TIL: looks like strtotime() ignores fractional seconds when converting an ISO 8601 timestamp:

$ php -r 'var_dump(strtotime("2021-09-21T22:19:51.689+00:00"));'
int(1632262791)
$ php -r 'var_dump(strtotime("2021-09-21 15:19:52"));'
int(1632262792)

    Also learned that floor(), ceil(), and round() return floats for some reason:

    $ php -r 'var_dump(ceil("1.1"));'
float(2)
$ php -r 'var_dump(floor("1.1"));'
float(1)
$ php -r 'var_dump(round("1.1"));'
float(1)

    And now that you've made me look at the manual, TIL that you can specify a negative precision for round():

    20:29 $ php  -a
Interactive shell

php > $foo = round(1234.567, -2);
php > var_export($foo);
1200.0
php >

    I've had a rule of thumb which is "don't start by assuming the guy before you was totally incompetent". I've got a more concise name for it now; Chesterton's Fence.

    I admit mine is a little more general and spills over into another rule of thumb of mine: "If you think you have to force it you're probably doing it wrong."

    7 days later

    TIL the interesting attribute of the number 6174: https://youtu.be/d8TRcZklX_Q

    For the fun of it, I came up with this PHP script, which I'm sure could be improved/streamlined. 🙂

    <?php
// Fun with 6174

if(empty($argv[1])) {
  die("Usage: ".basename(__FILE__)." nnnn\n    where 'nnnn' is a 4-digit number\n");
}
if(!preg_match('/^[0-9]{4}$/', $argv[1])) {
  die("Usage: " . basename(__FILE__) . " nnnn\n    where 'nnnn' is a 4-digit number\n");
}
$digits = str_split($argv[1]);
$unique = array_unique($digits);
if(count($unique) == 1) {
  die("The number cannot use the same value for all 4 digits.\n");
}
while(true) { // Danger, Will Robinson!
  sort($digits);
  $asc = implode('', $digits);
  rsort($digits);
  $desc = implode($digits);
  $diff = $desc - $asc;
  $diff = str_pad($diff, 4, '0', STR_PAD_LEFT);
  echo "$desc - $asc = $diff\n";
  if($diff == '6174') {
    die("7641 - 1467 = 6174\n");
  }
  $digits = str_split($diff);
}

    Weedpacket Now try and find a number that gives the longest path to 6174.

    My test came up with a max of 7 iterations, and applied to 2184 numbers. A bunch of those are duplicates of a sort, since 1234 will be the same as 4321 or 1432, etc.

    PS: Further coding determined (assuming no bugs) there are 116 unique sets of 4 digits that have 7 iterations.

      5 days later

      TIL: You can use (at least some?) expressions in an SQL order by clause, e.g.:

      select predicate_type_id
from some_schmea.predicate_type
order by mod(predicate_type_id, 2), predicate_type_id

      Output:

      4
6
8
10
12
14
16
1
7
9
11
13
15
17
      2 months later

      TIL that not sanitizing inputs can have far-reaching and long-lasting effects.

      Maybe I knew that in 2013, even. #WhyDidntTheApacheGuys

      dalecosp Today I learned logging frameworks really shouldn't be given the power to download and run code from off the Internet.

        a month later

        TIL about the PostgreSQL replace() function. (Nothing earth-shaking, just never ran into a situation before that had me searching for how to do what it does -- always did text replacing within the PHP code, I guess.)

          5 days later

          TIL PostgreSQL exposes the physical position of a row in a table by giving each table a hidden ctid column. No, you can't use this as a substitute for a primary key; anything that causes rows to be rearranged will cause their ctids to change.

          Despite this, they remain stable for long enough if you have one of those tables that don't have a proper primary key and end up with duplicate rows where you want to get rid of the duplicates.

          DELETE FROM tablename
WHERE ctid NOT IN (
    SELECT MIN(ctid)
    FROM tablename
    GROUP BY tablename.*
)

          So, grouping all identical rows together, and selecting the minimum ctid from reach group. Those are the rows in the table that get kept (one for each group of duplicate rows). Delete all the other rows.

          If you have a primary key in your table (as you generally should) then you don't need this; group by the columns other than the primary key, and use max or min or something to pick a representative primary key value. That's assuming you don't mind any dependent rows in other tables being deleted as well. It's more likely you'd want them to repoint to the definitive copy of the row you're keeping.

            13 days later

            PHP 8 removed the each() function. (I'm in the process of updating a PHP5 app to 8. Got the API test suite to pass, but now I have to look at the things not covered by that test. 🙂 )

            3 months later

            TIL: var_export() returns null for a PHP resource (which led me down a false debug path until I realized this).

            php > $context = stream_context_create($test);
php > print_r($context);
Resource id #1
php > var_dump($context);
resource(1) of type (stream-context)
php > var_export($context);
NULL
              2 months later
              6 days later

              TIL: the strftime() function is deprecated in PHP 8.1. In order to avoid deprecation warnings in a legacy application that we recently upgraded to 8.1, I opted to use https://gist.github.com/bohwaz/42fc223031e2b2dd2585aab159a20f30 and then included it into the only class that needs it with:

              <?php

namespace Foo\Bar;

use Foo\this;
use Foo\That;

// Avoid strftime() deprection warnings
// This approach avoids having to change anything else in this class
require_once __DIR__ . '/strftime.php';
use function \PHP81_BC\strftime;

class TheProblemClass {
  // lots of stuff, including one method that does a whole bunch
  // of things with date/time format strings like you use in strftime()
  // but only ever calls strftime() on one line
}

              This seemed better -- at least regarding time/effort on my part -- than trying to figure out what I would need to change in the several places it used the date/time format string. 🙂

                2 months later

                Today I learned Queen Elizabeth II died

                Rest in peace

                  3 months later