def test_tempfile_not_run_on_prune(self):
a = ppg.TempFileGeneratingJob("A", lambda: write("A", "A"))
b = ppg.FileGeneratingJob("B", lambda: write("B", "B" + read("A")))
b.depends_on(a)
b.prune()
ppg.run_pipegraph()
assert not Path("B").exists()
assert not Path("A").exists()
def test_tempfile_still_run_if_needed_for_other(self):
a = ppg.TempFileGeneratingJob("A", lambda: write("A", "A"))
b = ppg.FileGeneratingJob("B", lambda: write("B", "B" + read("A")))
c = ppg.FileGeneratingJob("C", lambda: write("C", "C" + read("A")))
b.depends_on(a)
c.depends_on(a)
b.prune()
ppg.run_pipegraph()
assert not Path("B").exists()
assert Path("C").exists()
assert Path("C").read_text() == "CA"
assert not Path("A").exists()
def job_download(info):
"""Download the package defined in info"""
target_fn = f'/anysnake/bioconductor_download/{info["repo"]}/{info["name"]}_{info["version"]}.tar.gz'
def download():
p = Path(target_fn)
p.parent.mkdir(exist_ok=True, parents=False)
r = requests.get(info["url"], stream=True)
if r.status_code != 200:
raise ValueError("Error return on %s %s " % (info["url"], r.status_code))
with open(str(target_fn) + "_temp", "wb") as op:
for block in r.iter_content(1024 * 1024):
op.write(block)
shutil.move(str(target_fn) + "_temp", str(target_fn))
job = ppg.TempFileGeneratingJob(target_fn, download)
job.ignore_code_changes()
return job
def test_accepts(self):
import pathlib
write("aaa", "hello")
write("bbb", "hello")
write("ccc", "hello")
a = ppg.FileTimeInvariant(pathlib.Path("aaa"))
a1 = ppg.MultiFileInvariant([pathlib.Path("bbb"), "ccc"])
b = ppg.FileGeneratingJob(
pathlib.Path("b"),
lambda of: write(of, "bb" + read("aaa") + read("bbb") + read("ccc")),
)
b.depends_on(a)
b.depends_on(a1)
dd = Dummy()
def mf():
write("c", "cc" + read("g"))
write("d", "dd" + read("h") + dd.attr)
write("e", "ee" + read("i") + read("j"))
c = ppg.MultiFileGeneratingJob([pathlib.Path("c"), "d", pathlib.Path("e")], mf)
c.depends_on(b)
d = ppg.FunctionInvariant(pathlib.Path("f"), lambda x: x + 1)
c.depends_on(d)
e = ppg.ParameterInvariant(pathlib.Path("c"), "hello")
c.depends_on(e)
f = ppg.TempFileGeneratingJob(pathlib.Path("g"), lambda: write("g", "gg"))
c.depends_on(f)
def tmf():
write("h", "hh")
write("i", "ii")
g = ppg.MultiTempFileGeneratingJob([pathlib.Path("h"), "i"], tmf)
c.depends_on(g)
def tpf():
write("j", "jjjj")
write("k", "kkkk")
h = ppg.TempFilePlusGeneratingJob(pathlib.Path("j"), pathlib.Path("k"), tpf)
c.depends_on(h)
i = ppg.CachedDataLoadingJob(
pathlib.Path("l"), lambda: write("l", "llll"), lambda res: res
)
c.depends_on(i)
m = ppg.CachedAttributeLoadingJob(pathlib.Path("m"), dd, "attr", lambda: "55")
c.depends_on(m)
ppg.run_pipegraph()
assert read("aaa") == "hello"
assert read("b") == "bbhellohellohello"
assert read("c") == "ccgg"
assert read("d") == "ddhh55"
assert read("e") == "eeiijjjj"
assert not (os.path.exists("g"))
assert not (os.path.exists("h"))
assert not (os.path.exists("i"))
assert not (os.path.exists("j"))
assert read("k") == "kkkk"
def prepare_input(self):
# input_strategy returns a list of
# paired fastq files
# ie. [('A_R1_.fastq1', 'A_R2.fastq', ...), ...]
input_pairs = self.input_strategy()
any_r2 = any([len(x) > 1 for x in input_pairs])
# Single end - works from flat list
if self.pairing == "single":
if any_r2:
raise PairingError(
f"{self.name}: paired end lane defined as single end - you need to change the pairing parameter"
)
input_filenames = [str(f[0]) for f in input_pairs]
elif self.pairing == "paired_as_single":
input_filenames = [str(f) for fl in input_pairs for f in fl]
elif self.pairing == "only_first":
input_filenames = [str(f[0]) for f in input_pairs]
elif self.pairing == "only_second":
input_filenames = [str(f[1]) for f in input_pairs]
elif self.pairing == "paired":
if not any_r2:
raise PairingError(
f"Paired end lane, but no R2 reads found. Found files: {input_pairs}"
)
input_filenames = [(str(f[0]), str(f[1])) for f in input_pairs
] # throwing away all later...
else:
raise PairingError("unknown pairing") # pragma: no cover
if self.pairing == "paired":
flat_input_filenames = [f for fl in input_pairs for f in fl]
else:
flat_input_filenames = input_filenames
if hasattr(self.input_strategy, "dependencies"):
deps = self.input_strategy.dependencies
else:
deps = [ppg.FileChecksumInvariant(f) for f in flat_input_filenames]
output_filenames = self.get_aligner_input_filenames()
if self.pairing == "paired":
if hasattr(self.fastq_processor, "generate_aligner_input_paired"):
def prep_aligner_input():
import shutil
self.fastq_processor.generate_aligner_input_paired(
str(output_filenames[0]) + ".temp",
str(output_filenames[1]) + ".temp",
input_filenames,
self.reverse_reads,
)
shutil.move(
str(output_filenames[0]) + ".temp",
output_filenames[0])
shutil.move(
str(output_filenames[1]) + ".temp",
output_filenames[1])
job = ppg.MultiTempFileGeneratingJob(output_filenames,
prep_aligner_input)
job.depends_on(
self.fastq_processor.get_dependencies(
[str(x) for x in output_filenames]))
else:
def prep_aligner_input_r1():
import shutil
self.fastq_processor.generate_aligner_input(
str(output_filenames[0]) + ".temp",
[x[0] for x in input_filenames],
self.reverse_reads,
)
shutil.move(
str(output_filenames[0]) + ".temp",
output_filenames[0])
def prep_aligner_input_r2():
import shutil
self.fastq_processor.generate_aligner_input(
str(output_filenames[1]) + ".temp",
[x[1] for x in input_filenames],
self.reverse_reads,
)
shutil.move(
str(output_filenames[1]) + ".temp",
output_filenames[1])
jobR1 = ppg.TempFileGeneratingJob(output_filenames[0],
prep_aligner_input_r1)
jobR2 = ppg.TempFileGeneratingJob(output_filenames[1],
prep_aligner_input_r2)
jobR1.depends_on(
self.fastq_processor.get_dependencies(
str(output_filenames[0])))
jobR2.depends_on(
self.fastq_processor.get_dependencies(
str(output_filenames[1])))
job = ppg.JobList([jobR1, jobR2])
# needed by downstream code.
job.filenames = [output_filenames[0], output_filenames[1]]
else:
def prep_aligner_input(output_filename):
import shutil
self.fastq_processor.generate_aligner_input(
str(output_filename) + ".temp", input_filenames,
self.reverse_reads)
shutil.move(str(output_filename) + ".temp", output_filename)
job = ppg.TempFileGeneratingJob(output_filenames[0],
prep_aligner_input)
job.depends_on(
self.fastq_processor.get_dependencies(str(
output_filenames[0])))
job.depends_on(
deps,
ppg.ParameterInvariant(
self.name + "input_files",
tuple(sorted(input_filenames)) +
(self.reverse_reads, self.fastq_processor.__class__.__name__),
),
)
return job
