String conversion algorithm

[John van Terheijden]

Hi.

I'm trying to make a conversion algorithm that colors even and odd words in
a HTML string with <div> tags.

// input text with all sorts of HTML tags and whitespace
$str = "<h1>Title here</h1>
<p>This is a <strong>nice</strong> <img src="picture.gi f" /><br>
and some nice text </p>";

// color the words and echo
echo color_words($st r);

Would like the output to be:
<h1><div class="c1">Titl e</div> <div class="c2">here </div></h1>
<p><div class="c1">This </div> <div class="c2">is</div> <div
class="c1">a</div> <strong><div class="c2">nice </div></strong> <img
src="picture.gi f" /><br>
<div class="c1">and</div> <div class="c2">some </div> <div
class="c1">nice </div> <div class="c2">text </div> </p>

I tried it with some nested for and while loops, but kept getting infinite
loops, char. skipping problems, etc.

Maybe regular expressions is the answer, but I have very little (and bad)
experience with those.

I'm sure there are lots of you would could solve this in a couple of
seconds; please do so and help me out! :)

Thanks,

- John

[steve]

Re: String conversion algorithm

"John van Terheijde" wrote:[color=blue]
> Hi.
>
> I’m trying to make a conversion algorithm that colors even and
> odd words in
> a HTML string with <div> tags.
>
> // input text with all sorts of HTML tags and whitespace
> $str = "<h1>Title here</h1>
> <p>This is a <strong>nice</strong> <img
> src="picture.gi f" /><br>
> and some nice text </p>";
>
> // color the words and echo
> echo color_words($st r);
>
> Would like the output to be:
> <h1><div class="c1">Titl e</div> <div
> class="c2">here </div></h1>
> <p><div class="c1">This </div> <div
> class="c2">is</div> <div
> class="c1">a</div> <strong><div
> class="c2">nice </div></strong> <img
> src="picture.gi f" /><br>
> <div class="c1">and</div> <div
> class="c2">some </div> <div
> class="c1">nice </div> <div class="c2">text </div>
> </p>
>
> I tried it with some nested for and while loops, but kept getting
> infinite
> loops, char. skipping problems, etc.
>
> Maybe regular expressions is the answer, but I have very little[/color]
(and[color=blue]
> bad)
> experience with those.
>
> I’m sure there are lots of you would could solve this in a
> couple of
> seconds; please do so and help me out!
>
> Thanks,
>
> - John[/color]

I give you a non-exact solution:

$array = explode(" ", $str); //you can use "split" for finer
splitting
foreach ($array as $key => $word) {

//detect odd/even
if (intval($key/2) == $key/2) $word = "<color...> " . $word .
"<...>";
else $word = "the other color" . $word . "other color stuff"
}

$out = implode(" ", $array);
echo $out;

Hope this helps get you started

--
http://www.dbForumz.com/ This article was posted by author's request
Articles individually checked for conformance to usenet standards
Topic URL: http://www.dbForumz.com/PHP-String-c...ict133295.html
Visit Topic URL to contact author (reg. req'd). Report abuse: http://www.dbForumz.com/eform.php?p=445292

[Blackie]

Re: String conversion algorithm

this code has one small bug - works with empty words

try to convert this: "evenword1 oddword1 evenword2 oddword2" (notice two
spaces between oddword1 and evenword2)
it produces this result: "<evencolor>eve nword1</...>
<oddcolor>oddwo rd1</...> <evencolor></...> <oddcolor>evenw ord2</...>
<evencolor>oddw ord2</...>"
but we need this result: "<evencolor>eve nword1</...>
<oddcolor>oddwo rd1</...> <evencolor>even word2</...>
<oddcolor>oddwo rd2</...>"

so try small correcture:

[PHP]
$array = explode(" ", $str); //you can use "split" for finer splitting
$i = 1;
foreach ($array as $key => $word) if($word != "") {
if ($i%2 == 0) $word = "<color...> " . $word . "<...>";
else $word = "the other color" . $word . "other color stuff"
$i++;
}
[/PHP]


"steve" <UseLinkToEmail @dbForumz.com> pí¹e v diskusním pøíspìvku
news:10gamvi5cm kpb68@news.supe rnews.com...[color=blue]
> "John van Terheijde" wrote:[color=green]
> > Hi.
> >
> > I’m trying to make a conversion algorithm that colors even and
> > odd words in
> > a HTML string with <div> tags.
> >
> > // input text with all sorts of HTML tags and whitespace
> > $str = "<h1>Title here</h1>
> > <p>This is a <strong>nice</strong> <img
> > src="picture.gi f" /><br>
> > and some nice text </p>";
> >
> > // color the words and echo
> > echo color_words($st r);
> >
> > Would like the output to be:
> > <h1><div class="c1">Titl e</div> <div
> > class="c2">here </div></h1>
> > <p><div class="c1">This </div> <div
> > class="c2">is</div> <div
> > class="c1">a</div> <strong><div
> > class="c2">nice </div></strong> <img
> > src="picture.gi f" /><br>
> > <div class="c1">and</div> <div
> > class="c2">some </div> <div
> > class="c1">nice </div> <div class="c2">text </div>
> > </p>
> >
> > I tried it with some nested for and while loops, but kept getting
> > infinite
> > loops, char. skipping problems, etc.
> >
> > Maybe regular expressions is the answer, but I have very little[/color]
> (and[color=green]
> > bad)
> > experience with those.
> >
> > I’m sure there are lots of you would could solve this in a
> > couple of
> > seconds; please do so and help me out!
> >
> > Thanks,
> >
> > - John[/color]
>
> I give you a non-exact solution:
>
> $array = explode(" ", $str); //you can use "split" for finer
> splitting
> foreach ($array as $key => $word) {
>
> //detect odd/even
> if (intval($key/2) == $key/2) $word = "<color...> " . $word .
> "<...>";
> else $word = "the other color" . $word . "other color stuff"
> }
>
> $out = implode(" ", $array);
> echo $out;
>
> Hope this helps get you started
>
> --
> http://www.dbForumz.com/ This article was posted by author's request
> Articles individually checked for conformance to usenet standards
> Topic URL:[/color]
http://www.dbForumz.com/PHP-String-c...ict133295.html[color=blue]
> Visit Topic URL to contact author (reg. req'd). Report abuse:[/color]

http://www.dbForumz.com/eform.php?p=445292

[Blackie]

Re: String conversion algorithm

using some loop is good idea, it's simple :))
try something like this, you can enhance it to detect more separators (not
only space), to ignore HTML tags and so on...

[PHP]
$src = "evenword1 oddword1 evenword2 oddword2";
$tmp = $src;
$dst = "";
$i = 0;
do{
$pos = strpos($tmp, " ");
if($pos !== false) // separator found, we have "word"
{
$word = substr($tmp, 0, $pos); // extract word
if($word != "") // isn't it
empty??
{
if($i%2 == 0) $word = "<color1>".$wor d."</color1>"; // colorize it
else $word = "<color2>".$wor d."</color2>";
$i++; // increase word
counter
}
$dst .= $word.substr($t mp, $pos, 1); // append
colorized word and separator to destination
$tmp = substr($tmp, $pos+1); // and trim used
part from temporary
}
else // no other separator found, we have last word
{
if($tmp != "") // isn't it
empty??
{
if($i%2 == 0) $tmp = "<color1>".$tmp ."</color1>"; // colorize it
else $tmp = "<color2>".$tmp ."</color2>";
$i++; // increase word
counter
}
$dst .= $tmp;
unset($tmp);
break;
}
}while(1);

echo "<pre>";
echo "source: ".htmlspecialch ars($src)."<br> ";
echo "destinatio n: ".htmlspecialch ars($dst)."<br> ";
echo "we have ".$i." words";
echo "</pre>";
[/PHP]


"John van Terheijden" <vanterheijden_ avoidin_spam@ho me.nl> pí¹e v diskusním
pøíspìvku news:ce3frp$rkm $1@news5.tilbu1 .nb.home.nl...[color=blue]
> Hi.
>
> I'm trying to make a conversion algorithm that colors even and odd words[/color]
in[color=blue]
> a HTML string with <div> tags.
>
> // input text with all sorts of HTML tags and whitespace
> $str = "<h1>Title here</h1>
> <p>This is a <strong>nice</strong> <img src="picture.gi f" /><br>
> and some nice text </p>";
>
> // color the words and echo
> echo color_words($st r);
>
> Would like the output to be:
> <h1><div class="c1">Titl e</div> <div class="c2">here </div></h1>
> <p><div class="c1">This </div> <div class="c2">is</div> <div
> class="c1">a</div> <strong><div class="c2">nice </div></strong> <img
> src="picture.gi f" /><br>
> <div class="c1">and</div> <div class="c2">some </div> <div
> class="c1">nice </div> <div class="c2">text </div> </p>
>
> I tried it with some nested for and while loops, but kept getting infinite
> loops, char. skipping problems, etc.
>
> Maybe regular expressions is the answer, but I have very little (and bad)
> experience with those.
>
> I'm sure there are lots of you would could solve this in a couple of
> seconds; please do so and help me out! :)
>
> Thanks,
>
> - John
>
>[/color]

[John van Terheijden]

Re: String conversion algorithm

Thanks for the replies!

The given solutions didn't do anything with the fact that <tags> shouldn't
be colored. I fiddled some more with my initial for loop and came up with a
solution. The main difference between this and earlier attempts is that I
now haven't got nested for and while loops, so every time only 1 character
is handled.

function color($string) {
$whitespace = explode('-', " -\n-\r- -\0-\x0B");
$out = '<span>';
$color1 = true;
$inTag = false;
$inColor = true;
for ($i = 0; $i < strlen($string) ; $i++) {
$char = $string[$i];

// in a tag
if ($inTag) {
if ($char == '>') $inTag = false;
}

// not in a tag
else {

// starting a tag
if ($char == '<') {
$inTag = true;
if ($inColor) {
$out .= '</span>';
$inColor = false;
}
}

// in whitespace
elseif (in_array($char , $whitespace)) {
if ($inColor) {
$out .= '</span>';
$inColor = false;
}
}

// in "normal" text
else {
if (!$inColor) {
if ($color1) $out .= '<span class="c1">';
else $out .= '<span class="c2">';
$color1 = !$color1;
$inColor = true;
}
}
}
$out .= $char;
}
$out .= '</span>';
return $out;
}


Thanks,

- John

"John van Terheijden" <vanterheijden_ avoidin_spam@ho me.nl> schreef in
bericht news:ce3frp$rkm $1@news5.tilbu1 .nb.home.nl...[color=blue]
> Hi.
>
> I'm trying to make a conversion algorithm that colors even and odd words[/color]
in[color=blue]
> a HTML string with <div> tags.
>
> // input text with all sorts of HTML tags and whitespace
> $str = "<h1>Title here</h1>
> <p>This is a <strong>nice</strong> <img src="picture.gi f" /><br>
> and some nice text </p>";
>
> // color the words and echo
> echo color_words($st r);
>
> Would like the output to be:
> <h1><div class="c1">Titl e</div> <div class="c2">here </div></h1>
> <p><div class="c1">This </div> <div class="c2">is</div> <div
> class="c1">a</div> <strong><div class="c2">nice </div></strong> <img
> src="picture.gi f" /><br>
> <div class="c1">and</div> <div class="c2">some </div> <div
> class="c1">nice </div> <div class="c2">text </div> </p>
>
> I tried it with some nested for and while loops, but kept getting infinite
> loops, char. skipping problems, etc.
>
> Maybe regular expressions is the answer, but I have very little (and bad)
> experience with those.
>
> I'm sure there are lots of you would could solve this in a couple of
> seconds; please do so and help me out! :)
>
> Thanks,
>
> - John
>
>[/color]

[Michael Fesser]

Re: String conversion algorithm

.oO(John van Terheijden)
[color=blue]
>The given solutions didn't do anything with the fact that <tags> shouldn't
>be colored. I fiddled some more with my initial for loop and came up with a
>solution. [...][/color]

But there are some bugs, the beginning and end of your teststring look
like this:

<span></span><h1><span class="c1">Titl e</span> ...
^^^^^^^^^^^^^

.... <span class="c2">text </span> </p></span>
^^^^^^^

I've played around a bit with regular expressions:

function color_words($st ring) {
$color = 2;
$pattern = '#[^\s<>]+(?![^<]*>)#e';
$replace = '"<span class=\"c".($co lor = 3-$color)."\">$0</span>"';
return preg_replace($p attern, $replace, $string);
}

With my teststrings it works, but it's quite possible that there are
situations where it'll fail, you have to test it.

Micha

[John van Terheijden]

Re: String conversion algorithm

Hi Michael.

Thanks for the reply.

You were right about the "bugs". The opening and closing <span> </span> were
needed in earlier attempts. I solved it better now:
---
function color($string) {
$whitespace = explode('-', " -\n-\r-\t-\0-\x0B");
$color1 = true;
$inTag = false;
$inColor = false;

for ($i = 0; $i < strlen($string) ; $i++) {
$char = $string[$i];

// in a tag
if ($inTag) {
if ($char == '>') $inTag = false;
}

// not in a tag
else {

// starting a tag
if ($char == '<') {
$inTag = true;
if ($inColor) {
$out .= '</span>';
$inColor = false;
}
}

// in whitespace
elseif (in_array($char , $whitespace)) {
if ($inColor) {
$out .= '</span>';
$inColor = false;
}
}

// in "normal" text
else {
if (!$inColor) {
if ($color1) $out .= '<span class="c1">';
else $out .= '<span class="c2">';
$color1 = !$color1;
$inColor = true;
}
}
}
$out .= $char;
}
if ($inColor) $out .= '</span>';
return $out;
}
---

I tested your code and with my test HTML it works just fine. I think I'll
use your code now, for it's nice and short (and prob. faster) and switch
back to mine when changes or additional stuff is needed :)

Thanks!

- John

"Michael Fesser" <netizen@gmx.ne t> schreef in bericht
news:38qcg0helf jm8hhd16bjo0gqb 3b90ivbpr@4ax.c om...[color=blue]
> .oO(John van Terheijden)
>[color=green]
> >The given solutions didn't do anything with the fact that <tags>[/color][/color]
shouldn't[color=blue][color=green]
> >be colored. I fiddled some more with my initial for loop and came up with[/color][/color]
a[color=blue][color=green]
> >solution. [...][/color]
>
> But there are some bugs, the beginning and end of your teststring look
> like this:
>
> <span></span><h1><span class="c1">Titl e</span> ...
> ^^^^^^^^^^^^^
>
> ... <span class="c2">text </span> </p></span>
> ^^^^^^^
>
> I've played around a bit with regular expressions:
>
> function color_words($st ring) {
> $color = 2;
> $pattern = '#[^\s<>]+(?![^<]*>)#e';
> $replace = '"<span class=\"c".($co lor = 3-$color)."\">$0</span>"';
> return preg_replace($p attern, $replace, $string);
> }
>
> With my teststrings it works, but it's quite possible that there are
> situations where it'll fail, you have to test it.
>
> Micha[/color]

[Michael Fesser]

Re: String conversion algorithm

.oO(John van Terheijden)
[color=blue]
>You were right about the "bugs". The opening and closing <span> </span> were
>needed in earlier attempts. I solved it better now:[/color]

Seems to work, but there are still some little issues:
[color=blue]
>function color($string) {
> $whitespace = explode('-', " -\n-\r-\t-\0-\x0B");
> $color1 = true;
> $inTag = false;
> $inColor = false;[/color]

You should add an

$out = '';

at the beginning, or you'll get a notice on a line with

$out .= ...

(if error_reporting is set to E_ALL, which should be the case on
development systems).
[color=blue]
> for ($i = 0; $i < strlen($string) ; $i++) {
> $char = $string[$i];[/color]

Use curly braces for accessing single chars in a string:

$char = $string{$i};

Using array-brackets for this is still allowed for backwards
compatibility, but deprecated.
[color=blue]
>I tested your code and with my test HTML it works just fine. I think I'll
>use your code now, for it's nice and short (and prob. faster) [...][/color]

I have done a little test on my "server" (P100, 64MB RAM ;) ): With your
given teststring the regular expression method was 2-3 times faster,
with a longer string (a complete webpage) it was still 2 times faster.

Micha

[John van Terheijden]

Re: String conversion algorithm

You're right about the notice and the curly braces. I didn't know about
them, it worked with [], so I thought that was the way..

Thanks for the testing!

- John

"Michael Fesser" <netizen@gmx.ne t> schreef in bericht
news:la2dg0tpnl s0vte488gsmupck vpuav3r9l@4ax.c om...[color=blue]
> .oO(John van Terheijden)
>[color=green]
> >You were right about the "bugs". The opening and closing <span> </span>[/color][/color]
were[color=blue][color=green]
> >needed in earlier attempts. I solved it better now:[/color]
>
> Seems to work, but there are still some little issues:
>[color=green]
> >function color($string) {
> > $whitespace = explode('-', " -\n-\r-\t-\0-\x0B");
> > $color1 = true;
> > $inTag = false;
> > $inColor = false;[/color]
>
> You should add an
>
> $out = '';
>
> at the beginning, or you'll get a notice on a line with
>
> $out .= ...
>
> (if error_reporting is set to E_ALL, which should be the case on
> development systems).
>[color=green]
> > for ($i = 0; $i < strlen($string) ; $i++) {
> > $char = $string[$i];[/color]
>
> Use curly braces for accessing single chars in a string:
>
> $char = $string{$i};
>
> Using array-brackets for this is still allowed for backwards
> compatibility, but deprecated.
>[color=green]
> >I tested your code and with my test HTML it works just fine. I think I'll
> >use your code now, for it's nice and short (and prob. faster) [...][/color]
>
> I have done a little test on my "server" (P100, 64MB RAM ;) ): With your
> given teststring the regular expression method was 2-3 times faster,
> with a longer string (a complete webpage) it was still 2 times faster.
>
> Micha[/color]