequence 1 by mistake) when running `approveformyorg` in the second and subsequent times. - The default value is preferably `0` to reuse the same parameter `sequence` in CLI commands to query the details of approved chaincode definitions. Signed-off-by: Tatsuya Sato <Tatsuya.Sato@hal.hitachi.com> * rm DB interface from privacyenabledstate pkg (hyperledger#1242) rm privacyenabledstate.DB interface As part of FAB-10794, we would be making changes to LoadCommittedVersions(). Before doing that, we remove unnecessary interface from the privacyenabledstate which includes LoadCommittedVersions() API. After we have removed privacyenabledstate.DB interface, the filename db.go does not make sense. Hence, we rename it to types.go as it holds all private data related types. Further, db_test.go contains tests that should be in common_storage_db_test.go. Hence, we also move those test to the correct location and rename db_test.go to types_test.go. Further, we have renamed CommonStorageDB to DB, removed DBProvider interface, and removed the term CommonStorage from variables. FAB-17880 Signed-off-by: senthil <cendhu@gmail.com> rename common_storage_db.go to db.go Signed-off-by: senthil <cendhu@gmail.com> * Fix some typos Signed-off-by: Matthew Sykes <sykesmat@us.ibm.com> * Reset golang.org/x/sys to commit ef0ce1748380 A breaking change was introduced to the golang.org/x/sys/windows package API that causes compilation failures in docker packages indirectly consumed by Fabric. While the docker code has been updated to accommodate the broken API, the change is not available in any release. Instead of using the development branch of docker, this change points the golang.org/x/sys dependency to the commit prior to the API change. Signed-off-by: Matthew Sykes <sykesmat@us.ibm.com> * Update vendored fabric-config version and associated integration test (hyperledger#1265) * Update vendored fabric-config version Pull in recent changes to the API. FAB-17893 Signed-off-by: Will Lahti <wtlahti@us.ibm.com> * Update configtx integration test to use new APIs FAB-17893 #done Signed-off-by: Will Lahti <wtlahti@us.ibm.com> * Enable stateleveldb for scanning for snapshot files generation This commit - Introduces a function in the statedb interface for getting access to an iterator that can be used for scanning the entire state, including private data hashes for a particular channel - Provides the implementation of the above mentioned function for the leveldb based statedb. FAB-17885 Signed-off-by: manish <manish.sethi@gmail.com> * Replace link to removed topic with link to relavent image Signed-off-by: NIKHIL E GUPTA <negupta@us.ibm.com> * Update master references for v1.4.7 Update bootstrap.sh and docs for v1.4.7 release Signed-off-by: David Enyeart <enyeart@us.ibm.com> * minor refactoring of pvtstatepurgemgmt (hyperledger#1252) * rm PurgeMgr and expiryKeeper interface As we have only one implementation of both PurgeMgr and expiryKeeper, we remove these interfaces. FAB-17884 Signed-off-by: senthil <cendhu@gmail.com> * minor rearrangement of types All types are moved to the top and kept in order. Signed-off-by: senthil <cendhu@gmail.com> * rm the term bookkeeping To be explicit, we remove the term bookkeeping from the pvtstatepurgemgmt and replace it with expiryKeeper or expiryInfo as appropriate Signed-off-by: senthil <cendhu@gmail.com> * Merge first component of create channel tutorial Signed-off-by: NIKHIL E GUPTA <negupta@us.ibm.com> * Clarify error message when lscc chaincode install fails during build (hyperledger#1276) When an LSCC chaincode install fails due to an error building the chaincode, the chaincode will remain installed on the peer. This commit updates the error message to indicate this behavior since it is (for better or worse) working as designed. FAB-17784 Signed-off-by: Will Lahti <wtlahti@us.ibm.com> * [DocUpdate]configtxlator decode/common.ConfigUpdate decode to json is using decode/common.ConfigUpdate routine Signed-off-by: david <david-khala@hotmail.com> Change-Id: I85d4578007a6dcb04a35561f657b8290b2cb7177 * Remove adding PWD to path in tutorials Signed-off-by: NIKHIL E GUPTA <negupta@us.ibm.com> * [FAB-17865] Update Fabric upgrade doc for dropping CouchDB The "upgrade-dbs" peer command has been enhanced to automatically drop the CouchDB state database as of v2.2.0. Update the upgrade document accordingly. Signed-off-by: Wenjian Qiao <wenjianq@gmail.com> * cleanup code Signed-off-by: hello2mao <hello2mao@gmail.com> * Should use Debugf instead of Debug Signed-off-by: bjzhang <bjzhang1991@gmail.com> * Add export to deploy CC tutorial Signed-off-by: NIKHIL E GUPTA <negupta@us.ibm.com> * FAB-17839 Ch.Part.API: List channels REST handler (hyperledger#1232) * FAB-17839 Ch.Part.API: List REST handler Develop the following handler functionality. List all channels: `GET /participation/v1/channels`. List a single handler: `GET /participation/v1/channels/my-channel`. Define the interface the Registrar will implement, place skeleton methods in it. Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: I4ce65b406ccbbbe8e1dbd4a95680c3561f5fc7c9 * fix gofmt Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: I4836ff6dc3101c1ceaeb98db8234ee732f511717 * Review comments: split types to leaf package Split the types used by packages multichannel & channelparticipation to a leaf package, in order to minimize dependencies. Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: I0b9ea17fd85f1f5bf1aba3ea8be80fc7b19e190f * Review comments: improve ChannelManagement interface Streamline the ChannelManagement interface return types and type names. Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: I67e105f9e7e315c1ce6f5b16898de4aa94e609eb * Review comments: use gorilla/mux to route requests Instead of parsing and matching the path, use gorilla/mux to route requests and identify the channel-id. Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: I6739fe027fab523a0a26a25838a36e42cc7f1469 * Review comments: use gorilla/mux to match headers Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: I179680a0556fef7110e69c0612687757844466e0 * Review comments: streamline types and return values Change return values to non-pointer type. Streamline ChannelList: - Drop Size, - Change Channels slice to carry app-channels only, - Change SystemChannel to ChannelInfoShort. Drop BlockHash from ChannelInfo. Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: I24c6ecccb58d95c307f03332ef0a4d2c06a063aa * Review comments: rename methods Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: Ib5b88ab5bc63e4c3ed3ef1cd44c6a17e755b8b91 * Address issues flagged by shellcheck A few new shellcheck issues have crept into the scripts. Make the changes recommended by static analysis. Signed-off-by: Matthew Sykes <sykesmat@us.ibm.com> * Add integration-test-prereqs target and update doc The `docker-test-prereqs` target used in the doc was never committed. Replace it with `integration-test-prereqs`. Signed-off-by: Matthew Sykes <sykesmat@us.ibm.com> * [DocUpdate] chaincode package tutorial - speak out the source tar name required in package Signed-off-by: david <david-khala@hotmail.com> Change-Id: I9843a22ffd474558d30bf7d085665d5deeeb1620 * [FAB-17793] Generate db name to namespace mapping for state couchdb (hyperledger#1268) This PR generates namespace to namespaceDBInfo mapping for state couchdb and stores the mapping data in channel’s metadata db. Due to couchdb's length restriction on db names, the mapping is needed to drop all the databases for a channel as well as snapshot support. Signed-off-by: Wenjian Qiao <wenjianq@gmail.com> * [FAB-17844] Copy symlinks as-is in external builder output Currently, the external builder code does not check for symlinks in build output when copying them. This results in the resolved files being copied as files instead of symlinks. Node.js chaincode relies on certain symlinks that npm creates in the node_modules tree. There may be other chaincode scenarios that rely on symlinks. This commit changes the external builder code so that it tests for symlinks, and copies them as symlinks instead of copying them as files into the destination directory. Signed-off-by: Simon Stone <sstone1@uk.ibm.com> * Fix the installed binary list in the document Signed-off-by: 5pecia1 <pdpxpd@gmail.com> * FAB-17889 Ch.Part.API: Registrar channel list (hyperledger#1290) * FAB-17889 Ch.Part.API: Registrar channel list Allow the registrar to produce a list of all channels. Note that all access to the system channel name is now under lock, because with the introduction of this API it is no longer immutable. Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: I4993372eb9961af528cd35d868e53dec6d53ae74 * Review comments: simplify code Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: I2499fcd9b21c2689829d275f4c28a2104d0d460a * Review comments: simplify more Signed-off-by: Yoav Tock <tock@il.ibm.com> Change-Id: Idcd63309c513e3d0848a3bd9e3b885687781b809 * fix test network docs typo Signed-off-by: zhuzeyu <zhuzeyu0409@gmail.com> * Change default CouchDB maxRetriesOnStartup to 10 Change default maxRetriesOnStartup from 12 to 10. Also fix retries so that there is no sleep after the last attempt, and improve the retry message to make it clear how many attempts and when retries are exhausted. The changes result in peer start failure after 2 minutes rather than 16 minutes. Signed-off-by: David Enyeart <enyeart@us.ibm.com> * Fix script help text in the test network document Signed-off-by: 5pecia1 <pdpxpd@gmail.com> * Fix Second-Chance Algorithm issue. New added item's referenced bit should be set to false when cache is not full. Signed-off-by: Hongbin Mao <hello2mao@gmail.com> * Errors should be checked when orderer grpc server is serving requests Signed-off-by: wuyingjun <wuyingjun@cmss.chinamobile.com> * Add ExportConfigHistory() API to confighistoryDB (hyperledger#1288) This commit adds ExportConfigHistory() API to the confighistoryDB retriever. It utilizes snapshot file writer to dump keys & values stored in the confighistoryDB to a snapshot file. Signed-off-by: senthil <cendhu@gmail.com> * [FAB-17434] Add `queryapproved` command to query the details of the approved chaincode definition (hyperledger#1243) Currently, Fabric peer does not provide a function to query the details of approved chaincode definitions. On the other hand, Fabric admins will need to confirm/use the information after approval for their operations. This patch adds `queryapproved` command for querying the details of the chaincode definition approved by the organization itself to peer CLI, so that Fabric peer can provide a function to query them. Signed-off-by: Tatsuya Sato <Tatsuya.Sato@hal.hitachi.com> * Improve error message when CLI cannot send to orderer (hyperledger#1308) from: "could not send" to: "could not send to orderer node" This change will assist users with troubleshooting error messages from peer CLI. Signed-off-by: David Enyeart <enyeart@us.ibm.com> * Generate statedb snapshot files This commit introduces a function in the statedb that exports the public state and private state hashes from statedb into a separate set of snapshot files in a deterministic format. FAB-17902 Signed-off-by: manish <manish.sethi@gmail.com> * [FAB-17435] Add integration tests for `queryapproved` command (hyperledger#1309) Currently, Fabric peer does not provide a function to query the details of approved chaincode definitions. On the other hand, Fabric admins will need to confirm/use the information after approval for their operations. Other patches add `queryapproved` command for querying the details of the approved chaincode definition to Fabric peer CLI (Refer to FAB-17433 and FAB-17434). This patch adds integration tests for the command. Signed-off-by: Tatsuya Sato <Tatsuya.Sato@hal.hitachi.com> * [FAB-17436] Add documents for `queryapproved` command (hyperledger#1310) Currently, Fabric peer does not provide a function to query the details of approved chaincode definitions. On the other hand, Fabric admins will need to confirm/use the information after approval for their operations. Other patches add `queryapproved` command for querying the details of the approved chaincode definition to Fabric peer CLI (Refer to FAB-17433 and FAB-17434). This patch adds some documents on how to use the CLI command. Signed-off-by: Tatsuya Sato <Tatsuya.Sato@hal.hitachi.com> * Provide more context in identity deserialization warning msg (hyperledger#1256) FAB-14099 #done Signed-off-by: Will Lahti <wtlahti@us.ibm.com> * Update Prereqs for Fabric users Signed-off-by: pama-ibm <pama@ibm.com> * Contain the dependency on bccsp This commit - Removes the dependency on bccsp package from much deeper code and contains the dependency only at the level of top level - kvledger - Though the refactoring is straight forward, still this commit adds a test to make sure that the code that generates and validates the merkle tree for the range queries is backward compatible. This test was long due that can help catch any incompatible changes in the code or the hash function used Signed-off-by: manish <manish.sethi@gmail.com> Co-authored-by: Senthil Nathan N <cendhu@users.noreply.github.com> Co-authored-by: manish <manish.sethi@gmail.com> Co-authored-by: wen <wenjianq@gmail.com> Co-authored-by: Svetoslav <betasve@users.noreply.github.com> Co-authored-by: Tatsuya Sato <Tatsuya.Sato@hal.hitachi.com> Co-authored-by: Will Lahti <wtlahti@us.ibm.com> Co-authored-by: Matthew Sykes <sykesmat@us.ibm.com> Co-authored-by: Brett Logan <brett.t.logan@ibm.com> Co-authored-by: pama-ibm <pama@ibm.com> Co-authored-by: Jason Yellick <jyellick@us.ibm.com> Co-authored-by: Jay Guo <guojiannan1101@gmail.com> Co-authored-by: NIKHIL E GUPTA <negupta@us.ibm.com> Co-authored-by: David Enyeart <enyeart@us.ibm.com> Co-authored-by: david <david-khala@hotmail.com> Co-authored-by: hello2mao <hello2mao@gmail.com> Co-authored-by: bjzhang <bjzhang1991@gmail.com> Co-authored-by: Yoav Tock <tock@il.ibm.com> Co-authored-by: Simon Stone <sstone1@uk.ibm.com> Co-authored-by: 5pecia1 <pdpxpd@gmail.com> Co-authored-by: zhuzeyu <zhuzeyu0409@gmail.com> Co-authored-by: wuyingjun <wuyingjun@cmss.chinamobile.com>

betasve requested a review from a team as a code owner April 12, 2020 19:57

wlahti suggested changes Apr 13, 2020

View reviewed changes

cmd/configtxgen/main_test.go Outdated Show resolved Hide resolved

cmd/configtxgen/main_test.go Outdated Show resolved Hide resolved

cmd/configtxgen/main_test.go Outdated Show resolved Hide resolved

betasve force-pushed the conftxgen-main-test-coverage branch from 45867b3 to c7241d1 Compare April 14, 2020 18:51

wlahti suggested changes Apr 14, 2020

View reviewed changes

betasve force-pushed the conftxgen-main-test-coverage branch from c7241d1 to 4ad1153 Compare April 15, 2020 20:04

wlahti reviewed Apr 17, 2020

View reviewed changes

betasve force-pushed the conftxgen-main-test-coverage branch from 4ad1153 to b3f0e91 Compare April 17, 2020 20:46

mastersingh24 changed the title ~~Improve test coverege for errors~~ Improve test coverage for errors Apr 18, 2020

wlahti previously approved these changes Apr 20, 2020

View reviewed changes

caod123 suggested changes Apr 20, 2020

View reviewed changes

Improve test coverege for errors

3c71c85

Signed-off-by: Svetoslav Blyahov <svetlio.blyahoff@gmail.com>

betasve dismissed wlahti’s stale review via 3c71c85 April 28, 2020 20:14

betasve force-pushed the conftxgen-main-test-coverage branch from b3f0e91 to 3c71c85 Compare April 28, 2020 20:14

betasve requested a review from caod123 April 28, 2020 20:15

sykesm approved these changes Apr 30, 2020

View reviewed changes

caod123 approved these changes May 12, 2020

View reviewed changes

caod123 merged commit 42d6581 into hyperledger:master May 12, 2020

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Improve test coverage for errors #1068

Improve test coverage for errors #1068

betasve commented Apr 12, 2020

wlahti left a comment

betasve commented Apr 14, 2020

wlahti left a comment

betasve commented Apr 15, 2020

wlahti Apr 17, 2020

betasve Apr 17, 2020

wlahti left a comment

caod123 left a comment

caod123 Apr 20, 2020

caod123 Apr 20, 2020

caod123 Apr 20, 2020

caod123 Apr 20, 2020

caod123 Apr 20, 2020

caod123 Apr 20, 2020

caod123 Apr 20, 2020

lindluni commented Apr 22, 2020

azure-pipelines bot commented Apr 22, 2020

wlahti commented May 12, 2020

caod123 commented May 12, 2020

Improve test coverage for errors #1068

Improve test coverage for errors #1068

Conversation

betasve commented Apr 12, 2020

Type of change

Description

wlahti left a comment

Choose a reason for hiding this comment

betasve commented Apr 14, 2020

wlahti left a comment

Choose a reason for hiding this comment

betasve commented Apr 15, 2020

Choose a reason for hiding this comment

Choose a reason for hiding this comment

wlahti left a comment

Choose a reason for hiding this comment

caod123 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lindluni commented Apr 22, 2020

azure-pipelines bot commented Apr 22, 2020

wlahti commented May 12, 2020

caod123 commented May 12, 2020