Collections, strings, numbers and dates

Pathshala’s accountant wants April’s fees per batch from 12 instalment rows, each amount in Indian digit grouping such as 1,23,456.78, and each payment time shown in India although the server stored it in UTC. Laravel ships a tool for each part of that job: collections for lists, Str for strings, Number for numbers and Carbon for dates.

This lesson builds the fee report with collections, formats its strings and numbers for Indian readers, and shows its dates in India’s time. Later parts stream inputs too large for memory, write the short Indian forms that Number lacks, and test the day boundaries that a UTC timestamp crosses.

Collections turn rows into a report, one method at a time

A collectioncollection: A Laravel object that wraps an array and offers chainable methods such as filter, map, groupBy and sum, each returning a new collection. collect() makes one, and Eloquent queries return them. wraps an array in an object whose chaining methods each return a new collection, so a report reads top to bottom as the steps it takes. collect($rows) makes one, and Eloquent queries return them too. The run builds Pathshala’s report from 12 rows:

<?php

// Turns Pathshala's instalment rows into a fee report with one collection pipeline, printing what each step gives.
require __DIR__.'/vendor/autoload.php';
$app = require_once __DIR__.'/bootstrap/app.php';
$app->make(Illuminate\Contracts\Console\Kernel::class)->bootstrap();

$rupees = fn (int $paise) => number_format($paise / 100, 2);
$rows = collect(require __DIR__.'/instalments.php');

echo 'rows: '.$rows->count().PHP_EOL;

// 1. Keep the rows that were paid, then add a month key to each.
$paid = $rows->filter(fn (array $row) => $row['status'] === 'paid')
    ->map(fn (array $row) => $row + ['month' => substr($row['due_on'], 0, 7)]);
echo 'paid rows: '.$paid->count().PHP_EOL;

// 2. Group by batch, then by month, and total each group.
$report = $paid->groupBy(['batch', 'month'])
    ->map(fn ($months) => $months->map(fn ($group) => $group->sum('amount_paise')));

echo PHP_EOL.'collected, by batch and month (rupees):'.PHP_EOL;
foreach ($report as $batch => $months) {
    foreach ($months as $month => $paise) {
        echo '  '.str_pad($batch, 16).$month.'  '.str_pad($rupees($paise), 9, ' ', STR_PAD_LEFT).PHP_EOL;
    }
}

// 3. One total per batch, with reduce.
$byBatch = $paid->groupBy('batch')->map(fn ($group) => $group->reduce(fn (int $sum, array $row) => $sum + $row['amount_paise'], 0));
echo PHP_EOL.'collected per batch: '.$byBatch->map($rupees)->toJson().PHP_EOL;

// 4. Who owes money: overdue rows keyed by student, with their totals.
$overdue = $rows->where('status', 'overdue')->groupBy('student')->map->sum('amount_paise');
echo 'overdue per student: '.$overdue->map($rupees)->toJson().PHP_EOL;

// 5. keyBy keeps the LAST row for each key.
echo 'keyBy(student) on 12 rows keeps '.$rows->keyBy('student')->count().' rows, Asha\'s due_on: '.$rows->keyBy('student')['Asha Verma']['due_on'].PHP_EOL;

<?php

// Twelve fee instalments of Pathshala's two batches for April to June 2031: invented students, amounts in paise.
return [
    ['batch' => 'Maths Morning', 'student' => 'Asha Verma',   'due_on' => '2031-04-05', 'amount_paise' => 118000, 'status' => 'paid'],
    ['batch' => 'Maths Morning', 'student' => 'Kabir Singh',  'due_on' => '2031-04-05', 'amount_paise' => 118000, 'status' => 'paid'],
    ['batch' => 'Physics Evening', 'student' => 'Meera Iyer', 'due_on' => '2031-04-10', 'amount_paise' => 95000,  'status' => 'paid'],
    ['batch' => 'Physics Evening', 'student' => 'Rohan Das',  'due_on' => '2031-04-10', 'amount_paise' => 95000,  'status' => 'overdue'],
    ['batch' => 'Maths Morning', 'student' => 'Asha Verma',   'due_on' => '2031-05-05', 'amount_paise' => 118000, 'status' => 'paid'],
    ['batch' => 'Maths Morning', 'student' => 'Kabir Singh',  'due_on' => '2031-05-05', 'amount_paise' => 118000, 'status' => 'overdue'],
    ['batch' => 'Physics Evening', 'student' => 'Meera Iyer', 'due_on' => '2031-05-10', 'amount_paise' => 95000,  'status' => 'paid'],
    ['batch' => 'Physics Evening', 'student' => 'Rohan Das',  'due_on' => '2031-05-10', 'amount_paise' => 95000,  'status' => 'overdue'],
    ['batch' => 'Maths Morning', 'student' => 'Asha Verma',   'due_on' => '2031-06-05', 'amount_paise' => 118000, 'status' => 'pending'],
    ['batch' => 'Maths Morning', 'student' => 'Kabir Singh',  'due_on' => '2031-06-05', 'amount_paise' => 118000, 'status' => 'pending'],
    ['batch' => 'Physics Evening', 'student' => 'Meera Iyer', 'due_on' => '2031-06-10', 'amount_paise' => 95000,  'status' => 'pending'],
    ['batch' => 'Physics Evening', 'student' => 'Rohan Das',  'due_on' => '2031-06-10', 'amount_paise' => 95000,  'status' => 'pending'],
];

"""A fee report from twelve instalment rows with one collection pipeline: filter, map, groupBy, sum, reduce and keyBy,
each step printed. Files: the folder report/ next to this file (instalments.php holds the rows)."""

from session import run

run(
    """
    php report_probe.php
    """,
    app=True,
    overlay=("report",),
)

$ php report_probe.php
rows: 12
paid rows: 5

collected, by batch and month (rupees):
  Maths Morning   2031-04   2,360.00
  Maths Morning   2031-05   1,180.00
  Physics Evening 2031-04     950.00
  Physics Evening 2031-05     950.00

collected per batch: {"Maths Morning":"3,540.00","Physics Evening":"1,900.00"}
overdue per student: {"Rohan Das":"1,900.00","Kabir Singh":"1,180.00"}
keyBy(student) on 12 rows keeps 4 rows, Asha's due_on: 2031-06-05

Step through what each method left in the collection:

Five of the twelve rows were paid. groupBy(['batch', 'month']) nested the groups by batch and month, and the inner sum gave 2,360.00 rupees for Maths Morning in April. reduce() folds a group into one value with a closure of your own and gave the same per-batch totals. Writing map->sum(...) is a higher-order messagehigher-order message: A short form for a collection method that calls the same method on each item: map->sum('amount_paise') means map with a closure that calls sum('amount_paise') on every item., short for a closure that calls sum on each item.

For an input too large for memory, lazy collectionslazy collection: A collection that produces its items one at a time from a generator instead of holding an array, so its methods can work through an input larger than memory. offer the same methods but read one item at a time.

Str and Number format strings and numbers for readers

Str holds Laravel’s helpers for strings, and Str::of() starts a fluent stringfluent string: The object Str::of() returns: it holds a string and offers Laravel's string helpers as chainable methods, so trimming, changing case and cutting can be written one after another. whose methods chain like a collection’s. Number formats numbers through PHP’s intl extension, whose locale data decides the grouping:

"""Strings, numbers and dates for a fee reminder: Str and fluent strings, Number with the en and en_IN locales, and
Carbon storing UTC while showing Asia/Kolkata. Files: the folder formats/ next to this file."""

from session import run

run(
    """
    php formats_probe.php
    """,
    app=True,
    overlay=("formats",),
)

$ php formats_probe.php
-- strings
name as typed                     '  asha   VERMA '
Str::of()->squish()->title()      Asha Verma
Str::slug(title)                  fee-reminder-april-2031-batch
Str::limit(text, 24)              Your April instalment fo...
Str::mask(phone, *, 3, 5)         +91*****00001
Str::headline(column name)        Amount Paise

-- numbers
Number::format(en)                123,456.78
Number::format(en_IN)             1,23,456.78
Number::currency(INR, en_IN)      ₹1,23,456.78
Number::currency(INR, en)         ₹123,456.78
Number::percentage(0.875 * 100)   87.5%
Number::forHumans(12345678)       12 million
Number::abbreviate(12345678)      12M

-- dates
app timezone                      UTC
stored (UTC)                      2031-04-05 09:30:00
shown in Asia/Kolkata             2031-04-05 03:00 PM
due date + 10 days                2031-04-15
days between 5 Apr and 15 May     40

The run squished a name typed as ' asha VERMA ' into Asha Verma, masked five digits of a phone number and cut a reminder to 24 characters. The same amount printed 123,456.78 in the en locale and 1,23,456.78 in en_IN, and the currency kept that grouping. Number::forHumans gives the short international form, 12 million.

Carbon stores UTC and shows the reader’s time

A new Laravel 13 project runs in UTC (config('app.timezone') printed UTC), and Carbon, the library behind now() and every date cast, keeps dates in that zone. Store in UTC and convert only when you show a value:

Carbon also does the calendar arithmetic, such as adding 10 days to a due date or counting the 40 days between two due dates in the run. India is 5 hours 30 minutes ahead of UTC, so dates near midnight change their day in India, and Carbon formats dates in Hindi as well. A daily report must cut its days at India’s midnight.

Lazy collections and cursors for large inputs, with memory and queries measured Pro: opens with a Pro, Premium or Ultimate pass

Short Indian amounts, Hindi slugs and dates, and the day boundaries of a UTC timestamp, recorded Pro: opens with a Pro, Premium or Ultimate pass

Practise

Build Pathshala’s monthly collection report in your own project:

1. Put the 12 instalment rows of this lesson in a collection, and print the total paid per batch and month with groupBy() and sum().
2. Add a column that shows each total with Number::currency(..., in: 'INR', locale: 'en_IN').
3. Store one payment at 09:30 UTC and print it in Asia/Kolkata with Carbon.

The questions below check the first three sections:

Interview questions

Warm-up: what is the difference between map() and each() on a Laravel collection?

map() returns a new collection with the result of the callback for every item, and leaves the original as it was. each() runs the callback for its side effect, returns the same collection, and stops early if the callback returns false. Use map() to transform values and each() to act on them.

These three take the lesson’s tools into harder cases:

Key takeaways

• Collections chain small steps into a report: filter, map, groupBy and sum produced four batch-month totals from twelve rows, and keyBy keeps only the last row of a repeated key.
• Str and fluent strings clean and shorten text, and Number formats numbers with the locale’s grouping, so en_IN prints 1,23,456.78.
• Carbon keeps dates in UTC, the app’s time zone, and converts them only when a value is shown in India’s time.