Port USFM code from Machine up to commit a9058ce

machine/corpora/dictionary_text_corpus.py

@@ -1,5 +1,6 @@
 from typing import Iterable, Optional, overload
 
+from ..scripture.verse_ref import Versification
 from .text import Text
 from .text_corpus import TextCorpus
 
@@ -21,6 +22,7 @@ def __init__(self, *args, **kwargs) -> None:
             texts = args[0]
         self._texts = {t.id: t for t in texts}
         self._is_tokenized = False
+        self._versification = None
 
     @property
     def texts(self) -> Iterable[Text]:
@@ -34,6 +36,14 @@ def is_tokenized(self) -> bool:
     def is_tokenized(self, value: bool) -> None:
         self._is_tokenized = value
 
+    @property
+    def versification(self) -> Optional[Versification]:
+        return self._versification
+
+    @versification.setter
+    def versification(self, value: Versification) -> None:
+        self._versification = value
+
     def __getitem__(self, id: str) -> Optional[Text]:
         return self._texts.get(id)
 

machine/corpora/flatten.py

@@ -1,6 +1,7 @@
 from itertools import chain
 from typing import Generator, Iterable, List, Optional, cast, overload
 
+from ..scripture.verse_ref import Versification
 from .alignment_collection import AlignmentCollection
 from .alignment_corpus import AlignmentCorpus
 from .alignment_row import AlignmentRow
@@ -54,6 +55,10 @@ def texts(self) -> Iterable[Text]:
     def is_tokenized(self) -> bool:
         return all(c.is_tokenized for c in self._corpora)
 
+    @property
+    def versification(self) -> Optional[Versification]:
+        return self._corpora[0].versification if len(self._corpora) > 0 else None
+
     def count(self, include_empty: bool = True) -> int:
         return sum(c.count(include_empty) for c in self._corpora)
 

machine/corpora/parallel_text_row.py

@@ -18,6 +18,8 @@ def __init__(
         source_flags: TextRowFlags = TextRowFlags.SENTENCE_START,
         target_flags: TextRowFlags = TextRowFlags.SENTENCE_START,
     ) -> None:
+        if not text_id:
+            raise ValueError("A text_id must be set.")
         if len(source_refs) == 0 and len(target_refs) == 0:
             raise ValueError("Either a source or target ref must be set.")
         self._text_id = text_id

machine/corpora/paratext_backup_text_corpus.py

@@ -10,7 +10,7 @@
 
 
 class ParatextBackupTextCorpus(ScriptureTextCorpus):
-    def __init__(self, filename: StrPath, include_markers: bool = False) -> None:
+    def __init__(self, filename: StrPath, include_markers: bool = False, include_all_text: bool = False) -> None:
         with ZipFile(filename, "r") as archive:
             parser = ZipParatextProjectSettingsParser(archive)
             settings = parser.parse()
@@ -28,6 +28,7 @@ def __init__(self, filename: StrPath, include_markers: bool = False) -> None:
                         sfm_entry.filename,
                         versification,
                         include_markers,
+                        include_all_text,
                     )
                 )
 

machine/corpora/paratext_project_settings_parser_base.py

@@ -73,8 +73,16 @@ def parse(self) -> ParatextProjectSettings:
             post_part = naming_elem.get("PostPart")
             if post_part:
                 suffix = post_part
-        biblical_terms = settings_tree.getroot().findtext("BiblicalTermsListSetting", "")
-        parts = biblical_terms.split(":", 2)
+        biblical_terms_list_setting = settings_tree.getroot().findtext("BiblicalTermsListSetting", "")
+        if biblical_terms_list_setting is None:
+            # Default to Major::BiblicalTerms.xml to mirror Paratext behavior
+            biblical_terms_list_setting = "Major::BiblicalTerms.xml"
+        parts = biblical_terms_list_setting.split(":", 2)
+        if len(parts) != 3:
+            raise ValueError(
+                f"The BiblicalTermsListSetting element in Settings.xml in project {full_name}"
+                f" is not in the expected format (i.e., Major::BiblicalTerms.xml) but is {biblical_terms_list_setting}."
+            )
 
         return ParatextProjectSettings(
             name, full_name, encoding, versification, stylesheet, prefix, form, suffix, parts[0], parts[1], parts[2]

machine/corpora/paratext_text_corpus.py

@@ -8,7 +8,7 @@
 
 
 class ParatextTextCorpus(ScriptureTextCorpus):
-    def __init__(self, project_dir: StrPath, include_markers: bool = False) -> None:
+    def __init__(self, project_dir: StrPath, include_markers: bool = False, include_all_text: bool = False) -> None:
         parser = FileParatextProjectSettingsParser(project_dir)
         settings = parser.parse()
 
@@ -17,7 +17,14 @@ def __init__(self, project_dir: StrPath, include_markers: bool = False) -> None:
         texts: List[UsfmFileText] = []
         for sfm_filename in Path(project_dir).glob(f"{settings.file_name_prefix}*{settings.file_name_suffix}"):
             texts.append(
-                UsfmFileText(settings.stylesheet, settings.encoding, sfm_filename, versification, include_markers)
+                UsfmFileText(
+                    settings.stylesheet,
+                    settings.encoding,
+                    sfm_filename,
+                    versification,
+                    include_markers,
+                    include_all_text,
+                )
             )
 
         super().__init__(versification, texts)

machine/corpora/scripture_ref_usfm_parser_handler.py

@@ -0,0 +1,174 @@
+from abc import ABC
+from enum import Enum, auto
+from typing import List, Optional, Sequence
+
+from ..scripture.scripture_element import ScriptureElement
+from ..scripture.scripture_ref import ScriptureRef
+from ..scripture.verse_ref import VerseRef, are_overlapping_verse_ranges
+from .corpora_utils import merge_verse_ranges
+from .usfm_parser_handler import UsfmParserHandler
+from .usfm_parser_state import UsfmParserState
+from .usfm_token import UsfmAttribute
+
+
+class ScriptureTextType(Enum):
+    NONVERSE = auto()
+    VERSE = auto()
+    NOTE = auto()
+
+
+class ScriptureRefUsfmParserHandler(UsfmParserHandler, ABC):
+    def __init__(self) -> None:
+        self._cur_verse_ref: VerseRef = VerseRef()
+        self._cur_elements_stack: List[ScriptureElement] = []
+        self._cur_text_type_stack: List[ScriptureTextType] = []
+        self._duplicate_verse: bool = False
+
+    @property
+    def _current_text_type(self) -> ScriptureTextType:
+        return ScriptureTextType.NONVERSE if len(self._cur_text_type_stack) == 0 else self._cur_text_type_stack[-1]
+
+    def end_usfm(self, state: UsfmParserState) -> None:
+        self._end_verse_text_wrapper(state)
+
+    def chapter(self, state: UsfmParserState, number: str, marker: str, alt_number: str, pub_number: str) -> None:
+        self._end_verse_text_wrapper(state)
+        self._update_verse_ref(state.verse_ref, marker)
+
+    def verse(
+        self, state: UsfmParserState, number: str, marker: str, alt_number: Optional[str], pub_number: Optional[str]
+    ) -> None:
+        if state.verse_ref == self._cur_verse_ref:
+            self._end_verse_text_wrapper(state)
+            # ignore duplicate verses
+            self._duplicate_verse = True
+        elif are_overlapping_verse_ranges(number, self._cur_verse_ref.verse):
+            # merge overlapping verse ranges in to one range
+            verse_ref: VerseRef = self._cur_verse_ref.copy()
+            verse_ref.verse = merge_verse_ranges(number, self._cur_verse_ref.verse)
+            self._update_verse_ref(verse_ref, marker)
+        else:
+            self._end_verse_text_wrapper(state)
+            self._update_verse_ref(state.verse_ref, marker)
+            self._start_verse_text_wrapper(state)
+
+    def start_para(
+        self,
+        state: UsfmParserState,
+        marker: str,
+        unknown: Optional[bool],
+        attributes: Optional[Sequence[UsfmAttribute]],
+    ) -> None:
+        if self._cur_verse_ref.is_default:
+            self._update_verse_ref(state.verse_ref, marker)
+        if not state.is_verse_text:
+            self._start_parent_element(marker)
+            self._start_non_verse_text_wrapper(state)
+
+    def end_para(self, state: UsfmParserState, marker: str) -> None:
+        if self._current_text_type == ScriptureTextType.NONVERSE:
+            self._end_parent_element()
+            self._end_non_verse_text_wrapper(state)
+
+    def start_row(self, state: UsfmParserState, marker: str) -> None:
+        if self._current_text_type == ScriptureTextType.NONVERSE:
+            self._start_parent_element(marker)
+
+    def end_row(self, state: UsfmParserState, marker: str) -> None:
+        if self._current_text_type == ScriptureTextType.NONVERSE:
+            self._end_parent_element()
+
+    def start_cell(self, state: UsfmParserState, marker: str, align: str, colspan: int) -> None:
+        if self._current_text_type == ScriptureTextType.NONVERSE:
+            self._start_parent_element(marker)
+            self._start_non_verse_text_wrapper(state)
+
+    def end_cell(self, state: UsfmParserState, marker: str) -> None:
+        if self._current_text_type == ScriptureTextType.NONVERSE:
+            self._end_parent_element()
+            self._end_non_verse_text_wrapper(state)
+
+    def start_sidebar(self, state: UsfmParserState, marker: str, category: str) -> None:
+        self._start_parent_element(marker)
+
+    def end_sidebar(self, state: UsfmParserState, marker: str, closed: bool) -> None:
+        self._end_parent_element()
+
+    def start_note(self, state: UsfmParserState, marker: str, caller: str, category: Optional[str]) -> None:
+        self._next_element(marker)
+        self._start_note_text_wrapper(state)
+
+    def end_note(self, state: UsfmParserState, marker: str, closed: bool) -> None:
+        self._end_note_text_wrapper(state)
+
+    def ref(self, state: UsfmParserState, marker: str, display: str, target: str) -> None: ...
+
+    def _start_verse_text(self, state: UsfmParserState, scripture_refs: Optional[List[ScriptureRef]]) -> None: ...
+
+    def _end_verse_text(self, state: UsfmParserState, scripture_refs: List[ScriptureRef]) -> None: ...
+
+    def _start_non_verse_text(self, state: UsfmParserState, scripture_ref: ScriptureRef) -> None: ...
+
+    def _end_non_verse_text(self, state: UsfmParserState, scripture_ref: ScriptureRef) -> None: ...
+
+    def _start_note_text(self, state: UsfmParserState, scripture_ref: ScriptureRef) -> None: ...
+
+    def _end_note_text(self, state: UsfmParserState, scripture_ref: ScriptureRef) -> None: ...
+
+    def _start_verse_text_wrapper(self, state: UsfmParserState) -> None:
+        self._duplicate_verse = False
+        self._cur_text_type_stack.append(ScriptureTextType.VERSE)
+        self._start_verse_text(state, self._create_verse_refs())
+
+    def _end_verse_text_wrapper(self, state: UsfmParserState) -> None:
+        if not self._duplicate_verse and self._cur_verse_ref.verse_num != 0:
+            self._end_verse_text(state, self._create_verse_refs())
+            self._cur_text_type_stack.pop()
+
+    def _start_non_verse_text_wrapper(self, state: UsfmParserState) -> None:
+        self._cur_text_type_stack.append(ScriptureTextType.NONVERSE)
+        self._start_non_verse_text(state, self._create_non_verse_ref())
+
+    def _end_non_verse_text_wrapper(self, state: UsfmParserState) -> None:
+        self._end_non_verse_text(state, self._create_non_verse_ref())
+        self._cur_text_type_stack.pop()
+
+    def _start_note_text_wrapper(self, state: UsfmParserState) -> None:
+        self._cur_text_type_stack.append(ScriptureTextType.NOTE)
+        self._start_note_text(state, self._create_non_verse_ref())
+
+    def _end_note_text_wrapper(self, state: UsfmParserState) -> None:
+        self._end_note_text(state, self._create_non_verse_ref())
+        self._cur_text_type_stack.pop()
+
+    def _update_verse_ref(self, verse_ref: VerseRef, marker: str) -> None:
+        if not are_overlapping_verse_ranges(verse_ref, self._cur_verse_ref):
+            self._cur_elements_stack.clear()
+            self._cur_elements_stack.append(ScriptureElement(0, marker))
+        self._cur_verse_ref = verse_ref.copy()
+
+    def _next_element(self, marker: str) -> None:
+        prev_elem: ScriptureElement = self._cur_elements_stack.pop()
+        self._cur_elements_stack.append(ScriptureElement(prev_elem.position + 1, marker))
+
+    def _start_parent_element(self, marker: str) -> None:
+        self._next_element(marker)
+        self._cur_elements_stack.append(ScriptureElement(0, marker))
+
+    def _end_parent_element(self) -> None:
+        self._cur_elements_stack.pop()
+
+    def _create_verse_refs(self) -> List[ScriptureRef]:
+        return (
+            [ScriptureRef(v) for v in self._cur_verse_ref.all_verses()]
+            if self._cur_verse_ref.has_multiple
+            else [ScriptureRef(self._cur_verse_ref)]
+        )
+
+    def _create_non_verse_ref(self) -> ScriptureRef:
+        verse_ref = (
+            list(self._cur_verse_ref.all_verses())[-1] if self._cur_verse_ref.has_multiple else self._cur_verse_ref
+        )
+        # No need to reverse unlike in Machine, elements are already added in correct order
+        path = [e for e in self._cur_elements_stack if e.position > 0]
+        return ScriptureRef(verse_ref, path)

machine/corpora/scripture_text.py

@@ -1,6 +1,7 @@
-from typing import Generator, List, Optional
+from typing import Generator, List, Optional, Union
 
 from ..scripture import ENGLISH_VERSIFICATION
+from ..scripture.scripture_ref import ScriptureElement, ScriptureRef
 from ..scripture.verse_ref import VerseRef, Versification
 from ..utils.context_managed_generator import ContextManagedGenerator
 from .corpora_utils import gen, get_scripture_text_sort_key
@@ -20,19 +21,46 @@ def versification(self) -> Versification:
     def get_rows(self) -> ContextManagedGenerator[TextRow, None, None]:
         seg_list: List[TextRow] = []
         out_of_order = False
-        prev_verse_ref = VerseRef()
+        prev_scr_ref = ScriptureRef()
         with super().get_rows() as rows:
             for row in rows:
-                verse_ref: VerseRef = row.ref
+                scr_ref: ScriptureRef = row.ref
                 seg_list.append(row)
-                if not out_of_order and verse_ref < prev_verse_ref:
+                if not out_of_order and scr_ref < prev_scr_ref:
                     out_of_order = True
-                prev_verse_ref = verse_ref
+                prev_scr_ref = scr_ref
         if out_of_order:
             seg_list.sort(key=lambda r: r.ref)
         return ContextManagedGenerator(gen(seg_list))
 
     def _create_rows(
+        self, ref: Union[List[ScriptureRef], VerseRef], text: str = "", is_sentence_start: bool = True
+    ) -> Generator[TextRow, None, None]:
+        if isinstance(ref, VerseRef):
+            yield from self._create_rows_verse_ref(ref, text, is_sentence_start)
+        else:
+            yield from self._create_rows_scripture_ref(ref, text, is_sentence_start)
+
+    def _create_rows_scripture_ref(
+        self, scripture_refs: List[ScriptureRef], text: str = "", is_sentence_start: bool = True
+    ) -> Generator[TextRow, None, None]:
+        if len(scripture_refs) > 1:
+            first_verse = True
+            for sref in scripture_refs:
+                if first_verse:
+                    flags: TextRowFlags = TextRowFlags.IN_RANGE | TextRowFlags.RANGE_START
+                    if is_sentence_start:
+                        flags |= TextRowFlags.SENTENCE_START
+                    yield super()._create_row(text, sref, flags)
+                    first_verse = False
+                else:
+                    yield self._create_empty_row(sref, TextRowFlags.IN_RANGE)
+        else:
+            yield super()._create_row(
+                text, scripture_refs[0], TextRowFlags.SENTENCE_START if is_sentence_start else TextRowFlags.NONE
+            )
+
+    def _create_rows_verse_ref(
         self, verse_ref: VerseRef, text: str = "", is_sentence_start: bool = True
     ) -> Generator[TextRow, None, None]:
         if verse_ref.has_multiple:
@@ -42,13 +70,33 @@ def _create_rows(
                     flags = TextRowFlags.IN_RANGE | TextRowFlags.RANGE_START
                     if is_sentence_start:
                         flags |= TextRowFlags.SENTENCE_START
-                    yield self._create_row(text, vref, flags)
+                    yield super()._create_row(text, ScriptureRef(vref), flags)
                     first_verse = False
                 else:
-                    yield self._create_empty_row(vref, TextRowFlags.IN_RANGE)
+                    yield self._create_empty_row(ScriptureRef(vref), TextRowFlags.IN_RANGE)
+        else:
+            yield super()._create_row(
+                text, ScriptureRef(verse_ref), TextRowFlags.SENTENCE_START if is_sentence_start else TextRowFlags.NONE
+            )
+
+    def _create_row(
+        self,
+        ref: Union[ScriptureRef, VerseRef],
+        text: str,
+        is_sentence_start: bool,
+        elements: Optional[List[ScriptureElement]] = None,
+    ) -> TextRow:
+        if isinstance(ref, VerseRef):
+            return super()._create_row(
+                text,
+                ScriptureRef(ref, elements),
+                TextRowFlags.SENTENCE_START if is_sentence_start else TextRowFlags.NONE,
+            )
         else:
-            yield self._create_row(
-                text, verse_ref, TextRowFlags.SENTENCE_START if is_sentence_start else TextRowFlags.NONE
+            return super()._create_row(
+                text,
+                ref,
+                TextRowFlags.SENTENCE_START if is_sentence_start else TextRowFlags.NONE,
             )
 
     def _create_verse_ref(self, chapter: str, verse: str) -> VerseRef: