4 releases
0.1.3 | Aug 15, 2020 |
---|---|
0.1.2 | Aug 12, 2020 |
0.1.1 | Aug 11, 2020 |
0.1.0 | Aug 11, 2020 |
#39 in #proposal
145KB
1K
SLoC
smlr
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.
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:
input: samples/file03.txt | uniq -i -c file03.txt |
smlr -cbf file03.txt |
---|---|---|
|
|
|
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.
Dependencies
~1.7–2.7MB
~50K SLoC