如何使用 Eigen 和 Bazel 構建一個簡單的 C++ 演示？

Question

如何在使用Bazel （版本 0.25.2）構建的 C++ 項目中使用Eigen ？ 我喜歡使用http_archive或git_repository獲取 Eigen 依賴項。

我嘗試了以下方法：

主文件

#include <iostream>
#include <Eigen/Dense>

using Eigen::MatrixXd;

int main() {
    MatrixXd m(2, 2);
    m(0, 0) = 3;
    m(1, 0) = 2.5;
    m(0, 1) = -1;
    m(1, 1) = m(1, 0) + m(0, 1);
    std::cout << m << std::endl;
}

工作空間

workspace(name = "EigenDemo")

load("@bazel_tools//tools/build_defs/repo:http.bzl", "http_archive")

# Eigen
http_archive(
    name = "eigen",
    build_file = "//:eigen.BUILD",
    sha256 = "3a66f9bfce85aff39bc255d5a341f87336ec6f5911e8d816dd4a3fdc500f8acf",
    url = "https://bitbucket.org/eigen/eigen/get/c5e90d9.tar.gz",
)

建造

cc_binary(
    name = "EigenDemo",
    srcs = ["main.cpp"],
    deps = [
        "@eigen",
    ],
)

特征構建

# Copyright 2017 Google Inc. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS-IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

package(
    default_visibility = ["//visibility:public"],
)

cc_library(
    name = "eigen",
    hdrs = glob(
        ["Eigen/**"],
        exclude = [
            "Eigen/src/OrderingMethods/Amd.h",
            "Eigen/src/SparseCholesky/**",
            "Eigen/Eigen",
            "Eigen/IterativeLinearSolvers",
            "Eigen/MetisSupport",
            "Eigen/Sparse",
            "Eigen/SparseCholesky",
            "Eigen/SparseLU",
        ],
    ),
    defines = [
        "EIGEN_MPL_ONLY",
        "EIGEN_NO_DEBUG",
    ],
    includes = ["."],
)

錯誤輸出：

INFO: Analysed target //:EigenTest (0 packages loaded, 0 targets configured).
INFO: Found 1 target...
ERROR: /BazelDemos/EigenDemo/BUILD:1:1: C++ compilation of rule '//:EigenTest' failed (Exit 1) gcc failed: error executing command /usr/bin/gcc -U_FORTIFY_SOURCE -fstack-protector -Wall -Wunused-but-set-parameter -Wno-free-nonheap-object -fno-omit-frame-pointer '-std=c++0x' -MD -MF ... (remaining 40 argument(s) skipped)

Use --sandbox_debug to see verbose messages from the sandbox
main.cpp:2:10: fatal error: Eigen/Dense: No such file or directory
 #include <Eigen/Dense>
      ^~~~~~~~~~~~~
compilation terminated.
Target //:EigenTest failed to build
Use --verbose_failures to see the command lines of failed build steps.
INFO: Elapsed time: 0.258s, Critical Path: 0.10s
INFO: 0 processes.
FAILED: Build did NOT complete successfully

Answer 1

WORKSPACE文件中缺少strip_prefix ：

workspace(name = "EigenDemo")

load("@bazel_tools//tools/build_defs/repo:http.bzl", "http_archive")

# Eigen
http_archive(
    name = "eigen",
    build_file = "//:eigen.BUILD",
    sha256 = "3a66f9bfce85aff39bc255d5a341f87336ec6f5911e8d816dd4a3fdc500f8acf",
    url = "https://bitbucket.org/eigen/eigen/get/c5e90d9.tar.gz",
    strip_prefix="eigen-eigen-c5e90d9e764e"
)

Answer 2

Tensorflow 不再使用單獨的 eigen.BUILD 文件。 相反，您可以將以下內容放入您的 WORKSPACE 文件中（無需單獨的 eigen.BUILD 文件）：

http_archive(
    name = "com_gitlab_libeigen_eigen",
    sha256 = "0215c6593c4ee9f1f7f28238c4e8995584ebf3b556e9dbf933d84feb98d5b9ef",
    strip_prefix = "eigen-3.3.8",
    urls = [
        "https://gitlab.com/libeigen/eigen/-/archive/3.3.8/eigen-3.3.8.tar.bz2",
    ],
    build_file_content =
"""
# TODO(keir): Replace this with a better version, like from TensorFlow.
# See https://github.com/ceres-solver/ceres-solver/issues/337.
cc_library(
    name = 'eigen',
    srcs = [],
    includes = ['.'],
    hdrs = glob(['Eigen/**']),
    visibility = ['//visibility:public'],
)
"""
)

（正如你可能從 TODO 中看出的那樣，我從 Ceres 撕下了這個。）

Answer 3

我使用了最新的rules_foreign_cc 。 有一個新的 cmake 規則。 這是我所擁有的：

工作空間

workspace(name="bazelt")
load("@bazel_tools//tools/build_defs/repo:http.bzl", "http_archive")

http_archive(
    name = "rules_foreign_cc",
    sha256 = "6041f1374ff32ba711564374ad8e007aef77f71561a7ce784123b9b4b88614fc",
    strip_prefix = "rules_foreign_cc-0.8.0",
    url = "https://github.com/bazelbuild/rules_foreign_cc/archive/0.8.0.tar.gz",
)

load("@rules_foreign_cc//foreign_cc:repositories.bzl", "rules_foreign_cc_dependencies")

# This sets up some common toolchains for building targets. For more details, please see
# https://bazelbuild.github.io/rules_foreign_cc/0.8.0/flatten.html#rules_foreign_cc_dependencies
rules_foreign_cc_dependencies()

_ALL_CONTENT = """\
filegroup(
    name = "all_srcs",
    srcs = glob(["**"]),
    visibility = ["//visibility:public"],
)
"""
http_archive(
  name="eigen",
  build_file_content = _ALL_CONTENT,
  strip_prefix = "eigen-3.4",
  urls=["https://gitlab.com/libeigen/eigen/-/archive/3.4/eigen-3.4.tar.bz2"],
  sha256 = "a6f7aaa7b19c289dfeb33281e1954f19bf2ba1c6cae2c182354d820f535abef8",
)

建造

load("@rules_foreign_cc//foreign_cc:defs.bzl", "cmake")
cmake(
    name = "eigen",
    cache_entries = {
        "CMAKE_C_FLAGS": "-fPIC",
    },
    lib_source = "@eigen//:all_srcs",
    out_headers_only = True,
    includes = ["eigen3/",],
    install = True,
)
cc_binary(
    name = "bazelt",
    srcs = [
        "main.cpp",
    ],
    visibility = ["//visibility:public"],
    deps = [":eigen",],
)

注意 cmake 規則中的"includes"參數。 默認情況下，cmake 規則會將頭文件放在緩存中的“include”目錄中。 所以你需要的只是相對路徑。 和 main.cpp

#include <iostream>
#include <Eigen/Dense>

int main()
{
  Eigen::MatrixXd m = Eigen::MatrixXd::Random(4,6);
  std::cout << m << '\n';
  return 0;
}