Brand-Brothers

• Forms
• Equipment
• Technical (Equipment )
• Fotobooth
• Sets and availabilty

Forms

Handleiding

Deze handleiding beschrijft de verschillende parameters die je kunt gebruiken bij het configureren van een bier selectie formulier.

Het form wordt aangeroepen door middel van een URL en alle parameters worden in de URL meegegeven.

Voorbeeld:

https://....../.../brand-brothers/forms/?eigenaar=naam&merk=Texels&set=Texels-vuurtorenbar&minbeers=3&beerfilter=amstel&options=1,2,3,4,25,26&t=1&maxbeers=3

Spaties in URL

In de gehele URL moeten spaties worden gecodeerd als %20.

Parameters

De parameters zoals in het voorbeeld getoond, worden hieronder beschreven. Elke parameter is rood gemarkeerd.

minbeers

• Beschrijving: Het minimaal aantal bieren dat geselecteerd moet worden.
• Werking: Als er minder bieren zijn geselecteerd dan het minimum, krijg je een waarschuwing wanneer je probeert het formulier te versturen.

maxbeers

• Beschrijving: Het maximaal aantal bieren dat geselecteerd kan worden.
• Werking: Als je meer bieren selecteert dan het maximum, verschijnt er onmiddellijk een waarschuwing.

Verborgen Velden

• Beschrijving: De volgende velden zijn verborgen en kunnen via de URL worden meegegeven. De verborgen velden worden dor het formulier meegestuurd en verschijnen in de email.
  • eigenaar: De eigenaar van het biermerk.
  • merk: De naam van het biermerk.
  • set: De naam van de set die in de titel van het formulier verschijnt.

t (Testmodus)

• Beschrijving: Hiermee kun je de testmodus van het formulier instellen. aan t kunnen de volgende waarden worden gegeven:
  • 0: Normale werking (dezelfde als geen waarde opgegeven).
  • 1: Alle verplichte velden zijn niet meer verplicht en het formulier wordt niet per e-mail verzonden.
  • 2: Alle verplichte velden zijn niet meer verplicht en het formulier wordt wel per e-mail verzonden

options

• Beschrijving: Een lijst van opties die als een komma-gescheiden lijst moeten worden opgegeven. Deze opties worden getoond op het formulier.
  

beerfilter

• Beschrijving: Een filter om bieren op naam te selecteren.
• Werking: Als de waarde 'none' is, wordt de gehele bierselectie niet getoond. Je kunt meerdere filters specificeren, bijvoorbeeld: beerfilter=amstel,brand.

Options aanpassen

Om options aan te passen kan deze URL worden gebruikt:

https://....../.../brand-brothers/forms/options

of nieuwe interface

https://....../.../brand-brothers/forms/options/index2.php

Optienummer

Elke optie heeft een nummer. Aan de hand van dit nummer wordt de optie op het formulieer geplaatst. De nummers hoeven niet uniek te zijn. Wanneer meerdere opties hetzelfde nummer hebben, worden alle opties op het fomrulier getoond wanneer deze optie wordt gekozen.

Optienaam

De naam van de optie is de naam die op de aanvraag verschijnt (email).

Optielabel

Het label is de de naam zoals die op het formulier verschijnt.

Extra naam

De extra naam is de (open) vraag die wordt gesteld indien de optie wordt geslecteerd. De naam is de naam zoals die op het aanvraag (email) verschijnt.

Extra label

Het extra label is de (open) vraag die wordt gesteld indien de optie wordt geslecteerd. Het label is de naam zoals die op het formulier verschijnt.

URL naar options (back-up)

De 'database' waarin de options worden opgeslagen kun je via de volgende URLs controleren:

https://.../.../brand-brothers/forms/options/options.js
https://.../.../brand-brothers/forms/options/options.json

Deze bestanden zou je ook als (extra) back-up kunnen gebruiken.

De twee bestanden bevatten dezelfde informatie, maar zijn om technische redenen op een andere wijze vormgegeven.

Options kunnen niet worden aangepast?

Als de opties niet kunnen worden aangepast, controleer dan de twee bestanden. Zijn de wijzigingen in deze bestanden wel aangepast dan zorgt de cache van de browser ervoor dat de wijzigingen niet (direct) worden getoond. Laadt de pagina dan opnieuw en leeg de browser cache (dit werkt op elke browser anders).

Beers aanpassen

(dit is nog nooit in productie getest)

https://............/brand-brothers/forms/beers

of nieuwe interface

https://............/brand-brothers/forms/beers/index2.php

Een beer heeft een ID en een naam zoals dat op het form te zien is. Zorg ervoor dat de ID's uniek blijven.

--

Equipment

Equipmentlist

(in ontwikkeling)

https://..../brand-brothers/equipment-list/folders.php

--

Technical (Equipment )

Equipment

folder -> equipment -> 

Performance

Plaatjes

Plaatjes staan in Amazon Cloud en zijn 200kb tot 500kb.

No Cache, no optimazation

Het laden van 5 items duurt 1 tot 2.5 seconden afhankelijk van de grootte van de plaatjes.

Async laden plaatjes

Laadtijden per 5 items tussen de 0.6 en 1.2 seconden zijn.  

Plaatjes kleiner maken (optimized for web)

Het laden van 5 items duurt 2 ms.. 200 to 500 maal sneller.

API Call optimization

API calls naar rentman kosten gemiddeld 120 ms per call (dat is snel); voor elk item heb je wel 3 call nodig (item data, plaatje data, calender data).
Dat is dus 360 ms per item.

Via cache is dat < 1 ms.

Mogleijk optimalisatie is parrallel async  verwerking.

Implementation
all getc api calls are cached. The cache has a predefiend TTL $cache_ttl = 3600
the raw return value of the API is json encoded written to file.

There is code to delete all the cache (not tested yet)

Rerender plaatje sop server

Let op dat de GD library op de server actief moet zijn.

Amazon picture load

TLS Handshake and Connection Setup: 100 ms
s3-eu-west-1.amazonaws.com - ping time 27 ms
server processing - 10 ms
Network tranfer speed 0.8 sec/KB

Cache of rentman API

all getc api calls are cached. The cache has a predefiend TTL $cache_ttl = 3600
the raw return value of the API is json encoded written to file.
There is code to delete all the cache (not tested yet)

Images cache

Images are cached as images (binaries) and are re-rendered at a lower resolution.
Images are cached under the name [equipment-id]-[image-modified].jpg
When equipment ids are not valid any more, there could be orphan images in the cache.

When the image is updated for an exisitng equipment id, there will be two files with the same equipment-id,
the one with the lowest image-modified timestamp can be deleted.
There is no code for this yet.

Performance test

• • Orginele laadtijd: 5,4 seconden (1.8 MB plaatjes)
  • Img cache v no cache 2600/3500 ms - 2 ms
  • API call cache  v no cache: 1700 ms - 2 ms
  • na cache plaatjes/ api calls 2 ms

Rechten op directory (Ubuntu)

// create new group
sudo groupadd shared-www

// add users to group
sudo usermod -aG shared-www www-data
sudo usermod -aG shared-www max

// set shared group as owner
sudo chown -R :shared-www /path/to/shared-directory
sudo chmod -R 770 /path/to/shared-directory

// Ensure that any new files or directories created inside the shared directory inherit the group ownership of shared-group.
sudo chmod g+s /path/to/shared-directory

Fotobooth

Code voor Arduino (1ste versie)

#include <Keyboard.h>

const int inputPin = 12; // Pin connected to Nayax output
int lastState = LOW;     // Previous state of the input pin
int count;

void setup() {
  pinMode(inputPin, INPUT);         // Set the pin as input
  digitalWrite(LED_BUILTIN, HIGH);  // initialize digital pin LED_BUILTIN as an output.
  pinMode(LED_BUILTIN, OUTPUT);     // turn the LED on (HIGH is the voltage level)
  Keyboard.begin();                 // Initialize the keyboard emulation
  Serial.begin(9600);
  count = 0;
  delay(5000);                       // Delay to give time to connect the Arduino
  digitalWrite(LED_BUILTIN, LOW);    // turn the LED off by making the voltage LOW
}

void loop() {

  int currentState = digitalRead(inputPin);  // Read the Nayax signal

  // Check if the state has changed from LOW to HIGH     
  if (currentState == HIGH && lastState == LOW) {
    count++;
    Serial.print("Pulse: ");
    Serial.println(count);   
    digitalWrite(LED_BUILTIN, HIGH);   // turn the LED on (HIGH is the voltage level)
    Keyboard.press(0x20);              // press space
    delay(100); 
    Keyboard.releaseAll();             // Release the key
    delay(1000);                        // Debounce delay to prevent multiple triggers
    digitalWrite(LED_BUILTIN, LOW);    // turn the LED off by making the voltage LOW
  }

  lastState = currentState;  // Update the last state
}

Aansluitschema

Arduino 5v - Optocoupler +
Arduino  GRND - weerstand 10K - pin12
Arduino Pin 12 - GRND Optocoupler
Optocoupler primair 12 volt via 10K weerstand

Pin 12 wordt met een pull down weerstand laag (0) gehouden. De (1) is 5V wordt via de optocoupler verbonden aan pin12 als er een signaal via de optocoupler binnenkomt. De stroom die door de 10K weerstand loopt als de optocoupler 'aan' staat, is 0.5 mA (2.5 mW).

De 10K weerstand voor de 12v primaire kant v/d optocoupler is proefondervindelijk vastgesteld. Bij 12v en 10K weerstand staat er 5v op de primare ingang en werkt de optocoupler goed. Meer spanning (lager weerstand kan) volgens de specs maar is onnodig.

Explanation primary side optocoupler

Primary (Input) Side: Connecting to Nayax

Explanation of Behavior

New code (v2) to ignore small pulses

#include <Keyboard.h>

const int inputPin = 12; // Pin connected to Nayax output
int lastState = LOW;     
int count;
unsigned long lastPulseTime = 0;
const unsigned long debounceDelay = 1000; // Minimum time between pulses (ms)
const unsigned long pulseValidationTime = 10; // Must stay HIGH at least this long (ms)

void setup() {
  pinMode(inputPin, INPUT);
  pinMode(LED_BUILTIN, OUTPUT);
  digitalWrite(LED_BUILTIN, HIGH);
  Keyboard.begin();
  Serial.begin(9600);
  count = 0;
  delay(5000);
  digitalWrite(LED_BUILTIN, LOW);
}

void loop() {
  int currentState = digitalRead(inputPin);

  if (currentState == HIGH && lastState == LOW) {
    // Wait a bit to check if the signal is stable
    delay(pulseValidationTime);
    if (digitalRead(inputPin) == HIGH) {
      unsigned long now = millis();
      if (now - lastPulseTime > debounceDelay) {
        count++;
        Serial.print("Pulse: ");
        Serial.println(count);
        digitalWrite(LED_BUILTIN, HIGH);
        Keyboard.press(0x20);
        delay(100);
        Keyboard.releaseAll();
        digitalWrite(LED_BUILTIN, LOW);
        lastPulseTime = now;
      }
    }
  }

  lastState = currentState;
}

Sets and availabilty

Step 1

Reformat plannedequipment into reservations per date

Processes equipment reservation data.

• Reads reservation records from JSON files in the "plannedequipment" directory.
• Reads equipment properties from JSON files in the "equipment" directory.
  

Calculates daily reservation totals for each equipment item.

• Covers a date range: current month plus a configurable number of months (default 9).
• Organizes data into a nested structure: year → month → day.

Computes daily available quantity: current quantity minus reservations.
Outputs results to JSON files in the "reservationperdate" directory.
Offers optional features:

• Suppression of days with zero reservations.
• Prevention of file creation when no reservation data exists.

Script

# createReservationperdatev2.py

import os
import json
from datetime import datetime, date, timedelta
import calendar

# Parameters
SUPPRESS_ZERO = True           # If True, skip storing day nodes with zero reservations.
SUPPRESS_EMPTY_FILE = True     # If True, do not create an output file if the "dates" part is empty.
ADDITIONAL_MONTHS = 9          # Number of additional months to process (beyond the current month)

# Define the directories
plannedequipment_dir = 'plannedequipment'
equipment_dir = 'equipment'
output_dir = 'reservationperdate'
os.makedirs(output_dir, exist_ok=True)

# Define the date range: from the first day of this month to the last day of the month ADDITIONAL_MONTHS later.
today = date.today()
start_date = date(today.year, today.month, 1)
end_month = today.month + ADDITIONAL_MONTHS
end_year = today.year + (end_month - 1) // 12
end_month = ((end_month - 1) % 12) + 1
last_day = calendar.monthrange(end_year, end_month)[1]
end_date = date(end_year, end_month, last_day)

# Define the list of equipment property keys to copy from the equipment JSON files.
equipment_keys = ["is_combination", "is_physical", "current_quantity"]

def add_day(nested_dates, current_date):
    """Ensure that nested_dates[year][month][day] exists and initialize it."""
    year = str(current_date.year)
    month = f"{current_date.month:02d}"
    day = f"{current_date.day:02d}"
    
    if year not in nested_dates:
        nested_dates[year] = {}
    if month not in nested_dates[year]:
        nested_dates[year][month] = {}
    if day not in nested_dates[year][month]:
        nested_dates[year][month][day] = {"reservations": 0, "available": None}

def remove_empty_containers(nested_dates):
    """Remove empty month and year nodes if they have no day entries."""
    years_to_remove = []
    for year, months in nested_dates.items():
        months_to_remove = []
        for month, days in months.items():
            if not days:  # If month dictionary is empty
                months_to_remove.append(month)
        for month in months_to_remove:
            del nested_dates[year][month]
        if not nested_dates[year]:
            years_to_remove.append(year)
    for year in years_to_remove:
        del nested_dates[year]

# Process each JSON file in the plannedequipment directory.
for filename in os.listdir(plannedequipment_dir):
    if not filename.endswith('.json'):
        continue

    # Build full paths for plannedequipment, equipment and output files.
    plan_file_path = os.path.join(plannedequipment_dir, filename)
    equipment_file_path = os.path.join(equipment_dir, filename)
    output_path = os.path.join(output_dir, filename)

    # Load the reservation data from the plannedequipment file.
    with open(plan_file_path, 'r', encoding='utf-8') as infile:
        plan_data = json.load(infile)

    # Initialize the nested dates dictionary.
    nested_dates = {}
    current_date = start_date
    while current_date <= end_date:
        add_day(nested_dates, current_date)
        current_date += timedelta(days=1)

    # Process each reservation record.
    for record in plan_data.get('data', []):
        start_str = record.get('planperiod_start')
        end_str = record.get('planperiod_end')
        if not start_str or not end_str:
            continue

        record_start = datetime.fromisoformat(start_str).date()
        record_end = datetime.fromisoformat(end_str).date()
        # Clamp the reservation period to our target range.
        period_start = max(record_start, start_date)
        period_end = min(record_end, end_date)
        if period_start > period_end:
            continue

        quantity = record.get('quantity', 0)
        current = period_start
        while current <= period_end:
            year = str(current.year)
            month = f"{current.month:02d}"
            day = f"{current.day:02d}"
            if year in nested_dates and month in nested_dates[year] and day in nested_dates[year][month]:
                nested_dates[year][month][day]["reservations"] += quantity
            current += timedelta(days=1)

    # Initialize equipment properties.
    item_properties = {}
    current_quantity = 0

    # Read the corresponding equipment file.
    if os.path.exists(equipment_file_path):
        with open(equipment_file_path, 'r', encoding='utf-8') as eq_file:
            equipment_data = json.load(eq_file)
        equipment_item = equipment_data.get('data', {})
        # Copy the desired properties.
        for key in equipment_keys:
            item_properties[key] = equipment_item.get(key, None)
        current_quantity = equipment_item.get('current_quantity', 0)
    else:
        print(f"Error: Equipment file '{filename}' not found in '{equipment_dir}'. Assuming current_quantity = 0 and default item properties.")
        for key in equipment_keys:
            item_properties[key] = None

    # Update each day with the available quantity and optionally suppress days with zero reservations.
    for year in list(nested_dates.keys()):
        for month in list(nested_dates[year].keys()):
            for day in list(nested_dates[year][month].keys()):
                day_data = nested_dates[year][month][day]
                reservations = day_data["reservations"]
                day_data["available"] = current_quantity - reservations
                if SUPPRESS_ZERO and reservations == 0:
                    del nested_dates[year][month][day]
            # If a month becomes empty after removing days, remove the month.
            if not nested_dates[year][month]:
                del nested_dates[year][month]
        # If a year becomes empty after removing months, remove the year.
        if not nested_dates[year]:
            del nested_dates[year]

    # If SUPPRESS_EMPTY_FILE is True and the nested dates dictionary is empty,
    # remove the output file if it exists and skip writing.
    if SUPPRESS_EMPTY_FILE and not nested_dates:
        if os.path.exists(output_path):
            os.remove(output_path)
            print(f"Removed existing file {output_path} because there are no reservations.")
        continue

    # Combine the item properties and the nested date reservation data.
    output_data = {
        "item": item_properties,
        "dates": nested_dates
    }

    # Write the result to the output directory.
    with open(output_path, 'w', encoding='utf-8') as outfile:
        json.dump(output_data, outfile, indent=2)

print("Reservation, availability, and item data have been processed and saved.")

Step 2

Download the set hiarchy from the API

(takes long, build in delays to prevent throttling)

# getSets.py

import json
import time
import requests
import sys

# Configuration variables (adjust these as needed)
BASE_URL = "https://api.rentman.net/"  # Replace with your actual base URL

FILTER_ON_TAGS = False  # Set to True if you want to apply tag filtering
API_TOKEN = "abc"  # Replace with your actual API token

def log_message(message, force=False):
    """Logs a message. You can modify this to log to a file if needed."""
    sys.stderr.write(message + "\n")
    if force:
        print(message)

def count_api_usage():
    """Stub function for counting API usage."""
    # Extend this function to track API usage if needed.
    pass

def filter_array_by_selections(items, selections):
    """
    Filters an array of dictionaries based on the presence of certain keys with truthy values.

    :param items: List of dictionaries to filter.
    :param selections: List of keys to check.
    :return: Filtered list of dictionaries.
    """
    filtered = []
    for item in items:
        for sel in selections:
            if item.get(sel):
                filtered.append(item)
                break
    return filtered

def make_request(method, endpoint, data=None):
    """
    Makes an HTTP request using the requests library.

    :param method: HTTP method ('GET' or 'POST').
    :param endpoint: API endpoint (appended to BASE_URL).
    :param data: Dictionary of data or query parameters.
    :return: Parsed JSON response or None if error occurs.
    """
    if data is None:
        data = {}

    url = BASE_URL + endpoint
    headers = {
        "Authorization": f"Bearer {API_TOKEN}",
        "Content-Type": "application/json"
    }

    try:
        if method.upper() == 'POST':
            response = requests.post(url, headers=headers, json=data)
        elif method.upper() == 'GET':
            response = requests.get(url, headers=headers, params=data)
        else:
            log_message(f"Unsupported HTTP method: {method}")
            return None
    except requests.RequestException as e:
        log_message(f"Request exception: {e}")
        return None

    if not response.ok:
        log_message(f"HTTP error: {response.status_code} | Response Body: {response.text}")
    try:
        return response.json()
    except json.JSONDecodeError:
        log_message("Error decoding JSON response.")
        return None

def get_request(endpoint, params=None):
    """
    Performs a GET request and counts API usage.

    :param endpoint: API endpoint.
    :param params: Query parameters as a dictionary.
    :return: Parsed JSON response.
    """
    count_api_usage()
    return make_request('GET', endpoint, params)

def equipmentsetscontent(filter_on_tags=FILTER_ON_TAGS):
    """
    Retrieves equipment sets content by fetching data in batches.

    :param filter_on_tags: If True, applies tag filtering.
    :return: Complete list of items.
    """
    all_items = []
    offset = 0
    batch_size = 300
    total_fetched = 0

    while True:
        params = {'offset': offset, 'limit': batch_size}
        response = get_request('equipmentsetscontent', params)

        if not response or 'data' not in response:
            print("<pre><h1>Error in API Request</h1>")
            print(response)
            sys.exit(1)

        items = response['data']
        all_items.extend(items)
        total_fetched += len(items)

        # If fewer than batch_size items are returned, no more records are available.
        if len(items) < batch_size:
            break

        offset += batch_size

        # If 3000 records have been fetched, pause for 1 second to avoid rate limits.
        if total_fetched >= 3000:
            log_message("Pausing for 1 second to avoid rate limits...", True)
            time.sleep(1)
            total_fetched = 0  # Reset the counter after the pause

    if filter_on_tags:
        all_items = filter_array_by_selections(all_items, ['hoofditem', 'set'])

    return all_items

# Main execution
if __name__ == "__main__":
    complete_data = equipmentsetscontent()
    print(json.dumps(complete_data, indent=4))

Step 3,

Convert the output from step 2 into sets-childs format (parrent with list of childs).

For each entry, the script removes the "/equipment/" prefix from both the parent and child equipment values using the helper function.

For every entry, after stripping the prefixes, a new dictionary is created for the child item (containing the stripped child equipment and its quantity). This child dictionary is then appended to the list associated with the corresponding parent key in the dictionary.

Once all entries are processed, the dictionary keys (i.e., the parent equipment numbers) are sorted in ascending order to organize the final output.

Prompt

Write a Python script that performs the following tasks:

• Define variables at the top of the script for an input file and an output file. The input file is "sets/sets.json" and the output file is "sets/sets-childs.json".
• Read the JSON data from the input file. The file contains a list of objects, each with fields such as "parent_equipment", "equipment", and "quantity". The "parent_equipment" and "equipment" fields are strings that start with the prefix "/equipment/".
• Create a helper function to remove the "/equipment/" prefix from these strings.
• Group the objects by the numeric part of the parent equipment (after stripping the prefix). For each group, store a list of child items, where each child is represented by a dictionary containing the stripped "equipment" number and its "quantity".
• Sort the groups by their parent equipment keys in ascending numerical order.
•  Write the resulting grouped and sorted data to the output file in JSON format with proper indentation.
• Please generate the complete Python code to achieve this.

# set-combined.py

import json
from collections import defaultdict

# Define file paths
input_file = "sets/sets.json"
output_file = "sets/sets-childs.json"

def strip_prefix(equipment_str):
    prefix = "/equipment/"
    return equipment_str[len(prefix):] if equipment_str.startswith(prefix) else equipment_str

# Load the input JSON file
with open(input_file, "r") as infile:
    data = json.load(infile)

# Group child items by parent equipment (after stripping the prefix)
sets_dict = defaultdict(list)
for entry in data:
    parent = strip_prefix(entry["parent_equipment"])
    child_item = {
        "equipment": strip_prefix(entry["equipment"]),
        "quantity": entry["quantity"]
    }
    sets_dict[parent].append(child_item)

# Sort the dictionary by parent keys in ascending numerical order
sorted_sets_dict = {k: sets_dict[k] for k in sorted(sets_dict.keys(), key=lambda x: int(x))}

# Write the output JSON file with sorted parent keys
with open(output_file, "w") as outfile:
    json.dump(sorted_sets_dict, outfile, indent=4)

print("The new JSON file has been created successfully.")

Step 4

Calculate set availability into reservationperdate-sets folder.

Prompt: prompt to create set reservations.docx

#!/usr/bin/env python3
# setReservationperdate.py

import os
import json
import math
from collections import defaultdict

# Directories for input and output files
SETS_FILE = os.path.join("sets", "sets-childs.json")
RESERVATION_DIR = "reservationperdate"
OUTPUT_DIR = "reservationperdate-sets"

# Ensure output directory exists
os.makedirs(OUTPUT_DIR, exist_ok=True)

def load_json(filepath):
    with open(filepath, "r") as f:
        return json.load(f)

def get_child_reservation(child_id):
    """
    Given a child equipment id, load its reservation file.
    Assumes filename is child_id + ".json" in the RESERVATION_DIR.
    Returns a dict with keys "item" and optionally "dates".
    """
    filepath = os.path.join(RESERVATION_DIR, f"{child_id}.json")
    if os.path.exists(filepath):
        return load_json(filepath)
    else:
        raise FileNotFoundError(f"Reservation file for equipment {child_id} not found at {filepath}")

def extract_date_availability(reservation):
    """
    Extract a mapping from (year, month, day) tuple to available quantity.
    If a reservation file has no "dates" node, returns an empty dict.
    """
    date_avail = {}
    dates = reservation.get("dates", {})
    for year, months in dates.items():
        for month, days in months.items():
            for day, data in days.items():
                date_avail[(year, month, day)] = data.get("available", 0)
    return date_avail

def nested_date_structure(date_dict):
    """
    Given a mapping with keys (year, month, day) and value (availability),
    build a nested dictionary: year -> month -> day -> {reservations: 0, available: value}
    """
    nested = defaultdict(lambda: defaultdict(dict))
    for (year, month, day), avail in date_dict.items():
        nested[year][month][day] = {"reservations": 0, "available": avail}
    return {year: dict(months) for year, months in nested.items()}

def main():
    # Load the sets definitions
    sets_def = load_json(SETS_FILE)
    
    # Process each set defined in sets-childs.json
    for set_id, children in sets_def.items():
        child_infos = []  # List of tuples: (child_id, required_quantity, full_stock, date_availability mapping)
        for child in children:
            child_id = child["equipment"]
            req_qty = child["quantity"]
            try:
                reservation = get_child_reservation(child_id)
            except FileNotFoundError as e:
                print(e)
                continue  # Skip this child if its reservation file is missing
            full_stock = reservation.get("item", {}).get("current_quantity", 0)
            date_avail = extract_date_availability(reservation)
            child_infos.append((child_id, req_qty, full_stock, date_avail))
        
        # Calculate the full availability (without date restrictions)
        available_sets_full = math.inf
        for (_, req_qty, full_stock, _) in child_infos:
            child_full_sets = full_stock // req_qty
            available_sets_full = min(available_sets_full, child_full_sets)
        if available_sets_full == math.inf:
            available_sets_full = 0

        # Gather the union of all dates from the children
        union_dates = set()
        for (_, _, full_stock, date_avail) in child_infos:
            union_dates.update(date_avail.keys())
        
        # Compute availability per date from the union of dates
        date_results = {}
        for date in union_dates:
            avail_for_date = math.inf
            for (_, req_qty, full_stock, date_avail) in child_infos:
                # Use date-specific available if provided, otherwise use full stock
                child_avail = date_avail.get(date, full_stock)
                child_sets = child_avail // req_qty
                avail_for_date = min(avail_for_date, child_sets)
            date_results[date] = avail_for_date
        
        # Only include dates where the available set count is lower than full availability
        filtered_dates = {date: avail for date, avail in date_results.items() if avail < available_sets_full}

        # Build output JSON structure following the reservation file structure
        output = {}
        output["item"] = {
            "is_combination": True,
            "current_quantity": available_sets_full
        }
        if filtered_dates:
            output["dates"] = nested_date_structure(filtered_dates)
        
        # Write the output JSON file for the set to the OUTPUT_DIR
        out_filepath = os.path.join(OUTPUT_DIR, f"{set_id}.json")
        with open(out_filepath, "w") as outfile:
            json.dump(output, outfile, indent=4)
        print(f"Set {set_id}: full availability {available_sets_full} written to {out_filepath}")

if __name__ == "__main__":
    main()

Add available to parents

Prompt: Calculate Parent Availability from Equipment JSON

Context
You are given a JSON file representing a list of equipment items. Each item has:

• A displayname and parent_equipment_number

• A list of children, and each child contains:

  • current_quantity: how many are in stock

  • quantity: how many are needed to build 1 parent

  • usage_by_day: reservations per day (structured as YYYY-MM → day → reserved quantity)

Task

Write a PHP function that:

1. Input: Takes an array decoded from this JSON structure.

2. Output: Adds two new keys to each parent item, inserted directly after the displayname key:

   • 'available': the maximum number of parents that can be assembled based on the current stock and child requirements.

     • For each child, calculate:
       floor(current_quantity / quantity)

     • The minimum across all children determines the parent’s availability.

   • 'available_by_day': a structure similar to usage_by_day, but showing the availability of the parent on specific days, accounting for child reservations.

     • For each day, subtract reserved quantity from current stock per child.

     • Then divide by the number of that child required per parent.

     • Again, take the minimum across children for that day.

     • If that value equals the overall 'available', omit the day.

     • If it drops below 0, set it to 0 (administrative error handling).

     • Output structure:

       {
         "2025-06": {
           "13": 3,
           "14": 2
         },
         "2025-07": {
           "1": 4
         }
       }

   • Ensure the final order of keys for each parent is:

     • parent_equipment_number

     • displayname

     • available

     • available_by_day

     • children

Function name: addParentAvailability

Code generated

/**
 * Enhances parent equipment data with:
 * 1. An overall 'available' count based on current stock of children.
 * 2. A per-day 'available_by_day' structure that shows date-based limitations.
 *
 * 'available' shows the max number of parents that can be built given stock.
 * 'available_by_day' shows reductions on specific dates due to reservations
 * on child equipment. Dates with full availability are omitted. Negative availability
 * (due to administrative error) is corrected to 0.
 *
 * @param array $equipmentSet The array of parent equipment items with their children.
 * @return array The modified array with the new availability keys added.
 */
function addParentAvailability(array $equipmentSet): array {
    foreach ($equipmentSet as &$parent) {
        $minAvailable = PHP_INT_MAX;
        $dailyAvailability = [];

        foreach ($parent['children'] as $child) {
            $neededPerParent = $child['quantity'];
            $childAvailable = floor($child['current_quantity'] / $neededPerParent);
            $minAvailable = min($minAvailable, $childAvailable);

            if (!isset($child['usage_by_day'])) continue;

            foreach ($child['usage_by_day'] as $ym => $days) {
                foreach ($days as $day => $usedQty) {
                    $available = floor(($child['current_quantity'] - $usedQty) / $neededPerParent);
                    if ($available < 0) $available = 0;

                    // If not yet set, initialize with high value
                    if (!isset($dailyAvailability[$ym][$day])) {
                        $dailyAvailability[$ym][$day] = $available;
                    } else {
                        $dailyAvailability[$ym][$day] = min($dailyAvailability[$ym][$day], $available);
                    }
                }
            }
        }

        // Clean up: remove days that match the full availability
        foreach ($dailyAvailability as $ym => &$days) {
            foreach ($days as $day => $value) {
                if ($value >= $minAvailable) {
                    unset($days[$day]);
                }
            }
            if (empty($days)) {
                unset($dailyAvailability[$ym]);
            }
        }

        // Rebuild parent structure with displayname -> available -> available_by_day -> rest
        $newParent = [];
        foreach ($parent as $key => $value) {
            $newParent[$key] = $value;
            if ($key === 'displayname') {
                $newParent['available'] = $minAvailable;
                $newParent['available_by_day'] = $dailyAvailability;
            }
        }

        $parent = $newParent;
    }

    return $equipmentSet;
}

Correction (last step)

 📦 Prompt: Calculate Parent Availability from Equipment JSON (Full Version with Bug Fix)

## 🧾 Context
You are given a JSON file representing a list of equipment items. Each item represents a **parent item**, and contains:
- A `displayname` and `parent_equipment_number`.
- A list of `children`, each with:
  - `current_quantity`: number in stock.
  - `quantity`: number required to build **1 parent**.
  - `usage_by_day`: reservations per date in the structure:
    ```json
    {
      "2025-06": {
        "13": 2,
        "14": 3
      }
    }
    ```

---

## 🎯 Goal
Write a PHP function that does the following:

---

### ✅ 1. Input
The function receives a decoded JSON structure (i.e., a PHP array), containing an array of parent items, each with `children`.

---

### ✅ 2. Calculates `available`
For each parent item:
- Determine how many complete parents can be built from the current stock of all required children.
- For each child:
  ```php
  floor(current_quantity / quantity_required_per_parent)
  ```
- The **minimum** result across all children is the number of complete parent items that can be built.
- Add this number as a new key:
  ```php
  "available": <int>
  ```
- Insert it **immediately after `displayname`** in the parent’s object.

---

### ✅ 3. Calculates `available_by_day`
Also add a new key **after** `available`, named:

```json
"available_by_day": {
  "YYYY-MM": {
    "DD": <int>
  }
}
```

This shows for which dates the availability of parents is **lower than** the full availability due to **child reservations**.

#### Logic:
- Loop through **all dates** that appear in any child’s `usage_by_day`.
- For each date:
  - For each child:
    - Subtract reserved amount on that date from current_quantity.
    - Compute:
      ```php
      floor((current_quantity - reserved) / quantity_required)
      ```
    - If the result < 0, treat it as 0.
  - The **minimum** across all children determines the **parent’s availability for that date**.
  - If that value is **less than** the overall `available`, include it in `available_by_day`.
  - If the value is **equal to or greater than** `available`, skip it (don't output it).
- If no days are limited, the key `available_by_day` may be an empty object.

---

### ✅ 4. Output Structure
Each parent item will be rewritten to insert:
- `available`
- `available_by_day`

…right after `displayname`.

#### Example:
```json
{
  "parent_equipment_number": 1234,
  "displayname": "Cool Parent",
  "available": 2,
  "available_by_day": {
    "2025-06": {
      "13": 1,
      "14": 0
    }
  },
  "children": [ ... ]
}
```

---

### ✅ 5. Edge Cases
- If no usage data is available for any child, the daily availability is not included.
- If calculated availability is negative on any date, set it to `0` (administrative error).
- Never allow availability on any date to exceed the total `available`.

---

This prompt includes all requirements, handles bugs related to overestimation, and ensures robust availability calculations per parent.

Code:

/**
 * Enhances each parent item with:
 * - 'available': max number of parents that can be built from current stock.
 * - 'available_by_day': date-specific limits due to reservations in children.
 *
 * For 'available_by_day', only include dates where the availability is lower than 'available'.
 * Values below 0 are treated as 0.
 * Dates with full availability are skipped.
 *
 * @param array $equipmentSet Decoded JSON with parent items and their children.
 * @return array Modified data with availability info inserted.
 */
function addParentAvailability(array $equipmentSet): array {
    foreach ($equipmentSet as &$parent) {
        $minAvailable = PHP_INT_MAX;
        $dailyAvailability = [];
        $allDates = [];

        // Step 1: Compute total availability and gather all relevant dates
        foreach ($parent['children'] as $child) {
            $neededPerParent = $child['quantity'];
            $childAvailable = floor($child['current_quantity'] / $neededPerParent);
            $minAvailable = min($minAvailable, $childAvailable);

            if (!isset($child['usage_by_day'])) continue;

            foreach ($child['usage_by_day'] as $ym => $days) {
                foreach ($days as $day => $_) {
                    $allDates[$ym][$day] = true;
                }
            }
        }

        // Step 2: For every known date, compute availability from all children
        foreach ($allDates as $ym => $days) {
            foreach ($days as $day => $_) {
                $minForDay = PHP_INT_MAX;

                foreach ($parent['children'] as $child) {
                    $needed = $child['quantity'];
                    $stock = $child['current_quantity'];
                    $used = $child['usage_by_day'][$ym][$day] ?? 0;
                    $available = floor(($stock - $used) / $needed);
                    $available = max(0, $available); // prevent negative availability

                    $minForDay = min($minForDay, $available);
                }

                if ($minForDay < $minAvailable) {
                    $dailyAvailability[$ym][$day] = $minForDay;
                }
            }
        }

        // Step 3: Rebuild parent array, insert new keys after 'displayname'
        $newParent = [];
        foreach ($parent as $key => $value) {
            $newParent[$key] = $value;
            if ($key === 'displayname') {
                $newParent['available'] = $minAvailable;
                $newParent['available_by_day'] = $dailyAvailability;
            }
        }

        $parent = $newParent;
    }

    return $equipmentSet;
}

Open issues

1. At the moment we only read and cache equipment belonging to a client. Set can consist out of items not belonging to any client directly, so there is no data available (in equipment). This means that there's no data for the availabilty in the cache. This would mean that we need to load and cach more equipment and more plannedequipment.
   
   These are appr. 1000 equipments (current number is 1200).
2. The calendar format for each month is: 
   

   Array ( [1] => 1 [2] => 1 [3] => 1 [4] => 1 [5] => 1 [6] => 1 [7] => 1 [8] => 1 [9] => 1 [10] => 1 [11] => 1 [12] => 1 [13] => 1 [14] => 1 [15] => 1 [16] => 1 [17] => 1 [18] => 1 [19] => 1 [20] => 1 [21] => 1 [22] => 1 [23] => 1 [24] => 1 [25] => 1 [26] => 1 [27] => 1 [28] => 1 [29] => 1 [30] => 1 )

---