feat: sva extracted and seeded
This commit is contained in:
gosangam
@@ -0,0 +1,815 @@
|
||||
#!/usr/bin/env python3
|
||||
"""Extract Śrī Brahma-saṁhitā verses from other.html into structured JSON files."""
|
||||
|
||||
import json
|
||||
import os
|
||||
import re
|
||||
import sys
|
||||
|
||||
from bs4 import BeautifulSoup, Tag
|
||||
from indic_transliteration import sanscript
|
||||
from indic_transliteration.sanscript import transliterate
|
||||
|
||||
# --- ScaGoudy → IAST Unicode mapping ---
|
||||
SCAGOUDY_TO_IAST = {
|
||||
"ä": "ā",
|
||||
"Ä": "Ā",
|
||||
"å": "ṛ",
|
||||
"Å": "Ṛ",
|
||||
"ñ": "ṣ",
|
||||
"Ñ": "Ṣ",
|
||||
"ö": "ṭ",
|
||||
"Ö": "Ṭ",
|
||||
"ë": "ṇ",
|
||||
"Ë": "Ṇ",
|
||||
"ò": "ḍ",
|
||||
"Ò": "Ḍ",
|
||||
"ç": "ś",
|
||||
"Ç": "Ś",
|
||||
"ì": "ṅ",
|
||||
"ï": "ñ",
|
||||
"ü": "ū",
|
||||
"Ü": "Ū",
|
||||
"é": "ī",
|
||||
"É": "Ī",
|
||||
"à": "ṁ",
|
||||
"ù": "ḥ",
|
||||
"è": "ṝ",
|
||||
"û": "ṁl",
|
||||
"ÿ": "ḷ",
|
||||
}
|
||||
|
||||
# --- Text number to title mapping ---
|
||||
TEXT_TITLES = {n: f"Text {n}" for n in range(1, 63)}
|
||||
|
||||
# --- CSS class constants ---
|
||||
BOOK_HEADER_CLASS = "p12"
|
||||
VERSE_BOOKMARK_CLASS = "p49"
|
||||
VERSE_HEADING_CLASS = "p50"
|
||||
TEXT_LABEL_CLASS = "p52"
|
||||
PADA_TRANSLIT_CLASS = "p5"
|
||||
PADA_TRANSLIT_CONT_CLASS = "p66"
|
||||
SYNONYMS_HEADER_CLASS = "p165"
|
||||
SYNONYM_TEXT_CLASS = "p54"
|
||||
TRANSLATION_LABEL_CLASS = "p55"
|
||||
TRANSLATION_TEXT_CLASS = "p56"
|
||||
PURPORT_LABEL_CLASS_ALT = "p57"
|
||||
PURPORT_BODY_CLASS = "p48"
|
||||
CROSS_REF_CLASS = "p58"
|
||||
QUOTED_VERSE_CLASS = "p53"
|
||||
COMMENTATOR_CLASS = "p44"
|
||||
SECTION_HEADING_CLASS = "p31"
|
||||
INTRO_BODY_CLASS = "p18"
|
||||
FOREWORD_BODY_CLASS = "p6"
|
||||
SIGNATURE_CLASS = "p164"
|
||||
END_MARKER_CLASS = "p166"
|
||||
|
||||
|
||||
def decode_scagoudy(text: str) -> str:
|
||||
"""Convert ScaGoudy-encoded text to proper IAST Unicode."""
|
||||
for src, dst in SCAGOUDY_TO_IAST.items():
|
||||
text = text.replace(src, dst)
|
||||
return text
|
||||
|
||||
|
||||
def decode_html_content(tag: Tag) -> str:
|
||||
"""Extract inner HTML from a tag, decode ScaGoudy, preserve formatting tags."""
|
||||
inner = "".join(str(child) for child in tag.children)
|
||||
inner = decode_scagoudy(inner)
|
||||
# Remove span tags with CSS classes (artifacts from RTF conversion)
|
||||
inner = re.sub(
|
||||
r'<span[^>]*class="[^"]*"[^>]*>(.*?)</span>', r"\1", inner, flags=re.DOTALL
|
||||
)
|
||||
# Clean up Apple-tab-span leftovers (tab characters)
|
||||
inner = inner.replace("\t", " ")
|
||||
# Normalize whitespace
|
||||
inner = re.sub(r"\s+", " ", inner).strip()
|
||||
return inner
|
||||
|
||||
|
||||
def decode_text(tag: Tag) -> str:
|
||||
"""Extract plain text from a tag and decode ScaGoudy."""
|
||||
text = tag.get_text()
|
||||
return decode_scagoudy(text).strip()
|
||||
|
||||
|
||||
def normalize_devanagari_anusvara(text: str) -> str:
|
||||
"""Replace explicit class nasals before same-class consonants with anusvāra.
|
||||
|
||||
Only normalize ङ् and ञ् — these are almost always written with anusvāra.
|
||||
Keep ण्ड, न्त, म्ब etc. as explicit conjuncts (standard spelling).
|
||||
"""
|
||||
replacements = [
|
||||
# ङ् before velars (क, ख, ग, घ)
|
||||
(r"ङ्([कखगघ])", r"ं\1"),
|
||||
# ञ् before palatals (च, छ, ज, झ)
|
||||
(r"ञ्([चछजझ])", r"ं\1"),
|
||||
]
|
||||
for pattern, repl in replacements:
|
||||
text = re.sub(pattern, repl, text)
|
||||
return text
|
||||
|
||||
|
||||
def join_virama_words(text: str) -> str:
|
||||
"""Join Devanagari words where the previous word ends with virāma (्).
|
||||
|
||||
Handles both virāma+consonant and virāma+vowel cases.
|
||||
Uses [ \\t]+ instead of \\s+ to avoid joining across newlines.
|
||||
"""
|
||||
# Join virāma + space + consonant (क U+0915 through ह U+0939)
|
||||
text = re.sub(r"्[ \t]+([क-ह])", r"्\1", text)
|
||||
|
||||
# Join virāma + space + independent vowel
|
||||
_VOWEL_TO_MATRA = {
|
||||
"अ": "", # inherent vowel — just drop virāma and the अ
|
||||
"आ": "ा",
|
||||
"इ": "ि",
|
||||
"ई": "ी",
|
||||
"उ": "ु",
|
||||
"ऊ": "ू",
|
||||
"ऋ": "ृ",
|
||||
"ॠ": "ॄ",
|
||||
"ऌ": "ॢ",
|
||||
"ॡ": "ॣ",
|
||||
"ए": "े",
|
||||
"ऐ": "ै",
|
||||
"ओ": "ो",
|
||||
"औ": "ौ",
|
||||
}
|
||||
|
||||
def _virama_vowel_replace(m: re.Match[str]) -> str:
|
||||
vowel = m.group(1)
|
||||
return _VOWEL_TO_MATRA.get(vowel, vowel)
|
||||
|
||||
text = re.sub(r"्[ \t]+([अ-औॠॡ])", _virama_vowel_replace, text)
|
||||
return text
|
||||
|
||||
|
||||
def iast_to_devanagari(iast_text: str) -> str:
|
||||
"""Convert IAST transliteration to Devanagari script."""
|
||||
lines = iast_text.split("\n")
|
||||
result_lines = []
|
||||
for line in lines:
|
||||
# Remove hyphens between words for Devanagari
|
||||
clean = line.replace("-", "")
|
||||
# Normalize ṁ (U+1E41) to ṃ (U+1E43) for proper anusvāra
|
||||
clean = clean.replace("ṁ", "ṃ")
|
||||
deva = transliterate(clean, sanscript.IAST, sanscript.DEVANAGARI)
|
||||
deva = normalize_devanagari_anusvara(deva)
|
||||
deva = join_virama_words(deva)
|
||||
result_lines.append(deva)
|
||||
return "\n".join(result_lines)
|
||||
|
||||
|
||||
def parse_synonyms(text: str) -> list:
|
||||
"""Parse synonym text into structured array.
|
||||
|
||||
Input format: "word—meaning; word—meaning; word—meaning"
|
||||
"""
|
||||
result = []
|
||||
entries = text.split(";")
|
||||
for entry in entries:
|
||||
entry = entry.strip()
|
||||
if not entry:
|
||||
continue
|
||||
if "—" in entry:
|
||||
parts = entry.split("—", 1)
|
||||
elif "--" in entry:
|
||||
parts = entry.split("--", 1)
|
||||
else:
|
||||
continue
|
||||
|
||||
if len(parts) == 2:
|
||||
word = parts[0].strip()
|
||||
meaning = parts[1].strip()
|
||||
# Clean word: remove trailing/leading punctuation
|
||||
word = word.strip("., ")
|
||||
if word and meaning:
|
||||
result.append({"word": word, "meanings": [meaning]})
|
||||
|
||||
return result
|
||||
|
||||
|
||||
def pair_padas(padas: list) -> str:
|
||||
"""Combine pādas into 2-line verse format.
|
||||
|
||||
If the number of pādas is odd, the first one is treated as a
|
||||
standalone line (e.g. speaker attribution like 'uvāca').
|
||||
|
||||
Empty strings in the list act as stanza-break sentinels and are
|
||||
rendered as blank lines in the output.
|
||||
"""
|
||||
if not padas:
|
||||
return ""
|
||||
|
||||
# Split pādas into stanza groups separated by "" sentinels
|
||||
groups: list[list[str]] = [[]]
|
||||
for p in padas:
|
||||
if p == "":
|
||||
groups.append([])
|
||||
else:
|
||||
groups[-1].append(p)
|
||||
|
||||
result_lines = []
|
||||
for group in groups:
|
||||
if not group:
|
||||
continue
|
||||
|
||||
lines = []
|
||||
i = 0
|
||||
|
||||
# If odd number of pādas, first one is standalone (speaker attribution)
|
||||
if len(group) % 2 == 1:
|
||||
lines.append(group[0])
|
||||
i = 1
|
||||
|
||||
# Pair remaining pādas
|
||||
while i < len(group) - 1:
|
||||
lines.append(group[i] + " " + group[i + 1])
|
||||
i += 2
|
||||
|
||||
# Handle any trailing single pāda
|
||||
if i < len(group):
|
||||
lines.append(group[i])
|
||||
|
||||
result_lines.append("\n".join(lines))
|
||||
|
||||
return "\n\n".join(result_lines)
|
||||
|
||||
|
||||
def get_css_class(tag: Tag) -> str:
|
||||
"""Get the primary CSS class of a tag."""
|
||||
classes = tag.get("class", [])
|
||||
return classes[0] if classes else ""
|
||||
|
||||
|
||||
def is_empty(tag: Tag) -> bool:
|
||||
"""Check if a tag is empty or whitespace-only."""
|
||||
text = tag.get_text().strip()
|
||||
return not text or text == "\xa0"
|
||||
|
||||
|
||||
def write_json(data, filepath):
|
||||
"""Write JSON with proper Unicode and formatting."""
|
||||
os.makedirs(os.path.dirname(filepath), exist_ok=True)
|
||||
with open(filepath, "w", encoding="utf-8") as f:
|
||||
json.dump(data, f, ensure_ascii=False, indent=2)
|
||||
print(f" Written: {filepath}")
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Boundary detection
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def find_bs_boundaries(all_paragraphs: list) -> tuple[int, int]:
|
||||
"""Find the start and end paragraph indices of the Brahma-saṁhitā content.
|
||||
|
||||
Start: <p class="p12"> containing "Brahma-sa" (ScaGoudy encoded)
|
||||
End: next <p class="p166"> after start (start of next book)
|
||||
"""
|
||||
start_idx = None
|
||||
end_idx = len(all_paragraphs)
|
||||
|
||||
for i, p in enumerate(all_paragraphs):
|
||||
cls = get_css_class(p)
|
||||
if cls == BOOK_HEADER_CLASS and start_idx is None:
|
||||
text = p.get_text().strip()
|
||||
# ScaGoudy-encoded: "Çré Brahma-saàhitä"
|
||||
if "Brahma-sa" in text:
|
||||
start_idx = i
|
||||
continue
|
||||
if start_idx is not None and cls == END_MARKER_CLASS:
|
||||
end_idx = i
|
||||
break
|
||||
|
||||
if start_idx is None:
|
||||
print("ERROR: Could not find 'Śrī Brahma-saṁhitā' header in other.html!")
|
||||
sys.exit(1)
|
||||
|
||||
return start_idx, end_idx
|
||||
|
||||
|
||||
def parse_verse_bookmark(text: str) -> int | None:
|
||||
"""Parse a verse bookmark like 'Bs 5.N'.
|
||||
|
||||
Returns text number (1-62), or None.
|
||||
"""
|
||||
text = text.strip()
|
||||
m = re.match(r"Bs\s+5\.(\d+)", text)
|
||||
if m:
|
||||
return int(m.group(1))
|
||||
return None
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Section boundary detection within BS range
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def find_section_boundaries(all_paragraphs: list, start_idx: int, end_idx: int) -> dict:
|
||||
"""Find key section boundaries within the BS content.
|
||||
|
||||
Returns dict with section names and their start indices.
|
||||
"""
|
||||
sections = {}
|
||||
|
||||
for i in range(start_idx, end_idx):
|
||||
p = all_paragraphs[i]
|
||||
cls = get_css_class(p)
|
||||
text = p.get_text().strip()
|
||||
|
||||
# Introduction bookmark
|
||||
if cls == VERSE_BOOKMARK_CLASS and "Bs Introduction" in text:
|
||||
sections["introduction_bookmark"] = i
|
||||
# Foreword bookmark
|
||||
elif cls == VERSE_BOOKMARK_CLASS and "Bs Foreword" in text:
|
||||
sections["foreword_bookmark"] = i
|
||||
# First verse bookmark (Bs 5.1)
|
||||
elif cls == VERSE_BOOKMARK_CLASS and parse_verse_bookmark(decode_text(p)) == 1:
|
||||
sections["first_verse"] = i
|
||||
break
|
||||
|
||||
# Find the chapter heading (p31 with "Chapter Five")
|
||||
for i in range(start_idx, end_idx):
|
||||
p = all_paragraphs[i]
|
||||
cls = get_css_class(p)
|
||||
if cls == SECTION_HEADING_CLASS:
|
||||
text = p.get_text().strip()
|
||||
if "Chapter Five" in text or "Chapter" in text:
|
||||
# Check if this is the chapter title, not intro/foreword headers
|
||||
if i > sections.get("foreword_bookmark", 0):
|
||||
sections["chapter_heading"] = i
|
||||
break
|
||||
|
||||
return sections
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Intro segment extraction
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def extract_intro_segment(
|
||||
all_paragraphs: list, start_idx: int, end_idx: int, body_classes: tuple
|
||||
) -> list[str]:
|
||||
"""Extract paragraphs for an intro segment.
|
||||
|
||||
Decodes ScaGoudy, preserves HTML formatting, skips empty paragraphs.
|
||||
"""
|
||||
result = []
|
||||
|
||||
for i in range(start_idx, end_idx):
|
||||
p = all_paragraphs[i]
|
||||
cls = get_css_class(p)
|
||||
|
||||
if is_empty(p):
|
||||
continue
|
||||
|
||||
# Skip bookmark and header classes
|
||||
if cls in (VERSE_BOOKMARK_CLASS, BOOK_HEADER_CLASS, SECTION_HEADING_CLASS):
|
||||
continue
|
||||
|
||||
# Include specified body classes and signature class
|
||||
if cls in body_classes or cls == SIGNATURE_CLASS:
|
||||
html = decode_html_content(p)
|
||||
if html:
|
||||
result.append(html)
|
||||
|
||||
return result
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Verse extraction state machine
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def extract_verses(all_paragraphs: list, verses_start: int, verses_end: int) -> list:
|
||||
"""Extract verses from the BS Chapter Five section using a state machine.
|
||||
|
||||
States: IDLE → VERSE_START → TRANSLITERATION → SYNONYMS →
|
||||
TRANSLATION → POST_TRANSLATION → PURPORT → (back to IDLE)
|
||||
|
||||
Returns list of verse dicts.
|
||||
"""
|
||||
verses = []
|
||||
current_verse = None
|
||||
state = "IDLE"
|
||||
|
||||
for i in range(verses_start, verses_end):
|
||||
p = all_paragraphs[i]
|
||||
cls = get_css_class(p)
|
||||
|
||||
# --- Verse bookmark (p49) ---
|
||||
if cls == VERSE_BOOKMARK_CLASS:
|
||||
# Finalize previous verse if any
|
||||
if current_verse and state != "IDLE":
|
||||
_finalize_verse(current_verse)
|
||||
verses.append(current_verse)
|
||||
|
||||
text = decode_text(p)
|
||||
verse_num = parse_verse_bookmark(text)
|
||||
if verse_num is not None:
|
||||
reference = f"Bs 5.{verse_num}"
|
||||
|
||||
current_verse = {
|
||||
"book_title": "Śrī Brahma-saṁhitā",
|
||||
"chapter": 5,
|
||||
"verse": verse_num,
|
||||
"reference": reference,
|
||||
"devanagari": "",
|
||||
"transliteration": "",
|
||||
"_padas": [],
|
||||
"synonyms": [],
|
||||
"translation": "",
|
||||
"purport": [],
|
||||
}
|
||||
state = "VERSE_START"
|
||||
continue
|
||||
|
||||
if not current_verse:
|
||||
continue
|
||||
|
||||
# --- Verse heading (p50) - just confirms the text number ---
|
||||
if cls == VERSE_HEADING_CLASS:
|
||||
continue
|
||||
|
||||
# --- TEXT label (p52) - transition to transliteration ---
|
||||
if cls == TEXT_LABEL_CLASS:
|
||||
state = "TRANSLITERATION"
|
||||
continue
|
||||
|
||||
# --- Pāda transliteration lines (p5 and p66) ---
|
||||
if cls in (PADA_TRANSLIT_CLASS, PADA_TRANSLIT_CONT_CLASS) and state in (
|
||||
"VERSE_START",
|
||||
"TRANSLITERATION",
|
||||
):
|
||||
state = "TRANSLITERATION"
|
||||
line = decode_text(p)
|
||||
# Remove trailing period
|
||||
line = line.rstrip(".")
|
||||
line = line.strip()
|
||||
if line:
|
||||
# p66 marks the start of a new stanza group — insert a blank
|
||||
# sentinel so _finalize_verse preserves the line break.
|
||||
if cls == PADA_TRANSLIT_CONT_CLASS and current_verse["_padas"]:
|
||||
current_verse["_padas"].append("")
|
||||
current_verse["_padas"].append(line)
|
||||
continue
|
||||
|
||||
# --- SYNONYMS header (p165) ---
|
||||
if cls == SYNONYMS_HEADER_CLASS:
|
||||
state = "SYNONYMS"
|
||||
continue
|
||||
|
||||
# --- Synonyms text (p54) ---
|
||||
if cls == SYNONYM_TEXT_CLASS and current_verse:
|
||||
state = "SYNONYMS"
|
||||
syn_text = decode_text(p)
|
||||
syn_text = syn_text.strip()
|
||||
parsed = parse_synonyms(syn_text)
|
||||
current_verse["synonyms"].extend(parsed)
|
||||
continue
|
||||
|
||||
# --- TRANSLATION label (p55) — check text content to distinguish from PURPORT ---
|
||||
if cls == TRANSLATION_LABEL_CLASS and current_verse:
|
||||
inner_text = p.get_text().strip().upper()
|
||||
if "TRANSLATION" in inner_text:
|
||||
state = "TRANSLATION"
|
||||
elif "PURPORT" in inner_text:
|
||||
state = "PURPORT"
|
||||
continue
|
||||
|
||||
# --- PURPORT label via p57 ---
|
||||
if cls == PURPORT_LABEL_CLASS_ALT and current_verse:
|
||||
inner_text = p.get_text().strip().upper()
|
||||
if "PURPORT" in inner_text:
|
||||
state = "PURPORT"
|
||||
elif "TRANSLATION" in inner_text:
|
||||
state = "TRANSLATION"
|
||||
continue
|
||||
|
||||
# --- Translation text (p56) ---
|
||||
if cls == TRANSLATION_TEXT_CLASS and current_verse and state == "TRANSLATION":
|
||||
html = decode_html_content(p)
|
||||
# Remove bold wrapper if present (the whole translation is bold)
|
||||
html = re.sub(r"^<b>(.*)</b>$", r"\1", html, flags=re.DOTALL)
|
||||
html = html.strip()
|
||||
current_verse["translation"] = html
|
||||
state = "POST_TRANSLATION"
|
||||
continue
|
||||
|
||||
# --- Purport paragraphs (p48, p6) ---
|
||||
if (
|
||||
cls in (PURPORT_BODY_CLASS, FOREWORD_BODY_CLASS)
|
||||
and current_verse
|
||||
and state in ("PURPORT", "POST_TRANSLATION")
|
||||
):
|
||||
state = "PURPORT"
|
||||
html = decode_html_content(p)
|
||||
if html:
|
||||
current_verse["purport"].append(html)
|
||||
continue
|
||||
|
||||
# --- Cross-references (p58) — treat as purport paragraph ---
|
||||
if (
|
||||
cls == CROSS_REF_CLASS
|
||||
and current_verse
|
||||
and state in ("PURPORT", "POST_TRANSLATION")
|
||||
):
|
||||
state = "PURPORT"
|
||||
html = decode_html_content(p)
|
||||
if html:
|
||||
current_verse["purport"].append(html)
|
||||
continue
|
||||
|
||||
# --- Quoted verse in purport (p53) — treat as purport paragraph ---
|
||||
if cls == QUOTED_VERSE_CLASS and current_verse and state == "PURPORT":
|
||||
html = decode_html_content(p)
|
||||
if html:
|
||||
current_verse["purport"].append(html)
|
||||
continue
|
||||
|
||||
# --- Commentator text in purport (p44) — treat as purport paragraph ---
|
||||
if cls == COMMENTATOR_CLASS and current_verse and state == "PURPORT":
|
||||
html = decode_html_content(p)
|
||||
if html:
|
||||
current_verse["purport"].append(html)
|
||||
continue
|
||||
|
||||
# --- Section heading within purport (p31) ---
|
||||
if cls == SECTION_HEADING_CLASS and current_verse and state == "PURPORT":
|
||||
html = decode_html_content(p)
|
||||
if html:
|
||||
current_verse["purport"].append(html)
|
||||
continue
|
||||
|
||||
# Finalize last verse
|
||||
if current_verse and state != "IDLE":
|
||||
_finalize_verse(current_verse)
|
||||
verses.append(current_verse)
|
||||
|
||||
return verses
|
||||
|
||||
|
||||
def _finalize_verse(verse: dict):
|
||||
"""Build transliteration and Devanagari from pādas. Remove internal fields."""
|
||||
padas = verse.get("_padas", [])
|
||||
|
||||
# Transliteration: one pāda per line, blank sentinels become blank lines
|
||||
trans_lines = []
|
||||
for p in padas:
|
||||
if p == "":
|
||||
trans_lines.append("") # blank line between stanza groups
|
||||
else:
|
||||
trans_lines.append(p)
|
||||
verse["transliteration"] = "\n".join(trans_lines)
|
||||
|
||||
# Devanagari: pair pādas into 2-line verse structure, then convert
|
||||
if padas:
|
||||
paired = pair_padas(padas)
|
||||
verse["devanagari"] = iast_to_devanagari(paired)
|
||||
|
||||
# Clean up internal field
|
||||
if "_padas" in verse:
|
||||
del verse["_padas"]
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Main
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def main():
|
||||
base_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
|
||||
html_path = os.path.join(base_dir, "public", "html", "other.html")
|
||||
output_base = os.path.join(base_dir, "lib", "jsons", "books", "bs")
|
||||
|
||||
if not os.path.exists(html_path):
|
||||
print(f"ERROR: {html_path} not found!")
|
||||
sys.exit(1)
|
||||
|
||||
print(f"Reading {html_path}...")
|
||||
with open(html_path, "r", encoding="utf-8") as f:
|
||||
soup = BeautifulSoup(f, "html.parser")
|
||||
|
||||
all_paragraphs = soup.find_all("p")
|
||||
print(f"Total paragraphs in file: {len(all_paragraphs)}")
|
||||
|
||||
# --- Find BS boundaries ---
|
||||
bs_start, bs_end = find_bs_boundaries(all_paragraphs)
|
||||
print(f"BS content: paragraphs {bs_start} to {bs_end}")
|
||||
|
||||
# --- Find section boundaries ---
|
||||
sections = find_section_boundaries(all_paragraphs, bs_start, bs_end)
|
||||
print(f"\nSection boundaries found:")
|
||||
for name, idx in sorted(sections.items(), key=lambda x: x[1]):
|
||||
print(f" [{idx}] {name}")
|
||||
|
||||
intro_bm = sections.get("introduction_bookmark")
|
||||
foreword_bm = sections.get("foreword_bookmark")
|
||||
chapter_heading = sections.get("chapter_heading")
|
||||
first_verse = sections.get("first_verse")
|
||||
|
||||
if first_verse is None:
|
||||
print("ERROR: Could not find first verse bookmark (Bs 5.1)!")
|
||||
sys.exit(1)
|
||||
|
||||
# --- Extract intro segments ---
|
||||
print("\n--- Extracting Intro Segments ---")
|
||||
|
||||
intro_segments = []
|
||||
|
||||
# 1. Introduction segment
|
||||
if intro_bm is not None:
|
||||
intro_end = foreword_bm if foreword_bm is not None else first_verse
|
||||
intro_paras = extract_intro_segment(
|
||||
all_paragraphs, intro_bm, intro_end, (INTRO_BODY_CLASS,)
|
||||
)
|
||||
intro_segment = {
|
||||
"slug": "introduction",
|
||||
"title": "Introduction",
|
||||
"type": "intro-segment",
|
||||
"book": "bs",
|
||||
"order": 1,
|
||||
"paragraphs": intro_paras,
|
||||
}
|
||||
intro_segments.append(intro_segment)
|
||||
print(f" Introduction: {len(intro_paras)} paragraphs")
|
||||
|
||||
# 2. Foreword segment
|
||||
if foreword_bm is not None:
|
||||
foreword_end = chapter_heading if chapter_heading is not None else first_verse
|
||||
foreword_paras = extract_intro_segment(
|
||||
all_paragraphs, foreword_bm, foreword_end, (FOREWORD_BODY_CLASS,)
|
||||
)
|
||||
foreword_segment = {
|
||||
"slug": "foreword",
|
||||
"title": "Foreword",
|
||||
"type": "intro-segment",
|
||||
"book": "bs",
|
||||
"order": 2,
|
||||
"paragraphs": foreword_paras,
|
||||
}
|
||||
intro_segments.append(foreword_segment)
|
||||
print(f" Foreword: {len(foreword_paras)} paragraphs")
|
||||
|
||||
# --- Extract verses ---
|
||||
print("\n--- Extracting Verses ---")
|
||||
|
||||
# Verse section runs from the first verse bookmark to bs_end
|
||||
verses = extract_verses(all_paragraphs, first_verse, bs_end)
|
||||
print(f"Extracted {len(verses)} verses (expected 62: Texts 1-62)")
|
||||
|
||||
# --- Write intro.json ---
|
||||
intro_meta = {
|
||||
"book": "bs",
|
||||
"title": "Śrī Brahma-saṁhitā",
|
||||
"author": "Śrīla Bhaktisiddhānta Sarasvatī Ṭhākura",
|
||||
"year": 1932,
|
||||
"description": "Chapter Five of the Hymns of Brahmā, discovered by Śrī Caitanya Mahāprabhu. Commentary by Śrīla Bhaktisiddhānta Sarasvatī Ṭhākura.",
|
||||
"totalChapters": 1,
|
||||
"chapters": [
|
||||
{
|
||||
"number": 5,
|
||||
"title": "Chapter Five",
|
||||
"totalVerses": 62,
|
||||
}
|
||||
],
|
||||
}
|
||||
write_json(intro_meta, os.path.join(output_base, "intro.json"))
|
||||
|
||||
# --- Write intro segments ---
|
||||
print("\n--- Writing Intro Segments ---")
|
||||
for seg in intro_segments:
|
||||
filepath = os.path.join(output_base, "intro", f"{seg['slug']}.json")
|
||||
write_json(seg, filepath)
|
||||
|
||||
# --- Write verse JSON files ---
|
||||
print("\n--- Writing Verse Files ---")
|
||||
for verse in verses:
|
||||
text_num = verse["verse"]
|
||||
filename = f"{str(text_num).zfill(2)}.json"
|
||||
filepath = os.path.join(output_base, "chapters", "05", "verses", filename)
|
||||
write_json(verse, filepath)
|
||||
|
||||
# --- Summary ---
|
||||
print(f"\n{'=' * 60}")
|
||||
print(f" EXTRACTION SUMMARY")
|
||||
print(f"{'=' * 60}")
|
||||
print(f" Book: Śrī Brahma-saṁhitā")
|
||||
print(f" Intro segments: {len(intro_segments)}")
|
||||
for seg in intro_segments:
|
||||
print(
|
||||
f" {seg['order']}. {seg['title']}: {len(seg['paragraphs'])} paragraphs"
|
||||
)
|
||||
print(f" Verses extracted: {len(verses)}")
|
||||
texts_found = sorted([v["verse"] for v in verses])
|
||||
if texts_found:
|
||||
print(
|
||||
f" Texts: {texts_found[0]}-{texts_found[-1]} ({len(texts_found)} total)"
|
||||
)
|
||||
else:
|
||||
print(" Texts: None")
|
||||
|
||||
total_purport_paras = sum(len(v["purport"]) for v in verses)
|
||||
total_synonyms = sum(len(v["synonyms"]) for v in verses)
|
||||
print(f" Total purport ¶s: {total_purport_paras}")
|
||||
print(f" Total synonyms: {total_synonyms}")
|
||||
print(f" Output directory: {output_base}/")
|
||||
|
||||
# --- Quality checks ---
|
||||
print(f"\n--- Quality Checks ---")
|
||||
|
||||
# Check all verses have expected fields
|
||||
issues = []
|
||||
for verse in verses:
|
||||
ref = verse["reference"]
|
||||
if not verse["transliteration"]:
|
||||
issues.append(f" {ref}: missing transliteration")
|
||||
if not verse["devanagari"]:
|
||||
issues.append(f" {ref}: missing devanagari")
|
||||
if not verse["synonyms"]:
|
||||
issues.append(f" {ref}: missing synonyms")
|
||||
if not verse["translation"]:
|
||||
issues.append(f" {ref}: missing translation")
|
||||
# Not all verses necessarily have purport, but most do
|
||||
if not verse["purport"]:
|
||||
issues.append(f" {ref}: missing purport (may be intentional)")
|
||||
|
||||
if issues:
|
||||
print(" ISSUES FOUND:")
|
||||
for issue in issues:
|
||||
print(issue)
|
||||
else:
|
||||
print(
|
||||
" All verses have transliteration, devanagari, synonyms, and translation ✓"
|
||||
)
|
||||
|
||||
# Check expected verse count
|
||||
expected_texts = set(range(1, 63))
|
||||
found_texts = set(v["verse"] for v in verses)
|
||||
missing = expected_texts - found_texts
|
||||
extra = found_texts - expected_texts
|
||||
if missing:
|
||||
print(f" MISSING texts: {sorted(missing)}")
|
||||
if extra:
|
||||
print(f" UNEXPECTED texts: {sorted(extra)}")
|
||||
if not missing and not extra:
|
||||
print(f" All 62 texts found ✓")
|
||||
|
||||
# --- Samples ---
|
||||
print(f"\n--- Sample: Text 1 (first) ---")
|
||||
t1 = [v for v in verses if v["verse"] == 1]
|
||||
if t1:
|
||||
v = t1[0]
|
||||
print(f" Reference: {v['reference']}")
|
||||
print(f" Transliteration: {v['transliteration'][:80]}...")
|
||||
print(f" Devanagari: {v['devanagari'][:80]}...")
|
||||
print(f" Synonyms: {len(v['synonyms'])} entries")
|
||||
if v["synonyms"]:
|
||||
print(f" First: {v['synonyms'][0]}")
|
||||
trans_preview = re.sub(r"<[^>]+>", "", v["translation"])[:100]
|
||||
print(f" Translation: {trans_preview}...")
|
||||
print(f" Purport ¶s: {len(v['purport'])}")
|
||||
|
||||
print(f"\n--- Sample: Text 29 (middle) ---")
|
||||
t29 = [v for v in verses if v["verse"] == 29]
|
||||
if t29:
|
||||
v = t29[0]
|
||||
print(f" Reference: {v['reference']}")
|
||||
print(f" Transliteration: {v['transliteration'][:80]}...")
|
||||
print(f" Devanagari: {v['devanagari'][:80]}...")
|
||||
print(f" Synonyms: {len(v['synonyms'])} entries")
|
||||
if v["synonyms"]:
|
||||
print(f" First: {v['synonyms'][0]}")
|
||||
print(f" Last: {v['synonyms'][-1]}")
|
||||
trans_preview = re.sub(r"<[^>]+>", "", v["translation"])[:100]
|
||||
print(f" Translation: {trans_preview}...")
|
||||
print(f" Purport ¶s: {len(v['purport'])}")
|
||||
|
||||
print(f"\n--- Sample: Text 62 (last) ---")
|
||||
t62 = [v for v in verses if v["verse"] == 62]
|
||||
if t62:
|
||||
v = t62[0]
|
||||
print(f" Reference: {v['reference']}")
|
||||
print(f" Transliteration: {v['transliteration'][:80]}...")
|
||||
print(f" Devanagari: {v['devanagari'][:80]}...")
|
||||
print(f" Synonyms: {len(v['synonyms'])} entries")
|
||||
trans_preview = re.sub(r"<[^>]+>", "", v["translation"])[:100]
|
||||
print(f" Translation: {trans_preview}...")
|
||||
print(f" Purport ¶s: {len(v['purport'])}")
|
||||
|
||||
# --- Intro segment paragraph counts ---
|
||||
print(f"\n--- Intro Segment Details ---")
|
||||
for seg in intro_segments:
|
||||
print(f" {seg['title']}: {len(seg['paragraphs'])} paragraphs")
|
||||
if seg["paragraphs"]:
|
||||
first_preview = re.sub(r"<[^>]+>", "", seg["paragraphs"][0])[:100]
|
||||
print(f" First ¶: {first_preview}...")
|
||||
last_preview = re.sub(r"<[^>]+>", "", seg["paragraphs"][-1])[:100]
|
||||
print(f" Last ¶: {last_preview}...")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
Reference in New Issue
Block a user