Build R documentation using Docker (#5188)

* rename dockerfile Signed-off-by: harupy <17039389+harupy@users.noreply.github.com> * build R doc using docker Signed-off-by: harupy <17039389+harupy@users.noreply.github.com> * update .Rbuildignore Signed-off-by: harupy <17039389+harupy@users.noreply.github.com> * fix volume target Signed-off-by: harupy <17039389+harupy@users.noreply.github.com> * use working_directory Signed-off-by: harupy <17039389+harupy@users.noreply.github.com> * add comment
mlflow · Dec 22, 2021 · 4deb9b7 · 4deb9b7
1 parent 852f567
commit 4deb9b7
Show file tree

Hide file tree

Showing 10 changed files with 113 additions and 141 deletions.
diff --git a/.circleci/config.yml b/.circleci/config.yml
@@ -2,56 +2,11 @@ version: 2.1
 
 jobs:
   build_doc_r:
-    docker:
-      - image: ubuntu:20.04
+    machine:
+      image: ubuntu-2004:202111-01
 
     steps:
       - checkout
-      - run:
-          name: Install dev tools
-          environment:
-            DEBIAN_FRONTEND: noninteractive
-          command: |
-            apt-get update --yes
-            apt-get install sudo git wget curl jq software-properties-common apt-transport-https --yes
-
-      - run:
-          name: Install Java
-          command: |
-            sudo apt-get install default-jdk --yes
-            java -version
-
-      - run:
-          name: Install R
-          command: |
-            # How To Install R on Ubuntu 20.04:
-            # https://www.digitalocean.com/community/tutorials/how-to-install-r-on-ubuntu-20-04
-
-            sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9
-            sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/'
-            sudo apt update -y
-            sudo apt install -y r-base libssl-dev libxml2-dev libcurl4-openssl-dev
-            R --version
-
-      - run:
-          name: Install pandoc
-          command: |
-            # Install a recent version of pandoc
-            TEMP_DEB="$(mktemp)"
-            wget -O "$TEMP_DEB" 'https://github.com/jgm/pandoc/releases/download/2.7.2/pandoc-2.7.2-1-amd64.deb'
-            sudo dpkg -i "$TEMP_DEB"
-            rm -f "$TEMP_DEB"
-
-      - run:
-          name: Dump R dependencies
-          working_directory: mlflow/R/mlflow
-          command: |
-            Rscript .dump-r-dependencies.R
-
-      - restore_cache:
-          keys:
-            - r-cache-{{ checksum "mlflow/R/mlflow/R-version" }}
-
       - run:
           name: Build documentation
           working_directory: docs
@@ -80,11 +35,6 @@ jobs:
 
             exit $failed
 
-      - save_cache:
-          key: r-cache-{{ checksum "mlflow/R/mlflow/R-version" }}-{{ checksum "mlflow/R/mlflow/depends.Rds" }}
-          paths:
-            - /usr/local/lib/R/site-library
-
       - store_artifacts:
           path: << pipeline.git.revision >>.patch
 

diff --git a/docs/build-rdoc.sh b/docs/build-rdoc.sh
@@ -1,28 +1,16 @@
 #!/usr/bin/env bash
 
 set -ex
+
 pushd ../mlflow/R/mlflow
 
-# `gert` requires `libgit2`:
-# https://github.com/r-lib/gert#installation
-sudo add-apt-repository ppa:cran/libgit2
-sudo apt-get update
-sudo apt-get install --yes libssh2-1-dev libgit2-dev
+image_name="mlflow-r-dev"
+docker build -f Dockerfile.dev -t $image_name .
+docker run \
+  --rm \
+  -v $(pwd):/mlflow/mlflow/R/mlflow \
+  -v $(pwd)/../../../docs/source:/mlflow/docs/source \
+  $image_name \
+  Rscript -e 'source(".build-doc.R", echo = TRUE)'
 
-Rscript -e 'install.packages("devtools", repos = "https://cloud.r-project.org")'
-Rscript -e 'devtools::install_dev_deps(dependencies = TRUE)'
-# Install Rd2md from source as a temporary fix for the rendering of code examples, until
-# a release is published including the fixes in https://github.com/quantsch/Rd2md/issues/1
-# Note that this commit is equivalent to commit 6b48255 of Rd2md master
-# (https://github.com/quantsch/Rd2md/tree/6b4825579a2df8a22898316d93729384f92a756b)
-# with a single extra commit to fix rendering of \link tags between methods in R documentation.
-Rscript -e 'devtools::install_git("https://github.com/smurching/Rd2md", branch = "mlflow-patches")'
-Rscript -e 'install.packages("rmarkdown", repos = "https://cloud.r-project.org")'
-rm -rf man
-Rscript -e "roxygen2::roxygenise()"
-# remove mlflow-package doc temporarily because no rst doc should be generated for it.
-rm man/mlflow-package.Rd
-Rscript document.R
-# roxygenize again to make sure the previously removed mlflow-packge doc is available as R helpfile
-Rscript -e "roxygen2::roxygenise()"
 popd
diff --git a/docs/source/R-api.rst b/docs/source/R-api.rst
@@ -899,8 +899,8 @@ Arguments
 | Argument                      | Description                          |
 +===============================+======================================+
 | ``flavor``                    | An MLflow flavor object loaded by    |
-|                               | `mlflow_load_model <#mlflow-load-mod |
-|                               | el>`__                               |
+|                               | `mlflo                               |
+|                               | w_load_model <#mlflow-load-model>`__ |
 |                               | , with class loaded from the flavor  |
 |                               | field in an MLmodel file.            |
 +-------------------------------+--------------------------------------+
@@ -1309,13 +1309,16 @@ to be used by package authors to extend the supported MLflow models.
 Arguments
 ---------
 
-========= ===================================================================
-Argument  Description
-========= ===================================================================
-``model`` The loaded MLflow model flavor.
-``data``  A data frame to perform scoring.
-``...``   Optional additional arguments passed to underlying predict methods.
-========= ===================================================================
++-----------+---------------------------------------------------------+
+| Argument  | Description                                             |
++===========+=========================================================+
+| ``model`` | The loaded MLflow model flavor.                         |
++-----------+---------------------------------------------------------+
+| ``data``  | A data frame to perform scoring.                        |
++-----------+---------------------------------------------------------+
+| ``...``   | Optional additional arguments passed to underlying      |
+|           | predict methods.                                        |
++-----------+---------------------------------------------------------+
 
 ``mlflow_register_external_observer``
 =====================================
@@ -1726,15 +1729,21 @@ model types.
 Arguments
 ---------
 
-============== ==================================================================
-Argument       Description
-============== ==================================================================
-``model``      The model that will perform a prediction.
-``path``       Destination path where this MLflow compatible model will be saved.
-``model_spec`` MLflow model config this model flavor is being added to.
-``...``        Optional additional arguments.
-``conda_env``  Path to Conda dependencies file.
-============== ==================================================================
++----------------+----------------------------------------------------+
+| Argument       | Description                                        |
++================+====================================================+
+| ``model``      | The model that will perform a prediction.          |
++----------------+----------------------------------------------------+
+| ``path``       | Destination path where this MLflow compatible      |
+|                | model will be saved.                               |
++----------------+----------------------------------------------------+
+| ``model_spec`` | MLflow model config this model flavor is being     |
+|                | added to.                                          |
++----------------+----------------------------------------------------+
+| ``...``        | Optional additional arguments.                     |
++----------------+----------------------------------------------------+
+| ``conda_env``  | Path to Conda dependencies file.                   |
++----------------+----------------------------------------------------+
 
 ``mlflow_search_runs``
 ======================
@@ -1836,42 +1845,6 @@ Arguments
 |                               | the path of all static paths.        |
 +-------------------------------+--------------------------------------+
 
-``mlflow_set_experiment``
-=========================
-
-Set Experiment
-
-Sets an experiment as the active experiment. Either the name or ID of
-the experiment can be provided. If the a name is provided but the
-experiment does not exist, this function creates an experiment with
-provided name. Returns the ID of the active experiment.
-
-.. code:: r
-
-   mlflow_set_experiment(
-     experiment_name = NULL,
-     experiment_id = NULL,
-     artifact_location = NULL
-   )
-
-.. _arguments-43:
-
-Arguments
----------
-
-+-------------------------------+--------------------------------------+
-| Argument                      | Description                          |
-+===============================+======================================+
-| ``experiment_name``           | Name of experiment to be activated.  |
-+-------------------------------+--------------------------------------+
-| ``experiment_id``             | ID of experiment to be activated.    |
-+-------------------------------+--------------------------------------+
-| ``artifact_location``         | Location where all artifacts for     |
-|                               | this experiment are stored. If not   |
-|                               | provided, the remote server will     |
-|                               | select an appropriate default.       |
-+-------------------------------+--------------------------------------+
-
 ``mlflow_set_experiment_tag``
 =============================
 
@@ -1884,7 +1857,7 @@ metadata that can be updated.
 
    mlflow_set_experiment_tag(key, value, experiment_id = NULL, client = NULL)
 
-.. _arguments-44:
+.. _arguments-43:
 
 Arguments
 ---------
@@ -1916,6 +1889,42 @@ Arguments
 |                               | the current tracking URI.            |
 +-------------------------------+--------------------------------------+
 
+``mlflow_set_experiment``
+=========================
+
+Set Experiment
+
+Sets an experiment as the active experiment. Either the name or ID of
+the experiment can be provided. If the a name is provided but the
+experiment does not exist, this function creates an experiment with
+provided name. Returns the ID of the active experiment.
+
+.. code:: r
+
+   mlflow_set_experiment(
+     experiment_name = NULL,
+     experiment_id = NULL,
+     artifact_location = NULL
+   )
+
+.. _arguments-44:
+
+Arguments
+---------
+
++-------------------------------+--------------------------------------+
+| Argument                      | Description                          |
++===============================+======================================+
+| ``experiment_name``           | Name of experiment to be activated.  |
++-------------------------------+--------------------------------------+
+| ``experiment_id``             | ID of experiment to be activated.    |
++-------------------------------+--------------------------------------+
+| ``artifact_location``         | Location where all artifacts for     |
+|                               | this experiment are stored. If not   |
+|                               | provided, the remote server will     |
+|                               | select an appropriate default.       |
++-------------------------------+--------------------------------------+
+
 ``mlflow_set_tag``
 ==================
 

diff --git a/mlflow/R/mlflow/.Rbuildignore b/mlflow/R/mlflow/.Rbuildignore
@@ -20,6 +20,8 @@ Reference_Manual_mlflow.md
 ^depends\.Rds
 ^R-version
 ^\.utils\.R$
+^\.install-deps\.R$
 ^\.build-package\.R$
+^\.build-doc\.R$
 ^build-package\.sh$
-^Dockerfile\.build$
+^Dockerfile\.dev$
diff --git a/mlflow/R/mlflow/.build-doc.R b/mlflow/R/mlflow/.build-doc.R
@@ -0,0 +1,14 @@
+# Install Rd2md from source as a temporary fix for the rendering of code examples, until
+# a release is published including the fixes in https://github.com/quantsch/Rd2md/issues/1
+# Note that this commit is equivalent to commit 6b48255 of Rd2md master
+# (https://github.com/quantsch/Rd2md/tree/6b4825579a2df8a22898316d93729384f92a756b)
+# with a single extra commit to fix rendering of \link tags between methods in R documentation.
+devtools::install_git("https://github.com/smurching/Rd2md", ref = "mlflow-patches")
+install.packages("rmarkdown", repos = "https://cloud.r-project.org")
+unlink("man", recursive = TRUE)
+roxygen2::roxygenise()
+# remove mlflow-package doc temporarily because no rst doc should be generated for it.
+file.remove("man/mlflow-package.Rd")
+source("document.R", echo = TRUE)
+# roxygenize again to make sure the previously removed mlflow-packge doc is available as R helpfile
+roxygen2::roxygenise()
diff --git a/mlflow/R/mlflow/.build-package.R b/mlflow/R/mlflow/.build-package.R
@@ -1,10 +1,5 @@
 source(".utils.R")
 
-# Increase the timeout length for `utils::download.file` because the default value (60 seconds)
-# could be too short to download large packages such as h2o.
-options(timeout=300)
-# Install dependencies required for the submission check.
-devtools::install_deps(".", dependencies = TRUE)
 # Bundle up the package into a .tar.gz file. This file will be submitted to CRAN.
 package_path <- devtools::build(".", path = ".")
 # Run the submission check against the built package.

diff --git a/mlflow/R/mlflow/.install-deps.R b/mlflow/R/mlflow/.install-deps.R
@@ -0,0 +1,5 @@
+# Increase the timeout length for `utils::download.file` because the default value (60 seconds)
+# could be too short to download large packages such as h2o.
+options(timeout=300)
+install.packages("devtools", dependencies = TRUE)
+devtools::install_dev_deps(dependencies = TRUE)
diff --git a/mlflow/R/mlflow/Dockerfile.build b/mlflow/R/mlflow/Dockerfile.build
diff --git a/mlflow/R/mlflow/Dockerfile.dev b/mlflow/R/mlflow/Dockerfile.dev
@@ -0,0 +1,13 @@
+FROM rocker/r-ver:4.1.2
+
+WORKDIR /mlflow/mlflow/R/mlflow
+RUN apt-get update -y
+RUN apt-get install git wget libxml2-dev libgit2-dev -y
+# pandoc installed by `apt-get` is too old and contains a bug.
+RUN TEMP_DEB=$(mktemp) && \
+    wget --directory-prefix $TEMP_DEB https://github.com/jgm/pandoc/releases/download/2.16.2/pandoc-2.16.2-1-amd64.deb && \
+    dpkg --install $(find $TEMP_DEB -name '*.deb') && \
+    rm -rf $TEMP_DEB
+COPY DESCRIPTION .
+COPY .install-deps.R .
+RUN Rscript -e 'source(".install-deps.R", echo = TRUE)'
diff --git a/mlflow/R/mlflow/build-package.sh b/mlflow/R/mlflow/build-package.sh
@@ -1,5 +1,6 @@
 #!/usr/bin/env bash
 set -ex
 
-docker build -f Dockerfile.build -t r-build-package .
-docker run --rm --workdir /app -v $(pwd):/app r-build-package Rscript -e 'source(".build-package.R", echo = TRUE)'
+image_name="mlflow-r-dev"
+docker build -f Dockerfile.dev -t $image_name .
+docker run --rm -v $(pwd):/mlflow/mlflow/R/mlflow $image_name Rscript -e 'source(".build-package.R", echo = TRUE)'