how to removie html tags from a string

[Ra71sh]

Hi,

I store the comments as a text in database, but for special characters as an HTML tag.
While fetching it in a text file, i just need the comments and no HTML tag.
Is there any way to remove this using C program.

Pls help.

-Ratish

[donbock]

Are you confident the html file is well-formed? If so, you could delete all text enclosed in angle brackets. Notice that horrible things will happen if the input file isn't well-formed.

What are you supposed to do with character entity references (such as "&lt") and numeric entity references (such as "&#931) -- pass them through or expand them?

Is this an assignment that you have to do in a particular way; or are you happy with any approach that works? Try opening the file with a browser and then saving it as a text file.

[Ra71sh]

Actually the text is stored in database while formatting from the application screen.
But I need to provide a report with the comments input for which I need to remove the html tags. I am using a C program to fetch the data from database.
But it carries the html tags. I need to remove all the HTML tags that is in these comments.

[JosAH]

Actually the text is stored in database while formatting from the application screen.
But I need to provide a report with the comments input for which I need to remove the html tags. I am using a C program to fetch the data from database.
But it carries the html tags. I need to remove all the HTML tags that is in these comments.

Yes, you already wrote that: copy all characters on a line until you scan a '<'; stop copying but keep on scanning until you see a '>'. Repeat until you've reached the end of the line.

As already mentioned character combinations such as '&lt;' and uglier pass unharmed.

kind regards,

Jos

[Ra71sh]

that is fine but my concern is for the cases where someone has entered some text e.g. points where he has used <a>, <b>, or a>, b> etc.
How would I handle this.
Also, in cases where I have special handling like &amp; or #3688 etc what would I do?

[JosAH]

that is fine but my concern is for the cases where someone has entered some text e.g. points where he has used <a>, <b>, or a>, b> etc.
How would I handle this.
Also, in cases where I have special handling like &amp; or #3688 etc what would I do?

You have to write a complete fault tolerant html parser then. As a corollary I understand that your database contains incorrect html data? If so the GIGO prinicples rears its ugly head (Garbage In Garbage Out).

kind regards,

Jos

[donbock]

that is fine but my concern is for the cases where someone has entered some text e.g. points where he has used <a>, <b>, or a>, b> etc.
How would I handle this.
Also, in cases where I have special handling like &amp; or #3688 etc what would I do?

In a well-formed html file, the input characters "<" and ">" would be replaced by "&lt" and "&gt". Is that happening for you? If not, then your file is malformed and you will have great difficulty parsing it.

Regarding "&#3688" and its kin, don't ask us ... what do you think should happen? Are you emitting a text file? If so, then you are limited to printable characters. What is a meaningful and useful way to handle nonprintable characters in your context?

Do you want to see the html page exactly as it would appear on a browser page -- with all the mark-ups? If so, then open the file with a browser and save or print to a pdf file.

[donbock]

A malformed html file can confuse the html parser:
html parse error

[JosAH]

A malformed html file can confuse the html parser:
html parse error

That is so funny. ;-)

kind regards,

Jos