Need help with regex expression

daesdaemar · 01-14-2009, 07:52 PM

I am trying to use find and replace in MS Word to get rid of footnote references in a text file. The references are in the form [n] or [nn] where n represents a numerical digit.

I presume that the bracket symbols, [] must be protected in the regex lingo because everything I try doesn't work.

I have tried both [?] and[*] (wildcards) and they don't work. I would appreciate any help.

Nate the great · 01-14-2009, 08:01 PM

MSWord uses nonstandard terms for regular expressions.

This one will work: (\[*\])

But I recommend that you use this one: (\[?\]) add more '?' as needed to catch all the desired terms

Andybaby · 01-14-2009, 08:42 PM

nate i assume when you use \ in a regex that means to look for the leading character instead of it being used as part of the formatting? if so i think that makes it so it finally makes sense to me.

I find that Notepad++ works great for removing junk lines in the text.
in notepad++
go to search > find
search for a term that appears in the lines you want to remove.
check mark line, hit find all.
go to search > delete bookmarked lines

Nate the great · 01-14-2009, 08:46 PM

Quote:

Originally Posted by Andybaby

nate i assume when you use \ in a regex that means to look for the leading character instead of it being used as part of the formatting? if so i think that makes it so it finally makes sense to me.

I find that Notepad++ works great for removing junk lines in the text.
in notepad++
go to search > find
search for a term that appears in the lines you want to remove.
check mark line, hit find all.
go to search > delete bookmarked lines

The characters [] have a special meaning in regular expressions. A slash '\' is placed before each one to force the use of the non-special meaning.

Anyway, I tested the expressions I provided above. They work correctly in MSWord.

Andybaby · 01-14-2009, 09:02 PM

Quote:

Originally Posted by Nate the great

The characters [] have a special meaning in regular expressions. A slash '\' is placed before each one to force the use of the non-special meaning.

thank you, they finally make sense to me.

Patricia · 01-14-2009, 09:03 PM

It sometimes is worth looking to see whether the footnotes are in a particular style. If so, you can search and delete the instances of the style.

Jellby · 01-15-2009, 04:04 AM

In vim it could be: \[\d\+\]

\[, \]: literal [ and ]
\d: a digit
\+: repeated one or more times.

daesdaemar · 01-15-2009, 11:46 AM

Quote:

Originally Posted by Nate the great

MSWord uses nonstandard terms for regular expressions.

This one will work: (\[*\])

But I recommend that you use this one: (\[?\]) add more '?' as needed to catch all the desired terms

Thanks for the reply, Nate. Works great.

delphidb96 · 01-15-2009, 06:29 PM

Sooo.... Does anyone understand how to get Word to replace the string 'p. ###', where ### can be anything from 1 to 999, with a space? Without having to do a separate find/replace on 3-digit, then 2-digit, then 1-digit numeric strings?

Derek

Nate the great · 01-15-2009, 06:36 PM

Quote:

Originally Posted by delphidb96

Sooo.... Does anyone understand how to get Word to replace the string 'p. ###', where ### can be anything from 1 to 999, with a space? Without having to do a separate find/replace on 3-digit, then 2-digit, then 1-digit numeric strings?

Derek

Look for this:

(p. [0-9]{1,3})

delphidb96 · 01-15-2009, 06:42 PM

Nate,

Thanks!

Derek

daesdaemar · 01-18-2009, 01:44 PM

Nate,

You're a wonder at these expressions. Do you have any suggestions for where I might teach myself this stuff? I use primarily Word for my editing.

mtravellerh · 01-18-2009, 02:53 PM

Quote:

Originally Posted by Jellby

In vim it could be: \[\d\+\]

\[, \]: literal [ and ]
\d: a digit
\+: repeated one or more times.

\d+ is also working (for pagenumbers, for example)

Andy's way also works fine and is fast but can lead to errors (overkill) if you're not conscientious. You really have to be sure that there is nothing else on that line you're looking for ever!

=X= · 01-18-2009, 03:27 PM

Quote:

Originally Posted by daesdaemar

Do you have any suggestions for where I might teach myself this stuff? I use primarily Word for my editing.

Okay I'm not Nate so parding me for intruding, now to answer your question

MS Word does not use standard regular notation and in addition it's implementation is inferior to true RegEx. But it is still powerful and conceptually still make use of the regular expression concepts

I'd recommend reading http://www.regular-expressions.info/

These two articles are pretty good. They got me started when I decided to write the BookCreator tool
http://office.microsoft.com/en-us/he...873041033.aspx
http://office.microsoft.com/en-us/he...873051033.aspx

Also there are scores of Perl and Javascript tutorials that use regular expressions. The syntax will be a little different but they still try to solve the same problems and you will learn from the exercise of converting the syntax.

Reglar expressions have a steep learning curve but once your over the first hump, you'll realize its the best invention since coffee. ... Yea I know the saying is sliced bread but give me coffee and regluar expressions --hmmmm I'm in heaven.

=X=

HarryT · 01-19-2009, 04:29 AM

Quote:

Originally Posted by Nate the great

MSWord uses nonstandard terms for regular expressions.

This one will work: (\[*\])

But I recommend that you use this one: (\[?\]) add more '?' as needed to catch all the desired terms

If, as stated, the footnote reference is numeric, then:

\[[0-9]*\]

is safer in that it will only catch numeric references.

01-14-2009, 07:52 PM	#1
daesdaemar Addict Posts: 210 Karma: 4282 Join Date: Oct 2008 Location: Florida Device: Sony 505, Kindle 3, iPad 3	Need help with regex expression I am trying to use find and replace in MS Word to get rid of footnote references in a text file. The references are in the form [n] or [nn] where n represents a numerical digit. I presume that the bracket symbols, [] must be protected in the regex lingo because everything I try doesn't work. I have tried both [?] and[*] (wildcards) and they don't work. I would appreciate any help.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Regular Expression Help	Azhad	Calibre	86	09-27-2011 02:37 PM
Regular Expression Help	smartmart	Calibre	5	10-17-2010 05:19 AM
Need Help Creating a Regular Expression	Worm	Calibre	9	08-18-2010 01:20 PM
Help with the regular expression	Dysonco	Calibre	9	03-22-2010 10:45 PM
help with regex expression	daesdaemar	Workshop	4	02-19-2010 07:38 AM

01-14-2009, 08:01 PM	#2
Nate the great Sir Penguin of Edinburgh Posts: 12,375 Karma: 23555235 Join Date: Apr 2007 Location: DC Metro area Device: Shake a stick plus 1	MSWord uses nonstandard terms for regular expressions. This one will work: (\[*\]) But I recommend that you use this one: (\[?\]) add more '?' as needed to catch all the desired terms

01-14-2009, 08:42 PM	#3
Andybaby Wizard Posts: 1,279 Karma: 1002683 Join Date: Nov 2008 Location: New York Device: PRS-700	nate i assume when you use \ in a regex that means to look for the leading character instead of it being used as part of the formatting? if so i think that makes it so it finally makes sense to me. I find that Notepad++ works great for removing junk lines in the text. in notepad++ go to search > find search for a term that appears in the lines you want to remove. check mark line, hit find all. go to search > delete bookmarked lines

01-14-2009, 09:03 PM	#6
Patricia Reader Posts: 11,504 Karma: 8720163 Join Date: May 2007 Location: South Wales, UK Device: Sony PRS-500, PRS-505, Asus EEEpc 4G	It sometimes is worth looking to see whether the footnotes are in a particular style. If so, you can search and delete the instances of the style.

01-15-2009, 04:04 AM	#7
Jellby frumious Bandersnatch Posts: 7,515 Karma: 18512745 Join Date: Jan 2008 Location: Spaniard in Sweden Device: Cybook Orizon, Kobo Aura	In vim it could be: \[\d\+\] \[, \]: literal [ and ] \d: a digit \+: repeated one or more times.

01-15-2009, 06:29 PM	#9
delphidb96 Wizard Posts: 2,999 Karma: 300001 Join Date: Jan 2007 Location: Citrus Heights, California Device: TWO Kindle 2s, one each Bookeen Cybook Gen3, Sony PRS-500, Axim X51V	Sooo.... Does anyone understand how to get Word to replace the string 'p. ###', where ### can be anything from 1 to 999, with a space? Without having to do a separate find/replace on 3-digit, then 2-digit, then 1-digit numeric strings? Derek

01-15-2009, 06:42 PM	#11
delphidb96 Wizard Posts: 2,999 Karma: 300001 Join Date: Jan 2007 Location: Citrus Heights, California Device: TWO Kindle 2s, one each Bookeen Cybook Gen3, Sony PRS-500, Axim X51V	Nate, Thanks! Derek

01-18-2009, 01:44 PM	#12
daesdaemar Addict Posts: 210 Karma: 4282 Join Date: Oct 2008 Location: Florida Device: Sony 505, Kindle 3, iPad 3	Nate, You're a wonder at these expressions. Do you have any suggestions for where I might teach myself this stuff? I use primarily Word for my editing.