interscript-maps 2.0.5

data/maps/un-ara-Arab-Latn-1971.imp

@@ -0,0 +1,137 @@
+metadata {
+  authority_id: ungegn
+  id: 1971
+  language: iso-639-2:ara
+  source_script: Arab
+  destination_script: Latn
+  name: 1971 "Beirut system"
+  url: https://unstats.un.org/unsd/geoinfo/UNGEGN/docs/2nd-uncsgn-docs/E_Conf61_4_Add1_e.pdf
+  creation_date: 1971
+  confirmation_date: 2018-06
+  description: |
+    The current United Nations recommended romanization
+    system was approved in 2017 (resolution XI/3), based on
+    the system adopted by Arabic experts at the conference
+    held in Beirut in 2007, the Unified Arabic
+    Transliteration System, taking into account the
+    practical amendments and corrections carried out and
+    agreed upon by the representatives of the Arabic-
+    speaking countries at the Fourth Arab Conference on
+    Geographical Names, held in Beirut in 2008, and some
+    clarifications and amendments agreed in Riyadh in 20171.
+    Previously, the United Nations had approved a
+    romanization system in 1972 (resolution II/8), based on the
+    system adopted by Arabic experts at the conference
+    held at Beirut in 1971 with the practical amendments carried out
+    and agreed upon by the representatives of the Arabic-speaking
+    countries at their conference. The table was published in volume
+    II of the conference report.
+    In UN resolution XI/3 it is specifically stated that the
+    system was recommended for the “romanization of the
+    geographical names within those Arabic-speaking countries
+    where this system is officially adopted”. There is
+    evidence of its partial implementation in Jordan, Oman and
+    Saudi Arabia. The UNGEGN Working Group on Romanization
+    Systems intends to continue monitoring the UN system’s
+    implementation across Arabic-speaking countries.
+    In some countries there exist local romanization schemes
+    or practices. The geographical names of Algeria, Djibouti,
+    Mauritania, Morocco and Tunisia are generally rendered in
+    the traditional manner which conforms to the principles of
+    the French orthography.
+    The previous UN-approved system is still found in
+    considerable international usage.
+    Arabic is written from right to left. The Arabic script
+    usually omits vowel points and diacritical marks from
+    writing which makes it difficult to obtain uniform results
+    in the romanization of Arabic. It is essential to identify
+    correctly the words which appear in any particular name
+    and to know the standard Arabic-script spelling including
+    the relevant vowels. One must also take into account
+    dialectal and idiosyncratic deviations. The romanization
+    is generally reversible though there may be some ambiguous
+    letter sequences (dh, kh, sh, th) which may also point to
+    combinations of Arabic characters in addition to the
+    respective single characters.
+  notes:
+    - |
+      ث is t͟h (th with sub marcon)
+      خ is k͟h (kh with sub marcon)
+      ذ is d͟h (dh with sub marcon)
+      ش is s͟h (sh with sub marcon)
+      ظ is z͟h (zh with sub marcon)
+      غ is g͟h (gh witg sub marcon)
+      The previous UN 1972 System had the following differences:
+      the character (ظ) was romanized as z̧ instead of d͟h;
+      the cedilla (¸) was used instead of sub-macron (_) in all characters with sub-macrons.  - |
+}
+
+tests {
+  # Examples taken from:
+  # https://unstats.un.org/unsd/geoinfo/UNGEGN/docs/2nd-uncsgn-docs/E_Conf61_4_Add1_e.pdf
+  # page 31 (38 digital)
+  test "خَيبَر", "K͟haybar"
+  test "ظَهران", "Z͟hahrān"
+  test "القُدس", "Al Quds"
+  test "شَرم الشَيْخ", "S͟harm as͟h S͟hayk͟h"
+}
+
+dependency "un-ara-Arab-Latn-2017", as: arablatn
+
+stage {
+
+
+  # CHARACTERS
+  parallel {
+
+    # sun letters
+    sub boundary + "\u0627\u0644\u062b", "at͟h t͟h" # الث
+    sub boundary + "\u0627\u0644\u0630", "ad͟h d͟h" # الذ
+    sub boundary + "\u0627\u0644\u0634", "as͟h s͟h" # الش
+    sub boundary + "\u0627\u0644\u0638", "az͟h z͟h" # الظ
+
+    # shadda
+    sub "\u062e\u0651", "k͟hk͟h" # خ
+    sub "\u0630\u0651", "d͟hd͟h" # ذ
+    sub "\u0634\u0651", "s͟h" # ش
+    sub "\u0638\u0651", "z͟hz͟h" # ظ
+    sub "\u063a\u0651", "g͟hg͟h" # غ
+
+    sub "\u062b", "t͟h" # ث
+    sub "\ufe9b", "t͟h" # ﺛ
+    sub "\ufe9c", "t͟h" # ﺜ
+    sub "\ufe9a", "t͟h" # ﺚ
+
+    sub "\u062e", "k͟h" # خ
+    sub "\ufea7", "k͟h" # ﺧ
+    sub "\ufea8", "k͟h" # ﺨ
+    sub "\ufea6", "k͟h" # ﺦ
+
+    sub "\u063a", "g͟h" # غ
+    sub "\ufecf", "g͟h" # ﻏ
+    sub "\ufed0", "g͟h" # ﻐ
+    sub "\ufece", "g͟h" # ﻎ
+
+    sub "\u0630", "d͟h" # ذ
+    sub "\ufeac", "d͟h" # ﺬ
+
+    sub "\u0634", "s͟h" # ش
+    sub "\ufeb7", "s͟h" # ﺷ
+    sub "\ufeb8", "s͟h" # ﺸ
+    sub "\ufeb6", "s͟h" # ﺶ
+
+    sub "\u0638", "z͟h" # ظ
+    sub "\ufec7", "z͟h" # ﻇ
+    sub "\ufec8", "z͟h" # ﻈ
+    sub "\ufec6", "z͟h" # ﻆ
+  }
+
+  run map.arablatn.stage.main
+
+  # POSTRULES
+  sub " At͟h T͟h", " at͟h T͟h" # الث
+  sub " Ad͟h D͟h", " ad͟h D͟h" # الذ
+  sub " As͟h S͟h", " as͟h S͟h" # الش
+  sub " Az͟h Z͟h", " az͟h Z͟h" # الظ
+
+}

data/maps/un-ara-Arab-Latn-1972.imp

@@ -0,0 +1,155 @@
+metadata {
+  authority_id: ungegn
+  id: 1972
+  language: iso-639-2:ara
+  source_script: Arab
+  destination_script: Latn
+  name: ROMANIZATION OF ARABIC -- UNGEGN 1972 System
+  url: http://www.eki.ee/wgrs/obs_rom_vers/rom1_ar_v4_0.pdf
+  creation_date: 1972
+  confirmation_date: 2018-06
+  description: |
+    The United Nations recommended romanization
+    system was approved in 1972 (resolution II/8),
+    based on the system adopted by Arabic experts at
+    the conference held at Beirut in 1971 with the
+    practical amendments carried out and agreed upon
+    by the representatives of the Arabic-speaking
+    countries at their conference. The table was
+    published in volume II of the conference report1
+    . In the UN resolution it was specifically
+    pointed out that the system was recommended "for
+    the romanization of the geographical names within
+    those Arabic-speaking countries where this system
+    is officially acknowledged". It cannot be
+    definitely ascertained which of the
+    Arabicspeaking countries have adopted this system
+    officially, especially since 2007 when there are
+    efforts by the Arabic Division to promote a
+    modification of the UN system (ADEGN
+    romanization, see the section on other
+    romanization systems below), with varying
+    success2 . Judging by the use of names in
+    international cartographic products which rely
+    mostly on national sources it appears that the UN
+    system or its modification is more or less
+    current in Iraq, Kuwait, Libya, Saudi Arabia3 ,
+    United Arab Emirates and Yemen, there and in some
+    other countries the system is often used without
+    diacritical marks. For the geographical names of
+    the Syrian Arab Republic international maps
+    favour the UN system while the local usage seems
+    to prefer a French-oriented romanization. Also in
+    Egypt and Sudan there exist local romanization
+    schemes or practices side by side with the UN
+    system. The geographical names of Algeria,
+    Djibouti, Mauritania, Morocco and Tunisia are
+    generally rendered in the traditional manner
+    which conforms to the principles of the French
+    orthography. Resolution 7 of the Seventh UN
+    Conference on the Standardization of Geographical
+    Names (1998) recommended that "the League of Arab
+    States should, through its specialized
+    structures, continue its efforts to organize a
+    conference with a view to considering the
+    difficulties encountered in applying the amended
+    Beirut system of 1972 for the romanization of
+    Arabic script, and submit, as soon as possible, a
+    solution to the United Nations Group of Experts
+    on Geographical Names". At the Eighth UN
+    Conference on the Standardization of Geographical
+    Names (2002), the Arabic Division of the UN Group
+    of Experts announced that it had finalised
+    proposed modifications to the UN recommended
+    romanization system. These proposals would be
+    submitted to the League of Arab States for
+    approval. Arabic is written from right to left.
+    The Arabic script usually omits vowel points and
+    diacritical marks from writing which makes it
+    difficult to obtain uniform results in the
+    romanization of Arabic. It is essential to
+    identify correctly the words which appear in any
+    particular name and to know the standard Arabic-
+    script spelling including proper pointing. One
+    must also take into account dialectal and
+    idiosyncratic deviations. The romanization is
+    generally reversible though there are some
+    ambiguous letter sequences (dh, kh, sh, th) which
+    may also point to combinations of Arabic
+    characters in addition to the respective single
+    characters.
+  notes:
+    - |
+      The previous UN 1972 System had the following differences:
+      the character (ظ) was romanized as z̧ instead of d͟h;
+      ح, ص, ض the cedilla (¸) was used instead of sub-macron (_) in all characters with sub-macrons.  - |
+      When the definite article al precedes a word beginning with one of the "sun letters" (t,
+      th, d, dh, r, z, s, sh, ş, ḑ, ţ, z, l, n ̧ ) the l of the definite article is assimilated with the first
+      consonant of the word: ash-Sh الشارقة āriqah.
+}
+
+tests {
+  # Examples taken from:
+  # https://unstats.un.org/unsd/geoinfo/geonames/
+  test "مِصر", "Mişr"
+  test "قَطَر", "Qaţar"
+  test "الجُمهُورِيَّة العِراقِيَّة", "Al Jumhūrīyah al ‘Irāqīyah"
+  test "جُمهُورِيَّة مِصر العَرَبِيَّة", "Jumhūrīyat Mişr al ‘Arabīyah"
+  test "الرِيَاض", "Ar Riyāḑ"
+  test "الشارِقة", "Ash Shāriqah"
+}
+
+dependency "un-ara-Arab-Latn-2017", as: arablatn
+
+stage {
+
+
+  # CHARACTERS
+  parallel {
+
+    sub boundary + "\u0627\u0644\u0635", "aş ş" # الص
+    sub boundary + "\u0627\u0644\u0636", "aḑ ḑ" # الض
+    sub boundary + "\u0627\u0644\u0637", "aţ ţ" # الط
+
+    sub "\u062d\u0651", "ḩḩ" # ح
+    sub "\u0635\u0651", "şş" # ص
+    sub "\u0636\u0651", "ḑḑ" # ض
+    sub "\u0637\u0651", "ţţ" # ط
+    sub "\u0638\u0651", "z̧z̧" # ظ
+
+    sub "\u062d", "ḩ" # ح
+    sub "\ufea3", "ḩ" # ﺣ
+    sub "\ufea4", "ḩ" # ﺤ
+    sub "\ufea2", "ḩ" # ﺢ
+
+    sub "\u0635", "ş" # ص
+    sub "\ufebb", "ş" # ﺻ
+    sub "\ufebc", "ş" # ﺼ
+    sub "\ufeba", "ş" # ﺺ
+
+    sub "\u0636", "ḑ" # ض
+    sub "\ufebf", "ḑ" # ﺿ
+    sub "\ufec0", "ḑ" # ﻀ
+    sub "\ufebe", "ḑ" # ﺾ
+
+    sub "\u0637", "ţ" # ط
+    sub "\ufec3", "ţ" # ﻃ
+    sub "\ufec4", "ţ" # ﻄ
+    sub "\ufec2", "ţ" # ﻂ
+
+    sub "\u0638", "z̧" # ظ
+    sub "\ufec7", "z̧" # ﻇ
+    sub "\ufec8", "z̧" # ﻈ
+    sub "\ufec6", "z̧" # ﻆ
+  }
+
+  run map.arablatn.stage.main
+
+  # POSTRULES
+  sub " Aş Ş", " aş Ş" # الص
+  sub " Aḑ Ḑ", " aḑ Ḑ" # الض
+  sub " Aţ Ţ", " aţ Ţ" # الط
+
+compose
+
+}

data/maps/un-ara-Arab-Latn-2017.imp

@@ -0,0 +1,375 @@
+metadata {
+  authority_id: ungegn
+  id: 2017
+  language: iso-639-2:ara
+  source_script: Arab
+  destination_script: Latn
+  name: ROMANIZATION OF ARABIC -- UNGEGN 2017 System
+  url: http://www.eki.ee/wgrs/rom1_ar.pdf
+  creation_date: 2017
+  confirmation_date: 2018-06
+  description: |
+    The current United Nations recommended romanization
+    system was approved in 2017 (resolution XI/3), based on
+    the system adopted by Arabic experts at the conference
+    held in Beirut in 2007, the Unified Arabic
+    Transliteration System, taking into account the
+    practical amendments and corrections carried out and
+    agreed upon by the representatives of the Arabic-
+    speaking countries at the Fourth Arab Conference on
+    Geographical Names, held in Beirut in 2008, and some
+    clarifications and amendments agreed in Riyadh in 20171.
+    Previously, the United Nations had approved a
+    romanization system in 1972 (resolution II/8), based on the
+    system adopted by Arabic experts at the conference
+    held at Beirut in 1971 with the practical amendments carried out
+    and agreed upon by the representatives of the Arabic-speaking
+    countries at their conference. The table was published in volume
+    II of the conference report.
+    In UN resolution XI/3 it is specifically stated that the
+    system was recommended for the “romanization of the
+    geographical names within those Arabic-speaking countries
+    where this system is officially adopted”. There is
+    evidence of its partial implementation in Jordan, Oman and
+    Saudi Arabia. The UNGEGN Working Group on Romanization
+    Systems intends to continue monitoring the UN system’s
+    implementation across Arabic-speaking countries.
+    In some countries there exist local romanization schemes
+    or practices. The geographical names of Algeria, Djibouti,
+    Mauritania, Morocco and Tunisia are generally rendered in
+    the traditional manner which conforms to the principles of
+    the French orthography.
+    The previous UN-approved system is still found in
+    considerable international usage.
+    Arabic is written from right to left. The Arabic script
+    usually omits vowel points and diacritical marks from
+    writing which makes it difficult to obtain uniform results
+    in the romanization of Arabic. It is essential to identify
+    correctly the words which appear in any particular name
+    and to know the standard Arabic-script spelling including
+    the relevant vowels. One must also take into account
+    dialectal and idiosyncratic deviations. The romanization
+    is generally reversible though there may be some ambiguous
+    letter sequences (dh, kh, sh, th) which may also point to
+    combinations of Arabic characters in addition to the
+    respective single characters.
+  notes:
+    - |
+      When the definite article al precedes a word beginning with
+      one of the "sun letters" (t, th, d, dh, r, z, s, sh, s̱, ḏ, ṯ,
+      d͟h, l, n) the l of the definite article is assimilated with
+      the first consonant of the word: الشارقة Ash Shāriqah.
+    - |
+      The definite article is always written with a capital
+      initial: الزيتون Az Zaytūn, البلد Al Balad, منية الضنية Minyat Aḏ
+      Ḏinniyyah.
+    - |
+      Nunation is unlikely to be found in geographical names and
+      the last letter remains silent: جبل = جبلٌ Jabal (not Jabalun).
+    - |
+      In order to disambiguate certain character sequences a
+      middle dot (·) may be used: سهيلة S·haylah (cf. شيلة Shaylah), دهيب
+      D·hayb (cf. ذيب Dhayb), أدهم Ad·ham (cf. أذم Adham).
+    - |
+      ta' marboota should be transliterated to 'ah' if it's in
+      a definite article, or at the end of the sentence
+      otherwise it should be transliterated to 'at'
+      to handle words starting with AL and ending with ta' marboota
+      which is pronounced as "ah" not "at" divided into multiple
+      regex because lookbehind in ruby doesn't support variable length
+    - |
+      مَكّة should be transliterated to makkah, shadda above ك
+      is to double the consonant, same applies to all arabic letters
+}
+
+tests {
+  # Examples taken from:
+  # https://unstats.un.org/unsd/geoinfo/geonames/
+  test "مِصر", "Mis̱r"
+  test "قَطَر", "Qaṯar"
+  test "المَغرِب", "Al Maghrib"
+  test "الجُمهُورِيَّة العِراقِيَّة", "Al Jumhūrīyah al ‘Irāqīyah"
+  test "جُمهُورِيَّة العِراق", "Jumhūrīyat al ‘Irāq"
+  test "جُمهُورِيَّة مِصر العَرَبِيَّة", "Jumhūrīyat Mis̱r al ‘Arabīyah"
+  test "بَغداد", "Baghdād"
+  test "تُونِس", "Tūnis"
+  test "السُعُودِيَّة", "As Su‘ūdīyah"
+  test "اليَمَن", "Al Yaman"
+  test "السُودان", "As Sūdān"
+  test "الجَزائِر", "Al Jazā'ir"
+  test "الجُمهُورِيَّة اللُبنانِيَّة", "Al Jumhūrīyah al Lubnānīyah"
+  test "أسمَرة", "Asmarah"
+  test "جِدَّة", "Jiddah"
+  test "مَكَّة", "Makkah"
+  test "الرِيَاض", "Ar Riyāḏ"
+}
+
+stage {
+  # CHARACTERS
+  parallel {
+
+    # Tool used for Unicode finding:
+    # https://www.branah.com/unicode-converter
+
+    # pointing
+    sub "\u064e", "a" # َ fatha
+    sub "\u064e", "", after: "\u0629" # َ fatha followed by ta' marboota
+    sub "\u064e", "", after: "a" + any("ht") # َ fatha followed by ta' marboota, handling different order of conversion
+    sub "\u0650", "i" # ِ kasra
+    sub "\u064f", "u" # ُ damma
+    sub "\u0652", "" # ْ sokoon, see note A below
+
+
+    # special pointed letters
+    sub "\u0639\u064e", "‘a" # عَ
+    sub "\u0639\u0650", "‘i" # عِ
+    sub "\u0639\u064f", "‘ū" # عُ
+    # handle MacOS regex difference
+    sub "\u0639\u064f\u0648", "‘ū" # عُو damma followed by و
+
+    sub "\u0650\u064a", "ī" # ـِي kasra followed by ي
+    sub "\u0650\u064a\u0651\u064e", "īy" # ـِيَّ
+    sub "\u0650\u064a", "iy", after: any("\u064e\u064f") # ـِي kasra followed by ي
+    sub "\u064f\u0648", "ū" # ـُو damma followed by و
+    sub "\u064e\u0627", "ā" # ـَا fatha followed by ا
+    sub "\u064e\u0649", "á" # ـَى fatha followed by ى which is ا not ي
+    sub "\u064e\u0648\u0652", "aw" # ـَوْ
+    sub "\u064e\u064a\u0652", "ay" # ـَيْ
+    sub "\u0622", "ā" # آ
+
+    # (A) Marks absence of the vowel.
+    # (B) Marks doubling of the consonant.
+
+    # Sun letters
+    sub boundary + "\u0627\u0644\u062a", "at t" # الت
+    sub boundary + "\u0627\u0644\u062b", "ath th" # الث
+    sub boundary + "\u0627\u0644\u062f", "ad d" # الد
+    sub boundary + "\u0627\u0644\u0630", "adh dh" # الذ
+    sub boundary + "\u0627\u0644\u0631", "ar r" # الر
+    sub boundary + "\u0627\u0644\u0632", "az z" # الز
+    sub boundary + "\u0627\u0644\u0633", "as s" # الس
+    sub boundary + "\u0627\u0644\u0634", "ash sh" # الش
+    sub boundary + "\u0627\u0644\u0635", "as̱ s̱" # الص
+    sub boundary + "\u0627\u0644\u0636", "aḏ ḏ" # الض
+    sub boundary + "\u0627\u0644\u0637", "aṯ ṯ" # الط
+    sub boundary + "\u0627\u0644\u0638", "ad͟h d͟h" # الظ
+    sub boundary + "\u0627\u0644\u0644", "al l" # الل
+    sub boundary + "\u0627\u0644\u0646", "an n" # الن
+
+    # TODO: shorten this fragment
+    # ta' marboota
+    sub "\u0629", "at" # ة in the middle of the sentence
+    sub "\u0629" + line_end, "ah"
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+    sub "\u0629", "ah", before: boundary + "\u0627\u0644" + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff") + any("\u0600".."\u06ff")
+
+    # shadda
+
+    sub "\u0628\u0651", "bb" # ب
+    sub "\u062a\u0651", "tt" # ت
+    sub "\u062b\u0651", "thth" # ث
+    sub "\u062c\u0651", "jj" # ج
+    sub "\u062d\u0651", "ẖẖ" # ح
+    sub "\u062e\u0651", "khkh" # خ
+    sub "\u062f\u0651", "dd" # د
+    sub "\u0630\u0651", "dhdh" # ذ
+    sub "\u0631\u0651", "rr" # ر
+    sub "\u0632\u0651", "zz" # ز
+    sub "\u0633\u0651", "ss" # س
+    sub "\u0634\u0651", "sh" # ش
+    sub "\u0635\u0651", "s̱s̱" # ص
+    sub "\u0636\u0651", "ḏḏ" # ض
+    sub "\u0637\u0651", "ṯṯ" # ط
+    sub "\u0638\u0651", "d͟hd͟h" # ظ
+    sub "\u063a\u0651", "ghgh" # غ
+    sub "\u0641\u0651", "ff" # ف
+    sub "\u0642\u0651", "qq" # ق
+    sub "\u0643\u0651", "kk" # ك
+    sub "\u0644\u0651", "ll" # ل
+    sub "\u0645\u0651", "mm" # م
+    sub "\u0646\u0651", "nn" # ن
+    sub "\u0647\u0651", "hh" # ه
+    sub "\u0648\u0651", "ww" # و
+    sub "\u064a\u0651", "yy" # ي
+
+    sub "\u0626", "'" # ئ
+
+
+    sub "\u0621", any(["’", ""]) # ء# see note A
+
+    sub "\u0623", "a" # أ
+    sub "\u0627", "ā" # ا
+
+    # See note B
+    sub boundary + "\u0627\u0644", "al " # ال
+    # '\uFE8E' : ''  # ﺎ
+
+    sub "\u0628", "b" # ب
+    sub "\uFE91", "b" # ﺑ
+    sub "\uFE92", "b" # ﺒ
+    sub "\uFE90", "b" # ﺐ
+
+    # See note C
+    sub "\u062a", "t" # ت
+    sub "\ufe97", "t" # ﺗ
+    sub "\ufe98", "t" # ﺘ
+    sub "\ufe96", "t" # ﺖ
+
+    sub "\u062b", "th" # ث
+    sub "\ufe9b", "th" # ﺛ
+    sub "\ufe9c", "th" # ﺜ
+    sub "\ufe9a", "th" # ﺚ
+
+    sub "\u062c", "j" # ج
+    sub "\ufe9f", "j" # ﺟ
+    sub "\ufea0", "j" # ﺠ
+    sub "\ufe9e", "j" # ﺞ
+
+    sub "\u062d", "ẖ" # ح
+    sub "\ufea3", "ẖ" # ﺣ
+    sub "\ufea4", "ẖ" # ﺤ
+    sub "\ufea2", "ẖ" # ﺢ
+
+    sub "\u062e", "kh" # خ
+    sub "\ufea7", "kh" # ﺧ
+    sub "\ufea8", "kh" # ﺨ
+    sub "\ufea6", "kh" # ﺦ
+
+    sub "\u062f", "d" # د
+    sub "\ufeaa", "d" # ﺪ
+
+    sub "\u0630", "dh" # ذ
+    sub "\ufeac", "dh" # ﺬ
+
+    sub "\u0631", "r" # ر
+    sub "\ufeae", "r" # ﺮ
+
+    sub "\u0632", "z" # ز
+    sub "\ufeb0", "z" # ﺰ
+
+    sub "\u0633", "s" # س
+    sub "\ufeb3", "s" # ﺳ
+    sub "\ufeb4", "s" # ﺴ
+    sub "\ufeb2", "s" # ﺲ
+
+    sub "\u0634", "sh" # ش
+    sub "\ufeb7", "sh" # ﺷ
+    sub "\ufeb8", "sh" # ﺸ
+    sub "\ufeb6", "sh" # ﺶ
+
+    sub "\u0635", "s̱" # ص
+    sub "\ufebb", "s̱" # ﺻ
+    sub "\ufebc", "s̱" # ﺼ
+    sub "\ufeba", "s̱" # ﺺ
+
+    sub "\u0636", "ḏ" # ض
+    sub "\ufebf", "ḏ" # ﺿ
+    sub "\ufec0", "ḏ" # ﻀ
+    sub "\ufebe", "ḏ" # ﺾ
+
+    sub "\u0637", "ṯ" # ط
+    sub "\ufec3", "ṯ" # ﻃ
+    sub "\ufec4", "ṯ" # ﻄ
+    sub "\ufec2", "ṯ" # ﻂ
+
+    sub "\u0638", "d͟h" # ظ
+    sub "\ufec7", "d͟h" # ﻇ
+    sub "\ufec8", "d͟h" # ﻈ
+    sub "\ufec6", "d͟h" # ﻆ
+
+    sub "\u0639", "‘" # ع
+    sub "\ufecb", "‘" # ﻋ
+    sub "\ufecc", "‘" # ﻌ
+    sub "\ufeca", "‘" # ﻊ
+
+    sub "\u063a", "gh" # غ
+    sub "\ufecf", "gh" # ﻏ
+    sub "\ufed0", "gh" # ﻐ
+    sub "\ufece", "gh" # ﻎ
+
+    sub "\u0641", "f" # ف
+    sub "\ufed3", "f" # ﻓ
+    sub "\ufed4", "f" # ﻔ
+    sub "\ufed2", "f" # ﻒ
+
+    sub "\u0642", "q" # ق
+    sub "\ufed7", "q" # ﻗ
+    sub "\ufed8", "q" # ﻘ
+    sub "\ufed6", "q" # ﻖ
+
+    sub "\u0643", "k" # ك
+    sub "\ufedb", "k" # ﻛ
+    sub "\ufedc", "k" # ﻜ
+    sub "\ufeda", "k" # ﻚ
+
+    sub "\u0644", "l" # ل
+    sub "\ufedf", "l" # ﻟ
+    sub "\ufee0", "l" # ﻠ
+    sub "\ufede", "l" # ﻞ
+
+    sub "\u0645", "m" # م
+    sub "\ufee3", "m" # ﻣ
+    sub "\ufee4", "m" # ﻤ
+    sub "\ufee2", "m" # ﻢ
+
+    sub "\u0646", "n" # ن
+    sub "\ufee7", "n" # ﻧ
+    sub "\ufee8", "n" # ﻨ
+    sub "\ufee6", "n" # ﻦ
+
+    # See note C
+    sub "\u0647", "h" # ه
+    sub "\ufeeb", "h" # ﻫ
+    sub "\ufeec", "h" # ﻬ
+    sub "\ufeea", "h" # ﻪ
+
+    sub "\u0648", "w" # و
+    sub "\ufeee", "w" # ﻮ
+
+    sub "\u064a", "y" # ي
+    sub "\ufef3", "y" # ﻳ
+    sub "\ufef4", "y" # ﻴ
+    sub "\ufef1", "y" # ﻱ
+
+    # (A) Not romanized word-initially.
+
+    # (B) Not romanized, but see romanizations accompanying alif (ا) in the table for vowels.
+
+    # (C) In certain endings, an original tā’ (ت) is written ة, i.e., like hā’ (ه) with two dots, and is known as tā’ marbūṯah. It is romanized h, except in the construct form of feminine nouns, where it is romanized t, instead.
+
+
+    # Vowels, diphthongs and diacritical marks
+    # (ـ stands for any consonant)
+  }
+
+
+  # POSTRULES
+  sub any("\u0061".."\uFFFF"), upcase, before: boundary, not_before: boundary + any("‘’'")
+  # don't capitalize defined article in the middle of a sentence
+  sub " At T", " at T" # الت
+  sub " Ath Th", " ath th" # الث
+  sub " Ad D", " ad D" # الد
+  sub " Adh Dh", " adh Dh" # الذ
+  sub " Ar R", " ar R" # الر
+  sub " Az Z", " az Z" # الز
+  sub " As S", " as S" # الس
+  sub " Ash Sh", " ash Sh" # الش
+  sub " As̱ S̱", " as̱ S̱" # الص
+  sub " Aḏ Ḏ", " aḏ Ḏ" # الض
+  sub " Aṯ Ṯ", " aṯ Ṯ" # الط
+  sub " Ad͟h D͟h", " ad͟h D͟h" # الظ
+  sub " Al L", " al L" # الل
+  sub " An N", " an N" # الن
+  sub " Al ", " al " # ال
+
+}