Skip to content

Latest commit

 

History

History
74 lines (55 loc) · 3.77 KB

regex.md

File metadata and controls

74 lines (55 loc) · 3.77 KB
Raw

Regex

Thought it may be a good idea to leave notes to myself and any other developers looking at this, to help understand what the regex is doing.

generate.js

genJSON()

@param

const re = new RegExp('\\*\\s+@param\\s+\\{(?<location>[a-zA-Z]+)\\|(?<type>[a-zA-Z]+)\\}\\s+\\[(?<name>\\w+)=(?<value>\\S+)\\]\\s+-\\s+(?<desc>.*)');

Once match is used on an applicable string, it will return named groups that we are able to assign exactly.

match.groups.location;
match.groups.type;
match.groups.name;
match.groups.value;
match.groups.desc;

Heres the same Regex with any JavaScript escaping and object initialization removed:

/\*\s+@param\s+\{(?<location>[a-zA-Z]+)\|(?<type>[a-zA-Z]+)\}\s+\[(?<name>\w+)=(?<value>\S+)\]\s+-\s+(?<desc>.*)

A Quick Summary of what this is doing: (Please note in the future its intended that each value is optional. Currently that is not the case.)

  • \*\s+ - Matches '*' and any whitespace character between one and unlimited times.
  • @param\s+ - Then matches '@param' exactly followed by any whitespace character between one and unlimited times.
  • \{ - Matches '{' escaped within Regex.
  • (?<location>...) - Sets up a Named Capture Group of 'location'
  • [a-zA-Z]+ Matches within 'location' characters between the index of a-z and A-Z one to unlimited times.
  • \| - Match '|' exactly.
  • (?<type>[a-zA-Z]+) - Again creates a Named Capture Group of 'type' where it matches the index a-z and A-Z one to unlimited times.
  • \}\s+\[ - Matches '}' followed by any whitespace character one to unlimited times, followed by '['
  • (?<name>\w+) - Named Capture Group Matching any word character one to unlimited times.
  • = - Matches '=' literally.
  • (?<value>\S+)\] - Named Capture Group Matching any Non-Whitespace character one to unlimited times, followed by ']' literally.
  • \s+-\s+ - Any Whitespace character one to unlimited times followed by '-' literally again followed by any whitespace character one to unlimited times.
  • (?<desc>.*) - Finally a Named Capture Group matching any single character one to unlimited times.

@response

const re = new RegExp('\\*\\s+@response\\s+(?<status>\\{\\d{1,3}\\})\\s+(?<type>\\[[a-zA-Z/]+\\])\\s+-\\s+(?<desc>.*)');

Once match is used on an applicable string, it will return named groups that need some small modifications to assign.

match.groups.status.replace("{","").replace("}","");
match.groups.type.replace("[","").replace("]","");
match.groups.desc.trim();

Heres the same Regex with any JavaScript escaping and object initialization removed.

\*\s+@response\s+(?<status>\{\d{1,3}\})\s+(?<type>\[[a-zA-Z/]+\])\s+-\s+(?<desc>.*)

A Quick Summary of what this is doing: (Please note in the future its intended that each value is optional. Currently that is not the case.)

  • \*\s+ - Matches '*' followed by any whitespace character one to unlimited times.
  • @response\s+ - Matches '@response' literally followed by any whitespace character one to unlimited times.
  • (?<status>\{\d{1,3}\}) - Named Capture Group 'status' matching a digit ([0-9]) between one and three times. Preceded by '{' literally and followed by '}' literally.
  • \s+(?<type>\[[a-zA-Z/]+\]) - Matches firstly any whitespace character between one and unlimited times. Followed by creating a Named Capture Group 'type' where inside it matches '[' literally, then the index a-z and A-Z AND '/' literally between one and unlimited times. Followed by ']' literally.
  • \s+-\s+ - Matches any whitespace character one to unlimited times, followed by '-' literally, then again followed by any whitespace character between one and unlimited times.
  • (?<desc>.*) - Finally creating a Named Capture Group 'desc' matching any single character an unlimited amount of times.