Unicode issues in blob/implementation/HmacSHA256Sign.java #192
Milestone
Comments
joostdenijs
pushed a commit
to joostdenijs/azure-sdk-for-java
that referenced
this issue
Jan 18, 2013
Dev: Fixed issue in storage convenience libraries with download resume etag lock, check for nulls on extended error parsing.
|
fix ready! |
|
Looks good. |
g2vinay
pushed a commit
to g2vinay/azure-sdk-for-java
that referenced
this issue
Mar 4, 2019
Updating changelog date and service version
sima-zhu
pushed a commit
to sima-zhu/azure-sdk-for-java
that referenced
this issue
Mar 21, 2019
Fix checkstyle
g2vinay
pushed a commit
to g2vinay/azure-sdk-for-java
that referenced
this issue
Jul 9, 2019
* inital commit for V3 port from vsts to github * Hash v2 for SDK v3 bumped version to 3.0.0-beta-3 for release * Modified Pom file to copy jar files to required location * updated sdk-version property * removed additions to pom * Removing unused code Changing version to 3.0.0-SNAPSHOT * Initial check in with reactor netty library * Initial commit for Non partitioned collection support for V3 async Java SDK * updated git ignore * updated git ignore * removed conflict markers in comments * removed unnecessary files * Removed public constructor that took PartitionKeyInternal * fixed javadoc errors * Work In Progress * Work in progress * Work In Progress, Added Reactor Netty Client and other interfaces for Reactor HTTP Client * Reverted test configuration: * Fixed Gateway Store Model Test * Removed unused classes, added headers to new classes. Added configuration to Reactor Netty Http Client * Implemented Reactor HTTP Client for HttpTransportClient * Fixed HttpTransportClient store response subscribe * Removed unused code, removed some rxnetty usages * Fixed Proxy Tests with correct log level for reactor netty client * Fixed Gateaway Address Cache Tests * Fixed test cases, optimized imports, and other refactoring changes * Removed propogating error, instead just throw the error * Storing Connection provider to close resources later * Fixed empty body issue with Flux not allowing null values * Disposing just connection provider on http client shutdown * increased used memory limit for document leak resource test. Fix it later * Incorporated port in Http Request * Replacing read and write time out handlers with idle state handlers * Removing experimental feature of pooled connection timeout * initial commit for fixing tetss (will break) * Removed timeout feature * Calling http client shutdown in HttpTransportClient * Fixed Document resource leak test by closing the channel when done * Experimenting changes * Commenting out old ssl provider * Setting ssl timeout explicitly * Trying out default ssl * Trying out default ssl * Trying out ssl config * Trying out default ssl * Using vanila ssl * fixed testsuitebase * Improving the configuration for http client * Merging v3_reactor_netty_client_experiment changes * Reverting Test Configurations change * Implemented code review suggestions as well as some optimizations and refactoring * Handling delete operation * Fixing Delete operation * Refactoring CosmosClient adding a Builder Removed cosmos configuration Adding getters for scripts Adding CosmosUser Changing Options to composition * WIP: Work in progress * - Added CosmosUser, get APIs for scripts - Removed CosmosConfiguration and moved builder to CosmosClient * Added retry utility for Backoff retry utility * - Ported Cosmos item changes * fixed more tests * Minor refactoring Adding listUsers and queryUsers * Compilation work. Work in progress * Refactoring in Reactor Netty Client: * fixed database tests * Fixed RxGatewayStoreModelTests * Fixed Store reader, consistency reader, writer, address selector test cases * fixed more tests * Replaced Func1 with java Function * - Fixed more tests - rx.internal tests run using asyncDocumentClient * Updated store and consistency reader writers, Collection caches, partition caches, helpers, address resolvers, global endpoint managers, retry policies with reactor-core changes * Updated query contexts, document service requests, cosmos client changes * Fixed bechmarks, and some tests in commons * fixed more tests * fixed more tests. commented out some tests for compilation * Implementing PR comments * Fixed Direct Implementation test cases * Refactoring internal tests to Reactor * Work in progress, fixing location cache test * Fixing Async benchmark after Mo's changes * Fixed Examples source code and test cases * Fixing javadoc warnings * Refactoring internal query tests to Reactor * Refactoring gateway tests to Reactor * Removed unused code, refactored to use internal test suite base * Fixed concurrency issue with internal test suite base, imported internal test suite base in rx.internal tests * Minor changes, removed invalid usage of TransportClient * Fixed gateway tests, removed verbose test subscriber, using basic test subscriber for now * Handling IO Exception for create attachment and media. Fixing Consistency and Retry Create Document tests * Fixing typo in doc * Work in progress * reverted changes to examples utils * Added paginator prototype code * Merging V3 * Code-complete draft port to reactor. The Direct TCP-related code compiles and is not yet testable because other significant portions of the code do not currently compile. * Code-complete draft port to reactor. The Direct TCP-related code compiles and is not yet testable because other significant portions of the code do not currently compile. * Renamed Reactor Transport Client to Transport Client, removed rx.Observable and rx.Single usages completely * fixed more tests made shared resources private * Addressed some error handling ommissions in RntbdRequestManager and verified locally on the V2 branch that the long test suite passes. * Fixed more tests to use getSharedDatabase and similar methods * Addressed some error handling ommissions in RntbdRequestManager and verified locally on the V2 branch that the long test suite passes. * Fixed buffer size issue * Added Cosmos Request options * Renamed COSMOS.PROTOCOL as cosmos.directModeProtocol and move ConfigsTests to the package that Configs is defined * Fixing tests and refactored the way to get values * Icreased Ssl handshake timeout for reactor netty client * Ensured that our dependency on org.hamcreate:hamcrest-all:1.3 is respected by mockito which brings in hacrest-all:1.1. Also: sorted dependency version properties and dependency lists. * Addressed RntbdTransportClient test faiures * Minor test fixes * We no longer skip the DocumentClientResourceLeakTest when Direct TCP is in use * We no longer permit VeryLargeDocumentQueryTest to fail when Direct TCP is in use. Also improved logger.info messages in DocumentClientResourceLeakTest * We no longer permit DCDocumentCrudTest.crossPartitionQuery to fail when Direct TCP is in use. * We no longer permit BackPressureCrossPartition.query to fail when Direct TCP is in use. * skipped tcp tests * removed unused imports * Resolves a long-standing TODO: Ensure that the Configs instance created by DirectHttpsClientUnderTest specifies Protocol.Https (now that Direct TCP is the default connection mode) * changes to ConsistencyLevel * TestSuiteBase.beforeMethod correctly logs the connection mode: Direct Tcp, Direct Https, or Gateway Https * Merged enable TCP default protocol changes * Fixed Address Resolver Tests * Disabling connection pool exhausted tests * Fixing Config Tests * removed changes to DocumentCrudTests edited manner in which ConsistencyLevel.toString is implemented * removed unused import * reverted changes to documentcrud tests * fixed failing test * made more ConsistencyLevel.name -> toString changes * changed line endings * Removed rx Composite Exception and used reactor Exceptions utility class to handle multiple exceptions * Made more changes to enums removed cosmos tests * more enum fixes * more enum changes * Test tweaks * Made more enum changes * fixed line endings in includepathtest * Fixed pagination and order by utils bugs * Swapped rx Http Client with Reactor Http Client in Cosmos Partition Key Tests * Removed rx java dependency * Fixing tests * Removed throws Exception and unused imports * Optimized imports in Microsoft style * Added parallel run code for bulk insert blocking method in Test Suite Base * Minor test fixes * Trying default Http Client * re-using https client * Disabling document resource leak test * Fixing up schedulers * fixed CosmosPartitionKeyTests and ConsistencyTests null pointer issues * Added logging to netty client and fixed logger category * Fixing spy wire test * fixed failing tests * Removing unused imports * Minor test changes * Removed usages of bodyAsString method from http response as its not working correctly, implemented toString to convert byteBuf flux to Mono string * Increased timeout * Replaced flux with mono across consistency layer to avoid confusion * Reverting test configurations changes * Fixed Parallel Document Query Context * Fixed Protocol tests * Replaced exceptions by Mono / Flux Error * Incresaed wait to capture requests * Added block instead of subscribe to capture response * Fixed ReadMyWrites test with emitting just one value * Modified the order of client builders and reverted flux / mono change for consistency stack * Fixed Read my writes workflow * Removed unnecessary stack trace * Fixed Benchmarks to use BaseSubscriber * Adding conflict api (Azure#192) Fixing Permission tests Making id method of Resource package private and adding id on individual resource classes to enable fluent composition * Removed unused ConfigsTests.java class * Fixed Read latency and write latency benchmarks * Fixed metrics version in parent pom.xml * Revised DocumentCrudTest and TestSuiteBase to address test breaks * Using Step Verifier, removed RetryAnalyzer * POM + test tweaks. * Eliminating some tests to test fast integration tests * Reverting the experiment * Fixed test break (see changes to CosmosPartitionKeyTests), tidied some test code (see changes to StoredProcedureCrudTest) and deleted two extraneous files * Removed unnecessary call * Code review comments
navalev
pushed a commit
to navalev/azure-sdk-for-java
that referenced
this issue
Dec 24, 2019
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Dev estimate: 1
Test estimate: 0.5
The following code exists in
HmacSHA256Sign:The input for the constructor is a byte array. This is unpredictable because if no encoding is specified, the constructor uses the platform's default charset. For example, if someone sets their platform's default charset to UTF-16, even ASCII characters will be un-parseable.
In the context of the code, it appears that the data should be interpreted using UTF-8. If that is the case, then, it should be made explicit.
The text was updated successfully, but these errors were encountered: