Plugin: csvstorage

• infos = Information about the csvstorage plugin is in keys below
• infos/author = Thomas Waser thoma.nosp@m.s.wa.nosp@m.ser@l.nosp@m.ibel.nosp@m.ektra.nosp@m..org
• infos/licence = BSD
• infos/provides = storage/csv
• infos/needs =
• infos/recommends =
• infos/placements = getstorage setstorage
• infos/status = maintained unittest nodep libc configurable limited
• infos/description = parses csv files

This plugin allows you to read and write CSV files within Elektra. It aims to be compatible with RFC 4180. Rows and columns are written in Elektra arrays (#0, #1,..). Using configuration you can give columns a name.

Configuration

delimiter Tells the plugin what delimiter is used in the file. The default delimiter is , and will be used if delimiter is not set.

header Tells the plugin to use the first line as a header if it's set to "colname". The columns will get the corresponding names. Skip the first line if it's set to "skip" or treat the first line as a record if it's set to "record". If header is not set, or set to "record", the columns get named #0,#1,... (array key naming)

columns If this key is set the plugin will yield an error for every file that doesn't have exactly the amount of columns as specified in columns.

columns/names Sets the column names. Only usable in combination with the columns key. The number of subkeys must match the number of columns. Conflicts with usage of header.

columns/index specify which column should be used to index records instead of the record number.

/export/<column name> only export column column name. Configuration can be give multiple times for different columns to export.

Examples

First line should determine the headers:

kdb mount test.csv /csv csvstorage "delimiter=;,header=colname,columns=2,columns/names,columns/names/#0=col0Name,columns/names/#1=col1Name"

Usage

The example below shows how you can use this plugin to read and write CSV files.

`` <h1>Mount plugin to cascading namespace/tests/csv`

We use the column names from the first line of the

config file as key names

sudo kdb mount config.csv /tests/csv csvstorage "header=colname,columns/names/#0=col0Name,columns/names/#1=col1Name"

Add some data

printf 'band,album
' >> kdb file /tests/csv printf 'Converge,All We Love We Leave Behind
' >> kdb file /tests/csv printf 'mewithoutYou,Pale Horses
' >> kdb file /tests/csv printf 'Kate Tempest,Everybody Down
' >> kdb file /tests/csv

kdb ls /tests/csv #> user/tests/csv/#0 #> user/tests/csv/#0/album #> user/tests/csv/#0/band #> user/tests/csv/#1 #> user/tests/csv/#1/album #> user/tests/csv/#1/band #> user/tests/csv/#2 #> user/tests/csv/#2/album #> user/tests/csv/#2/band #> user/tests/csv/#3 #> user/tests/csv/#3/album #> user/tests/csv/#3/band

The first array element contains the column names

kdb get /tests/csv/#0/band #> band kdb get /tests/csv/#0/album #> album

Retrieve data from the last entry

kdb get /tests/csv/#3/album #> Everybody Down kdb get /tests/csv/#3/band #> Kate Tempest

Change an existing item

kdb set /tests/csv/#1/album 'You Fail Me'

Retrieve the new item

kdb get /tests/csv/#1/album #> You Fail Me

The plugin stores the index of the last column

in all of the parent keys.

kdb get user/tests/csv/#0 #> #1 kdb get user/tests/csv/#1 #> #1 kdb get user/tests/csv/#2 #> #1 kdb get user/tests/csv/#3 #> #1

The configuration file reflects the changes

kdb file /tests/csv | xargs cat #> album,band #> You Fail Me,Converge #> Pale Horses,mewithoutYou #> Everybody Down,Kate Tempest

Undo changes to the key database

kdb rm -r /tests/csv sudo kdb umount /tests/csv

# Directory Values

By default the `csvstorage` plugin saves the name of the last column in each parent key. If you want to store a different value, you can do
so using the [Directory Value](@ref src_plugins_directoryvalue_README_md) plugin.

Mount plugin together with directoryvalue to cascading namespace /tests/csv

kdb mount config.csv /tests/csv csvstorage directoryvalue

Add some data

printf 'Schindler’s List,1993,8.9
' >> kdb file /tests/csv printf 'Léon: The Professional,1994,8.5
' >> kdb file /tests/csv

Retrieve data

kdb ls /tests/csv #> user/tests/csv/#0 #> user/tests/csv/#0/#0 #> user/tests/csv/#0/#1 #> user/tests/csv/#0/#2 #> user/tests/csv/#1 #> user/tests/csv/#1/#0 #> user/tests/csv/#1/#1 #> user/tests/csv/#1/#2

kdb get user/tests/csv/#0/#0 #> Schindler’s List kdb get user/tests/csv/#1/#2 #> 8.5

The plugin stores the index of the last column in the parent keys

kdb get user/tests/csv/#0 #> #2 kdb get user/tests/csv/#1 #> #2

Since we use the Directory Value plugin we can also change the data in a parent key

kdb set user/tests/csv/#0 'Movie – Year – Rating' kdb set user/tests/csv/#1 'It’s a Me.'

Retrieve data stored in parent keys

kdb get user/tests/csv/#0 #> Movie – Year – Rating kdb get user/tests/csv/#1 #> It’s a Me.

Undo changes to the key database

kdb rm -r /tests/csv sudo kdb umount /tests/csv

# Column as index

kdb mount config.csv /tests/csv csvstorage "delimiter=;,header=colname,columns/index=IMDB"

printf 'IMDB;Title;Year
' >> kdb file /tests/csv printf 'tt0108052;Schindler´s List;1993
' >> kdb file /tests/csv printf 'tt0110413;Léon: The Professional;1994
' >> kdb file /tests/csv

kdb ls /tests/csv #> user/tests/csv/tt0108052 #> user/tests/csv/tt0108052/IMDB #> user/tests/csv/tt0108052/Title #> user/tests/csv/tt0108052/Year #> user/tests/csv/tt0110413 #> user/tests/csv/tt0110413/IMDB #> user/tests/csv/tt0110413/Title #> user/tests/csv/tt0110413/Year

kdb get /tests/csv/tt0108052/Title #> Schindler´s List

kdb rm -r /tests/csv sudo kdb umount /tests/csv

# Export filter

kdb mount config.csv /tests/csv csvstorage "delimiter=;,header=colname,columns/index=IMDB"

printf 'IMDB;Title;Year
' >> kdb file /tests/csv printf 'tt0108052;Schindler´s List;1993
' >> kdb file /tests/csv printf 'tt0110413;Léon: The Professional;1994
' >> kdb file /tests/csv

kdb export /tests/csv csvstorage -c "delimiter=;,header=colname,columns/index=IMDB,export=,export/IMDB=,export/Title=" #> IMDB;Title #> tt0108052;Schindler´s List #> tt0110413;Léon: The Professional

kdb export /tests/csv csvstorage -c "delimiter=;,header=colname,columns/index=IMDB,export=,export/IMDB=,export/Year=" #> IMDB;Year #> tt0108052;1993 #> tt0110413;1994

kdb rm -r /tests/csv sudo kdb umount /tests/csv

```

Limitations

• Does not work on file streams (e.g. kdb import without file)