From 48e61df320e7571e527049cd3eef5cf96b8cb491 Mon Sep 17 00:00:00 2001 From: Mohammad Akhlaghi Date: Mon, 23 Mar 2020 03:16:06 +0000 Subject: Analysis and configuration file sections complete With this commit a description of these two important parts have been added to the project, along with several figures showing various parts of the files that are discussed. I also done some other restructuring of the figures and files to make things fit better into the the description of the paper. --- reproduce/analysis/config/INPUTS.conf | 15 +++++ reproduce/analysis/config/INPUTS.mk | 15 ----- reproduce/analysis/config/menke-demo-year.conf | 3 + reproduce/analysis/config/pdf-build.conf | 21 ++++++ reproduce/analysis/config/pdf-build.mk | 21 ------ reproduce/analysis/config/verify-outputs.conf | 3 + reproduce/analysis/config/verify-outputs.mk | 3 - reproduce/analysis/make/analysis-1.mk | 89 -------------------------- reproduce/analysis/make/demo-plot.mk | 59 +++++++++++++++++ reproduce/analysis/make/format.mk | 82 ++++++++++++++++++++++++ reproduce/analysis/make/top-make.mk | 5 +- 11 files changed, 186 insertions(+), 130 deletions(-) create mode 100644 reproduce/analysis/config/INPUTS.conf delete mode 100644 reproduce/analysis/config/INPUTS.mk create mode 100644 reproduce/analysis/config/menke-demo-year.conf create mode 100644 reproduce/analysis/config/pdf-build.conf delete mode 100644 reproduce/analysis/config/pdf-build.mk create mode 100644 reproduce/analysis/config/verify-outputs.conf delete mode 100644 reproduce/analysis/config/verify-outputs.mk delete mode 100644 reproduce/analysis/make/analysis-1.mk create mode 100644 reproduce/analysis/make/demo-plot.mk create mode 100644 reproduce/analysis/make/format.mk (limited to 'reproduce') diff --git a/reproduce/analysis/config/INPUTS.conf b/reproduce/analysis/config/INPUTS.conf new file mode 100644 index 0000000..b1cf546 --- /dev/null +++ b/reproduce/analysis/config/INPUTS.conf @@ -0,0 +1,15 @@ +# Input files necessary for this project. +# +# This file is read by the configure script and running Makefiles. +# +# Copyright (C) 2018-2020 Mohammad Akhlaghi +# +# Copying and distribution of this file, with or without modification, are +# permitted in any medium without royalty provided the copyright notice and +# this notice are preserved. This file is offered as-is, without any +# warranty. + +MK20DATA = menke20.xlsx +MK20MD5 = 8e4eee64791f351fec58680126d558a0 +MK20SIZE = 1.9MB +MK20URL = https://www.biorxiv.org/content/biorxiv/early/2020/01/18/2020.01.15.908111/DC1/embed/media-1.xlsx diff --git a/reproduce/analysis/config/INPUTS.mk b/reproduce/analysis/config/INPUTS.mk deleted file mode 100644 index b1cf546..0000000 --- a/reproduce/analysis/config/INPUTS.mk +++ /dev/null @@ -1,15 +0,0 @@ -# Input files necessary for this project. -# -# This file is read by the configure script and running Makefiles. -# -# Copyright (C) 2018-2020 Mohammad Akhlaghi -# -# Copying and distribution of this file, with or without modification, are -# permitted in any medium without royalty provided the copyright notice and -# this notice are preserved. This file is offered as-is, without any -# warranty. - -MK20DATA = menke20.xlsx -MK20MD5 = 8e4eee64791f351fec58680126d558a0 -MK20SIZE = 1.9MB -MK20URL = https://www.biorxiv.org/content/biorxiv/early/2020/01/18/2020.01.15.908111/DC1/embed/media-1.xlsx diff --git a/reproduce/analysis/config/menke-demo-year.conf b/reproduce/analysis/config/menke-demo-year.conf new file mode 100644 index 0000000..429b220 --- /dev/null +++ b/reproduce/analysis/config/menke-demo-year.conf @@ -0,0 +1,3 @@ +# This is the demonstration year showing the number of papers studied +# before 1997. +menke-demo-year = 1996 diff --git a/reproduce/analysis/config/pdf-build.conf b/reproduce/analysis/config/pdf-build.conf new file mode 100644 index 0000000..e2d59cc --- /dev/null +++ b/reproduce/analysis/config/pdf-build.conf @@ -0,0 +1,21 @@ +# Make the final PDF? +# ------------------- +# +# During the project's early phases, it is usually not necessary to build +# the PDF file (which makes a lot of output lines on the command-line and +# can make it hard to find the commands and possible errors (and their +# outputs). Also, in some cases, only the produced results may be of +# interest and not the final PDF, so LaTeX (and its necessary packages) may +# not be installed. +# +# If this variable is given any string, a PDF will be made with +# LaTeX. Otherwise, a notice will just printed that for now, no PDF will be +# created. +# +# Copyright (C) 2018-2020 Mohammad Akhlaghi +# +# Copying and distribution of this file, with or without modification, are +# permitted in any medium without royalty provided the copyright notice and +# this notice are preserved. This file is offered as-is, without any +# warranty. +pdf-build-final = yes diff --git a/reproduce/analysis/config/pdf-build.mk b/reproduce/analysis/config/pdf-build.mk deleted file mode 100644 index e2d59cc..0000000 --- a/reproduce/analysis/config/pdf-build.mk +++ /dev/null @@ -1,21 +0,0 @@ -# Make the final PDF? -# ------------------- -# -# During the project's early phases, it is usually not necessary to build -# the PDF file (which makes a lot of output lines on the command-line and -# can make it hard to find the commands and possible errors (and their -# outputs). Also, in some cases, only the produced results may be of -# interest and not the final PDF, so LaTeX (and its necessary packages) may -# not be installed. -# -# If this variable is given any string, a PDF will be made with -# LaTeX. Otherwise, a notice will just printed that for now, no PDF will be -# created. -# -# Copyright (C) 2018-2020 Mohammad Akhlaghi -# -# Copying and distribution of this file, with or without modification, are -# permitted in any medium without royalty provided the copyright notice and -# this notice are preserved. This file is offered as-is, without any -# warranty. -pdf-build-final = yes diff --git a/reproduce/analysis/config/verify-outputs.conf b/reproduce/analysis/config/verify-outputs.conf new file mode 100644 index 0000000..e580e04 --- /dev/null +++ b/reproduce/analysis/config/verify-outputs.conf @@ -0,0 +1,3 @@ +# To disable verification of output datasets set this variable to yes + +verify-outputs = diff --git a/reproduce/analysis/config/verify-outputs.mk b/reproduce/analysis/config/verify-outputs.mk deleted file mode 100644 index e580e04..0000000 --- a/reproduce/analysis/config/verify-outputs.mk +++ /dev/null @@ -1,3 +0,0 @@ -# To disable verification of output datasets set this variable to yes - -verify-outputs = diff --git a/reproduce/analysis/make/analysis-1.mk b/reproduce/analysis/make/analysis-1.mk deleted file mode 100644 index f739306..0000000 --- a/reproduce/analysis/make/analysis-1.mk +++ /dev/null @@ -1,89 +0,0 @@ -# Use the data from Menke 2020 (DOI:10.1101/2020.01.15.908111) as a -# demonstration analysis for this paper. This is a relevant paper because -# it provides good statistics about the status of reproducibility in -# scientific publications. -# -# Copyright (C) 2020 Mohammad Akhlaghi -# -# This Makefile 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 Makefile 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. See . - - - - -# Save the "Table 3" spreadsheet from the downloaded `.xlsx' file into a -# simple plain-text file that is easy to use. -a1dir = $(BDIR)/analysis-1 -mk20tab3 = $(a1dir)/menke20-table-3.txt -$(a1dir):; mkdir $@ -$(mk20tab3): $(indir)/menke20.xlsx | $(a1dir) - - # Set a base-name for the table-3 data. - base=$(basename $(notdir $<))-table-3 - - # Unfortunately XLSX I/O only works when the input and output are - # in the directory it is running. So first, we need to switch to - # the input directory, run it, then put our desired output where we - # want and delete the extra files. - topdir=$$(pwd) - cd $(indir) - xlsxio_xlsx2csv $(notdir $<) - cp $(notdir $<)."Table 3 All by journal by year".csv $$base.csv - rm $(notdir $<).*.csv - cd $$topdir - - # Read the necessary information. Note that we are dealing with a - # CSV (comma-separated value) file. But when there are commas in a - # string, quotation signs are put around it. The `FPAT' values is - # fully described in the GNU AWK manual. In short, it ensures that - # if there is a comma in the middle of double-quotes, it doesn't - # count as a delimter. - echo "# Column 1: YEAR [counter, i16] Year of journal's publication." > $@.tmp - echo "# Column 2: NUM_PAPERS [counter, i16] Number of studied papers in that journal." >> $@.tmp - echo "# Column 3: NUM_PAPERS_WITH_TOOLS [counter, i16] Number of papers with an identified tool." >> $@.tmp - echo "# Column 4: NUM_ID_TOOLS [counter, i16] Number of software/tools that were identified." >> $@.tmp - echo "# Column 5: JOURNAL_NAME [string, str150] Name of journal." >> $@.tmp - awk 'NR>1{printf("%-10d%-10d%-10d%-10d %s\n", $$2, $$3, $$3*$$NF, $$(NF-1), $$1)}' \ - FPAT='([^,]+)|("[^"]+")' $(indir)/$$base.csv >> $@.tmp - - # Set the temporary file as the final target. This was done so if - # there is any possible crash in the steps above, this rule is - # re-run (its final target isn't rebuilt). - mv $@.tmp $@ - - - - -############################ -# Recreate Figure 1C of Menke+20. -############################ - -# awk '!/^#/{all[$1]+=$2; id[$1]+=$3} END{for(year in all){print year, id[year]/all[year]}}' menke20-table-3.txt.tmp -############################ - - - - - -# Main LaTeX macro file -$(mtexdir)/analysis-1.tex: $(mk20tab3) | $(mtexdir) - - # Count the total number of papers in their study. - v=$$(awk '!/^#/{c+=$$2} END{print c}' $(mk20tab3)) - echo "\newcommand{\menkenumpapers}{$$v}" > $@ - - # Count how many unique journals there were in the study. Note that - # the `31' comes because we put 10 characters for each numeric - # column and separated the last numeric column from the string - # column with a space. If the number of numeric columns change in - # the future, the `31' also has to change. - v=$$(awk 'BEGIN{FIELDWIDTHS="41 10000"} !/^#/{print $$2}' \ - $(mk20tab3) | uniq | wc -l) - echo "\newcommand{\menkenumjournals}{$$v}" >> $@ diff --git a/reproduce/analysis/make/demo-plot.mk b/reproduce/analysis/make/demo-plot.mk new file mode 100644 index 0000000..caf77af --- /dev/null +++ b/reproduce/analysis/make/demo-plot.mk @@ -0,0 +1,59 @@ +# Second step of analysis: +# Data for plot of number/fraction of tools per year. +# +# Copyright (C) 2020 Mohammad Akhlaghi +# +# This Makefile 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 Makefile 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. See . + + + + +# Directory to host outputs +# ------------------------- +a2dir = $(texdir)/tools-per-year +$(a2dir):; mkdir $@ + + + + + +# Table for Figure 1C of Menke+20 +# ------------------------------- +a2mk20f1c = $(a2dir)/tools-per-year.txt +$(a2mk20f1c): $(mk20tab3) | $(a2dir) + + # Remove the (possibly) produced figure that is created from this + # table: it is created by LaTeX's TiKZ package, and includes + # multiple files with a fixed prefix. + rm -f $(tikzdir)/figure-tools-per-year* + + # Find the maximum number of papers. + awk '!/^#/{all[$$1]+=$$2; id[$$1]+=$$3} \ + END{ for(year in all) \ + print year, 100*id[year]/all[year], all[year] \ + }' $< \ + > $@ + + + + + +# Final LaTeX macro +$(mtexdir)/demo-plot.tex: $(a2mk20f1c) $(pconfdir)/menke-demo-year.conf + + # Find the first year (first column of first row) of data. + v=$$(awk 'NR==1{print $$1}' $(a2mk20f1c)) + echo "\newcommand{\menkefirstyear}{$$v}" > $@ + + # Find the number of papers in 1996. + v=$$(awk '$$1==$(menke-demo-year){print $$3}' $(a2mk20f1c)) + echo "\newcommand{\menkenumpapersdemocount}{$$v}" >> $@ + echo "\newcommand{\menkenumpapersdemoyear}{$(menke-demo-year)}" >> $@ diff --git a/reproduce/analysis/make/format.mk b/reproduce/analysis/make/format.mk new file mode 100644 index 0000000..868c411 --- /dev/null +++ b/reproduce/analysis/make/format.mk @@ -0,0 +1,82 @@ +# First step of analysis: +# Prepare the data, return basic values. +# +# As a demonstration analysis to go with the paper, we use the data from +# Menke 2020 (DOI:10.1101/2020.01.15.908111). This is a relevant paper +# because it provides interesting statistics about tools and methods used +# in scientific papers. +# +# Copyright (C) 2020 Mohammad Akhlaghi +# +# This Makefile is free software: you can redistribute it and/or modify it +# under the terms of the GNU General Public License as published by the +# Free Software Foundation, either version 3 of the License, or (at your +# option) any later version. +# +# This Makefile is distributed in the hope that it will be useful, but +# WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General +# Public License for more details. See . + + + + +# Save the "Table 3" spreadsheet from the downloaded `.xlsx' file into a +# simple plain-text file that is easy to use. +a1dir = $(BDIR)/analysis1 +mk20tab3 = $(a1dir)/menke20-table-3.txt +$(a1dir):; mkdir $@ +$(mk20tab3): $(indir)/menke20.xlsx | $(a1dir) + + # Set a base-name for the table-3 data. + base=$(basename $(notdir $<))-table-3 + + # Unfortunately XLSX I/O only works when the input and output are + # in the directory it is running. So first, we need to switch to + # the input directory, run it, then put our desired output where we + # want and delete the extra files. + topdir=$$(pwd) + cd $(indir) + xlsxio_xlsx2csv $(notdir $<) + cp $(notdir $<)."Table 3 All by journal by year".csv $$base.csv + rm $(notdir $<).*.csv + cd $$topdir + + # Read the necessary information. Note that we are dealing with a + # CSV (comma-separated value) file. But when there are commas in a + # string, quotation signs are put around it. The `FPAT' values is + # fully described in the GNU AWK manual. In short, it ensures that + # if there is a comma in the middle of double-quotes, it doesn't + # count as a delimter. + echo "# Column 1: YEAR [counter, i16] Year of journal's publication." > $@.tmp + echo "# Column 2: NUM_PAPERS [counter, i16] Number of studied papers in that journal." >> $@.tmp + echo "# Column 3: NUM_PAPERS_WITH_TOOLS [counter, i16] Number of papers with an identified tool." >> $@.tmp + echo "# Column 4: NUM_ID_TOOLS [counter, i16] Number of software/tools that were identified." >> $@.tmp + echo "# Column 5: JOURNAL_NAME [string, str150] Name of journal." >> $@.tmp + awk 'NR>1{printf("%-10d%-10d%-10d%-10d %s\n", $$2, $$3, $$3*$$NF, $$(NF-1), $$1)}' \ + FPAT='([^,]+)|("[^"]+")' $(indir)/$$base.csv >> $@.tmp + + # Set the temporary file as the final target. This was done so if + # there is any possible crash in the steps above, this rule is + # re-run (its final target isn't rebuilt). + mv $@.tmp $@ + + + + + +# Main LaTeX macro file +$(mtexdir)/format.tex: $(mk20tab3) + + # Count the total number of papers in their study. + v=$$(awk '!/^#/{c+=$$2} END{print c}' $(mk20tab3)) + echo "\newcommand{\menkenumpapers}{$$v}" > $@ + + # Count how many unique journals there were in the study. Note that + # the `31' comes because we put 10 characters for each numeric + # column and separated the last numeric column from the string + # column with a space. If the number of numeric columns change in + # the future, the `31' also has to change. + v=$$(awk 'BEGIN{FIELDWIDTHS="41 10000"} !/^#/{print $$2}' \ + $(mk20tab3) | uniq | wc -l) + echo "\newcommand{\menkenumjournals}{$$v}" >> $@ diff --git a/reproduce/analysis/make/top-make.mk b/reproduce/analysis/make/top-make.mk index 6dd322f..000c1fd 100644 --- a/reproduce/analysis/make/top-make.mk +++ b/reproduce/analysis/make/top-make.mk @@ -112,8 +112,9 @@ endif # wild-card like the configuration Makefiles). makesrc = initialize \ download \ + format \ + demo-plot \ verify \ - analysis-1 \ paper @@ -132,5 +133,5 @@ makesrc = initialize \ # But before that, we need to identify the phase for the Makefiles that are # run both in `./project prepare' and `./project make'. project-phase = make -include reproduce/analysis/config/*.mk +include reproduce/analysis/config/*.conf include $(foreach s,$(makesrc), reproduce/analysis/make/$(s).mk) -- cgit v1.2.1