# `smlr`

![smlr](smlr.png)

## Install

        cargo install smlr

## Help

This is the documentation for `smlr v0.1.3`.
Calling `smlr --help` will show the available options:

    smlr 0.1.3
    Wilfried Kopp <chevdor@gmail.com>
    Find similar lines in a file or stdin.

    USAGE:
        smlr [FLAGS] [OPTIONS] [FILE]

    FLAGS:
            --case-sensitive    Weather the checks consider the case or not
        -c, --count             Show the number of occurences
            --debug             Debug mode, slower
        -f, --full              Set the score to the max
            --help              Prints help information
        -h, --headers           Show the headers
        -b, --ignore-spaces     Ignore spaces
        -i, --invert            Invert the result
        -l, --line-numbers      How many of the next lines are considered in the match. A larger value requires more
                                processing.
            --no-index          Prevent saving index to disk
            --strict            = Max distance = 0
        -V, --version           Prints version information
        -v, --verbose           Sets the level of verbosity

    OPTIONS:
            --algo <algo>
                Algo used to calculate the distance [default: Levenshtein]  [possible values: Levenshtein,
                DamerauLevenshtein]
        -d, --distance <distance>                        The max distance [default: 3]
            --persistence-folder <persistence-folder>
                Location where the persistence files will be stored if the mode is custom [default: temp]

            --persistence-mode <persistence-mode>
                Location where the persistence files will be stored. In 'beside' mode, the current folder will be used.
                [default: temp]  [possible values: temp, custom, beside]
        -s <scope>
                How many of the next lines are considered in the match. A larger value requires more processing. [default:
                10]

    ARGS:
        <FILE>    Sets the input file to use. Alternatively, you may pipe <stdin>.

The rustdoc can be found [here](https://chevdor.gitlab.io/smlr/smlr).

## Overview

`smlr` is a command line utility that helps you find **similar** entries in text data.
It is doe not aim at replacing tools such as `uniq` or `awk`.

The following data is available in the `samples` folder.
Let’s consider a few examples:

## Performance

A test run on my iMac (fusion drive, no SSD) shows 11MB processed under 30s.

       $ TARGET=/var/log/install.log; du -h $TARGET | head -n 1; time smlr -lcb $TARGET > /tmp/output.txt
       11M    /var/log/install.log
       Finished release [optimized] target(s) in 0.01s
       Running `target/release/smlr -lcb /var/log/install.log`

       real    0m19.603s
       user    0m18.310s
       sys     0m1.202s

### Example \#1

Say we want to find duplicates in the following data:

<table>
<caption>Table uniq vs smlr</caption>
<colgroup>
<col style="width: 33%" />
<col style="width: 33%" />
<col style="width: 33%" />
</colgroup>
<thead>
<tr class="header">
<th>input: samples/file03.txt</th>
<th><code>uniq -i -c file03.txt</code></th>
<th><code>smlr -cbf file03.txt</code></th>
</tr>
</thead>
<tbody>
<tr class="odd">
<td><pre><code>Pizza
Ice Cream
Waffle
PiZZa
Waffle
pizza
Pizza
Pizaz
Peanuts
PIZZA
Beef Jerky
Popcorn
pizza
pizza</code></pre></td>
<td><pre><code>   1 Pizza
   1 Ice Cream
   1 Waffle
   1 PiZZa
   1 Waffle
   1 pizza
   1 Pizza
   1 Peanuts
   1 PIZZA
   1 Beef Jerky
   1 Popcorn
   2 pizza</code></pre></td>
<td><pre><code>7       Pizza
1       Ice Cream
2       Waffle
1       Peanuts
1       Beef Jerky
1       Popcorn</code></pre></td>
</tr>
</tbody>
</table>

Table uniq vs smlr

To my surprise, `uniq` seems to fail at finding some of the case insensitive matches.

The result from `uniq` is a bit suprising as I expected to find 4x pizza and not 2 here and 2 there. This issue can be easily solved using the `sort` command.

In the following example, we process the input using `sort` to group all duplicates. We then use `sort` again to show the most duplicates first.

    $ cat file03.txt | sort | uniq -i -c | sort -r
       3 pizza
       3 PiZZa
       2 Waffle
       1 Popcorn
       1 Peanuts
       1 PIZZA
       1 Ice Cream
       1 Beef Jerky

Still `uniq` has issue with duplicates with different case despite using `-i` but `sort` can help further.

    $ cat file03.txt | sort -f | uniq -i -c | sort -r
       7 PIZZA
       2 Waffle
       1 Popcorn
       1 Peanuts
       1 Ice Cream
       1 Beef Jerky

### Example \#2

If we introduce some variance on some fields in our input, `uniq` will fail on finding the duplicates.

Here is the data we work on:

    2020-01-01  Pizza
    2020-01-01  Ice Cream
    2020-01-02  Waffle
    2020-01-03  PiZZa
    2020-03-05  Waffle
    2020-03-05  pizza
    2020-03-06  pizza
    2020-03-06  Peanuts
    2020-03-06  PIZZA
    2020-03-06  Beef Jerky
    2020-03-07  Popcorn
    2020-03-07  pizza
    2020-03-08  pizza

Here we are stuck with out previous `sort+uniq` method. None of the following really work as expected:
- `cat file04.txt | sort -f | uniq -i -c -f 1`
- `cat file04.txt | sort -f | uniq -i -c -s 10`

The reason for the failure is that our `sort` trick no longer works. We call `awk` to the rescue:

    $ cat file04.txt | awk '{print $2}' | sort -f | uniq -c -i | sort -r
       7 PIZZA
       2 Waffle
       1 Popcorn
       1 Peanuts
       1 Ice
       1 Beef

### Example \#3

Here is a nasty example where `uniq`, even with helps of some other friendly commands, won’t be able to be helpful:

    2020-01-01  Pizza
    2020-01-01  Ice Cream
    2020-01-01  Ice  Cream
    2020-01-02  Waffle
    2020-01-03  PiZZa
    2020-03-05  Waflle
    2020-03-05  pizza
    2020-03-06  piiza
    2020-03-06  Peanuts
    2020-03-06  PlZZA
    2020-03-06  Beef Jerky
    2020-03-07  POPCORN
    2020-03-07  P0PCORN
    2020-03-07  pizza
    2020-03-08  pizza

This example contains a few typos on purpose. Those are the typos typically hard to spot (depending on your font and concentration level!).

Let see how `smlr` can handle that.

    $ time smlr -cbf file05.txt | sort -r
    7       2020-01-01  Pizza
    2       2020-03-07  POPCORN
    2       2020-01-02  Waffle
    2       2020-01-01  Ice Cream
    1       2020-03-06  Peanuts
    1       2020-03-06  Beef Jerky

    real    0m0.005s
    user    0m0.002s
    sys     0m0.004s

While `smlr` can do more, it has a cost in CPU and memory. Beware when parsing huge files!

## Install from GIT

        cargo install --git https://gitlab.com/chevdor/smlr.git --tag v0.1.3

Installing the version from \`master\` is not recommended for production: do \*NOT\* omit the \`--tag v0.1.3\` in the previous command.

## License

    MIT License

    Copyright (c) 2019-2020 Wilfried Kopp - Chevdor

    Permission is hereby granted, free of charge, to any person obtaining a copy
    of this software and associated documentation files (the "Software"), to deal
    in the Software without restriction, including without limitation the rights
    to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
    copies of the Software, and to permit persons to whom the Software is
    furnished to do so, subject to the following conditions:

    The above copyright notice and this permission notice shall be included in all
    copies or substantial portions of the Software.

    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
    IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
    FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
    AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
    LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
    OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
    SOFTWARE.