diff options
| author | Martin Czygan <martin.czygan@gmail.com> | 2021-02-02 01:54:12 +0100 |
|---|---|---|
| committer | Martin Czygan <martin.czygan@gmail.com> | 2021-02-02 01:54:12 +0100 |
| commit | 2495bdadfbf84b204961aa7f0539efba3e93b4c5 (patch) | |
| tree | 1bd64171a0cc25974bba0ea062a0313babf218fc /fuzzycat | |
| parent | 3afe6607ccc8eb7c62a21c7899617ce394d51987 (diff) | |
| download | fuzzycat-2495bdadfbf84b204961aa7f0539efba3e93b4c5.tar.gz fuzzycat-2495bdadfbf84b204961aa7f0539efba3e93b4c5.zip | |
make fmt
Diffstat (limited to 'fuzzycat')
| -rw-r--r-- | fuzzycat/cluster.py | 12 |
1 files changed, 7 insertions, 5 deletions
diff --git a/fuzzycat/cluster.py b/fuzzycat/cluster.py index a53f5a1..d859507 100644 --- a/fuzzycat/cluster.py +++ b/fuzzycat/cluster.py @@ -371,7 +371,8 @@ class Cluster: # them here; maybe offer TSV and JSON output and extra flag # XXX: this needs to be compressed (e.g. with 2B records, we # fill up disk too quickly) - data = bytes("{}\t{}\t{}\n".format(id, key, line.replace("\t", " ").strip()), + data = bytes("{}\t{}\t{}\n".format(id, key, + line.replace("\t", " ").strip()), encoding="utf-8") writer.write(data) if self.compress: @@ -405,10 +406,11 @@ class Cluster: if fast: env["LC_ALL"] = "C" if self.compress: - output = shellout("zstdcat -T0 {input} | LC_ALL=C TMPDIR={tmpdir} sort {opts} | zstd -T0 -c9 > {output}", - input=filename, - tmpdir=self.tmpdir, - opts=opts) + output = shellout( + "zstdcat -T0 {input} | LC_ALL=C TMPDIR={tmpdir} sort {opts} | zstd -T0 -c9 > {output}", + input=filename, + tmpdir=self.tmpdir, + opts=opts) else: subprocess.run(["sort"] + opts.split() + [filename], stdout=tf, env=env, check=True) output = tf.name |