Skip to content

Latest commit

 

History

History
304 lines (257 loc) · 10.9 KB

overwrite.md

File metadata and controls

304 lines (257 loc) · 10.9 KB
Raw

Overwrite - Manage Overwrites Associated with a CSV File

The overwrite utility allows you to manage a list of "overwrites" associated with a given CSV input file. Each overwrite entry is a tuple consisting of row, column, and new value, along with optional timestamp and author metadata.

Usage

overwrite <file> <command> [arguments] [options]

Commands

List
list
List all overwrites in the overwrites file.

Add
add <cell> <value>
Add an overwrite entry.
Option for either row-col style cells, or Excel A1 style. Examples:

  1. Overwrite the first column of the first non-header row:

    overwrite mydata.csv add B2 "new value"

    or

    overwrite mydata.csv add 1-1 "new value"

  2. Change the header in the second column to "ID #":

    overwrite mydata.csv add B1 "ID #"

    or

    overwrite mydata.csv add 0-1 "ID #"

Remove
remove <cell>
Remove an overwrite entry. Similar to add for examples, without value.

Clear
clear Remove any or all overwrites.

Bulk Operations bulk-add <datafile>
or bulk-remove <datafile>
Bulk operations add or remove overwrite entries from a CSV or JSON file.

Note: JSON not currently supported.

Options

  • -h, --help
    Show this help message.

  • --old-value
    For add or remove, only proceed if the current overwrite value matches the given value at the given cell.

  • --force

    • For add, proceed even if an overwrite for the specified cell already exists.
    • For remove, exit without error even if no overwrite for the specified cell already exists.

  • --no-timestamp
    For add, do not save a timestamp when adding an overwrite.

  • --all
    For remove, remove all overwrites and delete the SQLite file.

  • --A1
    For list, display addresses in A1-notation.

Description

Overwrite data for a given input file /path/to/my-data.csv is stored in the "overwrites" table of /path/to/.zsv/data/my-data.csv/overwrite.sqlite3.

For bulk operations, the data file must be a CSV with "row", "col," and "value" columns and may optionally include "old value," "timestamp," and/or "author."

Examples

Example CSV file (mydata.csv):

Arabian Fertile Crescent South Caucasus
Kuwait Iraq Armenia
Oman Jordan Azerbaijan
Qatar Lebanon Georgia
Yemen Palestine
Bahrain Syria
Saudi Arabia Israel
United Arab Emirates

Bulk Overwrite operations

Example layout of a bulk file bulkdata.csv:

row col value timestamp
1 2 Saudi Arabia 1733283235
1 7 Kuwait 1733283102
2 3 Israel 1733282194
2 7 Jordan 1733258125
2 2 Palestine 1733285225
2 5 Iraq 1733284211
3 3 Georgia 1733284010
3 4 Azerbaijan 1733285510

bulk-add would add overwrites to the overwrite file, where bulk-remove would remove matching overwrites from the overwrite file.

overwrite mydata.csv bulk-add bulkdata.csv
select mydata.csv --apply-overwrites

Output:

Arabian Fertile Crescent South Caucasus
Saudi Arabia Palestine Armenia
Oman Israel Georgia
Qatar Lebanon Azerbaijan
Yemen Iraq
Bahrain Syria
Saudi Arabia Jordan
Kuwait

Now we can list the overwrites

overwrite mydata.csv list

Output:

row,column,value,timestamp,author
1,2,Saudi Arabia,1733283235,
1,7,Kuwait,1733283102,
2,2,Palestine,1733285225,
2,3,Israel,1733282194,
2,5,Iraq,1733284211,
2,7,Jordan,1733258125,
3,3,Georgia,1733284010,
3,4,Azerbaijan,1733285510,

Applying bulk-remove

overwrite mydata.csv bulk-remove bulkdata.csv
overwrite mydata.csv list

Output:

row,column,value,timestamp,author

Basic Overwrite operations

To add an overwrite entry that changes the value in cell B2:

To add a value:

overwrite mydata.csv add B2 "Syria"
select mydata.csv --apply-overwrites

Output:

Arabian Fertile Crescent South Caucasus
Kuwait Syria Armenia
Oman Jordan Azerbaijan
Qatar Lebanon Georgia
Yemen Palestine
Bahrain Syria
Saudi Arabia Israel
United Arab Emirates

To remove the added value:

overwrite mydata.csv remove B2
select mydata.csv --apply-overwrites

Output:

Arabian Fertile Crescent South Caucasus
Kuwait Iraq Armenia
Oman Jordan Azerbaijan
Qatar Lebanon Georgia
Yemen Palestine
Bahrain Syria
Saudi Arabia Israel
United Arab Emirates

To force add a value, even if there is already a value in that cell. This overwrites the original value in B2 with the new selected value.

overwrite mydata.csv add B2 "Lebanon" --force
select mydata.csv --apply-overwrites

Output:

Arabian Fertile Crescent South Caucasus
Kuwait Lebanon Armenia
Oman Jordan India
Qatar Lebanon Georgia
Yemen Palestine
Bahrain Syria
Saudi Arabia Israel
United Arab Emirates

To remove/add a value, depending on the old value at the cell position. This will only trigger if the existing value is "Azerbaijan":

overwrite mydata.csv add C3 "India" --old-value "Azerbaijan"
select mydata.csv --apply-overwrites

Output:

Arabian Fertile Crescent South Caucasus
Kuwait Iraq Armenia
Oman Jordan India
Qatar Lebanon Georgia
Yemen Palestine
Bahrain Syria
Saudi Arabia Israel
United Arab Emirates

Alternatively, if the old value was not the same as the value in the table.

overwrite mydata.csv add C3 "Pakistan" --old-value "Armenia"
select mydata.csv --apply-overwrites

The output would remain the same, as no values would be changed.

To remove all overwrites and delete the SQLite file:

overwrite mydata.csv remove --all
select mydata.csv --apply-overwrites

Output:

Arabian Fertile Crescent South Caucasus
Kuwait Iraq Armenia
Oman Jordan Azerbaijan
Qatar Lebanon Georgia
Yemen Palestine
Bahrain Syria
Saudi Arabia Israel
United Arab Emirates
The table now looks like the original table.

File Storage

Overwrite data is stored in a dedicated SQLite database for each input file. The SQL operations are optimized for performance, by limiting the number of operations. The SQLite file gets automatically created when a new overwrite is initialized, and is organized based on the input filename.

Usage Details

zsv overwrite -h

Usage:
  overwrite <file> <command> [arguments] [options]

Commands (where <cell> can be <row>-<col> or an Excel-style address):
  list                   : Display all saved overwrite entries
  add <cell> <value>     : Add an overwrite entry
                           Example 1: overwrite the first column of the first
                           non-header row
                             overwrite mydata.csv add B2 "new value"
                               - or -
                             overwrite mydata.csv add 1-1 "new value"
                           Example 2: change the header in the second column
                           to "ID #"
                             overwrite mydata.csv add B1 "new value"
                               - or -
                             overwrite mydata.csv add 0-1 "ID #"
  remove <cell>          : Remove an overwrite entry
  clear                  : Remove any / all overwrites
  bulk-add <datafile>    : Bulk add overwrite entries from a CSV or JSON file
  bulk-remove <datafile> : Bulk remove overwrite entries from a CSV or JSON file

Options:
  -h,--help              : Show this help message
  --old-value <value>    : For `add` or `remove`, only proceed if the old value
                           matches the given value
  --force.               : For `add`, proceed even if an overwrite for the specified
                           cell already exists
                           For `remove`, exit without error even if no overwrite for
                           the specified cell already exists
  --no-timestamp.        : For `add`, don't save timestamp when adding an overwrite
  --all                  : For `remove`, remove all overwrites and delete sqlite file
  --A1                   : For `list`, Display addresses in A1-notation

Description:
  The  `overwrite`  utility  allows  you to manage a list of "overwrites" associated
  with a given CSV input file. Each overwrite entry is a tuple consisting  of  row,
  column,  original  value, and new value, along with optional timestamp and author
  metadata.

  Overwrite data for a given input file `/path/to/my-data.csv` is stored in the "over‐
  writes"  table  of  `/path/to/.zsv/data/my-data.csv/overwrite.sqlite3`.

  For bulk operations, the data file must be a CSV with "row", "col" and "value" columns
  and may optionally include "old value", "timestamp" and/or "author"