grep search expression

Terrence

Tagged: GREP

This topic has 5 replies, 4 voices, and was last updated 10 years, 1 month ago by Anonymous.

Viewing 3 reply threads

Author

Posts
- February 27, 2014 at 2:59 pm #67371
  
  Terrence
  Member
  
  HI all
  
  Anyone out there know the quick GREP search terms for finding an apostrophe followed by a space then an uppercase letter?
- February 27, 2014 at 3:33 pm #67375
  
  Terrence
  Member
  
  By jiggitty, give it up to the GREP newbie. I think I figured it out:
  
  ‘\s\u
  
  This came in handy: https://creativepro.com/grep-and-text-metacharacter-cheat-sheet.php
  
  .’\s\u apparently searches for any punctuation followed by an apostrophe followed by a space then an uppercase letter. It is finding instances where this happens between paragraphs, but I don’t mind.
- March 4, 2014 at 10:25 am #67446
  
  Milt Klingensmith
  Member
  
  Hi All,
  I’m trying to write a grep to convert improperly styled times on a calendar to the proper style.
  I want the grep to find “am.” and change it to “a.m.”
  But since there is the possibility of having an “am.” being part of a show title such as “Sam I am.” I wanted to use a look behind to exclude an “am.” that is preceded by an “I” and a blank space.
  
  The negative lookbehind grep I was writing looks like this:
  \s(am\.)(?<!I\s)
  
  I also tried a positive lookbehind for an am. preceded by a space and a digit to narrow the types of am.s that could be selected
  \s(am\.)(?<=\d)
  
  The first one still selects the “am.” following the “I” and the space.
  The second selects nothing.
  Any ideas on how to use a lookbehind to select only the “am.” following a digit and a space?
  - March 9, 2014 at 5:14 pm #67516
    
    Steve Jenks
    Member
    
    Hi Milt
    
    I think it could be as simple as ignoring the look behinds and simply searching for any occurrence of a digit followed by a space followed by ‘am.’
    
    If this is the case then if you searched
    (\d )(am.)
    
    and replaced with
    $1a.m.
    
    It should change just the times and leave the rest alone.
    
    Steve
- March 10, 2014 at 7:01 am #67520
  
  Milt Klingensmith
  Member
  
  Thanks for the reply, Steve.
  
  This is what I ended up writing that worked:
  
  (?<=\d)am\.|(?<=\d/s)am\.|(?<=\d)am(?=\,)|(?<=\d/s)am(?=\,)|(?<=\d)am(?=\-)|(?<=\d/s)am(?=\-)|(?<=\d)am(?= *)|(?<=\d\s)am(?= *)|(?<=\d)am(?=\;)|(?<=\d\s)am(?=\;)
  
  then replace with “a.m.”
  
  It picks up every variation of the incorrect “am”s
  - March 10, 2014 at 1:42 pm #67536
    
    Anonymous
    Inactive
    
    Glad you got it sorted Milt.
Author

Posts