modify usfm for chapter-level drafting to avoid import issues; move remarks to chapters

Author: mshannon-sil

machine/corpora/paratext_project_text_updater_base.py

@@ -29,6 +29,7 @@ def __init__(
     def update_usfm(
         self,
         book_id: str,
+        chapters: Optional[Sequence[int]] = None,
         rows: Optional[Sequence[UpdateUsfmRow]] = None,
         full_name: Optional[str] = None,
         text_behavior: UpdateUsfmTextBehavior = UpdateUsfmTextBehavior.PREFER_EXISTING,
@@ -61,7 +62,7 @@ def update_usfm(
         )
         try:
             parse_usfm(usfm, handler, self._settings.stylesheet, self._settings.versification)
-            return handler.get_usfm(self._settings.stylesheet)
+            return handler.get_usfm(self._settings.stylesheet, chapters)
         except Exception as e:
             error_message = (
                 f"An error occurred while parsing the usfm for '{book_id}'"

machine/corpora/update_usfm_parser_handler.py

@@ -334,27 +334,43 @@ def _end_embed_text(self, state: UsfmParserState, scripture_ref: ScriptureRef) -
         if embed_outside_of_block:
             self._end_update_block(state, [scripture_ref])
 
-    def get_usfm(self, stylesheet: Union[str, UsfmStylesheet] = "usfm.sty") -> str:
+    def get_usfm(
+        self, stylesheet: Union[str, UsfmStylesheet] = "usfm.sty", chapters: Optional[Sequence[int]] = None
+    ) -> str:
         if isinstance(stylesheet, str):
             stylesheet = UsfmStylesheet(stylesheet)
         tokenizer = UsfmTokenizer(stylesheet)
         tokens = list(self._tokens)
+        if chapters is not None:
+            tokens = self._get_incremental_draft_tokens(tokens, chapters)
         if len(self._remarks) > 0:
             remark_tokens: List[UsfmToken] = []
             for remark in self._remarks:
                 remark_tokens.append(UsfmToken(UsfmTokenType.PARAGRAPH, "rem"))
                 remark_tokens.append(UsfmToken(UsfmTokenType.TEXT, text=remark))
             if len(tokens) > 0:
-                index = 0
-                markers_to_skip = {"id", "ide", "rem"}
-                while tokens[index].marker in markers_to_skip:
-                    index += 1
-                    if len(tokens) > index and tokens[index].type == UsfmTokenType.TEXT:
-                        index += 1
-                for remark_token in reversed(remark_tokens):
-                    tokens.insert(index, remark_token)
+                for index, token in enumerate(tokens):
+                    if token.type == UsfmTokenType.CHAPTER:
+                        tokens[index + 1 : index + 1] = remark_tokens
         return tokenizer.detokenize(tokens)
 
+    def _get_incremental_draft_tokens(self, tokens: List[UsfmToken], chapters: Sequence[int]) -> List[UsfmToken]:
+        incremental_draft_tokens: List[UsfmToken] = []
+        in_chapter: bool = False
+        for index, token in enumerate(tokens):
+            if index == 0 and token.marker == "id":
+                incremental_draft_tokens.append(token)
+                continue
+            elif token.type == UsfmTokenType.CHAPTER:
+                if token.data and int(token.data) in chapters:
+                    in_chapter = True
+                    incremental_draft_tokens.append(token)
+                else:
+                    in_chapter = False
+            elif in_chapter:
+                incremental_draft_tokens.append(token)
+        return incremental_draft_tokens
+
     def _advance_rows(self, seg_scr_refs: Sequence[ScriptureRef]) -> Tuple[List[str], Optional[dict[str, object]]]:
         row_texts: List[str] = []
         row_metadata = None