egy_names 0.3.6

A production-grade Egyptian onomastic intelligence library for Dart & Flutter — generate, translate, annotate, split, and analyze Egyptian names.

0.3.6

• Version alignment with the other SDKs. Same book.
• Language-specific README on pub.dev.

0.3.5

• Fix: English and Arabic-variant key collisions — a rare misspelling could shadow a common lemma's own canonical spelling or steal its English key. The higher corpus-share lemma always wins now.
• Fix: detect_gender/detect_religion now key off the first personal, non-lineage token instead of a whole-name majority vote; a father's or family's community can no longer outvote the person.
• Fix: multi-word compound lemmas (kunya "Abu X", "Ahmed Saad-El-Din") are recognized as one token by split/detect_gender/detect_religion, not two meaningless fragments.
• Fix: is_valid/generate now exclude non-personal catalog rows (e.g. "الله") and low-confidence/fabricated filler entries.
• Fix: corrected ~100 malformed compound spellings and one mislabeled gender in the catalog.
• New: shared logic_config.json — every threshold/rule list is now data, synced across SDKs from one source.

0.3.4

• Catalog: Youakeem is an English spelling of يواقيم.

0.3.3

• Resolve names.json.gz from the installed package, not the app working directory.

0.3.2

• Edge-case hardening, whitespace safety, and full stress-test validation.

0.3.1

• Expanded canonical dataset to 44,626 Egyptian names derived from 15.88M+ records.
• Added 14-dimensional onomastic intelligence features.
• Bilingual authentic Egyptian pet names (أسماء الدلع) with Arabic lemmas, full Egyptian Tashkeel, English transliteration, and IPA phonetics.
• Added structured descriptions for iconic Egyptian public figures (الاسماء المشهورة) in Arabic and English.
• Added Semitic/Coptic root extraction, origin stratification, and sociological generation trends.
• Added Modern Standard & Egyptian Colloquial IPA phonetic transcriptions.

0.1.1

• Comprehensive onomastic dataset with 33,117 Egyptian name lemmas and 134,000+ lookup keys.
• Culturally authentic name generation with slot-weighted sampling.
• Bidirectional Arabic <-> English name translation.
• Dynamic programming segmentation for concatenated Arabic names.
• Orthographic correction with 54,000+ correction pairs and phonetic alif maqsura normalization.
• Full diacritization (tashkeel) with compound name support.
• Linguistic and demographic metadata annotation (gender, religion, frequency, etymology).
• Patronymic chain analysis, generational slot role identification, and uniqueness scoring.