feat: Hebrew honorifics + post-nominals, Devanagari titles (#269 follow-up)

Hebrew titles גברת/פרופ'/פרופ׳/פרופסור/עו"ד/עו״ד/הרב join מר as
plain titles (Israeli convention: family follows), each abbreviation
in both geresh/gershayim spellings per the ד"ר precedent. Hebrew
gains the suffix category Arabic deferred: ז"ל and שליט"א (both
spellings) -- post-nominals ubiquitous in genealogical data, never
names, mid-word quotes inert in extraction.

Devanagari becomes a #269 script: श्री/श्रीमती/डॉ (डॉ. matches via
edge-period normalization). Latin sri/shri deliberately NOT added --
the transliterations collide with real given names (Sri Mulyani);
the native script cannot.

Deferred under the collision rule, recorded in the data-module
comments: bare רב (the ordinary word 'many') and בר as a particle
(Bar is a common modern Israeli given name; the surname spelling is
hyphenated anyway).

TDD: 15 rows red first across title/suffix/script categories.
Differential exit 0 unchanged.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

Author: derek73

docs/release_log.rst

@@ -8,6 +8,7 @@ Release Log
     - The CLI gains ``--locale CODE`` (``python -m nameparser --locale ru "..."``)
     - Add non-Latin vocabulary to the default lexicon (#269): Cyrillic, Greek, Arabic and Hebrew titles, conjunctions, and name particles -- native-script entries cannot collide with Latin-script names. Deferred pending vetting: Cyrillic ``мл``/``ст`` suffixes and the bare Greek ``κ`` title (it collides with the initial+surname shape). Behavior note: ``محمد بن سلمان`` now chains ``بن`` onto the family name where 1.x read it as a middle name
     - Add the Arabic-script bound given names (#269 follow-up): ``عبد``, the kunya pair ``أبو``/``ابو``, and ``أم``/``ام`` join the following word into the given name like their transliterations (``abdul``, ``abu``, ``umm``) always did -- ``عبد الرحمن محمد`` now parses given ``عبد الرحمن``, family ``محمد``, where 1.x split it into given + middle (``tests/v2/cases.py`` rows ``arabic_bound_given_*``). Not present in the differential corpus
+    - Add Hebrew honorifics, Hebrew post-nominals, and Devanagari titles (#269 follow-up): the Israeli honorifics ``גברת``, ``פרופ'``/``פרופ׳``, ``פרופסור``, ``עו"ד``/``עו״ד``, and ``הרב`` as plain titles; the post-nominals ``ז"ל``/``ז״ל`` and ``שליט"א``/``שליט״א`` as suffixes (both gershayim spellings each); and a new #269 script -- Devanagari ``श्री``, ``श्रीमती``, and ``डॉ`` (``डॉ.`` matches via edge-period normalization). Latin ``sri``/``shri`` deliberately NOT added (they collide with real given names; the native script cannot). Deferred: bare ``רב`` (ordinary word "many") and ``בר`` as a particle (Bar is a common modern given name). Not present in the differential corpus
     - Add Arabic honorific titles and the conjunction ``و`` (#269 follow-up): the doctor/professor/hajj/sheikha/engineer forms (``الدكتور``/``الدكتورة``/``دكتور``/``دكتورة``, ``الأستاذ``/``الأستاذة``/``أستاذ``/``أستاذة``, ``الحاج``/``الحاجة``, ``الشيخة``, ``مهندس``) as given-name titles -- Arabic honorifics precede the given name, like ``الشيخ``. Deferred under the collision rule: bare ``سيد``/``شيخ``/``أمير``/``سلطان`` (all common given names), the ``د.`` abbreviation (bare ``د`` would swallow initials), and the Ottoman post-nominals ``باشا``/``بك``/``أفندي`` (survive as family names). Not present in the differential corpus
 
     **Behavior Changes (draft)**

nameparser/config/suffixes.py

@@ -26,6 +26,14 @@
     # literally instead of going through nickname/suffix disambiguation).
     'ret',
     'vet',
+
+    # #269 follow-up: Hebrew post-nominals, both gershayim spellings
+    # (ASCII '"' and U+05F4); the mid-word quote is inert in
+    # extraction, like the ד"ר title. Neither is ever a name.
+    'ז"ל',      # "of blessed memory" (deceased), ASCII quote
+    'ז״ל',      # same, U+05F4 gershayim
+    'שליט"א',   # honorific for a living rabbi, ASCII quote
+    'שליט״א',   # same, U+05F4 gershayim
 }
 """
 

nameparser/config/titles.py

@@ -748,4 +748,25 @@
     "גב'",
     'ד״ר',
     'גב׳',
+    # #269 follow-up: the rest of the common Israeli honorifics, same
+    # plain-title bucket (family follows) and the same dual geresh/
+    # gershayim spelling rule as above. Deferred under the collision
+    # rule: bare 'רב' (also the ordinary word "many"); 'בר' as a
+    # particle (Bar is a common modern given name, and the surname
+    # spelling is hyphenated anyway).
+    'גברת',       # Mrs./Ms., full form
+    "פרופ'",      # professor abbreviation, ASCII apostrophe
+    'פרופ׳',      # professor abbreviation, U+05F3 geresh
+    'פרופסור',    # professor, full form
+    'עו"ד',       # advocate/lawyer, ASCII quote
+    'עו״ד',       # advocate/lawyer, U+05F4 gershayim
+    'הרב',        # "the rabbi" (article form; bare רב deferred)
+
+    # #269 follow-up: Devanagari (hi/mr). NO Latin twins on purpose:
+    # transliterated sri/shri collide with real given names (Sri
+    # Mulyani); the native-script forms cannot. "डॉ." matches via the
+    # edge-period normalization, like Latin "Dr.".
+    'श्री',        # Shri (Mr.)
+    'श्रीमती',     # Shrimati (Mrs.)
+    'डॉ',         # Dr. abbreviation
 }

tests/v2/test_locales.py

@@ -449,6 +449,33 @@ def test_non_interference_all_packs_combined() -> None:
     # Hebrew "מר" title (plain title, not FIRST_NAME_TITLES -- like
     # 'mr', the following name reads as family).
     ("מר דוד לוי", "title", "מר"),
+    # Hebrew title/suffix sweep (#269 follow-up): plain titles (Israeli
+    # convention: family follows, the מר precedent); both geresh/
+    # gershayim spellings ship where an abbreviation carries one.
+    ("גברת רות כהן", "title", "גברת"),
+    ("פרופ' דוד לוי", "title", "פרופ'"),
+    ("פרופ׳ דוד לוי", "title", "פרופ׳"),
+    ("פרופסור רות כהן", "title", "פרופסור"),
+    ('עו"ד דוד לוי', "title", 'עו"ד'),
+    ("עו״ד דוד לוי", "title", "עו״ד"),
+    ("הרב עובדיה יוסף", "title", "הרב"),
+    # post-nominal Hebrew suffixes: ז"ל ("of blessed memory") and
+    # שליט"א (honorific for living rabbis) -- suffix words, exact
+    # token incl. the mid-word gershayim (inert in extraction, like
+    # the ד"ר title)
+    ('משה כהן ז"ל', "suffix", 'ז"ל'),
+    ("משה כהן ז״ל", "suffix", "ז״ל"),
+    ('הרב משה פיינשטיין שליט"א', "suffix", 'שליט"א'),
+    ("הרב משה פיינשטיין שליט״א", "title", "הרב"),
+    # Devanagari (hi/mr) titles -- a new #269 script: श्री/श्रीमती are
+    # the Mr./Mrs. analogs, डॉ the Dr. abbreviation (edge-period
+    # normalization makes "डॉ." match too). NO Latin twins on purpose:
+    # transliterated sri/shri collide with real given names (Sri
+    # Mulyani); the native script cannot.
+    ("श्री राम शर्मा", "title", "श्री"),
+    ("श्रीमती सीता शर्मा", "title", "श्रीमती"),
+    ("डॉ राम शर्मा", "title", "डॉ"),
+    ("डॉ. राम शर्मा", "title", "डॉ."),
     # Geresh/gershayim gate (#269 step 2): probed live against
     # extract_delimited's _open_ok/_close_ok boundary rules. Both the
     # ASCII-quote spelling and the typographic Unicode spelling of