Expert needed! Trying to cycle through array and replace with predefined regex rules

gregchet

PHP ARRAY/REGEX PRO NEEDED!! HEEEEELP!

In short, I'm trying to re-create a Poker Stars hand history convertor I made a while back. I used to use ereg_replace on a multi-line variable of text, but there were little annoying problems that kept popping up. I decided to split the multi-line variable into an array so that each line was now its own variable, and I figure this will make the searching and replacing much more accurate.

This is eventually all going to be database driven, but for this example I've turned it into an array. If I can get help on this, I'll have no problem converting it over to the database version.

I'm basically trying to take a list of predefined REGEX rules (and their replacements), and then if these rules much the individual lines in the newly created array, replace them (so every rule should be checked against every line, and replaced if it matches).

I must be missing something here, possibly due to the fact that I've been sleep deprived for quite some time now 😛 (but no, not because of this convertor, hehe)

If any pros wanna help me out with this, it would be GREATLY appreciated.

<?php
// This is an include that I use within another file, and the "$new_hand" variable is already populated with a multi-line block of text

$rule[0] = "PokerStars Game #([0-9]): Tournament #([0-9]), Freeroll Hold'em No Limit - Level ([0-9a-zA-Z]) (([0-9])/([0-9])) - ([0-9])/([0-9])/([0-9]) - ([0-9])🙁[0-9])🙁[0-9]*) (ET)";

$rule_replacement[0] = "-------------------- Hand Convertor v1.0 -- Poker Stars Tournament NL Texas Hold'em, Freeroll - Level \3 (\4/\5) Tournament #\2, Game #\1, ";

$rule[1] = "TESTER PokerStars Game #([0-9]): Tournament #([0-9]), Freeroll Hold'em No Limit - Level ([0-9a-zA-Z]) (([0-9])/([0-9])) - ([0-9])/([0-9])/([0-9]) - ([0-9])🙁[0-9])🙁[0-9]*) (ET)";

$rule_replacement[1] = "TESTER -------------------- Hand Convertor v1.0 -- Poker Stars Tournament NL Texas Hold'em, Freeroll - Level \3 (\4/\5) Tournament #\2, Game #\1, ";

// Split the multi-line variable into an array so that each line is now its own variable
$hand_test = split("\n", $new_hand);

$rule_count = count($rule);

foreach($hand_test as $ht) {

$count = 0;

while ($count <= $rule_count) {
	$skibble[$count] = eregi_replace($rule[$count], $rule_replacement[$count], $ht);
	$count++;
}

}
?>

At this point I'd like the "$skibble" variable to contain the old "$new_hand" information, except with the replaced, formatted text.

I'm sure I'm going to be incredibly embarassed when I see what I've done wrong here.

Weedpacket

If you used the [man]PCRE[/man] functions instead, then you could use [man]preg_grep[/man] to retrieve all lines that match a pattern or any one of an array of patterns. The regexps would be pretty much the same: just stick ... ooh ... # at the start of each, and #i at the end. (The "#" are there to separate the pattern itself from the modifiers on the end - in this case the "i" which indicates that matching should be case-insensitive. Oh, it does mean that you'll need to escape any occurrences of "#" in the pattern as "#".)

gregchet

I'm not sure if this would work for what I need it for, as the elements in the "$new_hand" variable need to be in the same order once they've been put into the new array.

For example, the original array might be:

$hand_test[0] = "NAME folds"
$hand_test[1] = "NAME checks"
$hand_test[2] = "NAME checks"
$hand_test[3] = " THE TURN "
$hand_test[4] = "NAME checks"
$hand_test[5] = "NAME folds"
$hand_test[6] = "NAME folds"

But with the method you mentioned, wouldn't it end up with all the same matches grouped together? Like so:

$skibble[0] = "NAME folds"
$skibble[1] = "NAME folds"
$skibble[2] = "NAME folds"
$skibble[3] = "NAME checks"
$skibble[4] = "NAME checks"
$skibble[5] = "NAME checks"
$skibble[6] = " THE TURN "

When I would need it to be:

$skibble[0] = "NAME folds"
$skibble[1] = "NAME checks"
$skibble[2] = "NAME checks"
$skibble[3] = " THE TURN "
$skibble[4] = "NAME checks"
$skibble[5] = "NAME folds"
$skibble[6] = "NAME folds"

Would it be possible to get a small scaled-down example that I could convert over? I know that's asking a lot, but I've been going nuts trying to figure this out, and my brain is a tad fried at this point 😛

Weedpacket

Um - I'll have to check that (I know with one pattern they'll be in the
right order... ).

Okay; so preg_grep() is for filtering an existing array, keeping only those lines
that match a certain pattern. preg_replace() it is, then.

I'll use your example arrays.

$hand_test = array(
"NAME folds",
"NAME checks",
"NAME checks",
"*** THE TURN ***",
"NAME checks",
"NAME folds",
"NAME folds");

// I tend to do it this way: one array for both search strings and replacements,
// instead of two separate arrays.
$patterns = array(
'/(.*) folds$/' => '<i><b>$1</b> folds</i>',
'/(.*) checks$/' => '<i><b>$1</b> checks</i>',
'/THE TURN/' => '<b>$0</b>');

$search = array_keys($patterns);
$replace = array_values($patterns);
$skibble = preg_replace($search, $replace, $hand_test);

print_r($skibble);

gregchet

Umm, ok, first, I LOVE YOU!! hehe... Not literally of course, but good lord, this is the most progress I've made on this in a while.

I tried making a substitution to reflect a real example, but I'm getting an error message when I try to run it. I'm guessing the syntax for ereg_replace and preg_replace is different? (the syntax for the actual REGEX matching code, that is)

I switched the \1 references to $1 references, but are there other changes that should be made?

New Code

$hand_test = array(
"PokerStars Game #5106862273: Tournament #25456538, Freeroll Hold'em No Limit - Level XIV (2000/4000) - 2006/05/31 - 02:10:52 (ET)",
"NAME folds",
"NAME checks",
"NAME checks",
" THE TURN ",
"NAME checks",
"NAME folds",
"NAME folds");

// I tend to do it this way: one array for both search strings and replacements,
// instead of two separate arrays.
$patterns = array(
'/PokerStars Game #([0-9]): Tournament #([0-9]), Freeroll Hold'em No Limit - Level ([0-9a-zA-Z]) (([0-9])/([0-9])) - ([0-9])/([0-9])/([0-9]) - ([0-9])🙁[0-9])🙁[0-9]) (ET)$/' => '-------------------- Hand Convertor v1.0 -- Poker Stars Tournament NL Texas Hold'em, Freeroll - Level $3 ($4/$5) Tournament #$2, Game #$1, ',
'/(.) checks$/' => '$1 checks',
'/THE TURN/' => '$0');

$search = array_keys($patterns);
$replace = array_values($patterns);
$skibble = preg_replace($search, $replace, $hand_test);

foreach($skibble as $sk) {
echo "$sk ";
}
exit;

The error message

Parse error: parse error, unexpected T_STRING, expecting ')' in /PATH_TO_FILE.inc.php on line 21

Oooh, I just realized that I'm probably going to have to escape the commas in the array now, since that's what separates the array variables. I'll go try that now.

gregchet

WOOHOO! So much progress!!

I was right, it was an escaping issue, but I'm having another very weird problem now.

When I assign the array manually, it works fine, but when I try to run the search and replace on the array I create from my original scalar, it doesn't work.

NOTE: The two snippets of green code are identical in both examples.

Manual code that works

[INDENT]$hand_test = array(
"PokerStars Game #5106862273: Tournament #25456538, Freeroll Hold'em No Limit - Level XIV (2000/4000) - 2006/05/31 - 02:10:52 (ET)",
"NAME folds",
"NAME checks",
"NAME checks",
" THE TURN ",
"NAME checks",
"NAME folds",
"NAME folds");

$patterns = array(
'/PokerStars Game #([0-9]): Tournament #([0-9]), Freeroll Hold\'em No Limit - Level ([0-9a-zA-Z]) (([0-9])\/([0-9])) - ([0-9])\/([0-9])\/([0-9]) - ([0-9])🙁[0-9])🙁[0-9]) (ET)$/' => '-------------------- Hand Convertor v1.0 -- Poker Stars Tournament NL Texas Hold\'em, Freeroll - Level $3 ($4/$5) Tournament #$2, Game #$1 --------------------',
'/(.) checks$/' => '$1 checks',
'/THE TURN/' => '$0');

$search = array_keys($patterns);
$replace = array_values($patterns);
$skibble = preg_replace($search, $replace, $hand_test);

foreach($skibble as $sk) {
echo "$sk ";
}

exit;[/INDENT]

----------
Scalar to Array code that doesn't work

[INDENT]$hand_test = split("\n", $new_hand);

exit;[/INDENT]

However, when I just print out the new "$hand_test" array like below, it looks just fine to me, so it seems like $hand_test[0] should be the exact same as the first array entry in the manual example above, but for some reason it doesn't match.

[INDENT]$hand_test = split("\n", $new_hand);

foreach($hand_test as $ht) {
echo "$ht ";
}
exit;[/INDENT]

Outputs:

[INDENT]PokerStars Game #5106862273: Tournament #25456538, Freeroll Hold'em No Limit - Level XIV (2000/4000) - 2006/05/31 - 02:10:52 (ET)

Table '25456538 16' 9-max Seat #2 is the button

Seat 1: Doktor B.S. (109536 in chips) is sitting out

(etc.)[/INDENT]

Alternatively, if I just use the print_r function instead of the foreach in the last example, like so, this is what it shows:

[INDENT]$hand_test = split("\n", $new_hand);

print_r($hand_test);
exit;[/INDENT]

Outputs (hacked off after [5] to save space here)

Array ( [0] => PokerStars Game #5106862273: Tournament #25456538, Freeroll Hold'em No Limit - Level XIV (2000/4000) - 2006/05/31 - 02:10:52 (ET) [1] => Table '25456538 16' 9-max Seat #2 is the button [2] => Seat 1: Doktor B.S. (109536 in chips) is sitting out [3] => Seat 2: spammond (144198 in chips) [4] => Seat 3: LouisVutton (201503 in chips) [5] => Seat 4: Jester2Dope (85436 in chips)

Is there something I'm missing here?

Ok, it's 4:30am, I think it's time I stopped working on this and got some sleep 😛

gregchet

Ok, so I lied, I wasn't quite going to sleep yet, hehe.

I figured out the problem, but I'm wondering if there's a more code-efficient solution?

Basically, there was extra whitespace on the end of the lines after the explode, so I used rtrim to remove it, like so:

Old

$hand_test = split("\n", $new_hand);

New

$hand_test = split("\n", $new_hand);

$hand_test_count = count($hand_test);

for ($count = 0; $count <= $hand_test_count-1; $count++) {
$hand_test_temp = rtrim($hand_test[$count]);
$hand_test[$count] = $hand_test_temp;
}

rtrim didn't seem to want to work on the array, so I had to assign the array variables to a scalar temporarily, rtrim, and them put the new variable back into the array.

Is there a more efficient way to rtrim the array?

gregchet

Ok, I think I'm gonna cry now.

After all this work, I've discovered that this isn't even gonna frickin' work for what I need it for 🙁

Let's say that 5 of the lines in my array are:

Seat 8: Big_Figga (185578 in chips)
Seat 9: squirt1bad (17162 in chips)
Doktor B.S. posts the ante 200
spammond posts the ante 200

LouisVutton posts the ante 200

If I use the preg_replace method mentioned in my 2nd last post everything works fine, however, there's no way to add an extra to the beginning of the first instance of the "posts the ante" line, is there? So that it would show up like this...

Seat 8: Big_Figga (185578 in chips)
Seat 9: squirt1bad (17162 in chips)

Doktor B.S. posts the ante 200
spammond posts the ante 200

LouisVutton posts the ante 200

Unfortunately, I need to be able to add the extra to the first instance so that it formats properly (otherwise everything's just punched together line after line).

My second issue is, I don't seem to be able to use a scalar variable within an array.

For example, I used to have the following variable setup so that I could use the "$name_replace" variable in any place that it needed to regex search for the player's name (this was done so that if I ever needed to add another allowable character to a player's name, I could just update the variable and all the replace patterns would be updated automatically).

$name_replace = "([a-zA-Z0-9_.&\$\'[:space:]-]*)";

But this doesn't seem to work now that I'm using the array method.

For example, I can't get this to work:

$name_replace = "([a-zA-Z0-9_.&\$\'[:space:]-]*)";

$patterns = array(

'/Seat ([0-9]): $name_replace (([0-9]) in chips)$/' => 'Seat $1:  $2  ($3 in chips)',);

But instead I have to include the name search manually, like so:

$patterns = array(

'/Seat ([0-9]): B[/B] (([0-9]) in chips)$/' => 'Seat $1:  $2  ($3 in chips)',);

So every search patter will have this code instead of the $name_replace variable, and if I ever need to make a change to the allowable characters in a player's name I'll have to update every single rule now, instead of just adding the character to the $name_replace variable and having it update everything else on the fly.

Does anyone have any suggestions for these two problems? Or am I just screwed? hehe

gregchet

Sweet! Ok, I was able to figure out the $name_replace problem.

All I did was switch the quotes around the array variables from single quotes to double quotes, and then removed the slashes from the singles quotes that were in the array variables, like so:

Old

'/Seat ([0-9]): B[/B] (([0-9]) in chips) is sitting out$/' => 'Seat $1:  $2  ($3 in chips) -- sitting out',
'/Seat ([0-9]): B (([0-9]) in chips)$/' => 'Seat $1:  $2  ($3 in chips)',

-----

New

"/Seat ([0-9]): $name_replace (([0-9]) in chips) is sitting out$/" => "Seat $1:  $2  ($3 in chips) -- sitting out",
"/Seat ([0-9]): $name_replace (([0-9]) in chips)$/" => "Seat $1:  $2  ($3 in chips)",

Was there a particular reason the array info was in single quotes in your example? Just want to make sure I'm not gonna break anything else by switching these to double quotes.

gregchet

Does anyone know if this is possible?

Let's say that 5 of the lines in my array are:

Seat 8: Big_Figga (185578 in chips)
Seat 9: squirt1bad (17162 in chips)
Doktor B.S. posts the ante 200
spammond posts the ante 200

LouisVutton posts the ante 200

If I use the preg_replace method mentioned in my 2nd last post everything works fine, however, there's no way to add an extra to the beginning of the first instance of the "posts the ante" line, is there? So that it would show up like this...

Seat 8: Big_Figga (185578 in chips)
Seat 9: squirt1bad (17162 in chips)

Doktor B.S. posts the ante 200
spammond posts the ante 200

LouisVutton posts the ante 200

Unfortunately, I need to be able to add the extra to the first instance so that it formats properly (otherwise everything's just punched together line after line).

Weedpacket

Little bit of advice; use [php]...[/code] around PHP code, and [code]...[/code] around other stuff that relies on whitespace formatting. Makes things much more legible.

I'm wondering if you even need to use split (incidentally, [man]explode[/man] would have been a better choice, there) at all. If none of the patterns will match anything that runs across multiple lines, then everything could be kept in the one string (and maybe exploded at the end if that's still necessary).

With that then using a second preg_replace() to tack " " onto the start of "post the ante" lines (um, "/.*posts the ante/" => " $0") can be done. Usually I'd use str_replace for that, but preg_replace() has an additional parameter to limit the number of times it is applied (in this case, it should only replace the first match).

And on a third incidental note: tags have been deprecated since last century. 🙂

gregchet

I actually did have this setup in one big multi-line variable originally, but there were a lot of weird problems happening with the formatting (I believe they were whitespace issues). This is one of the reasons I started the total re-code in the first place, plus I wanted to have it split up so that I could manipulate the data easier.

I switched over to explode this morning after doing a bit of research (since I don't need to use REGEX and it'll be faster), and I've also got the other issues figured out now too. I have the resulting array import into a database, and then I'm just manipulating it there (like searching for the first instance of the 'posts the ante' line to add the extra break). Just have so much more I can do with it all databased (especially since the rules and replacements are all going to be databased soon too, along with other settings for the software).

I probably should've jumped back on all the forums that I posted on last night/this morning and said not to worry about giving any more info, but I've been zoned out working on this thing all day today, hehe. Which reminds me, I should probably get my *ss to bed here 😛

And yes, I realize font tags are virtually extinct, but this just on my personal site for now and some of my family members are using archaic browsers, so it doesn't make much sense for me to switch it over to css quite yet (but will be doing so before I actually release the software).

Thanks again for all your help! The 2nd post you made with the example code got me off and runnin', and as soon as I decided to database the array information I've been flying, updating code and doing all the grunt work today to get this all nice and clean, hehe.

Thanks again! T'was greatly appreciated.