[RESOLVED] Problems with reading file

daltman1967

SO far, I've tried everything... fread, fgets, get_file_contents, and I'm still getting nothing. Here's my current code:

$file_handle = fopen("menu_pt1.txt", "rb");
while (!feof($file_handle))
{
$contents .= fread($file_handle, 4096);
}
fclose($file_handle);
echo "s ".$contents;

menu_pt1.txt is in the same directory as this file, so there's no problem with finding it.

I get no output whatsoever. I've tried and tried and tried.

What I want to do is to read three separate sections of text files. These files contain program code. I then want to insert new code between the pieces, and then write the resulting output to a new file. This is driving me batty, and I have no idea what I'm doing wrong.

Does the contents of the file make a difference? Any ideas? Here are the first few lines of the first file I'm trying to read:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html>
<head>
<script language="javascript" type="text/javascript">
.
.
.

I'm totally lost here. 🙁

bradgrafelman

For one, it'd be simpler to read the entire file into a variable using [man]file_get_contents/man.

Now, if you're not seeing any output (even static values, e.g. the "s" in the echo statement you posted above), then that seems to suggest a fatal error or something is occurring. Do you have error_reporting set to E_ALL and display_errors set to On?

daltman1967

I tried using file_get_contents() and got the same results - nothing but the "s"

I don't know anything about error_reporting, E_ALL and display_errors.

sneakyimp

Try reading the manual on [man]error_reporting[/man]. RTFM!

You may also want to try checking [man]is_readable[/man] before trying to access the file.

bradgrafelman

daltman1967;10959756 wrote:
I tried using file_get_contents() and got the same results - nothing but the "s"

Well that's entirely different than no output at all. Instead that would suggest that it's some sort of permissions problem perhaps.

daltman1967;10959756 wrote:
I don't know anything about error_reporting, E_ALL and display_errors.

You're developing PHP scripts but don't know anything about error reporting? Sounds like you've got your learning priorities backwards, then. 😉

For errors other than parse errors, you can modify error reporting settings from within the script itself using [man]ini_set/man and [man]error_reporting/man. Since a parse error means that your script was never parsed properly, using PHP functions obviously isn't possible. In that case, you'll need to modify the PHP directives via other methods, such as by modifying the php.ini file or by using a .htaccess file (assuming you're using Apache and PHP was installed as a module rather than a CGI binary).

Either way, you need to set error_reporting and display_errors (or log_errors, whichever you prefer - usually you'd want display_errors in your development environment and log_errors in your production environment) so you can see the PHP error messages. Otherwise, you might as well list every possible thing that could go wrong, arrange them nicely on your wall, and buy a decent set of darts...

daltman1967

Ok, I figured part of this out. I AM getting an output, just not what I expected. What's happening is that the echo function is outputting the results of the file - i.e., the buttons and so forth which would be on the page if this file were being displayed normally in the browser, but it's stopping at a certain point.

What I expected to see was text -

rather than the effect of this text. Since I am getting an output, is there any way to 'print' it as text as opposed to the commands?

sneakyimp

echo htmlspecialchars($str);

That'll convert all the < and > chars and some other stuff. [man]htmlspecialchars[/man]. Or [man]htmlentities[/man]

bradgrafelman

In other words, echo is doing exactly what you told it to - output the exact contents of the file. Unless you changed the Content-Type HTTP header, however, your webserver (possibly instructed by a default setting in PHP) is telling your browser that the content being sent is HTML. Thus, your browser is going to parse it as HTML.

If you just want to display plain text, set the appropriate Content-Type header (e.g. "text/plain").

daltman1967

That worked great.

BTW, I guess I wasn't having a problem in the first place... I was getting the same "s", but the weird, cut off buttons also. I didn't realize that the buttons were actually the output that was coming after the "s".

bradgrafelman

You used "\t" in a double-quote delimited string, which is the escape sequence reserved for a horizontal tab. Few ways to fix that: 1) escape the backslash, 2) don't use double-quote delimited strings when you don't want to use the special escape sequences (\t, \n, etc.), or 3) use forward slashes (which will work regardless of which O/S you're using).

There's also the PATH_SEPARATOR constant to consider as well (which will reflect the default from the operating system that PHP is being executed on).

daltman1967

Absolutely great! Thanks, guys!

A major oops on the \ vs / character... I know better, but was distracted. I guess I'm still holding over the \ from DOS days.

The last, last problem with this whole thing seems to be a whitespace problem. I read the three pieces, and put them together, and then wrote the file. For now, the resulting file SHOULD be exactly equal to the original... yet it's not. It's giving me a Parse error: syntax error, unexpected '}' from the file. Visually, they look identical.

The problem comes here:

                   echo "<li><a href=\"...

err#1==> }
else
{
echo "<li><a href=\"...
echo "<li><a href=\"...
echo "<li><a href=\"...
echo "<li><a href=\"...
err#2==> }

the last line above err #1 is the bottom of file1; the last line above err #2 is the
bottom of file2; the rest after the } is the start of file 3.

I'm guessing that there's some whitespace or an unseen return character of some kind here which is blocking the proper parsing of the file. There should be a way to detect and eliminate this rogue... right?

bradgrafelman

Post that same snippet of code, only 1) use the board's [noparse]

..

[/noparse] bbcode tags, and 2) don't modify/omit any of it (with the exception of obfuscating (carefully, so as not to change the structure/format/syntax of the code) sensitive information if any) so that we can see exactly what you're working with.

EDIT: Also, before doing any of that, make sure it's not a simple error e.g. a missing semicolon on the line before.

daltman1967

Here is the code again:

          <?
            if (isset($_GET['lg']))
              { 
                echo "<li><a href=\"../body.php?lg=1\">Body</a></li>";
                echo "<li><a href=\"../brakes.php?lg=1\">Brakes</a></li>";
                echo "<li><a href=\"../electrical.php?lg=1\">Electrical</a></li>";
                echo "<li><a href=\"../engine.php?lg=1\">Engine</a></li>";
              &#65279;}
            else
              {
                echo "<li><a href=\"../body.php\">Body</a></li>";
                echo "<li><a href=\"../brakes.php\">Brakes</a></li>";
                echo "<li><a href=\"../electrical.php\">Electrical</a></li>";
                echo "<li><a href=\"../engine.php\">Engine</a></li>";
              &#65279;}
          ?>

I had already checked for missing semi's etc. Visually, the code is identical to the original file. There's no extra or missing that I can see; that's why I'm thinking it must be whitespace or other non-printing characters (eof, lf, cr, etc.).

daltman1967

What the heck is ? That must be the non-printing character that's causing the problem.

What's the best way to remove this?

bradgrafelman

While it doesn't do it correctly, note that the board attempted to escape some Unicode characters in your post. That's most likely what's causing your error.

EDIT: More info on which Unicode character that is can be found here.

EDIT2:

daltman1967 wrote:
What's the best way to remove this?

Delete those lines and re-type them? :p

daltman1967

This code:

&#65279;

shows as an HTML entity (decimal). I don't want to remove ALL & or other symbols from this code; only that one specific reference. It's not clear whether it's at the end of one and beginning of three, or at the end & beginning of two.

I'm looking for a way to detect and remove these two - and ONLY these two - extra, unwanted symbols.

Retyping the line manually simply isn't possible; the whole idea is to have php do all the work for me.

bradgrafelman

daltman1967 wrote:
I don't want to remove ALL & or other symbols from this code; only that one specific reference.

There is no "" in your code - that's just the board's software showing a Unicode character that it filtered out. If you want to do a find-and-replace on the Unicode character it represents, then you'd have to copy/type the character itself, not its HTML-encoded representation (unless you've got some nifty editor that can decode HTML entities in its find-and-replace feature).

daltman1967 wrote:
the whole idea is to have php do all the work for me

Well then you might want to get started on developing and implementing AI, but please take heed of the Terminator movie series before you do!