How to read from a file until a character is read then skip some characters and continue reading again

Question

How do I read from a file till a particular character is reached and then seek to the next character and continue reading in C++.

In my program I am using some HTML syntax and generating a .htm file... So in my C++ code I have added the tags.

But when I read from my .htm file I want it to not include the tags.

What I plan on doing is reading the file till '<' is encountered then just seek to the point till '>' is encountered and continue reading from there.

Just a suggestion, but depending on the size of the file (probably small because HTML), you could read the whole file in first. Makes it a bit easier to search. — ColWhi, Apr 02 '12 at 15:40

CashCow · Answer 1 · 2018-04-08T09:41:56.760

In general, to read a file until a particular character is reached you use std::getline and you set the second parameter to your terminator so if you are reading until a '<' character you can do

std::getline( infile, str, '<' );

you can then do the same with a > character

In your case if you are parsing HTML then there are probably specific parsers for it already. I think HTML1.1 is XML compliant but HTML1.0 isn't as it was not always necessary to close all your tags, so an XML parser will not necessarily work.

You would need to assume that open and close tags are not part of comments or quoted text and the methodology I described above would not promise you that so you'd need a full state machine.

score 5 · Accepted Answer · answered Apr 02 '12 at 15:41

First of all, you should be aware that doing this correctly is quite a bit trickier than you apparently think.

Just answering the question as you asked it, you can use istream::get to read a character at a time until you get a '<'. You can use ignore to ignore characters up to the next '>' in the stream.

Getting back to the first point, however, that generally won't work correctly. In particular, it's entirely possible for a tag to contain a string, and the string (in turn) contain a '>' that is not the close of the tag. As such, to have any hope of parsing the HTML correctly, you need to parse for strings inside the tags, and when you find them, skip across their contents rather than treating any '>' they might contain as ending the tag.

A tag containing an unescaped `>` would be invalid, if it is not a nested tag. I think the best hint is: there are always as many `<` as `>` in a valid html file. — devsnd, Apr 02 '12 at 15:46
According to [W3c](http://www.w3.org/TR/html-markup/syntax.html#normal-character-data) normal text can't contain a "<" character, but I see no such restriction on ">". — Jerry Coffin, Apr 02 '12 at 15:52

score 1 · Answer 3 · answered Apr 02 '12 at 15:35

Here are some guidelines.

You can read the file line by line with getLine from a ifstream , and keep each line in a std::string
You can use std::string.find() method to find < and > characters.
You can use std::string.substr() method to get substrings.
You can group the strings, if required, in a std::vector.

You're not going to get a full implementation here, but this should be enough to get you started.

score 1 · Answer 4 · answered Apr 02 '12 at 15:42

The following reads from standard input; modify/reaplace the calls to getchar() to read from somewhere else.

int c;

c = getchar();
while ( c != EOF ) {
    while ( c != '<' && c != EOF) {
        /* Do something with character outside tag? */
        c = getchar();
    }
    while ( c != '>' && c != EOF ) {
        /* Do something with character inside tag? */
        c = getchar();
    }
}

How to read from a file until a character is read then skip some characters and continue reading again

4 Answers4