exercism/elixir/log-parser

Log Parser

Welcome to Log Parser on Exercism's Elixir Track. If you need help running the tests or submitting your code, check out HELP.md. If you get stuck on the exercise, check out HINTS.md, but try and solve it without using those first :)

Introduction

Regular Expressions

Regular expressions in Elixir follow the PCRE specification (Perl Compatible Regular Expressions), similarly to other popular languages like Java, JavaScript, or Ruby.

The Regex module offers functions for working with regular expressions. Some of the String module functions accept regular expressions as arguments as well.

This exercise assumes that you already know regular expression syntax, including character classes, quantifiers, groups, and captures.

If you need a refresh your regular expression knowledge, check out one of those sources: [Regular-Expressions.info](https://www.regular-expressions.info), [Rex Egg](https://www.rexegg.com/), [RegexOne](https://regexone.com/), [Regular Expressions 101](https://regex101.com/), [RegExr](https://regexr.com/).

Sigils

The most common way to create regular expressions is using the ~r sigil.

~r/test/

Note that all Elixir sigils support different kinds of delimiters, not only /.

Matching

The =~/2 operator can be used to perform a regex match that returns boolean result. Alternatively, there are also match?/2 functions in the Regex module as well as the String module.

"this is a test" =~ ~r/test/
# => true

String.match?("Alice has 7 apples", ~r/\d{2}/)
# => false

Capturing

If a simple boolean check is not enough, use the Regex.run/3 function to get a list of all captures (or nil if there was no match). The first element in the returned list is always a match for the whole regular expression, and the following elements are matched groups.

Regex.run(~r/(\d) apples/, "Alice has 7 apples")
# => ["7 apples", "7"]

Modifiers

The behavior of a regular expression can be modified by appending special flags. When using a sigil to create a regular expression, add the modifiers after the second delimiter.

Common modifiers are:

• i - makes the match case-insensitive.
• u - enables Unicode specific patterns like \p and causes character classes like \w, \s etc. to also match Unicode.

"this is a TEST" =~ ~r/test/i
# => true

Instructions

After a recent security review you have been asked to clean up the organization's archived log files.

1. Identify garbled log lines

You need some idea of how many log lines in your archive do not comply with current standards. You believe that a simple test reveals whether a log line is valid. To be considered valid a line should begin with one of the following strings:

• [DEBUG]
• [INFO]
• [WARNING]
• [ERROR]

Implement the valid_line?/1 function to return true if the log line is valid.

LogParser.valid_line?("[ERROR] Network Failure")
# => true

LogParser.valid_line?("Network Failure")
# => false

2. Split the log line

Shortly after starting the log parsing project, you realize that one application's logs aren't split into lines like the others. In this project, what should have been separate lines, is instead on a single line, connected by fancy arrows such as <---> or <*~*~>.

In fact, any string that has a first character of <, a last character of >, and any combination of the following characters ~, *, =, and - in between can be used as a separator in this project's logs.

Implement the split_line/1 function that takes a line and returns a list of strings.

LogParser.split_line("[INFO] Start.<*>[INFO] Processing...<~~~>[INFO] Success.")
# => ["[INFO] Start.", "[INFO] Processing...", "[INFO] Success."]

3. Remove artifacts from log

You have found that some upstream processing of the logs has been scattering the text "end-of-line" followed by a line number (without an intervening space) throughout the logs.

Implement the remove_artifacts/1 function to take a string and remove all occurrence end-of-line text (case-insensitive) and return a clean log line.

Lines not containing end-of-line text should be returned unmodified.

Just remove the end of line string, there's no need to adjust the whitespaces.

LogParser.remove_artifacts("[WARNING] end-of-line23033 Network Failure end-of-line27")
# => "[WARNING]  Network Failure "

4. Tag lines with user names

You have noticed that some of the log lines include sentences that refer to users. These sentences always contain the string "User", followed by one or more whitespace characters, and then a user name. You decide to tag such lines.

Implement a function tag_with_user_name/1 that processes log lines:

• Lines that do not contain the string "User" remain unchanged.
• For lines that contain the string "User", prefix the line with [USER] followed by the user name.

LogParser.tag_with_user_name("[INFO] User Alice created a new project")
# => "[USER] Alice [INFO] User Alice created a new project"

You can assume that:

• Each occurrence of the string "User" is followed by one or more whitespace character and the user name.
• There is at most one occurrence of the string "User" on each line.
• User names are non-empty strings that do not contain whitespace.

Source

Created by

• @angelikatyborska