From 0debbf45fce5553db475476bd45f043436b02932 Mon Sep 17 00:00:00 2001 From: Ronald Tse Date: Mon, 16 Nov 2026 21:32:21 +0800 Subject: [PATCH] feat(WO20): exhaustive per-row tests for the eight thin odni maps (coverage depth) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The full letter tables of ICS 630-01 Annexes Q/P/R/S/N/U/C/H are transcribed from the held source PDFs into row-source provenance artifacts and generated into embedded tests with scripts/gen-row-tests.mjs. No map rules changed. 314 rows transcribed, all passing end-to-end; 15 skipped printed rows and 11 source-anchored findings documented in the artifacts (kaz "Ya" case slip; tgk е footnote semantics + see-Russian cells; che palochka findings incl. missing plain-ц rule and unimplemented footnotes 1/3; mkd dze: Latin S/s encoded instead of Ѕ/ѕ). --- maps/odni-aze-Cyrl-Latn-2015.isc | 52 ++++ maps/odni-che-Cyrl-Latn-2015.isc | 97 +++++++ maps/odni-kaz-Cyrl-Latn-2015.isc | 59 +++++ maps/odni-kir-Cyrl-Latn-2015.isc | 47 ++++ maps/odni-mkd-Cyrl-Latn-2015.isc | 88 +++++++ maps/odni-tat-Cyrl-Latn-2015.isc | 51 ++++ maps/odni-tgk-Cyrl-Latn-2015.isc | 62 +++++ maps/odni-uig-Cyrl-Latn-2015.isc | 54 ++++ .../row-sources/odni-aze-Cyrl-Latn-2015.json | 158 +++++++++++ .../row-sources/odni-che-Cyrl-Latn-2015.json | 158 +++++++++++ .../row-sources/odni-kaz-Cyrl-Latn-2015.json | 162 ++++++++++++ .../row-sources/odni-kir-Cyrl-Latn-2015.json | 142 ++++++++++ .../row-sources/odni-mkd-Cyrl-Latn-2015.json | 246 ++++++++++++++++++ .../row-sources/odni-tat-Cyrl-Latn-2015.json | 154 +++++++++++ .../row-sources/odni-tgk-Cyrl-Latn-2015.json | 138 ++++++++++ .../row-sources/odni-uig-Cyrl-Latn-2015.json | 146 +++++++++++ 16 files changed, 1814 insertions(+) create mode 100644 scripts/row-sources/odni-aze-Cyrl-Latn-2015.json create mode 100644 scripts/row-sources/odni-che-Cyrl-Latn-2015.json create mode 100644 scripts/row-sources/odni-kaz-Cyrl-Latn-2015.json create mode 100644 scripts/row-sources/odni-kir-Cyrl-Latn-2015.json create mode 100644 scripts/row-sources/odni-mkd-Cyrl-Latn-2015.json create mode 100644 scripts/row-sources/odni-tat-Cyrl-Latn-2015.json create mode 100644 scripts/row-sources/odni-tgk-Cyrl-Latn-2015.json create mode 100644 scripts/row-sources/odni-uig-Cyrl-Latn-2015.json diff --git a/maps/odni-aze-Cyrl-Latn-2015.isc b/maps/odni-aze-Cyrl-Latn-2015.isc index 179da2e..9eaeef8 100644 --- a/maps/odni-aze-Cyrl-Latn-2015.isc +++ b/maps/odni-aze-Cyrl-Latn-2015.isc @@ -56,6 +56,58 @@ metadata { tests { "Рашад Садыхов" -> "Rashad Sadykhov" +# Exhaustive per-row tests for odni-aze-Cyrl-Latn-2015, generated from the source +# document's table. +# Source: ICS 630-01 Annex P (ANNEX P - Azeri_Personal_Names_FLTS (U).pdf), +# p. 2 (letter table); held source +# interscript-private-references/ics-630-01/ANNEX P - +# Azeri_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every +# value cell cross-checked against the rendered page image. Cells "г, ҝ", +# "й, ј" and "у, ү" (one value cell naming two keys) transcribed as two rows +# each. The value cells for е and э print the Latin letter e; the text layer +# stores the lookalike Cyrillic е (U+0435) — values transcribed from the +# printed image. General Notes (below the table): double digraphs as a +# single digraph (шш -> sh); ъ and ь are not transliterated. +# Provenance artifact: scripts/row-sources/odni-aze-Cyrl-Latn-2015.json (38 rows). +# generated-by: maps/scripts/gen-row-tests.mjs (2026-11-16); regenerate from the artifact, do not hand-edit. + "а" -> "a" + "б" -> "b" + "в" -> "v" + "г" -> "g" + "ҝ" -> "g" + "ғ" -> "gh" + "д" -> "d" + "е" -> "e" + "ё" -> "yo" + "ә" -> "a" + "ж" -> "zh" + "з" -> "z" + "и" -> "i" + "й" -> "y" + "ј" -> "y" + "к" -> "k" + "л" -> "l" + "м" -> "m" + "н" -> "n" + "о" -> "o" + "ө" -> "o" + "п" -> "p" + "р" -> "r" + "с" -> "s" + "т" -> "t" + "у" -> "u" + "ү" -> "u" + "ф" -> "f" + "х" -> "kh" + "һ" -> "h" + "ч" -> "ch" + "ҹ" -> "j" + "ш" -> "sh" + "щ" -> "shch" + "ы" -> "y" + "э" -> "e" + "ю" -> "yu" + "я" -> "ya" } stage main { diff --git a/maps/odni-che-Cyrl-Latn-2015.isc b/maps/odni-che-Cyrl-Latn-2015.isc index f31efc4..45a1d85 100644 --- a/maps/odni-che-Cyrl-Latn-2015.isc +++ b/maps/odni-che-Cyrl-Latn-2015.isc @@ -60,6 +60,103 @@ metadata { tests { "Ильяс Ахмадкӏант" -> "Ilyas Akhmadkant" "Ильяс Ахмадк1ант" -> "Ilyas Akhmadkant" +# Exhaustive per-row tests for odni-che-Cyrl-Latn-2015, generated from the source +# document's table. +# Source: ICS 630-01 Annex C (ANNEX C - Chechen_Personal_Names_FLTS +# (U).pdf), p. 2 (letter table); held source +# interscript-private-references/ics-630-01/ANNEX C - +# Chechen_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every +# value cell cross-checked against the rendered page image. Two-column table +# read in document order. The palochka cell "|, ӏ, 1" (three keys, one value +# cell) transcribed as three rows: U+04C0, U+04CF and ASCII "1". Footnote 2 +# (verbatim): "This is a single apostrophe as found on regular US keyboards +# (ASCII character 39)...". Other notes (verbatim): digraph interpretation +# takes precedence over individual characters (кх -> q, not k + kh); +# diphthong digraphs combine (aй -> ay, oй -> oy, эй/eй -> ey). Skipped +# printed rows: е — printed value cell "e , ye¹"; footnote 1 (verbatim): +# "Add a preceding 'y' if the vowel is preceded by the sound of a 'y' +# consonant." — the value is position-dependent, un-expressible as a plain +# row. The map implements footnote 1 as a word-initial rule (boundary+е -> +# Ye/ye), so a plain-row test of "е" would take the word-initial path and +# fail. | ъ — printed value cell "' ²,³"; footnote 3: "A word-initial hard +# sign 'ъ' is not represented, but instead is left out of the +# transliteration." — context-dependent (word-initial dropped); the map +# encodes ъ -> ' unconditionally and does not implement footnote 3. Skipped +# rather than forced. | пӏ — printed value cell "p", but the map has no пӏ +# digraph rule: lowercase пӏ falls through to п + ӏ -> "p'" and uppercase Пӏ +# is encoded as "Ph" (a value this annex never prints). The printed row +# fails against the map as encoded — recorded as a finding, never forced +# green. | цӏ — printed key is the two-glyph sequence ц + palochka (U+0446 + +# U+04CF); the map encodes the single codepoint ҹ (U+04B9) -> ts and ҹ + ӏ +# (U+04B9 + U+04CF) -> ts, but not the printed ц + ӏ sequence (which falls +# through to "ts'"). Key-mismatch finding; row left untested. | ц — printed +# row ц -> ts FAILS end-to-end: the map encodes no plain ц rule at all (only +# the single codepoint ҹ U+04B9), so printed-key input passes through +# unchanged ("ац" -> "aц"). Finding, never forced green. | 1 (the palochka +# cell's ASCII-1 key) — printed value "' ²"; FAILS end-to-end standalone: +# the map's 1 -> ӏ normalization rules (before/after not_word conditions) do +# not fire at a string edge, so standalone "1" passes through unchanged ("1" +# -> "1"). Inside a word the chain works ("а1а" -> "a'a"); the standalone +# printed-key reading fails. Finding. | ь — the Other notes state the +# standalone soft sign is not represented; the map encodes ь -> empty, which +# the generator contract cannot express as a row (no printed value cell +# exists either). Findings (map-vs-source, map untouched): пӏ (п + +# palochka): printed value "p"; map lacks the lowercase digraph rule (falls +# through to "p'") and encodes uppercase Пӏ as "Ph", which no printed cell +# supports. | цӏ (ц + palochka, U+0446+U+04CF): printed value "ts"; map +# encodes the single-codepoint ҹ (U+04B9) instead of the printed glyph +# sequence, so printed-key input falls through to "ts'". | ц: printed row ц +# -> ts has NO map rule at all (plain U+0446 passes through unchanged); only +# ҹ (U+04B9) is encoded. | 1 (palochka-cell ASCII-1 key): map normalization +# rules do not fire at string edges; standalone "1" unchanged. | ъ: footnote +# 3 (word-initial ъ dropped) is not implemented in the map (ъ -> ' +# unconditionally); row skipped as context-dependent. | е: footnote 1 +# variant ye is implemented by the map as a word-initial rule; printed +# wording differs ("preceded by the sound of a 'y' consonant"); row skipped +# as context-dependent. | Uppercase Ы carries two sub rules in the map (Ы -> +# Y at the letter table and a second Ы -> Е further down); only the printed +# lowercase row ы -> e is testable and passes. The duplicate uppercase rule +# is recorded here as an observation; map untouched this round. +# Provenance artifact: scripts/row-sources/odni-che-Cyrl-Latn-2015.json (38 rows). +# generated-by: maps/scripts/gen-row-tests.mjs (2026-11-16); regenerate from the artifact, do not hand-edit. + "а" -> "a" + "б" -> "b" + "в" -> "v" + "г" -> "g" + "гӏ" -> "gh" + "д" -> "d" + "ж" -> "j" + "з" -> "z" + "и" -> "i" + "ий" -> "i" + "й" -> "y" + "к" -> "k" + "кӏ" -> "k" + "кх" -> "q" + "къ" -> "q" + "л" -> "l" + "м" -> "m" + "н" -> "n" + "о" -> "o" + "п" -> "p" + "р" -> "r" + "с" -> "s" + "т" -> "t" + "тӏ" -> "t" + "у" -> "u" + "ф" -> "f" + "х" -> "kh" + "хь" -> "h" + "хӏ" -> "h" + "ч" -> "ch" + "чӏ" -> "ch" + "ш" -> "sh" + "ы" -> "e" + "э" -> "e" + "ю" -> "yu" + "я" -> "ya" + "Ӏ" -> "'" + "ӏ" -> "'" } stage main { diff --git a/maps/odni-kaz-Cyrl-Latn-2015.isc b/maps/odni-kaz-Cyrl-Latn-2015.isc index f492392..311cec1 100644 --- a/maps/odni-kaz-Cyrl-Latn-2015.isc +++ b/maps/odni-kaz-Cyrl-Latn-2015.isc @@ -56,6 +56,65 @@ metadata { tests { "Бекзат Саттарханов" -> "Bekzat Sattarkhanov" +# Exhaustive per-row tests for odni-kaz-Cyrl-Latn-2015, generated from the source +# document's table. +# Source: ICS 630-01 Annex Q (ANNEX Q - Kazakh_Personal_Names_FLTS (U).pdf), +# p. 2 (letter table); held source +# interscript-private-references/ics-630-01/ANNEX Q - +# Kazakh_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every +# value cell cross-checked against the rendered page image. Cell "и, і" (one +# value cell naming two keys) transcribed as two rows. The value cells for е +# and э print the Latin letter e; the text layer stores the lookalike +# Cyrillic е (U+0435) — values transcribed from the printed image. Skipped +# printed rows: я — the printed value cell reads "Ya" with a capital Y +# (verified visually at 600 dpi); the parallel cells (yo, yu) are lowercase, +# so this is a case slip in the annex. The map encodes "ya"; a row test +# transcribed as printed ("Ya") fails. Recorded as a finding, not forced +# green. | ъ, ь — not table rows; the General Notes below the table state +# they are not transliterated (left out). The map encodes them as deletions. +# Findings (map-vs-source, map untouched): я: printed cell "Ya" (capital Y) +# vs map value "ya" — annex case slip; left untested. +# Provenance artifact: scripts/row-sources/odni-kaz-Cyrl-Latn-2015.json (39 rows). +# generated-by: maps/scripts/gen-row-tests.mjs (2026-11-16); regenerate from the artifact, do not hand-edit. + "а" -> "a" + "ә" -> "a" + "б" -> "b" + "в" -> "v" + "г" -> "g" + "ғ" -> "gh" + "д" -> "d" + "е" -> "e" + "ё" -> "yo" + "ж" -> "zh" + "з" -> "z" + "и" -> "i" + "і" -> "i" + "й" -> "y" + "к" -> "k" + "қ" -> "q" + "л" -> "l" + "м" -> "m" + "н" -> "n" + "ң" -> "ng" + "о" -> "o" + "ө" -> "o" + "п" -> "p" + "р" -> "r" + "с" -> "s" + "т" -> "t" + "у" -> "u" + "ү" -> "u" + "ұ" -> "u" + "ф" -> "f" + "х" -> "kh" + "һ" -> "h" + "ц" -> "ts" + "ч" -> "ch" + "ш" -> "sh" + "щ" -> "shch" + "ы" -> "y" + "э" -> "e" + "ю" -> "yu" } stage main { diff --git a/maps/odni-kir-Cyrl-Latn-2015.isc b/maps/odni-kir-Cyrl-Latn-2015.isc index ada757c..dde45aa 100644 --- a/maps/odni-kir-Cyrl-Latn-2015.isc +++ b/maps/odni-kir-Cyrl-Latn-2015.isc @@ -56,6 +56,53 @@ metadata { tests { "Гульжигит Калыков" -> "Guljigit Kalykov" +# Exhaustive per-row tests for odni-kir-Cyrl-Latn-2015, generated from the source +# document's table. +# Source: ICS 630-01 Annex R (ANNEX R - Kyrgyz_Personal_Names_FLTS (U).pdf), +# p. 2 (letter table); held source +# interscript-private-references/ics-630-01/ANNEX R - +# Kyrgyz_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every +# value cell cross-checked against the rendered page image. The value cells +# for е and э print the Latin letter e; the text layer stores the lookalike +# Cyrillic е (U+0435) — values transcribed from the printed image. General +# Notes: double digraphs as a single digraph; ъ and ь are not +# transliterated. +# Provenance artifact: scripts/row-sources/odni-kir-Cyrl-Latn-2015.json (34 rows). +# generated-by: maps/scripts/gen-row-tests.mjs (2026-11-16); regenerate from the artifact, do not hand-edit. + "а" -> "a" + "б" -> "b" + "в" -> "v" + "г" -> "g" + "д" -> "d" + "е" -> "e" + "ё" -> "yo" + "ж" -> "j" + "з" -> "z" + "и" -> "i" + "й" -> "y" + "к" -> "k" + "л" -> "l" + "м" -> "m" + "н" -> "n" + "ң" -> "ng" + "о" -> "o" + "ө" -> "o" + "п" -> "p" + "р" -> "r" + "с" -> "s" + "т" -> "t" + "у" -> "u" + "ү" -> "u" + "ф" -> "f" + "х" -> "kh" + "ц" -> "ts" + "ч" -> "ch" + "ш" -> "sh" + "щ" -> "shch" + "ы" -> "y" + "э" -> "e" + "ю" -> "yu" + "я" -> "ya" } stage main { diff --git a/maps/odni-mkd-Cyrl-Latn-2015.isc b/maps/odni-mkd-Cyrl-Latn-2015.isc index 642054f..dd65a3d 100644 --- a/maps/odni-mkd-Cyrl-Latn-2015.isc +++ b/maps/odni-mkd-Cyrl-Latn-2015.isc @@ -50,6 +50,94 @@ metadata { tests { "Билјана" -> "Biljana" "Душко" -> "Dushko" +# Exhaustive per-row tests for odni-mkd-Cyrl-Latn-2015, generated from the source +# document's table. +# Source: ICS 630-01 Annex H (ANNEX H - Macedonian_Personal_Names_FLTS +# (U).pdf), p. 2 (letter table); held source +# interscript-private-references/ics-630-01/ANNEX H - +# Macedonian_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, +# every value cell cross-checked against the rendered page image. Each +# printed cell gives both cases (e.g. "А а | A a"); transcribed as two rows +# per cell (upper -> upper value, lower -> lower value). The dze row’s +# letter cell prints "S s" — a substitute-glyph rendering of Cyrillic Ѕ ѕ in +# this annex (verified visually; the Macedonian alphabet has no Latin S +# between З and И, and the value Dz dz is the standard dze romanization) — +# the row is transcribed as Ѕ -> Dz / ѕ -> dz and, failing end-to-end, +# recorded as a finding. The text layer for the same cell stores "Ss"/"Dz +# dz". Skipped printed rows: ’ (apostrophe sign) — printed value cell "[Not +# represented]" — instruction cell, no romanization value. | Ѕ/ѕ (dze) — +# printed value cell "Dz dz" FAILS end-to-end on the Cyrillic key: the map +# encodes the annex’s substitute glyphs literally (Latin S/s, U+0053/U+0073 +# -> Dz/dz) while the actual Cyrillic dze letters Ѕ/ѕ (U+0405/U+0455) carry +# no rule and pass through unchanged ("Ѕ" -> "Ѕ"). Finding, never forced +# green. Findings (map-vs-source, map untouched): Ѕ/ѕ (U+0405/U+0455): +# printed row Dz/dz untestable — the map has no rule for the Cyrillic dze +# letters; it instead maps Latin S/s -> Dz/dz (the annex’s substituted +# letter cell taken literally), so Cyrillic dze passes through unencoded AND +# Latin-letter input is silently rewritten to Dz/dz, against the annex’s own +# convention of spelling Latin-letter names as-is. +# Provenance artifact: scripts/row-sources/odni-mkd-Cyrl-Latn-2015.json (60 rows). +# generated-by: maps/scripts/gen-row-tests.mjs (2026-11-16); regenerate from the artifact, do not hand-edit. + "А" -> "A" + "а" -> "a" + "Б" -> "B" + "б" -> "b" + "В" -> "V" + "в" -> "v" + "Г" -> "G" + "г" -> "g" + "Д" -> "D" + "д" -> "d" + "Ѓ" -> "Gj" + "ѓ" -> "gj" + "Е" -> "E" + "е" -> "e" + "Ж" -> "Zh" + "ж" -> "zh" + "З" -> "Z" + "з" -> "z" + "И" -> "I" + "и" -> "i" + "Ј" -> "J" + "ј" -> "j" + "К" -> "K" + "к" -> "k" + "Л" -> "L" + "л" -> "l" + "Љ" -> "Lj" + "љ" -> "lj" + "М" -> "M" + "м" -> "m" + "Н" -> "N" + "н" -> "n" + "Њ" -> "Nj" + "њ" -> "nj" + "О" -> "O" + "о" -> "o" + "П" -> "P" + "п" -> "p" + "Р" -> "R" + "р" -> "r" + "С" -> "S" + "с" -> "s" + "Т" -> "T" + "т" -> "t" + "Ќ" -> "Kj" + "ќ" -> "kj" + "У" -> "U" + "у" -> "u" + "Ф" -> "F" + "ф" -> "f" + "Х" -> "H" + "х" -> "h" + "Ц" -> "Ts" + "ц" -> "ts" + "Ч" -> "Ch" + "ч" -> "ch" + "Џ" -> "Dzh" + "џ" -> "dzh" + "Ш" -> "Sh" + "ш" -> "sh" } stage main { diff --git a/maps/odni-tat-Cyrl-Latn-2015.isc b/maps/odni-tat-Cyrl-Latn-2015.isc index ed95f48..e37e987 100644 --- a/maps/odni-tat-Cyrl-Latn-2015.isc +++ b/maps/odni-tat-Cyrl-Latn-2015.isc @@ -56,6 +56,57 @@ metadata { tests { "Рустам Абдрашитов" -> "Rustam Abdrashitov" +# Exhaustive per-row tests for odni-tat-Cyrl-Latn-2015, generated from the source +# document's table. +# Source: ICS 630-01 Annex S (ANNEX S - Tatar_Personal_Names_FLTS (U).pdf), +# p. 2 (letter table); held source +# interscript-private-references/ics-630-01/ANNEX S - +# Tatar_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every +# value cell cross-checked against the rendered page image. Cell "у, ү" (one +# value cell naming two keys) transcribed as two rows. The value cells for е +# and э print the Latin letter e; the text layer stores the lookalike +# Cyrillic е (U+0435) — values transcribed from the printed image. General +# Notes: double digraphs as a single digraph; ъ and ь are not +# transliterated. +# Provenance artifact: scripts/row-sources/odni-tat-Cyrl-Latn-2015.json (37 rows). +# generated-by: maps/scripts/gen-row-tests.mjs (2026-11-16); regenerate from the artifact, do not hand-edit. + "а" -> "a" + "ә" -> "a" + "б" -> "b" + "в" -> "v" + "г" -> "g" + "д" -> "d" + "е" -> "e" + "ё" -> "yo" + "ж" -> "zh" + "җ" -> "j" + "з" -> "z" + "и" -> "i" + "й" -> "y" + "к" -> "k" + "л" -> "l" + "м" -> "m" + "н" -> "n" + "ң" -> "ng" + "о" -> "o" + "ө" -> "o" + "п" -> "p" + "р" -> "r" + "с" -> "s" + "т" -> "t" + "у" -> "u" + "ү" -> "u" + "ф" -> "f" + "х" -> "kh" + "һ" -> "h" + "ц" -> "ts" + "ч" -> "ch" + "ш" -> "sh" + "щ" -> "shch" + "ы" -> "y" + "э" -> "e" + "ю" -> "yu" + "я" -> "ya" } stage main { diff --git a/maps/odni-tgk-Cyrl-Latn-2015.isc b/maps/odni-tgk-Cyrl-Latn-2015.isc index b6c7ef3..3ccbd7f 100644 --- a/maps/odni-tgk-Cyrl-Latn-2015.isc +++ b/maps/odni-tgk-Cyrl-Latn-2015.isc @@ -57,6 +57,68 @@ metadata { tests { "Парвона Ҷамшедов" -> "Parvona Jamshedov" +# Exhaustive per-row tests for odni-tgk-Cyrl-Latn-2015, generated from the source +# document's table. +# Source: ICS 630-01 Annex N (ANNEX N - Tajik_Person-Name_IC_Standard_FLTS +# (U).pdf), p. 2 (letter table); held source +# interscript-private-references/ics-630-01/ANNEX N - +# Tajik_Person-Name_IC_Standard_FLTS (U).pdf. Extraction: annex text layer, +# every value cell cross-checked against the rendered page image. Two-column +# table read in document order (left column а..п, right column р..я). The +# value cell for э prints the Latin letter e; the text layer stores the +# lookalike Cyrillic е (U+0435) — transcribed from the printed image. +# General Notes: double digraphs as a single digraph (шш -> sh). Skipped +# printed rows: е — printed value cell "e, ye¹"; note 1: "Transliterate as +# ‘ye’ at beginning of words, ‘e’ elsewhere" — the value is +# position-dependent, un-expressible as a plain row. The map encodes е -> e +# unconditionally (its word-initial interpretation of the note is not +# implemented), so a plain-row test would also diverge from the printed +# semantics. | ъ — printed value cell "²"; note 2: "Leave out +# entirely; do not transliterate" — instruction cell, no romanization value. +# | ц, щ, ь, ы — printed value cells "³"; note 3: "Consult +# Russian IC standard for this letter" — instruction cells; no Tajik value +# is printed. (The map carries Ts/Shch/Y values for ц/щ/ы from the Russian +# standard; not testable against THIS page.) Findings (map-vs-source, map +# untouched): е: printed note-1 semantics (ye word-initial, e elsewhere) vs +# map unconditional e — map does not implement the note; row left untested. +# | ц/щ/ы: map encodes Ts/Shch/Y although the Tajik annex prints no values +# for them (defers to the Russian IC standard) — map rules exceed this page; +# rows skipped, map untouched. +# Provenance artifact: scripts/row-sources/odni-tgk-Cyrl-Latn-2015.json (33 rows). +# generated-by: maps/scripts/gen-row-tests.mjs (2026-11-16); regenerate from the artifact, do not hand-edit. + "а" -> "a" + "б" -> "b" + "в" -> "v" + "г" -> "g" + "ғ" -> "gh" + "д" -> "d" + "ё" -> "yo" + "ж" -> "zh" + "з" -> "z" + "и" -> "i" + "ӣ" -> "i" + "й" -> "y" + "к" -> "k" + "қ" -> "q" + "л" -> "l" + "м" -> "m" + "н" -> "n" + "о" -> "o" + "п" -> "p" + "р" -> "r" + "с" -> "s" + "т" -> "t" + "у" -> "u" + "ӯ" -> "u" + "ф" -> "f" + "х" -> "kh" + "ҳ" -> "h" + "ч" -> "ch" + "ҷ" -> "j" + "ш" -> "sh" + "э" -> "e" + "ю" -> "yu" + "я" -> "ya" } stage main { diff --git a/maps/odni-uig-Cyrl-Latn-2015.isc b/maps/odni-uig-Cyrl-Latn-2015.isc index 49a18f0..ee453b2 100644 --- a/maps/odni-uig-Cyrl-Latn-2015.isc +++ b/maps/odni-uig-Cyrl-Latn-2015.isc @@ -56,6 +56,60 @@ metadata { tests { "Зордун Сабир" -> "Zordun Sabir" +# Exhaustive per-row tests for odni-uig-Cyrl-Latn-2015, generated from the source +# document's table. +# Source: ICS 630-01 Annex U (ANNEX U - Uyghur_Personal_Names_FLTS (U).pdf), +# p. 2 (letter table); held source +# interscript-private-references/ics-630-01/ANNEX U - +# Uyghur_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every +# value cell cross-checked against the rendered page image. The printed +# table has three columns: Cyrillic letter, Uyghur Arabic-script spelling, +# IC Standard (Latin). The middle column is the Arabic-script equivalent and +# is not part of this Cyrl->Latn map; documented here, not transcribed as +# rows. Distinctive values verified visually: ә -> e (not a), в -> w (not +# v), х -> Latin x (not kh). The value cells for е and ә print the Latin +# letter e and х the Latin letter x; the text layer stores the lookalike +# Cyrillic glyphs (U+0435 / U+0445) — values transcribed from the printed +# image. General Notes: double digraphs as a single digraph (шш -> sh). +# Skipped printed rows: ъ, ь — not table rows; the General Notes state they +# are not transliterated (left out). +# Provenance artifact: scripts/row-sources/odni-uig-Cyrl-Latn-2015.json (35 rows). +# generated-by: maps/scripts/gen-row-tests.mjs (2026-11-16); regenerate from the artifact, do not hand-edit. + "а" -> "a" + "ә" -> "e" + "б" -> "b" + "в" -> "w" + "г" -> "g" + "ғ" -> "gh" + "д" -> "d" + "е" -> "e" + "ё" -> "yo" + "ж" -> "j" + "җ" -> "zh" + "з" -> "z" + "и" -> "i" + "й" -> "y" + "к" -> "k" + "қ" -> "q" + "л" -> "l" + "м" -> "m" + "н" -> "n" + "ң" -> "ng" + "о" -> "o" + "ө" -> "o" + "п" -> "p" + "р" -> "r" + "с" -> "s" + "т" -> "t" + "у" -> "u" + "ү" -> "u" + "ф" -> "f" + "х" -> "x" + "һ" -> "h" + "ч" -> "ch" + "ш" -> "sh" + "ю" -> "yu" + "я" -> "ya" } stage main { diff --git a/scripts/row-sources/odni-aze-Cyrl-Latn-2015.json b/scripts/row-sources/odni-aze-Cyrl-Latn-2015.json new file mode 100644 index 0000000..ab78422 --- /dev/null +++ b/scripts/row-sources/odni-aze-Cyrl-Latn-2015.json @@ -0,0 +1,158 @@ +{ + "system": "odni-aze-Cyrl-Latn-2015", + "rows": [ + [ + "а", + "a" + ], + [ + "б", + "b" + ], + [ + "в", + "v" + ], + [ + "г", + "g" + ], + [ + "ҝ", + "g" + ], + [ + "ғ", + "gh" + ], + [ + "д", + "d" + ], + [ + "е", + "e" + ], + [ + "ё", + "yo" + ], + [ + "ә", + "a" + ], + [ + "ж", + "zh" + ], + [ + "з", + "z" + ], + [ + "и", + "i" + ], + [ + "й", + "y" + ], + [ + "ј", + "y" + ], + [ + "к", + "k" + ], + [ + "л", + "l" + ], + [ + "м", + "m" + ], + [ + "н", + "n" + ], + [ + "о", + "o" + ], + [ + "ө", + "o" + ], + [ + "п", + "p" + ], + [ + "р", + "r" + ], + [ + "с", + "s" + ], + [ + "т", + "t" + ], + [ + "у", + "u" + ], + [ + "ү", + "u" + ], + [ + "ф", + "f" + ], + [ + "х", + "kh" + ], + [ + "һ", + "h" + ], + [ + "ч", + "ch" + ], + [ + "ҹ", + "j" + ], + [ + "ш", + "sh" + ], + [ + "щ", + "shch" + ], + [ + "ы", + "y" + ], + [ + "э", + "e" + ], + [ + "ю", + "yu" + ], + [ + "я", + "ya" + ] + ], + "note": "ICS 630-01 Annex P (ANNEX P - Azeri_Personal_Names_FLTS (U).pdf), p. 2 (letter table); held source interscript-private-references/ics-630-01/ANNEX P - Azeri_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every value cell cross-checked against the rendered page image. Cells \"г, ҝ\", \"й, ј\" and \"у, ү\" (one value cell naming two keys) transcribed as two rows each. The value cells for е and э print the Latin letter e; the text layer stores the lookalike Cyrillic е (U+0435) — values transcribed from the printed image. General Notes (below the table): double digraphs as a single digraph (шш -> sh); ъ and ь are not transliterated." +} diff --git a/scripts/row-sources/odni-che-Cyrl-Latn-2015.json b/scripts/row-sources/odni-che-Cyrl-Latn-2015.json new file mode 100644 index 0000000..574c514 --- /dev/null +++ b/scripts/row-sources/odni-che-Cyrl-Latn-2015.json @@ -0,0 +1,158 @@ +{ + "system": "odni-che-Cyrl-Latn-2015", + "rows": [ + [ + "а", + "a" + ], + [ + "б", + "b" + ], + [ + "в", + "v" + ], + [ + "г", + "g" + ], + [ + "гӏ", + "gh" + ], + [ + "д", + "d" + ], + [ + "ж", + "j" + ], + [ + "з", + "z" + ], + [ + "и", + "i" + ], + [ + "ий", + "i" + ], + [ + "й", + "y" + ], + [ + "к", + "k" + ], + [ + "кӏ", + "k" + ], + [ + "кх", + "q" + ], + [ + "къ", + "q" + ], + [ + "л", + "l" + ], + [ + "м", + "m" + ], + [ + "н", + "n" + ], + [ + "о", + "o" + ], + [ + "п", + "p" + ], + [ + "р", + "r" + ], + [ + "с", + "s" + ], + [ + "т", + "t" + ], + [ + "тӏ", + "t" + ], + [ + "у", + "u" + ], + [ + "ф", + "f" + ], + [ + "х", + "kh" + ], + [ + "хь", + "h" + ], + [ + "хӏ", + "h" + ], + [ + "ч", + "ch" + ], + [ + "чӏ", + "ch" + ], + [ + "ш", + "sh" + ], + [ + "ы", + "e" + ], + [ + "э", + "e" + ], + [ + "ю", + "yu" + ], + [ + "я", + "ya" + ], + [ + "Ӏ", + "'" + ], + [ + "ӏ", + "'" + ] + ], + "note": "ICS 630-01 Annex C (ANNEX C - Chechen_Personal_Names_FLTS (U).pdf), p. 2 (letter table); held source interscript-private-references/ics-630-01/ANNEX C - Chechen_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every value cell cross-checked against the rendered page image. Two-column table read in document order. The palochka cell \"|, ӏ, 1\" (three keys, one value cell) transcribed as three rows: U+04C0, U+04CF and ASCII \"1\". Footnote 2 (verbatim): \"This is a single apostrophe as found on regular US keyboards (ASCII character 39)...\". Other notes (verbatim): digraph interpretation takes precedence over individual characters (кх -> q, not k + kh); diphthong digraphs combine (aй -> ay, oй -> oy, эй/eй -> ey). Skipped printed rows: е — printed value cell \"e , ye¹\"; footnote 1 (verbatim): \"Add a preceding 'y' if the vowel is preceded by the sound of a 'y' consonant.\" — the value is position-dependent, un-expressible as a plain row. The map implements footnote 1 as a word-initial rule (boundary+е -> Ye/ye), so a plain-row test of \"е\" would take the word-initial path and fail. | ъ — printed value cell \"' ²,³\"; footnote 3: \"A word-initial hard sign 'ъ' is not represented, but instead is left out of the transliteration.\" — context-dependent (word-initial dropped); the map encodes ъ -> ' unconditionally and does not implement footnote 3. Skipped rather than forced. | пӏ — printed value cell \"p\", but the map has no пӏ digraph rule: lowercase пӏ falls through to п + ӏ -> \"p'\" and uppercase Пӏ is encoded as \"Ph\" (a value this annex never prints). The printed row fails against the map as encoded — recorded as a finding, never forced green. | цӏ — printed key is the two-glyph sequence ц + palochka (U+0446 + U+04CF); the map encodes the single codepoint ҹ (U+04B9) -> ts and ҹ + ӏ (U+04B9 + U+04CF) -> ts, but not the printed ц + ӏ sequence (which falls through to \"ts'\"). Key-mismatch finding; row left untested. | ц — printed row ц -> ts FAILS end-to-end: the map encodes no plain ц rule at all (only the single codepoint ҹ U+04B9), so printed-key input passes through unchanged (\"ац\" -> \"aц\"). Finding, never forced green. | 1 (the palochka cell's ASCII-1 key) — printed value \"' ²\"; FAILS end-to-end standalone: the map's 1 -> ӏ normalization rules (before/after not_word conditions) do not fire at a string edge, so standalone \"1\" passes through unchanged (\"1\" -> \"1\"). Inside a word the chain works (\"а1а\" -> \"a'a\"); the standalone printed-key reading fails. Finding. | ь — the Other notes state the standalone soft sign is not represented; the map encodes ь -> empty, which the generator contract cannot express as a row (no printed value cell exists either). Findings (map-vs-source, map untouched): пӏ (п + palochka): printed value \"p\"; map lacks the lowercase digraph rule (falls through to \"p'\") and encodes uppercase Пӏ as \"Ph\", which no printed cell supports. | цӏ (ц + palochka, U+0446+U+04CF): printed value \"ts\"; map encodes the single-codepoint ҹ (U+04B9) instead of the printed glyph sequence, so printed-key input falls through to \"ts'\". | ц: printed row ц -> ts has NO map rule at all (plain U+0446 passes through unchanged); only ҹ (U+04B9) is encoded. | 1 (palochka-cell ASCII-1 key): map normalization rules do not fire at string edges; standalone \"1\" unchanged. | ъ: footnote 3 (word-initial ъ dropped) is not implemented in the map (ъ -> ' unconditionally); row skipped as context-dependent. | е: footnote 1 variant ye is implemented by the map as a word-initial rule; printed wording differs (\"preceded by the sound of a 'y' consonant\"); row skipped as context-dependent. | Uppercase Ы carries two sub rules in the map (Ы -> Y at the letter table and a second Ы -> Е further down); only the printed lowercase row ы -> e is testable and passes. The duplicate uppercase rule is recorded here as an observation; map untouched this round." +} diff --git a/scripts/row-sources/odni-kaz-Cyrl-Latn-2015.json b/scripts/row-sources/odni-kaz-Cyrl-Latn-2015.json new file mode 100644 index 0000000..a4800e7 --- /dev/null +++ b/scripts/row-sources/odni-kaz-Cyrl-Latn-2015.json @@ -0,0 +1,162 @@ +{ + "system": "odni-kaz-Cyrl-Latn-2015", + "rows": [ + [ + "а", + "a" + ], + [ + "ә", + "a" + ], + [ + "б", + "b" + ], + [ + "в", + "v" + ], + [ + "г", + "g" + ], + [ + "ғ", + "gh" + ], + [ + "д", + "d" + ], + [ + "е", + "e" + ], + [ + "ё", + "yo" + ], + [ + "ж", + "zh" + ], + [ + "з", + "z" + ], + [ + "и", + "i" + ], + [ + "і", + "i" + ], + [ + "й", + "y" + ], + [ + "к", + "k" + ], + [ + "қ", + "q" + ], + [ + "л", + "l" + ], + [ + "м", + "m" + ], + [ + "н", + "n" + ], + [ + "ң", + "ng" + ], + [ + "о", + "o" + ], + [ + "ө", + "o" + ], + [ + "п", + "p" + ], + [ + "р", + "r" + ], + [ + "с", + "s" + ], + [ + "т", + "t" + ], + [ + "у", + "u" + ], + [ + "ү", + "u" + ], + [ + "ұ", + "u" + ], + [ + "ф", + "f" + ], + [ + "х", + "kh" + ], + [ + "һ", + "h" + ], + [ + "ц", + "ts" + ], + [ + "ч", + "ch" + ], + [ + "ш", + "sh" + ], + [ + "щ", + "shch" + ], + [ + "ы", + "y" + ], + [ + "э", + "e" + ], + [ + "ю", + "yu" + ] + ], + "note": "ICS 630-01 Annex Q (ANNEX Q - Kazakh_Personal_Names_FLTS (U).pdf), p. 2 (letter table); held source interscript-private-references/ics-630-01/ANNEX Q - Kazakh_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every value cell cross-checked against the rendered page image. Cell \"и, і\" (one value cell naming two keys) transcribed as two rows. The value cells for е and э print the Latin letter e; the text layer stores the lookalike Cyrillic е (U+0435) — values transcribed from the printed image. Skipped printed rows: я — the printed value cell reads \"Ya\" with a capital Y (verified visually at 600 dpi); the parallel cells (yo, yu) are lowercase, so this is a case slip in the annex. The map encodes \"ya\"; a row test transcribed as printed (\"Ya\") fails. Recorded as a finding, not forced green. | ъ, ь — not table rows; the General Notes below the table state they are not transliterated (left out). The map encodes them as deletions. Findings (map-vs-source, map untouched): я: printed cell \"Ya\" (capital Y) vs map value \"ya\" — annex case slip; left untested." +} diff --git a/scripts/row-sources/odni-kir-Cyrl-Latn-2015.json b/scripts/row-sources/odni-kir-Cyrl-Latn-2015.json new file mode 100644 index 0000000..4dd3b9f --- /dev/null +++ b/scripts/row-sources/odni-kir-Cyrl-Latn-2015.json @@ -0,0 +1,142 @@ +{ + "system": "odni-kir-Cyrl-Latn-2015", + "rows": [ + [ + "а", + "a" + ], + [ + "б", + "b" + ], + [ + "в", + "v" + ], + [ + "г", + "g" + ], + [ + "д", + "d" + ], + [ + "е", + "e" + ], + [ + "ё", + "yo" + ], + [ + "ж", + "j" + ], + [ + "з", + "z" + ], + [ + "и", + "i" + ], + [ + "й", + "y" + ], + [ + "к", + "k" + ], + [ + "л", + "l" + ], + [ + "м", + "m" + ], + [ + "н", + "n" + ], + [ + "ң", + "ng" + ], + [ + "о", + "o" + ], + [ + "ө", + "o" + ], + [ + "п", + "p" + ], + [ + "р", + "r" + ], + [ + "с", + "s" + ], + [ + "т", + "t" + ], + [ + "у", + "u" + ], + [ + "ү", + "u" + ], + [ + "ф", + "f" + ], + [ + "х", + "kh" + ], + [ + "ц", + "ts" + ], + [ + "ч", + "ch" + ], + [ + "ш", + "sh" + ], + [ + "щ", + "shch" + ], + [ + "ы", + "y" + ], + [ + "э", + "e" + ], + [ + "ю", + "yu" + ], + [ + "я", + "ya" + ] + ], + "note": "ICS 630-01 Annex R (ANNEX R - Kyrgyz_Personal_Names_FLTS (U).pdf), p. 2 (letter table); held source interscript-private-references/ics-630-01/ANNEX R - Kyrgyz_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every value cell cross-checked against the rendered page image. The value cells for е and э print the Latin letter e; the text layer stores the lookalike Cyrillic е (U+0435) — values transcribed from the printed image. General Notes: double digraphs as a single digraph; ъ and ь are not transliterated." +} diff --git a/scripts/row-sources/odni-mkd-Cyrl-Latn-2015.json b/scripts/row-sources/odni-mkd-Cyrl-Latn-2015.json new file mode 100644 index 0000000..2014adb --- /dev/null +++ b/scripts/row-sources/odni-mkd-Cyrl-Latn-2015.json @@ -0,0 +1,246 @@ +{ + "system": "odni-mkd-Cyrl-Latn-2015", + "rows": [ + [ + "А", + "A" + ], + [ + "а", + "a" + ], + [ + "Б", + "B" + ], + [ + "б", + "b" + ], + [ + "В", + "V" + ], + [ + "в", + "v" + ], + [ + "Г", + "G" + ], + [ + "г", + "g" + ], + [ + "Д", + "D" + ], + [ + "д", + "d" + ], + [ + "Ѓ", + "Gj" + ], + [ + "ѓ", + "gj" + ], + [ + "Е", + "E" + ], + [ + "е", + "e" + ], + [ + "Ж", + "Zh" + ], + [ + "ж", + "zh" + ], + [ + "З", + "Z" + ], + [ + "з", + "z" + ], + [ + "И", + "I" + ], + [ + "и", + "i" + ], + [ + "Ј", + "J" + ], + [ + "ј", + "j" + ], + [ + "К", + "K" + ], + [ + "к", + "k" + ], + [ + "Л", + "L" + ], + [ + "л", + "l" + ], + [ + "Љ", + "Lj" + ], + [ + "љ", + "lj" + ], + [ + "М", + "M" + ], + [ + "м", + "m" + ], + [ + "Н", + "N" + ], + [ + "н", + "n" + ], + [ + "Њ", + "Nj" + ], + [ + "њ", + "nj" + ], + [ + "О", + "O" + ], + [ + "о", + "o" + ], + [ + "П", + "P" + ], + [ + "п", + "p" + ], + [ + "Р", + "R" + ], + [ + "р", + "r" + ], + [ + "С", + "S" + ], + [ + "с", + "s" + ], + [ + "Т", + "T" + ], + [ + "т", + "t" + ], + [ + "Ќ", + "Kj" + ], + [ + "ќ", + "kj" + ], + [ + "У", + "U" + ], + [ + "у", + "u" + ], + [ + "Ф", + "F" + ], + [ + "ф", + "f" + ], + [ + "Х", + "H" + ], + [ + "х", + "h" + ], + [ + "Ц", + "Ts" + ], + [ + "ц", + "ts" + ], + [ + "Ч", + "Ch" + ], + [ + "ч", + "ch" + ], + [ + "Џ", + "Dzh" + ], + [ + "џ", + "dzh" + ], + [ + "Ш", + "Sh" + ], + [ + "ш", + "sh" + ] + ], + "note": "ICS 630-01 Annex H (ANNEX H - Macedonian_Personal_Names_FLTS (U).pdf), p. 2 (letter table); held source interscript-private-references/ics-630-01/ANNEX H - Macedonian_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every value cell cross-checked against the rendered page image. Each printed cell gives both cases (e.g. \"А а | A a\"); transcribed as two rows per cell (upper -> upper value, lower -> lower value). The dze row’s letter cell prints \"S s\" — a substitute-glyph rendering of Cyrillic Ѕ ѕ in this annex (verified visually; the Macedonian alphabet has no Latin S between З and И, and the value Dz dz is the standard dze romanization) — the row is transcribed as Ѕ -> Dz / ѕ -> dz and, failing end-to-end, recorded as a finding. The text layer for the same cell stores \"Ss\"/\"Dz dz\". Skipped printed rows: ’ (apostrophe sign) — printed value cell \"[Not represented]\" — instruction cell, no romanization value. | Ѕ/ѕ (dze) — printed value cell \"Dz dz\" FAILS end-to-end on the Cyrillic key: the map encodes the annex’s substitute glyphs literally (Latin S/s, U+0053/U+0073 -> Dz/dz) while the actual Cyrillic dze letters Ѕ/ѕ (U+0405/U+0455) carry no rule and pass through unchanged (\"Ѕ\" -> \"Ѕ\"). Finding, never forced green. Findings (map-vs-source, map untouched): Ѕ/ѕ (U+0405/U+0455): printed row Dz/dz untestable — the map has no rule for the Cyrillic dze letters; it instead maps Latin S/s -> Dz/dz (the annex’s substituted letter cell taken literally), so Cyrillic dze passes through unencoded AND Latin-letter input is silently rewritten to Dz/dz, against the annex’s own convention of spelling Latin-letter names as-is." +} diff --git a/scripts/row-sources/odni-tat-Cyrl-Latn-2015.json b/scripts/row-sources/odni-tat-Cyrl-Latn-2015.json new file mode 100644 index 0000000..dc128c5 --- /dev/null +++ b/scripts/row-sources/odni-tat-Cyrl-Latn-2015.json @@ -0,0 +1,154 @@ +{ + "system": "odni-tat-Cyrl-Latn-2015", + "rows": [ + [ + "а", + "a" + ], + [ + "ә", + "a" + ], + [ + "б", + "b" + ], + [ + "в", + "v" + ], + [ + "г", + "g" + ], + [ + "д", + "d" + ], + [ + "е", + "e" + ], + [ + "ё", + "yo" + ], + [ + "ж", + "zh" + ], + [ + "җ", + "j" + ], + [ + "з", + "z" + ], + [ + "и", + "i" + ], + [ + "й", + "y" + ], + [ + "к", + "k" + ], + [ + "л", + "l" + ], + [ + "м", + "m" + ], + [ + "н", + "n" + ], + [ + "ң", + "ng" + ], + [ + "о", + "o" + ], + [ + "ө", + "o" + ], + [ + "п", + "p" + ], + [ + "р", + "r" + ], + [ + "с", + "s" + ], + [ + "т", + "t" + ], + [ + "у", + "u" + ], + [ + "ү", + "u" + ], + [ + "ф", + "f" + ], + [ + "х", + "kh" + ], + [ + "һ", + "h" + ], + [ + "ц", + "ts" + ], + [ + "ч", + "ch" + ], + [ + "ш", + "sh" + ], + [ + "щ", + "shch" + ], + [ + "ы", + "y" + ], + [ + "э", + "e" + ], + [ + "ю", + "yu" + ], + [ + "я", + "ya" + ] + ], + "note": "ICS 630-01 Annex S (ANNEX S - Tatar_Personal_Names_FLTS (U).pdf), p. 2 (letter table); held source interscript-private-references/ics-630-01/ANNEX S - Tatar_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every value cell cross-checked against the rendered page image. Cell \"у, ү\" (one value cell naming two keys) transcribed as two rows. The value cells for е and э print the Latin letter e; the text layer stores the lookalike Cyrillic е (U+0435) — values transcribed from the printed image. General Notes: double digraphs as a single digraph; ъ and ь are not transliterated." +} diff --git a/scripts/row-sources/odni-tgk-Cyrl-Latn-2015.json b/scripts/row-sources/odni-tgk-Cyrl-Latn-2015.json new file mode 100644 index 0000000..768620a --- /dev/null +++ b/scripts/row-sources/odni-tgk-Cyrl-Latn-2015.json @@ -0,0 +1,138 @@ +{ + "system": "odni-tgk-Cyrl-Latn-2015", + "rows": [ + [ + "а", + "a" + ], + [ + "б", + "b" + ], + [ + "в", + "v" + ], + [ + "г", + "g" + ], + [ + "ғ", + "gh" + ], + [ + "д", + "d" + ], + [ + "ё", + "yo" + ], + [ + "ж", + "zh" + ], + [ + "з", + "z" + ], + [ + "и", + "i" + ], + [ + "ӣ", + "i" + ], + [ + "й", + "y" + ], + [ + "к", + "k" + ], + [ + "қ", + "q" + ], + [ + "л", + "l" + ], + [ + "м", + "m" + ], + [ + "н", + "n" + ], + [ + "о", + "o" + ], + [ + "п", + "p" + ], + [ + "р", + "r" + ], + [ + "с", + "s" + ], + [ + "т", + "t" + ], + [ + "у", + "u" + ], + [ + "ӯ", + "u" + ], + [ + "ф", + "f" + ], + [ + "х", + "kh" + ], + [ + "ҳ", + "h" + ], + [ + "ч", + "ch" + ], + [ + "ҷ", + "j" + ], + [ + "ш", + "sh" + ], + [ + "э", + "e" + ], + [ + "ю", + "yu" + ], + [ + "я", + "ya" + ] + ], + "note": "ICS 630-01 Annex N (ANNEX N - Tajik_Person-Name_IC_Standard_FLTS (U).pdf), p. 2 (letter table); held source interscript-private-references/ics-630-01/ANNEX N - Tajik_Person-Name_IC_Standard_FLTS (U).pdf. Extraction: annex text layer, every value cell cross-checked against the rendered page image. Two-column table read in document order (left column а..п, right column р..я). The value cell for э prints the Latin letter e; the text layer stores the lookalike Cyrillic е (U+0435) — transcribed from the printed image. General Notes: double digraphs as a single digraph (шш -> sh). Skipped printed rows: е — printed value cell \"e, ye¹\"; note 1: \"Transliterate as ‘ye’ at beginning of words, ‘e’ elsewhere\" — the value is position-dependent, un-expressible as a plain row. The map encodes е -> e unconditionally (its word-initial interpretation of the note is not implemented), so a plain-row test would also diverge from the printed semantics. | ъ — printed value cell \"²\"; note 2: \"Leave out entirely; do not transliterate\" — instruction cell, no romanization value. | ц, щ, ь, ы — printed value cells \"³\"; note 3: \"Consult Russian IC standard for this letter\" — instruction cells; no Tajik value is printed. (The map carries Ts/Shch/Y values for ц/щ/ы from the Russian standard; not testable against THIS page.) Findings (map-vs-source, map untouched): е: printed note-1 semantics (ye word-initial, e elsewhere) vs map unconditional e — map does not implement the note; row left untested. | ц/щ/ы: map encodes Ts/Shch/Y although the Tajik annex prints no values for them (defers to the Russian IC standard) — map rules exceed this page; rows skipped, map untouched." +} diff --git a/scripts/row-sources/odni-uig-Cyrl-Latn-2015.json b/scripts/row-sources/odni-uig-Cyrl-Latn-2015.json new file mode 100644 index 0000000..ee301f3 --- /dev/null +++ b/scripts/row-sources/odni-uig-Cyrl-Latn-2015.json @@ -0,0 +1,146 @@ +{ + "system": "odni-uig-Cyrl-Latn-2015", + "rows": [ + [ + "а", + "a" + ], + [ + "ә", + "e" + ], + [ + "б", + "b" + ], + [ + "в", + "w" + ], + [ + "г", + "g" + ], + [ + "ғ", + "gh" + ], + [ + "д", + "d" + ], + [ + "е", + "e" + ], + [ + "ё", + "yo" + ], + [ + "ж", + "j" + ], + [ + "җ", + "zh" + ], + [ + "з", + "z" + ], + [ + "и", + "i" + ], + [ + "й", + "y" + ], + [ + "к", + "k" + ], + [ + "қ", + "q" + ], + [ + "л", + "l" + ], + [ + "м", + "m" + ], + [ + "н", + "n" + ], + [ + "ң", + "ng" + ], + [ + "о", + "o" + ], + [ + "ө", + "o" + ], + [ + "п", + "p" + ], + [ + "р", + "r" + ], + [ + "с", + "s" + ], + [ + "т", + "t" + ], + [ + "у", + "u" + ], + [ + "ү", + "u" + ], + [ + "ф", + "f" + ], + [ + "х", + "x" + ], + [ + "һ", + "h" + ], + [ + "ч", + "ch" + ], + [ + "ш", + "sh" + ], + [ + "ю", + "yu" + ], + [ + "я", + "ya" + ] + ], + "note": "ICS 630-01 Annex U (ANNEX U - Uyghur_Personal_Names_FLTS (U).pdf), p. 2 (letter table); held source interscript-private-references/ics-630-01/ANNEX U - Uyghur_Personal_Names_FLTS (U).pdf. Extraction: annex text layer, every value cell cross-checked against the rendered page image. The printed table has three columns: Cyrillic letter, Uyghur Arabic-script spelling, IC Standard (Latin). The middle column is the Arabic-script equivalent and is not part of this Cyrl->Latn map; documented here, not transcribed as rows. Distinctive values verified visually: ә -> e (not a), в -> w (not v), х -> Latin x (not kh). The value cells for е and ә print the Latin letter e and х the Latin letter x; the text layer stores the lookalike Cyrillic glyphs (U+0435 / U+0445) — values transcribed from the printed image. General Notes: double digraphs as a single digraph (шш -> sh). Skipped printed rows: ъ, ь — not table rows; the General Notes state they are not transliterated (left out)." +}