Skip to content

Latest commit

 

History

History
388 lines (380 loc) · 10.1 KB

oldtable.md

File metadata and controls

388 lines (380 loc) · 10.1 KB
Raw

oldtable

Purpose

Format standard input in the form of a table. Various styles of input and output are supported which can be mixed together.

This script has been replaced table which is much more flexible and handles more combinations of reading and writing formats. I already had this documentation page for the old script and thought I would just keep it around. I'm a bit of a packrat!

Syntax

usage: oldtable [-h] [-c | -s SEPARATOR | -f] [-H | -m | -b | -r | -j]
             [--headings] [--simple-headings] [--file FILENAME] [-v]

Options and arguments

Option Description Default
-c or --csv Expect CSV as input There is no default input style
-s SEPARATOR or --separator SEPARATOR Use a regular expression to separate input columns There is no default input style
-f or --fixed Expect fixed columns as input There is no default input style
-h or --html Print output as HTML There is no default output style
-m or --markdown Print output as markdown. Popular in Git, Slack There is no default output style
-b or --bbcode Print output as BBCode. Popular with message board systems There is no default output style
-j or --json Print output as JSON There is no default output style
-r or --rotate Print output in a rotated style There is no default output style
--headings Expect the first row of input to be headings Headings are not expected
--simple-headings Parse headings row very simply, not trying to retain blanks Headings are parsed in a way to try to preserve blanks embedded inside a heading string
--file FILENAME Specify a file from which to read By default the script reads from stdin if --file is not specified
-v or --verbose Enable debugging Debugging is not enabled

Example

Input: regular expression / Output: JSON

$ df | head -5 | oldtable -s \\s+ -j
[
  [
    "Filesystem",
    "1K-blocks",
    "Used",
    "Available",
    "Use%",
    "Mounted_on"
  ],
  [
    "udev",
    "8133756",
    "0",
    "8133756",
    "0%",
    "/dev"
  ],
  [
    "tmpfs",
    "1631580",
    "2132",
    "1629448",
    "1%",
    "/run"
  ],
  [
    "/dev/sdb2",
    "236102400",
    "31740828",
    "192345140",
    "15%",
    "/"
  ],
  [
    "tmpfs",
    "8157880",
    "34564",
    "8123316",
    "1%",
    "/dev/shm"
  ]
]

Note that this is a list of lists because the headings row were treated as just another row of input.

Input: regular expression with headings / Output: JSON

$ df | head -5 | oldtable -s \\s+ -j  --headings
[
  {
    "1K-blocks": "8133756",
    "Available": "8133756",
    "Filesystem": "udev",
    "Mounted_on": "/dev",
    "Use%": "0%",
    "Used": "0"
  },
  {
    "1K-blocks": "1631580",
    "Available": "1629448",
    "Filesystem": "tmpfs",
    "Mounted_on": "/run",
    "Use%": "1%",
    "Used": "2132"
  },
  {
    "1K-blocks": "236102400",
    "Available": "192345136",
    "Filesystem": "/dev/sdb2",
    "Mounted_on": "/",
    "Use%": "15%",
    "Used": "31740832"
  },
  {
    "1K-blocks": "8157880",
    "Available": "8123452",
    "Filesystem": "tmpfs",
    "Mounted_on": "/dev/shm",
    "Use%": "1%",
    "Used": "34428"
  }
]
$

By using the --headings option, we now get a list of dictionaries and the heading for each column is used for the keys of the dictionaries.

Input: regular expression with headings / Output: HTML

$ df | head -5 | oldtable -s \\s+ --html  --headings
<table border='1'>
<tbody>
<tr>
<th>Filesystem</th>
<th>1K-blocks</th>
<th>Used</th>
<th>Available</th>
<th>Use%</th>
<th>Mounted_on</th>
</tr>
<tr>
<td>udev</td>
<td>8133756</td>
<td>0</td>
<td>8133756</td>
<td>0%</td>
<td>/dev</td>
</tr>
<tr>
<td>tmpfs</td>
<td>1631580</td>
<td>2132</td>
<td>1629448</td>
<td>1%</td>
<td>/run</td>
</tr>
<tr>
<td>/dev/sdb2</td>
<td>236102400</td>
<td>31740804</td>
<td>192345164</td>
<td>15%</td>
<td>/</td>
</tr>
<tr>
<td>tmpfs</td>
<td>8157880</td>
<td>34536</td>
<td>8123344</td>
<td>1%</td>
<td>/dev/shm</td>
</tr>
</tbody>
</table>

Note that the <th> tags are used for the headings. The output isn't especially pretty because it's meant to be rendered in a browser window but it should look good.

Input: regular expression with headings / Output: BBCode

$ df | head -5 | oldtable -s \\s+ --b  --headings
[table]
[tbody]
[tr]
[th]Filesystem[/th]
[th]1K-blocks[/th]
[th]Used[/th]
[th]Available[/th]
[th]Use%[/th]
[th]Mounted_on[/th]
[/tr]
[tr]
[td]udev[/td]
[td]8133756[/td]
[td]0[/td]
[td]8133756[/td]
[td]0%[/td]
[td]/dev[/td]
[/tr]
[tr]
[td]tmpfs[/td]
[td]1631580[/td]
[td]2132[/td]
[td]1629448[/td]
[td]1%[/td]
[td]/run[/td]
[/tr]
[tr]
[td]/dev/sdb2[/td]
[td]236102400[/td]
[td]31738668[/td]
[td]192347300[/td]
[td]15%[/td]
[td]/[/td]
[/tr]
[tr]
[td]tmpfs[/td]
[td]8157880[/td]
[td]34432[/td]
[td]8123448[/td]
[td]1%[/td]
[td]/dev/shm[/td]
[/tr]
[/tbody]
[/table]
$

This output looks kind of like HTML but there are some differences in the tags. Internally, a shared piece of code is used to generate both output styles.

Input: regular expression with headings / Output: rotated

$ df | head -5 | oldtable -s \\s+ -r  --headings
000000 00 Filesystem                       udev
000000 01 1K-blocks                        8133756
000000 02 Used                             0
000000 03 Available                        8133756
000000 04 Use%                             0%
000000 05 Mounted_on                       /dev
000001 00 Filesystem                       tmpfs
000001 01 1K-blocks                        1631580
000001 02 Used                             2132
000001 03 Available                        1629448
000001 04 Use%                             1%
000001 05 Mounted_on                       /run
000002 00 Filesystem                       /dev/sdb2
000002 01 1K-blocks                        236102400
000002 02 Used                             31746856
000002 03 Available                        192339112
000002 04 Use%                             15%
000002 05 Mounted_on                       /
000003 00 Filesystem                       tmpfs
000003 01 1K-blocks                        8157880
000003 02 Used                             35592
000003 03 Available                        8122288
000003 04 Use%                             1%
000003 05 Mounted_on                       /dev/shm

My rotated style is a little hard to describe but it basically prints every column of every row on a separate line, indicating the row and column numbers. Since headings were specified, the heading also appears along on every line.

Input: regular expression without headings / Output: rotated

$ df | head -5 | oldtable -s \\s+ -r
000000 00 Filesystem
000000 01 1K-blocks
000000 02 Used
000000 03 Available
000000 04 Use%
000000 05 Mounted_on
000001 00 udev
000001 01 8133756
000001 02 0
000001 03 8133756
000001 04 0%
000001 05 /dev
000002 00 tmpfs
000002 01 1631580
000002 02 2132
000002 03 1629448
000002 04 1%
000002 05 /run
000003 00 /dev/sdb2
000003 01 236102400
000003 02 31746880
000003 03 192339088
000003 04 15%
000003 05 /
000004 00 tmpfs
000004 01 8157880
000004 02 34432
000004 03 8123448
000004 04 1%
000004 05 /dev/shm
$

Without the --headings option, the rotated output is a little different but not much.

Input: fixed width / Output: JSON

$ ls -l | drop 1 | oldtable --simple -f -j | head -40
[
  [
    "-rwxr-xr-x",
    "1",
    "mrbruno",
    "mrbruno",
    "1440",
    "Jan",
    "26",
    "07:15",
    "abspath"
  ],
  [
    "-rwxr-xr-x",
    "1",
    "mrbruno",
    "mrbruno",
    "2771",
    "Mar",
    "29",
    "18:04",
    "acp"
  ],
  [
    "-rwxr-xr-x",
    "1",
    "mrbruno",
    "mrbruno",
    "999",
    "Aug",
    "11",
    "2018",
    "addcrs"
  ],

While coming up with this example, I actually had to add the --simple-headings option to treat the ls output a little differently because the headings were so close together. I had code to try to retain embedded blanks in heading strings and the --simple-headings option disables that code.

Note this example makes use of my drop tool.

Input: CSV / Output: JSON

$ head -5 /home/mrbruno/repos/iyt/csv/20190407_102324_846368.csv
Category,Game,GameID,User,UserID,WinLoss,Moves,Color,Date,Time
Regular,Anti-Backgammon,15300064812762,Donna D,15200000663034,Win,189,white,18/12/16,16:58:00
Regular,Anti-Backgammon,15300065319531,Danny Bad Boy,15200000782400,Win,31,black,18/08/28,08:53:00
Regular,Anti-Backgammon,15300064893256,supermanwuvsme,15200003044703,Win,290,black,18/07/05,22:28:00
Regular,Anti-Backgammon,15300064965692,pandagirl,15200003408150,Win,11,white,18/05/20,17:58:00
$ head -5 /home/mrbruno/repos/iyt/csv/20190407_102324_846368.csv | oldtable --csv -j --headings
[
  {
    "Category": "Regular",
    "Color": "white",
    "Date": "18/12/16",
    "Game": "Anti-Backgammon",
    "GameID": "15300064812762",
    "Moves": "189",
    "Time": "16:58:00",
    "User": "Donna D",
    "UserID": "15200000663034",
    "WinLoss": "Win"
  },
  {
    "Category": "Regular",
    "Color": "black",
    "Date": "18/08/28",
    "Game": "Anti-Backgammon",
    "GameID": "15300065319531",
    "Moves": "31",
    "Time": "08:53:00",
    "User": "Danny Bad Boy",
    "UserID": "15200000782400",
    "WinLoss": "Win"
  },
  {
    "Category": "Regular",
    "Color": "black",
    "Date": "18/07/05",
    "Game": "Anti-Backgammon",
    "GameID": "15300064893256",
    "Moves": "290",
    "Time": "22:28:00",
    "User": "supermanwuvsme",
    "UserID": "15200003044703",
    "WinLoss": "Win"
  },
  {
    "Category": "Regular",
    "Color": "white",
    "Date": "18/05/20",
    "Game": "Anti-Backgammon",
    "GameID": "15300064965692",
    "Moves": "11",
    "Time": "17:58:00",
    "User": "pandagirl",
    "UserID": "15200003408150",
    "WinLoss": "Win"
  }
]
$

This input file is a list of games I've played.

Notes

  • There was no --file option until I had to debug an issue with the PyCharm debugger and had no way to feed input through a pipeline. Otherwise, there wouldn't be such an option!