diff --git a/README.md b/README.md index c5ba925edb..0296f61b3c 100644 --- a/README.md +++ b/README.md @@ -156,7 +156,7 @@ Curve supports deployment in private and public cloud environments, and can also
-One of them, CurveFS shared file storage system, can be elasticly scaled to public cloud storage, which can provide users with greater capacity elasticity, lower cost, and better performance experience. +One of them, CurveFS shared file storage system, can be elastically scaled to public cloud storage, which can provide users with greater capacity elasticity, lower cost, and better performance experience. diff --git a/developers_guide.md b/developers_guide.md index c5acbafaae..3f190b9fcc 100644 --- a/developers_guide.md +++ b/developers_guide.md @@ -18,15 +18,15 @@ The premise of participating in an open source project is to understand it, espe Through the study of the above materials, I believe that you already have an overall understanding of the Curve project, and there may also be some curiosity and doubts. At this time, you can deploy a Curve experience environment, which is conducive to a more intuitive perception of the Curve system. If you encounter a problem or want a new feature, you can track the relevant code. In the process, it is easy to understand the relevant modules. This is lots of contributors completed their first contributions. -The Curve community has multiple [communication channels](#Comminication), and there will be an online community meeting every two weeks. The regular meeting will synchronize the recent progress of the Curve project and answer your questions. You are also welcome to participate in the Curve community regular meeting. We will communicate at the meeting, so that we can quickly answer your questions, improve our understanding of the Curve project and synchronize what we are doing at this stage. +The Curve community has multiple [communication channels](#Communication), and there will be an online community meeting every two weeks. The regular meeting will synchronize the recent progress of the Curve project and answer your questions. You are also welcome to participate in the Curve community regular meeting. We will communicate at the meeting, so that we can quickly answer your questions, improve our understanding of the Curve project and synchronize what we are doing at this stage. ## How to participate in Curve After you have a certain understanding of the Curve project, if you are interested, you can choose the starting point to participate in the Curve. You can choose from the following aspects: -- Start by selecting a interested one from the [Issues](https://github.com/opencurve/curve/issues) of the Curve project. You can pay attention to issues with the good_first_issue tag which we have assessed as relatively good starters. +- Start by selecting a interested one from the [issues](https://github.com/opencurve/curve/issues) of the Curve project. You can pay attention to issues with the [good first issue](https://github.com/opencurve/curve/issues?q=is%3Aissue+is%3Aopen+sort%3Aupdated-desc+label%3A%22good+first+issue%22) tag which we have assessed as relatively good starters. - Based on a certain understanding of the Curve project, you can also choose from [Roadmap](https://github.com/opencurve/curve/wiki/Roadmap). -- Selecting form [Issues](https://github.com/opencurve/curveadm/issues) and [Roadmap](https://github.com/opencurve/curveadm/issues/92) of Curve Operation and maintenance tools [Curveadm](https://github.com/opencurve/curveadm). Here are [quick start guidelines](https://github.com/opencurve/curveadm/wiki/develop)。 +- Selecting form [issues](https://github.com/opencurve/curveadm/issues) and [Roadmap](https://github.com/opencurve/curveadm/issues/92) of Curve Operation and maintenance tools [Curveadm](https://github.com/opencurve/curveadm). Here are [quick start guidelines](https://github.com/opencurve/curveadm/wiki/develop)。 - In addition to the existing issues, you are also welcome to submit issues that you have discovered or new features you hope for and resolve them. - You can pay attention to the *TODO* in the existing Curve code, most of which are code logic optimization and features to be supplemented, choose the ones you are interested in and raise relevant issues to follow up and try to solve. @@ -117,7 +117,7 @@ Curve Ci use ```cpplint``` check what your changed. For PR we have the following requirements: -- The CURVE coding standard strictly follows the [Google C++ Open Source Project Coding Guide](https://google.github.io/styleguide/cppguide.html), but we use 4 spaces to indent, Clang-format will more helpful for you. Of course, CI will check what your changed. +- The Curve coding standard strictly follows the [Google C++ Open Source Project Coding Guide](https://google.github.io/styleguide/cppguide.html), but we use 4 spaces to indent, Clang-format will more helpful for you. Of course, CI will check what your changed. - The code must have test cases, excluding documentation, unit tests (incremental lines cover more than 80%, and incremental branches cover more than 70%); integration tests (merge statistics with unit tests, and meet the unit test coverage requirements). - Please fill in the description of the PR as detailed as possible, associate with the relevant issues, and the PR commit message can clearly see the resolved issues. After submitting to the Curve master branch, Curve CI will be triggered automatically. It is necessary to ensure that the CI is passed, and the Jenkins username and password of the CI is netease/netease, if the CI fails to run, you can log in to the Jenkins platform to view the reason for the failure. - After the CI is passed, the review can start, and each PR needs to get at least two LGTMs of Committer/Maintainer before merging. @@ -140,7 +140,7 @@ Repush will trigger CI, If github page have no reaction. Please wait. If CI is not stabled, comment ```recheck``` will trigger CI. -## Comminication +## Communication At present, the Curve community has multiple communication channels, please choose an appropriate and efficient communication method according to your needs: @@ -151,4 +151,4 @@ At present, the Curve community has multiple communication channels, please choo - **slack**: cloud-native.slack.com，channel #project_curve - **Curve User Group**: In order to facilitate instant communication with everyone, the Curve User Group is currently a WeChat group. Due to the large number of people in the group, it is necessary to add OpenCurve_bot WeChat first, and then invite into the group. In the user group, everyone can freely communicate about Curve and storage-related topics, and get immediate feedback on problems. - \ No newline at end of file + diff --git a/developers_guide_cn.md b/developers_guide_cn.md index 13f1f754f8..99fc185b47 100644 --- a/developers_guide_cn.md +++ b/developers_guide_cn.md @@ -24,9 +24,9 @@ Curve 社区有多个[沟通渠道](#社区交流)，每两周还会有线上的 在对 Curve 项目有一定了解后，如果感兴趣就可以选择入手点进行 Curve 开源项目的参与了，可以从以下几个方面进行选择： -- 从 Curve 项目的 [Issue](https://github.com/opencurve/curve/issues) 中选择感兴趣的问题入手，可以特别关注带有 good_first_issue tag 的 issue，这些是我们经过评估认为是相对好入门的一些问题。 +- 从 Curve 项目的 [issue](https://github.com/opencurve/curve/issues) 中选择感兴趣的问题入手，可以特别关注带有 [good first issue](https://github.com/opencurve/curve/issues?q=is%3Aissue+is%3Aopen+sort%3Aupdated-desc+label%3A%22good+first+issue%22) tag 的 issue，这些是我们经过评估认为是相对好入门的一些问题。 - 在对 Curve 项目有一定了解的基础上也可以从 [Roadmap](https://github.com/opencurve/curve/wiki/Roadmap_CN) 中进行选择。 -- 从 Curve 项目的运维工具 [Curveadm](https://github.com/opencurve/curveadm) 的 [Issue](https://github.com/opencurve/curveadm/issues) 和 [Roadmap](https://github.com/opencurve/curveadm/issues/92) 中进行选择，开发方式可参见[快速上手 CurveAdm 开发](https://github.com/opencurve/curveadm/wiki/develop)。 +- 从 Curve 项目的运维工具 [Curveadm](https://github.com/opencurve/curveadm) 的 [issue](https://github.com/opencurve/curveadm/issues) 和 [Roadmap](https://github.com/opencurve/curveadm/issues/92) 中进行选择，开发方式可参见 [快速上手 CurveAdm 开发](https://github.com/opencurve/curveadm/wiki/develop)。 - 除了已有的 issue，也欢迎将自己发现的问题或希望的新特性提出 issue 并进行解决。 - 可以关注现有 Curve 代码中的 *TODO* ，大部分为代码逻辑优化和待补充特性，选择感兴趣的提出相关 issue 跟进并尝试解决。 @@ -73,9 +73,9 @@ Curve CI 使用```cpplint```检查更改的代码, 对于 PR 我们有如下要求： -- CURVE编码规范严格按照[Google C++开源项目编码指南](https://zh-google-styleguide.readthedocs.io/en/latest/google-cpp-styleguide/contents/)来进行代码编写，但使用 4 空格进行缩进, 可使用 clang-format 进行格式化, CI 会检查相关更改代码是否符合规则. +- Curve编码规范严格按照[Google C++开源项目编码指南](https://zh-google-styleguide.readthedocs.io/en/latest/google-cpp-styleguide/contents/)来进行代码编写，但使用 4 空格进行缩进, 可使用 clang-format 进行格式化, CI 会检查相关更改代码是否符合规则. - 代码必须有测试，文档除外，单元测试（增量行覆盖80%以上，增量分支覆盖70%以上）；集成测试（与单元测试合并统计，满足单元测试覆盖率要求即可） -- 请尽可能详细的填写 PR 的描述，关联相关问题的 issuse，PR commit message 能清晰看出解决的问题，提交到 Curve master 分支之后会自动触发Curve CI，需保证 CI 通过，CI 的 Jenkins 用户名密码为 netease/netease，如遇到 CI 运行失败可以登录 Jenkins 平台查看失败原因。 +- 请尽可能详细的填写 PR 的描述，关联相关问题的 issue，PR commit message 能清晰看出解决的问题，提交到 Curve master 分支之后会自动触发Curve CI，需保证 CI 通过，CI 的 Jenkins 用户名密码为 netease/netease，如遇到 CI 运行失败可以登录 Jenkins 平台查看失败原因。 - CI 通过之后可开始进行 review，每个 PR 在合并之前都需要至少得到两个 Committer/Maintainer 的 LGTM。 - PR 代码需要一定量的注释来使代码容易理解，且所有注释和 review 意见和回复均要求使用英语。 @@ -149,4 +149,4 @@ CI 检查点有: - **slack**: cloud-native.slack.com，channel #project_curve - **Curve User Group**: 为了便于大家即时的沟通，Curve User Group 目前为微信群，由于群人数过多，需要先添加OpenCurve_bot微信，再邀请进群。在用户群里大家可以自由的沟通关于 Curve 和存储相关的话题，对于存在问题也可以较为即时的得到反馈。 - \ No newline at end of file + diff --git a/docs/cn/build_and_run.md b/docs/cn/build_and_run.md index db681fd068..6c9618c260 100644 --- a/docs/cn/build_and_run.md +++ b/docs/cn/build_and_run.md @@ -2,13 +2,10 @@ # 编译环境搭建 -请注意： -1. 如您只是想体验CURVE的部署流程和基本功能，**则不需要编译CURVE**，请参考[部署](https://github.com/opencurve/curveadm/wiki) -2. 本文档仅用来帮助你搭建CURVE代码编译环境，便于您参与CURVE的开发调试 +**请注意：** -**注意：** - -mk-tar.sh 和 mk-deb.sh 用于 curve v2.0 之前版本的编译打包，v2.0 版本之后不再维护。 +1. 如您只是想体验Curve的部署流程和基本功能，**则不需要编译Curve**，请参考 [部署](https://github.com/opencurve/curveadm/wiki) +2. 本文档仅用来帮助你搭建Curve代码编译环境，便于您参与Curve的开发调试 ## 使用Docker进行编译（推荐方式） @@ -20,10 +17,6 @@ mk-tar.sh 和 mk-deb.sh 用于 curve v2.0 之前版本的编译打包，v2.0 版 docker pull opencurvedocker/curve-base:build-debian9 ``` -```bash -docker pull opencurvedocker/curve-base:build-debian9 -``` - 方法二：手动构建docker镜像 使用工程目录下的 docker/debian9/compile/Dockerfile 进行构建，命令如下： @@ -32,7 +25,7 @@ docker pull opencurvedocker/curve-base:build-debian9 docker build -t opencurvedocker/curve-base:build-debian9 ``` -注意：上述操作不建议在CURVE工程目录执行，否则构建镜像时会把当前目录的文件都复制到docker镜像中，建议把Dockerfile拷贝到新建的干净目录下进行docker镜像的构建。 +**注意：** 上述操作不建议在Curve工程目录执行，否则构建镜像时会把当前目录的文件都复制到docker镜像中，建议把Dockerfile拷贝到新建的干净目录下进行docker镜像的构建。 ### 在docker镜像中编译 @@ -40,28 +33,29 @@ docker build -t opencurvedocker/curve-base:build-debian9 docker run -it opencurvedocker/curve-base:build-debian9 /bin/bash cd git clone https://github.com/opencurve/curve.git 或者 git clone https://gitee.com/mirrors/curve.git -# （可选步骤）将外部依赖替换为国内下载点或镜像仓库，可以加快编译速度： bash replace-curve-repo.sh +# （中国大陆可选）将外部依赖替换为国内下载点或镜像仓库，可以加快编译速度： bash replace-curve-repo.sh # curve v2.0 之前 bash mk-tar.sh （编译 curvebs 并打tar包） bash mk-deb.sh （编译 curvebs 并打debian包） -# curve v2.0 及之后 +# （当前）curve v2.0 及之后 编译 curvebs: cd curve && make build stor=bs dep=1 编译 curvefs: cd curve && make build stor=fs dep=1 ``` +**注意：** `mk-tar.sh` 和 `mk-deb.sh` 用于 curve v2.0 之前版本的编译打包，v2.0 版本之后不再维护。 ## 在物理机上编译 -CURVE编译依赖的包括： +Curve编译依赖的包括： | 依赖 | 版本 | |:-- |:-- | | bazel | 4.2.2 | | gcc | 支持c++11的兼容版本 | -CURVE的其他依赖项，均由bazel去管理，不可单独安装。 +Curve的其他依赖项，均由bazel去管理，不可单独安装。 -**注意** 4.* 版本的 bazel 均可以成功编译 curve 项目，其他版本不兼容。 +**注意：** 4.* 版本的 bazel 均可以成功编译 Curve 项目，其他版本不兼容。 4.2.2 为推荐版本。 ### 安装依赖 @@ -70,13 +64,13 @@ CURVE的其他依赖项，均由bazel去管理，不可单独安装。 ### 一键编译 -``` +```bash git clone https://github.com/opencurve/curve.git 或者 git clone https://gitee.com/mirrors/curve.git -# （可选步骤）将外部依赖替换为国内下载点或镜像仓库，可以加快编译速度： bash replace-curve-repo.sh +# （中国大陆可选）将外部依赖替换为国内下载点或镜像仓库，可以加快下载速度： bash replace-curve-repo.sh # curve v2.0 之前 bash mk-tar.sh （编译 curvebs 并打tar包） bash mk-deb.sh （编译 curvebs 并打debian包） -# curve v2.0 及之后 +# （当前）curve v2.0 及之后 编译 curvebs: cd curve && make build stor=bs dep=1 编译 curvefs: cd curve && make build stor=fs dep=1 ``` @@ -86,15 +80,24 @@ bash mk-deb.sh （编译 curvebs 并打debian包） ### 编译全部模块 仅编译全部模块，不进行打包 -``` +```bash bash ./build.sh ``` -### 编译对应模块的代码和运行测试 +### 列出所有测试模块 + +```bash +# curvebs +bazel query '//test/...' +# curvefs +bazel query '//curvefs/test/...' +``` + +### 编译对应模块的代码 编译对应模块，例如test/common目录下的common-test测试： -``` +```bash bazel build test/common:common-test --copt -DHAVE_ZLIB=1 --define=with_glog=true --compilation_mode=dbg --define=libunwind=true ``` @@ -171,7 +174,8 @@ cd etcd-v3.4.10-linux-amd64 && cp etcd etcdctl /usr/bin ``` #### 执行单个测试模块 -``` + +```bash ./bazel-bin/test/common/common-test ``` diff --git a/docs/cn/gmock.md b/docs/cn/gmock.md index 8120e8f498..4dde034b85 100644 --- a/docs/cn/gmock.md +++ b/docs/cn/gmock.md @@ -118,7 +118,7 @@ + 第6行WillOnce(action)是定义一次调用时所产生的行为，比如定义该方法返回怎么样的值等等。 + 第7行WillRepeatedly(action)的意思是缺省/重复行为。 -## Machers（匹配器） +## Matchers（匹配器） Matcher用于定义Mock类中的方法的形参的值（当然，如果你的方法不需要形参时，可以保持match为空）。 @@ -179,7 +179,7 @@ Matcher可以在ON_CALL() or EXPECT_CALL()内部使用，也可以直接判断 | ContainerEq(container) | 类型Eq(container)，就是输出结果有点不一样，这里输出结果会带上哪些个元素不被包含在另一个容器中 | | Pointwise(m, container) | | -所有这些Machers都在命名空间::testing中，使用时需引入这一命名空间 +所有这些Matchers都在命名空间::testing中，使用时需引入这一命名空间 例： using ::testing::Ge; @@ -197,7 +197,7 @@ Matcher可以在ON_CALL() or EXPECT_CALL()内部使用，也可以直接判断 | Field(&class::field, m) | argument.field (或 argument->field, 当argument是一个指针时)与匹配器m匹配, 这里的argument是一个class类的实例. | |:----------------------|:----------------------| -| Key(e) | 容器的key值符合e，e可以是一个值或者是一个Macher | +| Key(e) | 容器的key值符合e，e可以是一个值或者是一个Matcher | | Pair(m1, m2) | argument是一个std::pair，并且argument.first等于m1，argument.second等于m2. | | Property(&class::property, m) | argument.property()(或argument->property(),当argument是一个指针时)与匹配器m匹配, 这里的argument是一个class类的实例. | @@ -292,7 +292,7 @@ Side Effects 指的是这个action还将对测试起的作用，比如给指针 | Invoke(f) | 使用模拟函数的参数调用f, 这里的f可以是全局/静态函数或函数对象. | |:-------------------------|:----------------------| -| Invoke(object_pointer, &class::method) | 使用模拟函数的参数调用object_pointer对象的mothod方法. | +| Invoke(object_pointer, &class::method) | 使用模拟函数的参数调用object_pointer对象的method方法. | ### 复合动作 diff --git a/docs/en/build_and_run_en.md b/docs/en/build_and_run_en.md index 838a40b46a..f0c54f4f36 100644 --- a/docs/en/build_and_run_en.md +++ b/docs/en/build_and_run_en.md @@ -3,8 +3,8 @@ # Build compilation environment **Note:** -1. If you just want to experience the deployment and basic functions of CURVE, **you do not need to compile CURVE**, please refer to [deployment](https://github.com/opencurve/curveadm/wiki). -2. This document is only used to help you build the CURVE code compilation environment, which is convenient for you to participate in the development, debugging and run tests of CURVE. +1. If you just want to experience the deployment and basic functions of Curve, **you do not need to compile Curve**, please refer to [deployment](https://github.com/opencurve/curveadm/wiki). +2. This document is only used to help you build the Curve code compilation environment, which is convenient for you to participate in the development, debugging and run tests of Curve. ## Compile with docker (recommended) @@ -24,7 +24,7 @@ Use the Dockerfile in the project directory to build. The command is as follows: docker build -t opencurvedocker/curve-base:build-debian9 ``` -Note: The above operations are not recommended to be performed in the CURVE project directory, otherwise the files in the current directory will be copied to the docker image when building the image. It is recommended to copy the Dockerfile to the newly created clean directory to build the docker image. +**Note:** The above operations are not recommended to be performed in the Curve project directory, otherwise the files in the current directory will be copied to the docker image when building the image. It is recommended to copy the Dockerfile to the newly created clean directory to build the docker image. ### Compile in docker image @@ -32,27 +32,29 @@ Note: The above operations are not recommended to be performed in the CURVE proj docker run -it opencurvedocker/curve-base:build-debian9 /bin/bash cd git clone https://github.com/opencurve/curve.git or git clone https://gitee.com/mirrors/curve.git -# (Optional step) Replace external dependencies with domestic download points or mirror warehouses, which can speed up compilation： bash replace-curve-repo.sh +# (Mainland China optional) Replace external dependencies with domestic download points or mirror warehouses, which can speed up compilation： bash replace-curve-repo.sh # before curve v2.0 bash mk-tar.sh （compile curvebs and make tar package） bash mk-deb.sh （compile curvebs and make debian package） -# after curve v2.0 +# (current) after curve v2.0 compile curvebs: cd curve && make build stor=bs dep=1 compile curvefs: cd curve && make build stor=fs dep=1 ``` +**Note:** `mk-tar.sh` and `mk-deb.sh` are used for compiling and packaging curve v2.0. They are no longer maintained after v2.0. + ## Compile on a physical machine -CURVE compilation depends on: +Curve compilation depends on: | Dependency | Version | |:-- |:-- | | bazel | 4.2.2 | | gcc | Compatible version supporting C++11 | -Other dependencies of CURVE are managed by bazel and do not need to be installed separately. +Other dependencies of Curve are managed by bazel and do not need to be installed separately. -**Note** The 4.* version of bazel can successfully compile the curve project, other versions are not compatible. +**Note:** The 4.* version of bazel can successfully compile the curve project, other versions are not compatible. 4.2.2 is the recommended version. ### Installation dependency @@ -63,11 +65,11 @@ For dependencies, you can refer to the installation steps in [dockerfile](../../ ```bash git clone https://github.com/opencurve/curve.git or git clone https://gitee.com/mirrors/curve.git -# (Optional step) Replace external dependencies with domestic download points or mirror warehouses, which can speed up compilation： bash replace-curve-repo.sh +# (Mainland China optional) Replace external dependencies with domestic download points or mirror warehouses, which can speed up compilation： bash replace-curve-repo.sh # before curve v2.0 bash mk-tar.sh （compile curvebs and make tar package） bash mk-deb.sh （compile curvebs and make debian package） -# after curve v2.0 +# (current) after curve v2.0 compile curvebs: cd curve && make build stor=bs dep=1 compile curvefs: cd curve && make build stor=fs dep=1 ``` @@ -78,15 +80,24 @@ compile curvefs: cd curve && make build stor=fs dep=1 Only compile all modules without packaging -``` +```bash $ bash ./build.sh ``` -### Compile the corresponding module code and run the test +### List all test modules + +```bash +# curvebs +bazel query '//test/...' +# curvefs +bazel query '//curvefs/test/...' +``` + +### Compile the corresponding module code Compile corresponding modules, such as common-test in the `test/common` directory -``` +```bash $ bazel build test/common:common-test --copt -DHAVE_ZLIB=1 \ $ --define=with_glog=true --compilation_mode=dbg \ $ --define=libunwind=true @@ -170,7 +181,7 @@ $ cd etcd-v3.4.10-linux-amd64 && cp etcd etcdctl /usr/bin #### Execute a single test module -``` +```bash $ ./bazel-bin/test/common/common-test ```