Still celebrating National IT Professionals Day with 3 months of free Premium Membership. Use Code ITDAY17

x
?
Solved

Regular expression pattern macth to weed out words with HTML character codes & entities

Posted on 2016-10-29
7
Medium Priority
?
186 Views
Last Modified: 2016-11-18
I have a file that is riddled with words containing HTML character codes & entities such as: Müller, Étre or "Flying—High" -- however, there's also a lot of junk ones that need to corrected such as Teenage��������s or &sup1dsjÝhccbß

I'm using a perl script, so what would be a good regular expression to match these words with HTML codes & entities? Thanks
0
Comment
Question by:hadrons
[X]
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
  • 2
  • 2
  • 2
  • +1
7 Comments
 
LVL 27

Expert Comment

by:wilcoxon
ID: 41865196
If you just want to match words with escape sequences in the middle then this should work:
m{((?:\w+(?:&(?:#\d+|[a-z]+);)+)+\w*)}

Open in new window


That will match a word with one or more escapes (either in a row or separated by letters).  It will not match words with the escape at the beginning of the word.
0
 
LVL 29

Expert Comment

by:Dr. Klahn
ID: 41865213
I am not sure that there is a simple solution to this one.  But here's a start.  The numeric entities expressed in decimal can be matched by

&\#[0-9]+;

Open in new window


Unfortunately this doesn't match numeric entities expressed in hex, and it also does matches entities you probably want to retain that are expressed in numeric form rather than by the HTML name.  For example, it matches the numeric for the euro symbol, and you would probably not want to throw that character out.

&euro; <== interchangeable with ==> &#8364;

Open in new window


If this is a one-off situation, it might be easier to go through the file with a text editor to knock out the obvious and frequently unwanted entities, then see how the result looks.
0
 
LVL 27

Accepted Solution

by:
wilcoxon earned 2000 total points
ID: 41865245
I had forgotten that you can have hex.  This should also pick up hex (and correct to allow numbers in entity names):
m{((?:\w+(?:&(?:#[0-9a-f]+|[a-z]\w*);)+)+\w*)}

Open in new window


Isn't a semi-colon required to end an escape sequence?  Your "bad" example has &sup1dsj without a closing semi-colon.

If you want to keep some and remove others, there is no really easy way.  The simplest would be either a list of ones to keep or a hash map indicating which to keep and which to get rid of (but then you have to process them one by one).
0
Technology Partners: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

 
LVL 22

Expert Comment

by:Kim Walker
ID: 41865442
I have a file that is riddled with words containing HTML character codes...
If you only have one file, why not use a text editor for a one-time conversion. One of my favorite code editors, EditPad Pro, has the functionality to convert HTML character entities into unicode characters. A free trial version of EditPad Pro is available which is fully functional except for Spell Checker. I'm sure there are others out there that might be free.
0
 

Author Comment

by:hadrons
ID: 41865540
Hi, Kim, actually a bunch of files and it's ones I work with on a regular basis, but that info was unimportant to the question itself, so I just simplified the narrative - however, thanks for the EditPadPro - its something that could be useful in the future.

Dr. Klahn, I did use an expression similar to what you suggested: \&\#[0-9A-Za-z]+?; with various other expressions, but its somewhat limited.

But I'll follow up on Wilcoxon suggestion - it looks promising, but I have to wait until Monday. Thanks for the feedback all so far; Mike
0
 
LVL 22

Expert Comment

by:Kim Walker
ID: 41865595
Are you trying to convert the entities to their unicode character or are you trying to strip them out? You will not be able to convert them using a regex. But you can search for the regex match and then perform a lookup/replace from a reference table.
0
 

Author Comment

by:hadrons
ID: 41865785
The match is needed to isolate these words so I can grep them out and look them over
0

Featured Post

New benefit for Premium Members - Upgrade now!

Ready to get started with anonymous questions today? It's easy! Learn more.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

This article describes how to create custom column layout styles for Bootstrap. The article uses 5 columns to illustrate the concept, but the principle can be extended to any number of columns.
This article discusses how to create an extensible mechanism for linked drop downs.
In this Micro Tutorial viewers will learn how to create navigation buttons that change on rollover, using CSS (Continuation of the CSS Image Sprite tutorial) Create a parent ID for all the list items       - Specify position: absolute and display: block…
In this tutorial viewers will learn how to embed an audio file in a webpage using HTML5. Ensure your DOCTYPE declaration is set to HTML5: : The declaration should display (CODE) HTML5 is supported by the most recent versions of all major browsers…

704 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question