Finalizing refactoring of 'replicate results'

README.md

@@ -0,0 +1 @@
+doc/README.md

app/generate-timings/Main.hs

@@ -380,6 +380,7 @@ timeFilePoint taxaNumPadder strLenPadder counter tcmPath fp = do
             , "--output"
             , "/dev/null"
             ]
+    putStrLn commandStr
 
     p <- makeCleanProcess counter [] commandStr
 

ChangeLog.md → doc/CHANGELOG.md

@@ -1,5 +1,12 @@
 # Revision history for commutative-multi-string-alignment
 
+## 1.1.1 -- 2022-03-01
+
+* Refactoring replicate results script
+* Droppping `stack` build support
+
+
+
 ## 1.1.0 -- 2022-02-23
 
 * Added support for GHC `9.0.*` and `9.2.*`

doc/README.md

@@ -0,0 +1,40 @@
+ Efficient Implied Alignment
+=============================
+
+[![DOI:10.1186/s12859-020-03595-2](https://zenodo.org/badge/DOI/10.1186/s12859-020-03595-2.svg)](https://doi.org/10.1186/s12859-020-03595-2)
+
+This repository hosts the program `implied-align` implementing the algorithm described in the paper *Efficient Implied Alignment* and a script to replicate the results of the paper.
+
+### Installation of `implied-align`
+
+You can build and install `implied-align` from source using the Haskell build system `ghcup` via the supplied `makefile` from the source directory of this project:
+
+```
+$ make install
+```
+
+After the `make install` command has completed, the `implied-align` binary will be placed in this project's `bin` directory.
+
+### Running `implied-align`
+
+The `implied-align` program takes a number of command line arguments to specify inputs and outputs. For more information run the following command:
+
+```
+$ implied-align --help
+```
+
+### Replicating results of the paper
+
+For convience of replicating the results of the paper, a "replicate-results" script has been provided. All that is need to replicate the paper's results is to run the following command:
+
+```
+$ make replicate
+```
+
+This will create a `replicate-results` directory, with sub directories `csv`, `data`, `img`, `taxa`, and `tree`. 
+
+ - The `taxa` directory holds temporary files used in pruning the data-sets. 
+ - The `data` & `tree` directories hold the pruned input files for the data-sets.
+ - The `csv` directory holds the timing information measured during the replication of the results.
+ - The `img` directory holds the generted images from the results.
+

efficient-implied-alignment.cabal

@@ -3,7 +3,7 @@ category:       Algorithms
 build-type:     Simple
 
 name:           efficient-implied-alignment
-version:        1.1.0
+version:        1.1.1
 
 author:         Alex Washburn
 maintainer:     [email protected]
@@ -31,8 +31,8 @@ tested-with:
     GHC == 9.2.1
 
 extra-source-files:
-    ChangeLog.md
-    README.md
+    doc/CHANGELOG.md
+    doc/README.md
 
 source-repository head
   type: git

makefile

@@ -6,64 +6,96 @@
 #
 # # # # #
 
-bin-dir          := ./bin
-data-dir         := replicate-results
-executable       := $(bin-dir)/generate-timings
-generate         := $(bin-dir)/measure-scaling-performance.sh using
-biological-scale := [1%1,1%2,1%4,1%8,1%16,1%32]
-customized-scale := [1%1,1%2,1%4,1%8,1%16,1%32,1%64]
-customized-nodes := [4,8,16,32,64,128,256]
+dir-bin         := ./bin
+dir-data        := replicate-results
+bin-aln         := implied-align
+bin-aln-path    := $(dir-bin)/$(bin-aln)
+bin-gen         := generate-timings
+bin-gen-path    := $(dir-bin)/$(bin-gen)
+measure-script  := measure-scaling-performance.sh
+measure-dataset := $(dir-bin)/$(measure-script) using
+prerequisites   := $(bin-aln-path) $(bin-aln-path) ensure-python ensure-workspace
 
 
 # All synonyms for replicating the paper's results.
-all: results
-
 replicate: results
 
 reproduce: results
 
-results: fungi metazoa pathological
+results:   fungi metazoa pathological
+
 
+# Install dependencies required to replicate results.
+ensure-haskell:
+	@command -v ghcup      >/dev/null 2>&1 || \
+	    curl --proto '=https' --tlsv1.2 -sSf https://get-ghcup.haskell.org | sh
 
-# Install dependencies required to replicate results:
-ensure-Haskell: $(bin-dir)/generate-timings
-	@command -v ghcup &> /dev/null || curl --proto '=https' --tlsv1.2 -sSf https://get-ghcup.haskell.org | sh
+ensure-python:
+	@command -v python3    >/dev/null 2>&1 || apt-get install python3     --yes
+	@command -v pip3       >/dev/null 2>&1 || apt-get install python3-pip --yes
+	@pip3  show matplotlib >/dev/null 2>&1 || pip3 install --upgrade matplotlib
 
-ensure-Python:
-	@command -v python3    &> /dev/null || apt-get install python3     --yes
-	@command -v pip3       &> /dev/null || apt-get install python3-pip --yes
-	@pip3  show matplotlib &> /dev/null || pip3 install --upgrade matplotlib
+ensure-workspace:
+	@$(eval make-workspace := mkdir -p $(dir-data))
+	@$(make-workspace)/csv
+	@$(make-workspace)/data
+	@$(make-workspace)/img
+	@$(make-workspace)/taxa
+	@$(make-workspace)/tree
 
-ensure-R:
-	command -v &> /dev/null || sudo apt install r-base
 
-compile-binaries: ensure-Haskell
-	@echo "Copiling binaries"
-	@ghcup run \
-	    --ghc 9.2.1 \
-	    --cabal 3.6.2.0 \
-	    -- cabal update && \
-	       cabal install \
-	        --installdir=$(bin-dir) \
-	        --install-method=copy 
+# Build binaries
+require := ensure-haskell $(wildcard app/**/*.hs) $(wildcard src/**/*.hs)
+install := ghcup run --ghc 9.2.1 --cabal 3.6.2.0 -- \
+		cabal update &&	cabal install $(bin-gen) $(bin-aln) \
+		--installdir=$(dir-bin) --install-method=copy
 
-ensure-workspace:
-	@mkdir -p $(data-dir)/csv
-	@mkdir -p $(data-dir)/data
-	@mkdir -p $(data-dir)/img
-	@mkdir -p $(data-dir)/taxa
-	@mkdir -p $(data-dir)/tree
-
-fungi: compile-binaries ensure-Python ensure-R ensure-workspace
-	$(generate) \
-	    'fungi'        '11' '[25,50,100,200,400,800,1553]' $(biological-scale)
-
-metazoa: compile-binaries ensure-Python ensure-R ensure-workspace
-	$(generate) \
-	    'metazoa'      '11' '[25,50,100,200,400,800,1766]' $(biological-scale)
-
-pathological: compile-binaries ensure-Python ensure-R ensure-workspace
-	$(generate) \
-	    'pathological' '12' $(customized-nodes) $(customized-scale)
-	$(generate) \
-	    'pathological' '31' $(customized-nodes) $(customized-scale) '--no-generate'
+$(bin-aln-path): $(require)
+	@$(install)
+
+$(bin-gen-path): $(require)
+	@$(install)
+
+
+# Generate timing data of data sets
+biological-scale := '\[1%1,1%2,1%4,1%8,1%16,1%32\]'
+customized-scale := '\[1%1,1%2,1%4,1%8,1%16,1%32,1%64\]'
+customized-nodes := '\[4,8,16,32,64,128,256\]'
+
+fungi: $(prerequisites)
+	@$(MAKE) --no-print-directory measure \
+		name='fungi'   cost='11' sizes='\[25,50,100,200,400,800,1553\]' scale=$(biological-scale)
+
+metazoa: $(prerequisites)
+	@$(MAKE) --no-print-directory measure \
+		name='metazoa' cost='11' sizes='\[25,50,100,200,400,800,1766\]' scale=$(biological-scale)
+
+pathological: $(prerequisites)
+	@$(MAKE) --no-print-directory measure \
+		name='pathological' cost='12' sizes=$(customized-nodes) scale=$(customized-scale)
+
+	@$(MAKE) --no-print-directory measure \
+		name='pathological' cost='31' sizes=$(customized-nodes) scale=$(customized-scale) flags='--no-generate'
+
+measure:
+	./bin/generate-timings $(flags) \
+		--data data-sets/$(name).afasta \
+		--tree data-sets/$(name).tree \
+		--tcm  data-sets/tcm-$(cost).tcm \
+		--output  $(name)-$(cost) \
+		--leaves  $(sizes) \
+		--lengths $(scale)
+
+	python3 ./bin/plot-figure.py \
+		"replicate-results/csv/$(name)-$(cost).preorder.csv" \
+		"replicate-results/img/$(name)-preorder.eps"
+	python3 ./bin/plot-figure.py \
+		"replicate-results/csv/$(name)-$(cost).postorder.csv" \
+		"replicate-results/img/$(name)-postorder.eps"
+
+# Clean up after replicating results
+clean:
+	@rm -fr replicate-results
+	@rm -fr dist-newstyle
+	@rm -f  $(bin-aln-path)
+	@rm -f  $(bin-gen-path)

replicate-results.sh

@@ -1,6 +1,5 @@
 #!/bin/bash
 
-
 # # # # #
 #
 # Script to reproduce the results of the paper from a clean environment.
@@ -9,59 +8,4 @@
 #
 # # # # #
 
-
-# Install dependencies required to replicate results:
-#   * Load Python3 dependencies
-which python3         &> /dev/null || apt-get install python3     --yes
-which pip3            &> /dev/null || apt-get install python3-pip --yes
-pip3  show matplotlib &> /dev/null || pip3 install --upgrade matplotlib
-
-#   * Load R dependencies
-which R &> /dev/null || sudo apt install r-base
-
-#   * Load Haskell dependencies
-which ghcup &> /dev/null || curl --proto '=https' --tlsv1.2 -sSf https://get-ghcup.haskell.org | sh
-
-
-# Build binaries
-ghcup run \
-    --ghc 9.2.1 \
-    --cabal 3.6.2.0 \
-    -- cabal update && \
-       cabal install \
-        --installdir=./bin \
-        --install-method=copy 
-
-
-# Ensure output directories exist to place results data
-mkdir -p replicate-results/csv
-mkdir -p replicate-results/data
-mkdir -p replicate-results/img
-mkdir -p replicate-results/taxa
-mkdir -p replicate-results/tree
-
-
-# Run data sets to produce the results data
-function generate()
-{
-    ./bin/generate-timings $5 \
-        --data "data-sets/$1.afasta" \
-        --tree "data-sets/$1.tree" \
-        --tcm  "data-sets/tcm-${2}.tcm" \
-        --output "$1-$2" \
-        -n "$3" \
-        -k "$4"
-
-    python3 plot-figure.py \
-        "replicate-results/csv/${1}-${2}.preorder.csv" \
-        "replicate-results/img/${1}-preorder.eps"
-
-    python3 plot-figure.py \
-        "replicate-results/csv/${1}-${2}.postorder.csv" \
-        "replicate-results/img/${1}-postorder.eps"
-}
-
-generate 'fungi'        '11' '[25,50,100,200,400,800,1553]' '[1%1,1%2,1%4,1%8,1%16,1%32]'
-generate 'metazoa'      '11' '[25,50,100,200,400,800,1766]' '[1%1,1%2,1%4,1%8,1%16,1%32]'
-generate 'pathological' '12' '[4,8,16,32,64,128,256]'       '[1%1,1%2,1%4,1%8,1%16,1%32,1%64]'
-generate 'pathological' '31' '[4,8,16,32,64,128,256]'       '[1%1,1%2,1%4,1%8,1%16,1%32,1%64]' '--no-generate'
+make all