Dependencies on Data and Code

Overview

Teaching: 15 min
Exercises: 5 min

Questions

• How can I write a Makefile to update things when my scripts have changed rather than my input files?

Objectives

• Output files are a product not only of input files but of the scripts or code that created the output files.

• Recognize and avoid false dependencies.

Our Makefile now looks like this (fetch with git checkout 04-dependencies):

# Generate summary table.
results.txt : isles.dat abyss.dat last.dat
	python testzipf.py $^ > $@

# Count words.
.PHONY : dats
dats : isles.dat abyss.dat last.dat

isles.dat : books/isles.txt
	python countwords.py $< $@

abyss.dat : books/abyss.txt
	python countwords.py $< $@

last.dat : books/last.txt
	python countwords.py $< $@

.PHONY : clean
clean :
	rm -f *.dat
	rm -f results.txt

Our data files are a product not only of our text files but the script, countwords.py, that processes the text files and creates the data files. A change to countwords.py (e.g. to add a new column of summary data or remove an existing one) results in changes to the .dat files it outputs. So, let’s pretend to edit countwords.py, using touch, and re-run Make:

$ make dats
$ touch countwords.py
$ make dats

Nothing happens! Though we’ve updated countwords.py our data files are not updated because our rules for creating .dat files don’t record any dependencies on countwords.py.

We need to add countwords.py as a dependency of each of our data files also:

isles.dat : books/isles.txt countwords.py
	python countwords.py $< $@

abyss.dat : books/abyss.txt countwords.py
	python countwords.py $< $@

last.dat : books/last.txt countwords.py
	python countwords.py $< $@

If we pretend to edit countwords.py and re-run Make,

$ touch countwords.py
$ make dats

then we get:

python countwords.py books/isles.txt isles.dat
python countwords.py books/abyss.txt abyss.dat
python countwords.py books/last.txt last.dat

Dry run

make can show the commands it will execute without actually running them if we pass the -n flag:

$ touch countwords.py
$ make -n dats

This gives the same output to the screen as without the -n flag, but the commands are not actually run. Using this ‘dry-run’ mode is a good way to check that you have set up your Makefile properly before actually running the commands in it.

The following figure shows the dependencies embodied within our Makefile, involved in building the results.txt target, after adding countwords.py and testzipf.py as dependencies to their respective target files (i.e. how the Makefile should look after completing the rest of the exercises in this episode).

Why Don’t the .txt Files Depend on countwords.py?

.txt files are input files and have no dependencies. To make these depend on countwords.py would introduce a false dependency.

Intuitively, we should also add countwords.py as dependency for results.txt, as the final table should be rebuilt as we remake the .dat files. However, it turns out we don’t have to! Let’s see what happens to results.txt when we update countwords.py:

$ touch countwords.py
$ make results.txt

then we get:

python countwords.py books/abyss.txt abyss.dat
python countwords.py books/isles.txt isles.dat
python countwords.py books/last.txt last.dat
python testzipf.py abyss.dat isles.dat last.dat > results.txt

The whole pipeline is triggered, even the creation of the results.txt file! To understand this, note that according to the dependency figure, results.txt depends on the .dat files. The update of countwords.py triggers an update of the *.dat files. Thus, make sees that the dependencies (the .dat files) are newer than the target file (results.txt) and thus it recreates results.txt. This is an example of the power of make: updating a subset of the files in the pipeline triggers rerunning the appropriate downstream steps.

Updating One Input File

What will happen if you now execute:

$ touch books/last.txt
$ make results.txt

1. only last.dat is recreated
2. all .dat files are recreated
3. only last.dat and results.txt are recreated
4. all .dat and results.txt are recreated

Solution

3. only last.dat and results.txt are recreated.

Follow the dependency tree to understand the answer(s).

testzipf.py as a Dependency of results.txt.

What would happen if you added testzipf.py as dependency of results.txt, and why?

Solution

If you change the rule for the results.txt file like this:

results.txt : isles.dat abyss.dat last.dat testzipf.py
        python testzipf.py $^ > $@

testzipf.py becomes a part of $^, thus the command becomes

python testzipf.py abyss.dat isles.dat last.dat testzipf.py > results.txt

This results in an error from testzipf.py as it tries to parse the script as if it were a .dat file. Try this by running:

$ make results.txt

You’ll get

python testzipf.py abyss.dat isles.dat last.dat testzipf.py > results.txt
Traceback (most recent call last):
  File "testzipf.py", line 19, in <module>
    counts = load_word_counts(input_file)
  File "path/to/testzipf.py", line 39, in load_word_counts
    counts.append((fields[0], int(fields[1]), float(fields[2])))
IndexError: list index out of range
make: *** [results.txt] Error 1

We still have to add the testzipf.py script as dependency to results.txt. Given the answer to the challenge above, we cannot use $^ in the rule. We can however move testzipf.py to be the first dependency and then use $< to refer to it. In order to refer to the .dat files, we can just use *.dat for now (we will cover a better solution later on).

results.txt : testzipf.py isles.dat abyss.dat last.dat
	python $< *.dat > $@

Key Points

• Make results depend on processing scripts as well as data files.

• Dependencies are transitive: if A depends on B and B depends on C, a change to C will indirectly trigger an update to A.