Skip to content

oscaro/tools-io

Repository files navigation

oscaro-tools-io

Oscaro’s generic I/O tools collection.

Clojars Project

cljdoc badge

Usage

io

join-path

Join multiple parts of a path, like os.path.join in Python.

(join-path "foo" "bar") ; => "foo/bar"
(join-path "foo/" "bar") ; => "foo/bar"
(join-path "gs://mybucket" "bar") ; => "gs://mybucket/bar"

basename

(basename "/var/log/mysql/") ; => "mysql"
(basename "http://www.google.com/index.html") ; => "index.html"

parent

(parent "/var/log/mysql/") ; => "/var/log"
(parent "http://www.google.com/index.html") ; => "http://www.google.com"

splitext

(splitext "http://www.google.com/index.html") ; => ["http://www.google.com/index" "html"]
(splitext "archive.tar.gz") ; => ["archive.tar" "gz"]

read-text-file

return a lazy seq of string from a [protocol://]jsons[.gz] file. warning: the seq must be entirely consumed before the file is closed.

arguments:

  • filename: string

returns: an lazy seq of string

read-jsons-file

return a lazy seq of parsed json objects from a [protocol://]jsons[.gz] file. warning: the seq must be entirely consumed before the file is closed.

arguments:

  • filename: string

returns: an lazy seq of parsed objects

example

(doall (map println (read-jsons-file "sample.jsons.gz")))

read-edns-file

return a lazy seq of parsed edn objects from a [protocol://]edn[.gz] file. warning: the seq must be entirely consumed before the file is closed.

arguments:

  • filename: string

returns: an lazy seq of parsed objects

read-csv-file

return a lazy seq of parsed csv row as vector from a [protocol://]file.csv[.gz] file. see http://clojure.github.io/data.csv/ for options. warning: the seq must be entirely consumed before the file is closed.

arguments:

  • filename: string
  • args: options for data.csv {:separator (default \,) :quote (default \")}

returns: an lazy seq of parsed objects

read-jsons-files

return a lazy seq of parsed json objects from [protocol://]jsons[.gz] files. warning: the seq must be entirely consumed before every files are closed.

arguments:

  • [filenames]

returns: an lazy seq of parsed objects

example

(doall (map println (read-jsons-files ["part1.jsons.gz" "part1.jsons.gz"])))

read-edns-files

return a lazy seq of parsed json objects from [protocol://]jsons[.gz] files. warning: the seq must be entirely consumed before every files are closed.

arguments:

  • [filenames]

returns: an lazy seq of parsed objects

list-files

return a seq of filenames beginning with provided path.

arguments:

  • path
  • [options]

returns: seq of string

examples

(doall (map println (list-files "gs://my-bucket/dir/20160902/animals")))
;-> output:
;gs://my-bucket/dir/20160902/animals-aaaaaaaaaa.jsons.gz
;gs://my-bucket/dir/20160902/animals-aaaaaaaaab.jsons.gz
;gs://my-bucket/dir/20160902/animals-aaaaaaaaac.jsons.gz
;gs://my-bucket/dir/20160902/animals-aaaaaaaaad.jsons.gz
;gs://my-bucket/dir/20160902/animals-aaaaaaaaae.jsons.gz


(doall (map println (list-files "/home/alice/dir/20160902/animals")))
;-> output:
;/home/alice/dir/20160902/animals-aaaaaaaaaa.jsons.gz
;/home/alice/dir/20160902/animals-aaaaaaaaab.jsons.gz
;/home/alice/dir/20160902/animals-aaaaaaaaac.jsons.gz
;/home/alice/dir/20160902/animals-aaaaaaaaad.jsons.gz
;/home/alice/dir/20160902/animals-aaaaaaaaae.jsons.gz

list-dirs

return a seq of directory under the path directory.

arguments:

  • path
  • [options]

returns: seq of string

examples

(doall (map println (list-dirs "gs://my-bucket/dir/")))
;gs://my-bucket/dir/20160902/
;gs://my-bucket/dir/20160902/
;gs://my-bucket/dir/20160902/
;gs://my-bucket/dir/20160902/
;gs://my-bucket/dir/20160902/

load-config-file

read and parse a configuration file. edn, clj, json, js, yaml, yml supported.

note: if filename is a string, its searched first in resources, then locally

arguments:

  • filename (string or io/resource or io/file)

returns: an object

copy

Copy file from source to destination.

arguments

  • from
  • from-opts
  • to
  • to-opts
  • [copy-opts]: by defaults copy-opts = {buffer-size 1024}

example

Text file:

(copy
  "/tmp/windows-file.csv"
  {:encoding "windows-1252"}
  "gs://my-bucket/dir/uf8-file.csv"
  {:encoding "UTF-8" :mime-type "text/csv"}
  {:buffer-size 2048})

For binary file, you must use an 8-bits encoding:

(def byte-encoding "ISO-8859-1")
(copy
  "/tmp/workbook.xlsx"
  {:encoding byte-encoding}
  "gs://my-bucket/dir/uf8-file.csv"
  {:encoding byte-encoding :mime-type "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"}
  {:buffer-size 2048})

rm-rf

Recursively remove a directory.

(rm-rf "/path/to/my/directory")

with-tempfile

Create a temporary file and remove it at the end of the body.

(with-tempfile [filename]
  (println "There's a file called" filename "."))
(println "The file is now gone.")

with-tempdir

Create a temporary directory and remove it at the end of the body.

(with-tempdir [dirname]
  (println "There's a directory called" dirname "."))
(println "The directory is now gone.")

slurp

spit

exists?

Test if a file exists.

(exists? "https://oscaro.com") ;=> true
(exists? "local-file-that-do-not-exists") ;=> false

core

file-reader

return a file as map like {:stream clojure.java.io/reader}, with support for [protocol://]file and file.gz. you need to call (close! file) when you done.

arguments:

  • filename
  • [options]: by default options = {encoding "UTF-8"}

returns: an map with a :stream key

About compression

By default, tools.io supports gzip, bzip2 and framed lz4 compression algorithms and can be extended by implementing a custom protocol (see sources).

It also supports the following formats if you provide the required dependencies.

  • xz with org.tukaani/xz provided
  • zstd with com.github.luben/zstd-jni provided

clj-kondo

An exported clj-kondo with hooks is provided. You can import it using

clj-kondo --lint "$(clojure -Spath)" --copy-configs --skip-lint

or, if using lein,

clj-kondo --lint "$(lein classpath)" --copy-configs --skip-lint

License

Copyright © 2016-2024 Oscaro.com

Distributed under the Eclipse Public License either version 1.0 or (at your option) any later version.