Update infrastructure & tests (WiP)

ebl/atf_importer/domain/atf_conversions.py

@@ -40,7 +40,7 @@ def control_line(self, tree: Tree) -> str:
 class GetLineNumber(Visitor):
     number: str = ""
 
-    def ebl_atf_text_line__single_line_number(self, tree: Tree) -> str:
+    def ebl_atf_text_line__ebl_atf_common__single_line_number(self, tree: Tree) -> str:
         result = DepthFirstSearch().visit_topdown(tree, "")
         self.number += result
         return result

ebl/corpus/domain/chapter_transformer.py

@@ -1,13 +1,10 @@
-from typing import Iterable
-
 from lark.visitors import v_args
 from ebl.common.domain.period import Period
 
 from ebl.corpus.domain.line import Line
 from ebl.corpus.domain.manuscript_line import ManuscriptLine
 from ebl.corpus.domain.line_variant import LineVariant
 from ebl.corpus.domain.manuscript import (
-    Manuscript,
     Siglum,
 )
 from ebl.common.domain.manuscript_type import ManuscriptType
@@ -16,23 +13,28 @@
 
 
 class ChapterTransformer(LineTransformer):
-    def __init__(self, manuscripts: Iterable[Manuscript]):
-        self._manuscripts = {
-            manuscript.siglum: manuscript.id for manuscript in manuscripts
-        }
-
     def manuscript_label(self, children):
         return children
 
-    @v_args(inline=True)
-    def siglum(self, provenance, period, type_, disambiquator):
+    def get_siglum(self, provenance, period, type_, disambiquator):
         return Siglum(
             Provenance.from_abbreviation(provenance or ""),
             Period.from_abbreviation(period),
             ManuscriptType.from_abbreviation(type_ or ""),
             disambiquator or "",
         )
 
+    @v_args(inline=True)
+    def siglum(self, provenance=None, period=None, type_=None, disambiquator=None):
+        if period:
+            return self.get_siglum(provenance, period, type_, disambiquator)
+        else:
+            return self.standard_text_siglum(disambiquator)
+
+    @v_args(inline=True)
+    def ebl_atf_manuscript_line__siglum(self, provenance, period, type_, disambiquator):
+        return self.get_siglum(provenance, period, type_, disambiquator)
+
     @v_args(inline=True)
     def standard_text_siglum(self, disambiquator):
         return Siglum(
@@ -44,9 +46,7 @@ def standard_text_siglum(self, disambiquator):
 
     @v_args(inline=True)
     def manuscript_line(self, siglum, labels, line, *paratext):
-        return ManuscriptLine(
-            self._manuscripts[siglum], labels or (), line, tuple(paratext)
-        )
+        return ManuscriptLine(siglum, labels or (), line, tuple(paratext))
 
     @v_args(inline=True)
     def reconstruction(self, line, note, *parallels):

ebl/corpus/domain/parser.py

@@ -5,7 +5,11 @@
 from ebl.corpus.domain.manuscript import Manuscript
 from ebl.errors import DataError
 from ebl.transliteration.domain.dollar_line import DollarLine
-from ebl.transliteration.domain.atf_parsers.lark_parser import CHAPTER_PARSER, PARATEXT_PARSER
+from ebl.transliteration.domain.atf_parsers.lark_parser import (
+    CHAPTER_PARSER,
+    PARATEXT_PARSER,
+    MANUSCRIPT_PARSER,
+)
 from ebl.transliteration.domain.atf_parsers.lark_parser_errors import PARSE_ERRORS
 from ebl.transliteration.domain.note_line import NoteLine
 
@@ -14,7 +18,17 @@ def parse_chapter(
     atf: str, manuscripts: Iterable[Manuscript], start: Optional[str] = None
 ) -> Sequence[Line]:
     try:
-        tree = CHAPTER_PARSER.parse(atf)
+        tree = CHAPTER_PARSER.parse(atf, start=start)
+        return ChapterTransformer(manuscripts).transform(tree)
+    except PARSE_ERRORS as error:
+        raise DataError(error) from error
+
+
+def parse_manuscript(
+    atf: str, manuscripts: Iterable[Manuscript], start: Optional[str] = None
+) -> Sequence[Line]:
+    try:
+        tree = MANUSCRIPT_PARSER.parse(atf, start=start)
         return ChapterTransformer(manuscripts).transform(tree)
     except PARSE_ERRORS as error:
         raise DataError(error) from error

ebl/tests/corpus/test_parse_chapter.py

@@ -12,7 +12,11 @@
 )
 from ebl.common.domain.manuscript_type import ManuscriptType
 from ebl.common.domain.provenance import Provenance
-from ebl.corpus.domain.parser import parse_chapter, parse_paratext
+from ebl.corpus.domain.parser import (
+    parse_chapter,
+    parse_paratext,
+    parse_manuscript as _parse_manuscript,
+)
 from ebl.errors import DataError
 from ebl.tests.factories.corpus import ManuscriptFactory
 from ebl.transliteration.domain.labels import parse_labels
@@ -34,7 +38,7 @@
 
 
 def parse_siglum(siglum):
-    return parse_chapter(siglum, MANUSCRIPTS, "siglum")
+    return _parse_manuscript(siglum, MANUSCRIPTS, "siglum")
 
 
 @pytest.mark.parametrize("period", [Period.NEO_ASSYRIAN])
@@ -61,7 +65,7 @@ def test_parse_siglum_standard_text(disambiquator: str) -> None:
 
 
 def parse_manuscript(atf):
-    return parse_chapter(atf, MANUSCRIPTS, "manuscript_line")
+    return _parse_manuscript(atf, MANUSCRIPTS, "manuscript_line")
 
 
 @pytest.mark.parametrize(
@@ -224,21 +228,21 @@ def parse_chapter_line(atf):
 @pytest.mark.parametrize(
     "lines,expected",
     [
-        (["1. kur"], Line(LineNumber(1), (parse_line_variant("1. kur")[1],))),
+        (["1. kur"], Line(LineNumber(1), (parse_line_variant("1. kur")[0],))),
         (
             ["1. kur", "1. ra"],
             Line(
                 LineNumber(1),
-                (parse_line_variant("1. kur")[1], parse_line_variant("1. ra")[1]),
+                (parse_line_variant("1. kur")[0], parse_line_variant("1. ra")[0]),
             ),
         ),
         (
             [f"1. kur\n{MANUSCRIPTS[0].siglum} 1. kur", "1. ra"],
             Line(
                 LineNumber(1),
                 (
-                    parse_line_variant(f"1. kur\n{MANUSCRIPTS[0].siglum} 1. kur")[1],
-                    parse_line_variant("1. ra")[1],
+                    parse_line_variant(f"1. kur\n{MANUSCRIPTS[0].siglum} 1. kur")[0],
+                    parse_line_variant("1. ra")[0],
                 ),
             ),
         ),
@@ -256,15 +260,15 @@ def test_parse_chapter_line(lines, expected) -> None:
             ["#tr.en: translation", "1. kur"],
             Line(
                 LineNumber(1),
-                (parse_line_variant("1. kur")[1],),
+                (parse_line_variant("1. kur")[0],),
                 translation=(parse_translation_line("#tr.en: translation"),),
             ),
         ),
         (
             ["#tr.en: translation", "#tr.de: translation", "1. kur"],
             Line(
                 LineNumber(1),
-                (parse_line_variant("1. kur")[1],),
+                (parse_line_variant("1. kur")[0],),
                 translation=(
                     parse_translation_line("#tr.en: translation"),
                     parse_translation_line("#tr.de: translation"),

ebl/transliteration/domain/atf_parsers/lark_parser.py

@@ -44,8 +44,11 @@
 )
 PARATEXT_PARSER = Lark.open(ATF_GRAMMAR_PATH, **kwargs_lark, start="paratext")
 CHAPTER_PARSER = Lark.open(
-    ATF_GRAMMAR_PATH, **kwargs_lark, start="chapter", debug=True
-)  # ToDo: Remove debug
+    "lark_parser/ebl_atf_chapter.lark", **kwargs_lark, start="chapter"
+)
+MANUSCRIPT_PARSER = Lark.open(
+    "lark_parser/ebl_atf_manuscript_line.lark", **kwargs_lark, start="manuscript_line"
+)
 LINE_PARSER = Lark.open(ATF_GRAMMAR_PATH, **kwargs_lark)
 
 

ebl/transliteration/domain/atf_parsers/lark_parser/ebl_atf.lark

@@ -1,13 +1,14 @@
 %import .ebl_atf_text_line (text_line, any_word)
 %import .ebl_atf_parallel_line (parallel_line)
-%import .ebl_atf_manuscript_line (manuscript_line, paratext)
+%import .ebl_atf_manuscript_line (manuscript_line, paratext, siglum)
 %import .ebl_atf_translation_line (translation_line)
 %import .ebl_atf_note_line (note_line, markup)
 %import .ebl_atf_dollar_line (dollar_line)
 %import .ebl_atf_at_line (at_line)
 %import .ebl_atf_chapter (chapter)
 %import .ebl_atf_empty_line (empty_line)
 %import .ebl_atf_common (labels)
+%import .ebl_atf_control_line (control_line)
 %import .oracc_atf_lem_line (lem_line)
 
 ?start: line
@@ -20,6 +21,4 @@
      | parallel_line
      | translation_line
      | control_line
-     | lem_line
-
-!control_line.-2: ("=:" | "&" | "#") /.+/?
+     | lem_line

ebl/transliteration/domain/atf_parsers/lark_parser/ebl_atf_at_line.lark

@@ -1,4 +1,4 @@
-%import .ebl_atf_common (INT)
+%import common.INT
 %import .ebl_atf_common (status)
 %import .ebl_atf_common (object, OBJECT, generic_object, fragment)
 %import .ebl_atf_common (surface, SURFACE, generic_surface, face, edge)

ebl/transliteration/domain/atf_parsers/lark_parser/ebl_atf_chapter.lark

@@ -1,8 +1,8 @@
 %import common.CR
 %import common.LF
 
-%import .ebl_atf_manuscript_line (manuscript_line)
-%import .ebl_atf_text_line (text_line)
+%import .ebl_atf_manuscript_line.manuscript_line -> manuscript_line
+%import .ebl_atf_text_line.text_line -> text_line
 %import .ebl_atf_note_line (note_line)
 %import .ebl_atf_parallel_line (parallel_line)
 %import .ebl_atf_translation_line (translation_line)

ebl/transliteration/domain/atf_parsers/lark_parser/ebl_atf_control_line.lark

@@ -0,0 +1 @@
+!control_line.-2: ("=:" | "&" | "#") /.+/?

ebl/transliteration/domain/atf_parsers/lark_parser/ebl_atf_dollar_line.lark

@@ -1,4 +1,6 @@
-%import .ebl_atf_common (DIGIT, INT, LCASE_LETTER)
+%import common.DIGIT -> DIGIT
+%import common.INT -> INT
+%import common.LCASE_LETTER -> LCASE_LETTER
 %import .ebl_atf_common (free_text)
 %import .ebl_atf_common (object, OBJECT, generic_object, fragment)
 %import .ebl_atf_common (surface, SURFACE, generic_surface, face, edge)

ebl/transliteration/domain/atf_parsers/lark_parser/ebl_atf_manuscript_line.lark

@@ -2,11 +2,11 @@
 %import common.LF
 %import common.WS_INLINE
 
-%import .ebl_atf_common (labels)
-%import .ebl_atf_text_line (text_line)
-%import .ebl_atf_note_line (note_line)
-%import .ebl_atf_empty_line (empty_line)
-%import .ebl_atf_dollar_line (dollar_line)
+%import .ebl_atf_common.labels -> labels
+%import .ebl_atf_text_line.text_line -> text_line
+%import .ebl_atf_note_line.note_line -> note_line
+%import .ebl_atf_empty_line.empty_line -> empty_line
+%import .ebl_atf_dollar_line.dollar_line -> dollar_line
 
 %import .ebl_atf_abbreviations (PROVENANCE, PERIOD, TYPE)
 
@@ -15,7 +15,7 @@ manuscript_line: _WHITE_SPACE? siglum  [" " labels] manuscript_text paratext_lin
 ?manuscript_text: " " text_line | empty_line
 ?paratext_line: _NEWLINE _WHITE_SPACE? paratext
 siglum: [PROVENANCE] PERIOD [TYPE] [DISAMBIQUATOR]
-      | "Std" [DISAMBIQUATOR] -> standard_text_siglum
+      | "Std" [DISAMBIQUATOR]
 DISAMBIQUATOR: /[\S]+/
 
 _WHITE_SPACE: WS_INLINE

ebl/transliteration/domain/atf_parsers/lark_parser/ebl_atf_text_line.lark

@@ -1,7 +1,7 @@
 %import .ebl_atf_common (AKKADIAN_ALPHABET, GREEK_ALPHABET)
 %import .ebl_atf_common (VALUE_CHARACTER, LOGOGRAM_CHARACTER)
 %import .ebl_atf_common (UNCERTAIN, CORRECTION, COLLATION, NO_LONGER_VISIBLE)
-%import .ebl_atf_common (line_number)
+%import .ebl_atf_common.line_number -> line_number
 %import .legacy_atf (LEGACY_OPEN_HALF_BRACKET, LEGACY_CLOSE_HALF_BRACKET)
 %import .legacy_atf (LEGACY_ORACC_JOINER, LEGACY_ORACC_DISH_DIVIDER)
 %import .legacy_atf (legacy_uncertain_sign_prefix)

ebl/transliteration/domain/atf_parsers/lark_parser/ebl_atf_translation_line.lark

@@ -1,6 +1,6 @@
 %import common.LCASE_LETTER
-
-%import .ebl_atf_common (line_number, labels)
+%import .ebl_atf_common.line_number -> line_number
+%import .ebl_atf_common.labels -> labels
 %import .ebl_atf_note_line (_markup)
 
 translation_line: "#tr" [ "." TRANSLATION_LANGUAGE ] [".(" translation_extent ")"] ": " _markup

ebl/transliteration/domain/atf_parsers/lark_parser/legacy_atf.lark

@@ -10,6 +10,4 @@ legacy_single_ruling: " "* "ruling" [" "? ("!?" | "*" | "?" | "!")]
 
 
 //# ToDo:
-//# Implement here the following:
-//# oracc_atf_text_line__uncertain_sign
-//# etc. (check `atf_conversions.py`).
+//# check `atf_conversions.py` and implement all missing features.

ebl/transliteration/domain/text_line_transformer.py

@@ -136,11 +136,11 @@ def text_line(self, line_number, content):
         return TextLine.of_iterable(line_number, content)
 
     @v_args(inline=True)
-    def ebl_atf_text_line__line_number_range(self, start, end):
+    def ebl_atf_text_line__ebl_atf_common__line_number_range(self, start, end):
         return LineNumberRange(start, end)
 
     @v_args(inline=True)
-    def ebl_atf_text_line__single_line_number(
+    def ebl_atf_text_line__ebl_atf_common__single_line_number(
         self, prefix_modifier, number, prime, suffix_modifier
     ):
         return LineNumber(