project
#!/bin/sh
#
# High-level script to manage the project.
# Run `./project --help' for a description of how to use it.
#
# Copyright (C) 2019-2020 Mohammad Akhlaghi <mohammad@akhlaghi.org>
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
# Basic settings
# --------------
# Stop the script if there are any errors.
set -e
# Default option values
jobs=0 # 0 is for the default for the `configure.sh' script.
group=
debug=
host_cc=0
operation=
build_dir=
input_dir=
check_config=
make_targets=
software_dir=
clean_texdir=0
prepare_redo=0
all_highlevel=0
existing_conf=0
scriptname="./project"
minmapsize=10000000000
# Parse the options
# -----------------
#
# Separate command-line arguments from options. Then put the option value
# into the respective variable.
#
# Each option has two lines because we want to process both these formats:
# `--name=value' and `--name value'. The former (with `=') is a single
# command-line argument, so we just need to shift the counter by one. The
# latter (without `=') is two arguments, so we'll need two shifts.
#
# Note on the case strings: for every option, we need three lines: one when
# the option name and value are separate. Another when there is an equal
# between them, and finally one where the value is immediately after the
# short-format. This exact order is important. Otherwise, there will be a
# conflict between them.
print_help() {
# Print the output.
cat <<EOF
Usage: $scriptname configure [OPTIONS]
$scriptname make [OPTIONS]
Top-level script to manage the reproducible project. The high-level
operation is defined by the (mandatory) second argument:
configure - Configure project for this machine (e.g., build software).
make - Run the project (do analysis and build outputs).
shell - Execute the project's shell for interactive testing.
RECOMMENDATION: If this is the first time you are configuring this
template, please don't use the options and let the script explain each
parameter in full detail by simply running './project configure'.
Project 'make' special features.
./project make Build the project on one thread
./project make -jN Built the project in parallel on N threads.
./project make clean Clean all files generated by 'make' (not software).
./project make distclean Clean everything (including compiled software).
./project make dist Produce a LaTeX-ready-to-build distribution tarball
('tar.gz') of the project. This is ready to be
uploaded to servers like 'arXiv.org'.
./project make dist-lzip Similar to 'dist', but compress to '.tar.lz'.
./project make dist-zip Similar to 'dist', but compress to '.zip'.
./project make dist-software Build a .tar.gz tarball containing all
software source tarballs necessary for the
project.
With the options below you can modify the default behavior.
Configure options:
-b, --build-dir=STR Top directory to build the project in.
-e, --existing-conf Use (possibly existing) local configuration.
--host-cc Use host system's C compiler, don't build GCC.
-i, --input-dir=STR Directory containing input datasets (optional).
-m, --minmapsize=INT [Gnuastro] Minimum number of bytes to use RAM.
-s, --software-dir=STR Directory containing necessary software tarballs.
--check-config During configuration, show what is being built.
--clean-texdir Remove possibly existing build-time subdirectories
under the project's 'tex/' directory (can happen
when source is from arXiv for example).
--all-highlevel Build all high-level software (for development).
Data download:
./project make do-download Download WHO data and LaTeX customisation files.
./project make clean-download Remove downloaded files.
Poisson analysis:
./project make check-poisson Check that the Poisson model runs as expected.
./project make poisson Run the Poisson analysis.
./project make clean-poisson Remove Poisson analysis output files.
Final paper:
./project make clean-paper Remove temporary files for producing submittable pdf.
./project make clean-full clean-download + clean-poisson + clean-paper
Configure and Make options:
-g, --group=STR Build and run with write permissions for a group.
-j, --jobs=INT Number of threads to build/run the software.
-?, --help Print this help list.
Make options:
-d, --debug=FLAGS Print various types of debugging information.
-p, --prepare-redo Re-do preparation (only done automatically once).
Mandatory or optional arguments to long options are also mandatory or optional
for any corresponding short options.
Maneage URL: https://maneage.org
Report bugs to mohammad@akhlaghi.org
EOF
}
on_off_option_error() {
if [ "x$2" = x ]; then
echo "$scriptname: '$1' doesn't take any values."
else
echo "$scriptname: '$1' (or '$2') doesn't take any values."
fi
exit 1
}
check_v() {
if [ x"$2" = x ]; then
echo "$scriptname: option '$1' requires an argument."
echo "Try '$scriptname --help' for more information."
exit 1;
fi
}
func_operation_set() {
if [ x$operation = x ]; then
operation=$1
else
echo "Only one operation ('configure' or 'make') may be given."
exit 1
fi
}
while [ $# -gt 0 ]
do
case $1 in
# Main operation.
configure) func_operation_set $1; shift;;
make) func_operation_set $1; shift;;
shell) func_operation_set $1; shift;;
# Configure options:
-b|--builddir) build_dir="$2"; check_v "$1" "$build_dir"; shift;shift;;
-b=*|--build-dir=*) build_dir="${1#*=}"; check_v "$1" "$build_dir"; shift;;
-b*) build_dir=$(echo "$1" | sed -e's/-b//'); check_v "$1" "$build_dir"; shift;;
-e|--existing-conf) existing_conf=1; shift;;
-e*|--existing-conf=*) on_off_option_error --existing-conf -e;;
--host-cc) host_cc=1; shift;;
--host-cc=*) on_off_option_error --host-cc;;
-i|--input-dir) input_dir="$2"; check_v "$1" "$input_dir"; shift;shift;;
-i=*|--input-dir=*) input_dir="${1#*=}"; check_v "$1" "$input_dir"; shift;;
-i*) input_dir=$(echo "$1" | sed -e's/-i//'); check_v "$1" "$input_dir"; shift;;
-m|--minmapsize) minmapsize="$2"; check_v "$1" "$minmapsize"; shift;shift;;
-m=*|--minmapsize=*) minmapsize="${1#*=}"; check_v "$1" "$minmapsize"; shift;;
-m*) minmapsize=$(echo "$1" | sed -e's/-m//'); check_v "$1" "$minmapsize"; shift;;
-s|--software-dir) software_dir="$2"; check_v "$1" "$software_dir"; shift;shift;;
-s=*|--software-dir=*) software_dir="${1#*=}"; check_v "$1" "$software_dir"; shift;;
-s*) software_dir=$(echo "$1" | sed -e's/-s//'); check_v "$1" "$software_dir"; shift;;
--check-config) check_config=1; shift;;
--check-config=*) on_off_option_error --check-config;;
--clean-texdir) clean_texdir=1; shift;;
--clean-texdir=*) on_off_option_error --clean-texdir;;
--all-highlevel) all_highlevel=1; shift;;
--all-highlevel=*) on_off_option_error --all-highlevel;;
# Configure and Make options:
-g|--group) group="$2"; check_v group "$group"; shift;shift;;
-g=*|--group=*) group="${1#*=}"; check_v group "$group"; shift;;
-g*) group=$(echo "$1" | sed -e's/-g//'); check_v group "$group"; shift;;
-j|--jobs) jobs="$2"; check_v jobs "$jobs"; shift;shift;;
-j=*|--jobs=*) jobs="${1#*=}"; check_v jobs "$jobs"; shift;;
-j*) jobs=$(echo "$1" | sed -e's/-j//'); check_v jobs "$jobs"; shift;;
-'?'|--help) print_help; exit 0;;
-'?'*|--help=*) on_off_option_error --help -?;;
# Make options
# ------------
#
# Note that Make's `debug' can take values, but when called without any
# value, it is like giving it a value of `a'):
-d|--debug) if [ x"$2" = x ]; then debug=a; shift;
else debug="$2"; check_v debug "$debug"; shift;shift; fi;;
-d=*|--debug=*) debug="${1#*=}"; check_v debug "$debug"; shift;;
-d*) debug=$(echo "$1" | sed -e's/-d//'); check_v debug "$debug"; shift;;
-p|--prepare-redo) prepare_redo=1; shift;;
-p=*|--prepare-redo=*) on_off_option_error --prepare-redo; shift;;
# Unrecognized option:
-*) echo "$scriptname: unknown option '$1'"; exit 1;;
# Not an option, an argument (so its a Make target).
*) make_targets="$make_targets $1"; shift;;
esac
done
# Check configuration status
# --------------------------
if ! [ x$check_config = x ]; then
# Find the color option to pass to `ls'. Note that `--color' (for GNU
# Coreutils `ls') should be checked first because it also has `-G', but
# for something else.
if ls --color 2> /dev/null > /dev/null; then coloropt="--color=auto"
elif ls -G 2> /dev/null > /dev/null; then coloropt="-G"
else coloropt=""
fi
# Print a notice to let the user know what is happening.
cat <<EOF
This is an infinite loop that will print what software are being built at
every moment. It is actually a listing ('ls' command) of the temporary
directory where software source code are unpacked while they are being
built. If the project isn't being configured, the output (every second)
will either be empty (only a date) or a with an error about a non-existant
directory. This feature is thus only useful when the project's software are
being built.
EOF
# Run the infinite loop.
while true; do
echo;
echo "$(date) [[press CTRL-C to stop]]";
ls $coloropt .build/software/build-tmp || junk=1;
sleep 1;
done
exit 0
fi
# Basic group settings
# --------------------
if ! [ x$group = x ]; then
# Check if group is usable.
if ! sg "$group" "echo test &> /dev/null" &> /dev/null; then
echo "$scriptname: '$group' is not a usable group name on this system.";
echo "(TIP: you can use the 'groups' command to see your groups)"
exit 1
fi
# Set the group option for running Make.
gopt="reproducible_paper_group_name=$group"
fi
# Error when configuration isn't run
configuration_necessary() {
cat <<EOF
The project is either (1) not configured on this system, or (2) the
configuration wasn't successful.
(1) If it hasn't been configured at all, use the command below to configure
it (set a build directory and let it build its necessary software in it).
$ ./project configure
(2) If it has been configured, but the configuration failed in a step, you
can re-configure it using your previous settings with the command
below. All successful steps will be skipped, allowing a fast completion.
$ ./project configure -e
If there was a problem, please let us know by filling this online form:
http://savannah.nongnu.org/support/?func=additem&group=reproduce
EOF
exit 1
}
# Run operations in controlled environment
# ----------------------------------------
controlled_env() {
# Get the full address of the build directory:
bdir=`.local/bin/realpath .build`
# Remove all existing environment variables (with `env -i') and only
# use some pre-defined environment variables, then build the project.
envmake=".local/bin/env -i HOME=$bdir sys_rm=$(which rm) $gopt"
envmake="$envmake .local/bin/make --no-builtin-rules"
envmake="$envmake --no-builtin-variables -f $1"
if ! [ x"$debug" = x ]; then envmake="$envmake --debug=$debug"; fi
# Set the number of jobs. Note that for the `configure.sh' script the
# default value has to be 0, so the default is the maximum number of
# threads. But here, the default value is 1.
if ! [ x"$jobs" = x0 ]; then envmake="$envmake -j$jobs"; fi
# Run the project
if [ x"$group" = x ]; then
$envmake $make_targets
else
# Set the group and permission flags.
sg "$group" "umask $perms && $envmake $make_targets"
fi
}
# Do requested operation
# ----------------------
perms="u+r,u+w,g+r,g+w,o-r,o-w,o-x"
configscript=./reproduce/software/shell/configure.sh
case $operation in
# Build the project's software.
configure)
# Set executable flags
# --------------------
#
# In some scenarios (for example when using a tarball from arXiv),
# it may happen that the host server has removed the executable
# flags of all the files. In `README.md' we instruct the readers on
# setting the executable flag of this script. But we don't want the
# user to have to worry about any other file that needs an
# executable flag.
#
# Basically, all the project shell scripts need executable flags so
# to make sure they have them, we are activating the executable
# flags by default here every time './project configure' is run. If
# any other file in your project needs such flags, add them here.
chmod +x reproduce/software/shell/* reproduce/software/config/*.sh \
reproduce/analysis/bash/*
# If the user requested, clean the TeX directory from the extra
# (to-be-built) directories that may already be there (and will not
# allow the configuration to complete).
if [ x"$clean_texdir" = x1 ]; then
rm -rf tex/build tex/tikz
fi
# Variables to pass to the configuration script.
export jobs=$jobs
export host_cc=$host_cc
export build_dir=$build_dir
export input_dir=$input_dir
export scriptname=$scriptname
export minmapsize=$minmapsize
export software_dir=$software_dir
export existing_conf=$existing_conf
export all_highlevel=$all_highlevel
export reproducible_paper_group_name=$group
# Run the configuration script
if [ x"$group" = x ]; then
$configscript
else
# Set the group and permission flags.
sg "$group" "umask $perms && $configscript"
# Set the group writing permission for everything in the
# installed software directory. The common build process sets
# the writing permissions of the installed programs/libraries
# to `755'. So group members can't write over a file. This
# creates problems when another group member wants to update
# the software for example. We thus need to manually add the
# group writing flag to all installed software files.
echo "Enabling group writing permission on all installed software..."
.local/bin/chmod -R g+w .local/;
fi
;;
# Batch execution of the project.
make)
# Make sure the configure script has been completed properly
# (`configuration-done.txt' exists).
if ! [ -f .build/software/configuration-done.txt ]; then
configuration_necessary
fi
# Run data preparation phase (optionally build Makefiles with
# special values for optimizing the main `top-make.mk'). But note
# that data preparation is only done automatically the first time
# the project is built (when `.build/software/preparation-done.mk'
# doesn't yet exist). After that, if the user wants to re-do the
# preparation they have to use the `--prepare-redo' option.
if ! [ -f .build/software/preparation-done.mk ] \
|| [ x"$prepare_redo" = x1 ]; then
controlled_env reproduce/analysis/make/top-prepare.mk
fi
# Run the actual project.
controlled_env reproduce/analysis/make/top-make.mk
;;
shell)
# Make sure the configure script has been completed properly
# (`configuration-done.txt' exists).
if ! [ -f .build/software/configuration-done.txt ]; then
configuration_necessary
fi
# Run the project's own shell without inheriting any environment
# from the host.
bdir=`.local/bin/realpath .build`
instdir=$bdir/software/installed
.local/bin/env -i \
HOME=$bdir \
CCACHE_DISABLE=1 \
PATH=$instdir/bin \
LDFLAGS=-L$instdir/lib \
SHELL=$instdir/bin/bash \
CPPFLAGS=-I$instdir/include \
LD_LIBRARY_PATH=$instdir/lib \
OMPI_MCA_plm_rsh_agent=/bin/false \
PYTHONPATH=$instdir/lib/python/site-packages \
PYTHONPATH3=$instdir/lib/python/site-packages \
$instdir/bin/bash
;;
# Operation not specified.
*)
cat <<EOF
No operation defined!
Please run with '--help' for more information.
(TIP: available operations are: 'configure' or 'make').
EOF
exit 1
;;
esac