A4 for stet docx

.env

@@ -72,12 +72,6 @@ CHECK_ALL_BOOKS_FOR_LANGUAGE=true
 # they all are available by cloning, hence the value false.
 DOWNLOAD_ASSETS=false
 
-# Setting to false will cause interleave by verse layout to put each
-# TW word associated with the verse in a horizontal comma delimited
-# list, otherwise it will put each TW word in a list with one word per
-# line.
-TW_WORD_LIST_VERTICAL=false
-
 # File lock duration when cloning and potentially removing git repos
 LOCK_TIMEOUT_SECONDS=300
 

Makefile

@@ -1,5 +1,21 @@
+.PHONY: startdocker
+startdocker:
+	@if docker info >/dev/null 2>&1; then \
+		echo "Docker is already running."; \
+	else \
+		echo "Starting Docker..."; \
+		if [ "$$(uname)" = "Darwin" ]; then \
+			open -a Docker; \
+		else \
+			sudo systemctl start docker; \
+		fi; \
+		echo "Waiting for Docker to fully initialize..."; \
+		until docker info >/dev/null 2>&1; do sleep 1; done; \
+		echo "Docker is ready!"; \
+	fi
+
 .PHONY: checkvenv
-checkvenv:
+checkvenv: startdocker
 # raises error if environment is not active
 ifeq ("$(VIRTUAL_ENV)","")
 	@echo "Venv is not activated!"
@@ -119,7 +135,7 @@ test: clean-local-docker-output-dir
 	docker compose -f docker-compose.yml -f docker-compose.api-test.yml -f docker-compose.override.yml up  test-runner
 
 .PHONY: unit-tests
-unit-tests:
+unit-tests: startdocker
 	docker compose -f docker-compose.yml -f docker-compose.api-test.yml -f docker-compose.override.yml up  test-runner
 
 .PHONY: e2e-tests
@@ -133,7 +149,7 @@ e2e-docx-tests: clean-local-docker-output-dir
 
 
 .PHONY: frontend-tests
-frontend-tests:
+frontend-tests: startdocker
 	# NOTE If we are experiencing some issues with the docker
 	# compose running of frontend tests, we can still use the
 	# non-Dockerized approach successfully. Doing so requires that
@@ -184,7 +200,7 @@ clean-mypyc-artifacts:
 	find . ! -path .venv -type f -name "*.c" -exec rm -- {} +
 
 .PHONY: prune-docker-images-volumes
-prune-docker-images-volumes:
+prune-docker-images-volumes: checkvenv
 	docker system prune --volumes
 
 # https://radon.readthedocs.io/en/latest/commandline.html
@@ -301,10 +317,10 @@ local-run-celery:
 
 .PHONY: local-run-flower
 local-run-flower:
- 	celery --broker=redis:// --result-backend=redis:// flower
-# This is one to run after running local-e2e-tests or any tests which
-# has yielded HTML and PDFs that need to be checked for linking
-# correctness.
+	celery --broker=redis:// --result-backend=redis:// flower
+
+# Run after running local-e2e-tests or any tests which has yielded HTML
+# and PDFs that need to be checked for linking correctness.
 .PHONY: local-check-anchor-links
 local-check-anchor-links: checkvenv
 	python tests/e2e/test_anchor_linking.py

backend/doc/config.py

@@ -20,6 +20,8 @@ class Settings(BaseSettings):
 
     DATA_API_URL: HttpUrl
 
+    LANGUAGES_WHERE_NON_ULB_PREFERRED: Sequence[str] = ["fr"]
+
     # This is only used to see if a lang_code is in the collection
     # otherwise it is a heart language. Eventually the graphql data api may
     # provide gateway/heart boolean value.
@@ -177,7 +179,11 @@ class Settings(BaseSettings):
     BOOK_NAME_FMT_STR: str = "<h2 class='book-name'>{}</h2>"
     RESOURCE_TYPE_NAME_FMT_STR: str = "<h1 class='book-name'>{}</h1>"
     HR: str = "<hr/>"
-    TW_WORD_LIST_VERTICAL: bool = False
+    BIEL_TW_RESOURCE_URL_FMT_STR: str = (
+        "<span><a href='https://bibleineverylanguage.org/resources/languages/{}?resource-type=tw'>{}</a></span>"
+    )
+    TW_RESOURCE_URL_FMT_STR: str = "<span><a href='#{}-{}'>{}</a></span>"
+    LINK_RATHER_THAN_INCLUDE_TW_DEFINITIONS: bool = True
 
     DOWNLOAD_ASSETS: bool = False  # If true then download assets, else clone assets
 

backend/doc/domain/assembly_strategies/assemble_by_book.py

@@ -24,7 +24,7 @@
     TWBook,
     USFMBook,
 )
-from doc.domain.parsing import handle_split_chapter_into_verses
+from doc.domain.parsing import split_chapter_into_verses_with_formatting
 from doc.reviewers_guide.model import RGBook
 
 logger = settings.logger(__name__)
@@ -475,7 +475,7 @@ def assemble_usfm_by_verse_book_at_a_time(
             chapter_num,
             chapter,
         ) in usfm_book.chapters.items():
-            chapter.verses = handle_split_chapter_into_verses(usfm_book, chapter)
+            chapter.verses = split_chapter_into_verses_with_formatting(chapter)
             chapter_intros = get_chapter_intros(
                 tn_book,
                 tnc_book,
@@ -531,8 +531,10 @@ def assemble_usfm_by_verse_book_at_a_time(
                     # ulb, f10, show the second USFM content here
                     if usfm_book2:
                         usfm_book2_chapter = usfm_book2.chapters[chapter_num]
-                        usfm_book2_chapter.verses = handle_split_chapter_into_verses(
-                            usfm_book2, usfm_book2_chapter
+                        usfm_book2_chapter.verses = (
+                            split_chapter_into_verses_with_formatting(
+                                usfm_book2_chapter
+                            )
                         )
                         if (
                             usfm_book2_chapter.verses

backend/doc/domain/assembly_strategies/assemble_by_chapter.py

@@ -37,10 +37,9 @@
     TWBook,
     USFMBook,
 )
-from doc.domain.parsing import handle_split_chapter_into_verses
+from doc.domain.parsing import split_chapter_into_verses_with_formatting
 from doc.reviewers_guide.model import RGBook
 
-
 logger = settings.logger(__name__)
 
 
@@ -625,13 +624,13 @@ def assemble_usfm_by_verse_chapter_at_a_time(
             primary_verses = None
             secondary_verses = None
             if usfm_book and usfm_chapter:
-                usfm_chapter.verses = handle_split_chapter_into_verses(
-                    usfm_book, usfm_chapter
+                usfm_chapter.verses = split_chapter_into_verses_with_formatting(
+                    usfm_chapter
                 )
                 primary_verses = usfm_chapter.verses
             if usfm_book2 and usfm_book2_chapter:
-                usfm_book2_chapter.verses = handle_split_chapter_into_verses(
-                    usfm_book2, usfm_book2_chapter
+                usfm_book2_chapter.verses = split_chapter_into_verses_with_formatting(
+                    usfm_book2_chapter
                 )
                 secondary_verses = usfm_book2_chapter.verses
             if usfm_book and primary_verses:

backend/doc/domain/document_generator.py

@@ -51,6 +51,7 @@
 from doc.utils.docx_util import (
     add_internal_docx_links,
     generate_docx_toc,
+    override_and_clean_hyperlinks,
     style_superscripts,
 )
 from doc.utils.file_utils import (
@@ -558,6 +559,7 @@ def assemble_content(
     tw_books: Sequence[TWBook],
     bc_books: Sequence[BCBook],
     rg_books: Sequence[RGBook],
+    link_rather_than_include_tw_definitions: bool = settings.LINK_RATHER_THAN_INCLUDE_TW_DEFINITIONS,
 ) -> list[DocumentPart]:
     """
     Assemble and return the content from all requested resources according to the
@@ -651,7 +653,11 @@ def assemble_content(
         )
     t1 = time.time()
     logger.info("Time for interleaving document: %s", t1 - t0)
-    if tw_books and not document_request.layout_for_print:
+    if (
+        tw_books
+        and not link_rather_than_include_tw_definitions
+        and not document_request.layout_for_print
+    ):
         t0 = time.time()
         # Add the translation words definition section for each language requested.
         unique_tw_books = filter_unique_by_lang_code(tw_books)
@@ -778,6 +784,15 @@ def compose_docx_document(
         if part.add_page_break:
             add_page_break(doc)
     style_superscripts(doc, lift_half_points=2, color=None)
+    # html4doc defeats normal use of hyperlink inline styling in Word
+    # via a customized (otherwise standard) Hyperlink style, but
+    # this handles it by doing a pass over the document after the fact
+    # and forcing hyperlinks to render using a specific style we
+    # created to match the PO's desired look.
+    # PlainHyperlinkChar is a character style we created in
+    # template.docx, but its actual style ID is PlainHyperlinkChar0
+    # under the hood.
+    override_and_clean_hyperlinks(doc, "PlainHyperlinkChar0")
     t1 = time.time()
     logger.info("Time for converting HTML to Docx: %.2f seconds", t1 - t0)
     return doc