Step-by-Step Analysis Tutorial

In this tutorial, we will manually walk through the process of analyzing a verse, breaking it, and understanding how PyArud "sees" the poem.

The Verse

We will use a famous line from Al-Mutanabbi (Meter: Basit):

أَنَامُ مِلْءَ جُفُونِي عَنْ شَوَارِدِهَا ... وَيَسْهَرُ الْخَلْقُ جَرَّاهَا وَيَخْتَصِمُ

(I sleep with my eyelids full, ignoring its stray thoughts ... while the people stay awake because of them, arguing)

Step 1: The Arudi Conversion

Before looking for patterns, we must convert "Written Arabic" to "Spoken Arabic" (Arudi style).

from pyarud.arudi import ArudiConverter

converter = ArudiConverter()
sadr = "أَنَامُ مِلْءَ جُفُونِي عَنْ شَوَارِدِهَا"

arudi_text, pattern = converter.prepare_text(sadr)

print(f"Arudi Text: {arudi_text}")
print(f"Pattern:    {pattern}")

Output: - Arudi Text: أنام ملء جفوني عن شواردها (Notice: letters pronounced but not written are added, and vice versa. Here, it's mostly straight forward). - Pattern: 11011011101010110110

Step 2: Identifying the Meter

The pattern starts with 1101 (Fa'ilun) or 1010 (Mustaf'ilun via Khaban?).

Let's run it through the processor.

from pyarud.processor import ArudhProcessor

processor = ArudhProcessor()
verse = [
    ("أَنَامُ مِلْءَ جُفُونِي عَنْ شَوَارِدِهَا", "وَيَسْهَرُ الْخَلْقُ جَرَّاهَا وَيَخْتَصِمُ")
]

result = processor.process_poem(verse)
print(result['meter'])

Output: baseet

Step 3: Introducing a Defect

Let's intentionally break the rhythm to see how PyArud handles it. We will remove the word "ملء" (mil'a) from the first hemistich.

Broken Sadr: أَنَامُ جُفُونِي عَنْ شَوَارِدِهَا

broken_verse = [
    ("أَنَامُ جُفُونِي عَنْ شَوَارِدِهَا", "وَيَسْهَرُ الْخَلْقُ جَرَّاهَا وَيَخْتَصِمُ")
]

# Force 'baseet' because we know what it SHOULD be
result = processor.process_poem(broken_verse, meter_name="baseet")

# Inspect the Sadr
sadr_feet = result['verses'][0]['sadr_analysis']
for foot in sadr_feet:
    print(f"Expected: {foot['expected_pattern']:<10} | Got: {foot['actual_segment']:<10} | Status: {foot['status']}")

Output Analysis: 1. Foot 1: Mustaf'ilun (1010110). Our text أَنَامُ جُ (A-na-mu-ju) -> 11011. This doesn't match well. 2. PyArud will try to find the best match. It might flag the first foot as broken, or "eat" the wrong characters.

By forcing the meter (meter_name="baseet"), you tell PyArud: "I expect this structure. Tell me where the text deviates." This is powerful for educational tools or auto-correct features.