def __load_chapter_text_into_db(self, chapter_paths: List[dict]):
"""
Load chapters text from the `stories` files into the chapters table. Uses Windows 1252 if UTF-8 fails.
:param chapter_paths: List of chapter metadata including path, author id and chapter id
:return:
"""
warnings = []
self.logger.info("...loading data from chapters table...")
old_chapters, current, total = self.sql.read_table_with_total(
self.working_original, "chapters")
self.logger.info("...removing rows from existing chapters table...")
self.sql.execute(self.working_open_doors, "TRUNCATE TABLE chapters;")
self.logger.info("...loading text from chapter files...")
for old_chapter in old_chapters:
chapid = old_chapter['chapid']
chapter = [
chapter_path for chapter_path in chapter_paths
if chapter_path['chap_id'] == str(chapid)
]
if chapter:
file = chapter[0]['path']
try:
with open(file, 'r') as f:
raw = f.read()
except UnicodeDecodeError as err:
warnings.append(
f"Chapter with id {chapid} contains non-ASCII characters which are not valid "
f"UTF-8. Trying Windows 1252...")
with open(file, 'r', encoding='cp1252') as f:
raw = f.read()
text = normalize(raw)
if old_chapter['endnotes']:
text = text + f"\n\n\n<hr>\n{old_chapter['endnotes']}"
query = """
INSERT INTO chapters (id, position, title, text, story_id, notes)
VALUES (%s, %s, %s, %s, %s, %s);
"""
self.sql.execute(
self.working_open_doors, query,
(chapid, old_chapter['inorder'], old_chapter['title'],
text, old_chapter['sid'], old_chapter['notes']))
current = print_progress(current, total, "chapters converted")
# If there were any errors, display a warning for the user to check the affected chapters
if warnings:
self.logger.warning("\n".join(warnings))
self.logger.error(
make_banner(
'-',
"There were warnings; check the affected chapters listed above to make sure curly quotes "
"and accented characters are correctly displayed."))
return self.sql.execute_and_fetchall(self.working_open_doors,
"SELECT * FROM chapters;")
def convert_stories(self, language_code):
"""
Convert eFiction stories to the Open Doors format. Note that we leave all the tag columns (rating, relationships,
tags, categories etc) empty because they are all in the `tags` table and will be populated with AO3 tags when
this archive is processed in the ODAP.
:return: The Open Doors stories table as a dict.
"""
self.logger.info("Converting stories...")
old_stories, current, total = self.sql.read_table_with_total(
self.working_original, "stories")
for old_story in old_stories:
new_story = {
'id': old_story['sid'],
'title': key_find('title', old_story, '').strip(),
'summary': normalize(old_story['summary']),
'notes': key_find('storynotes', old_story, '').strip(),
'date': str(old_story['date']),
'updated': str(old_story['updated']),
'language_code': language_code
}
self.logger.debug(
f"Converting story metadata for '{new_story['title']}'")
query = f"""
INSERT INTO stories (id, title, summary, notes, date, updated, language_code)
VALUES {new_story['id'], new_story['title'], new_story['summary'],
new_story['notes'], new_story['date'], new_story['updated'], new_story['language_code']};
"""
self.sql.execute(self.working_open_doors, query)
self.logger.debug(f" tags...")
tags = self._convert_story_tags(old_story)
self._convert_tags_join(new_story, tags)
self.logger.debug(f" authors...")
self._convert_author_join(new_story, old_story['uid'])
# Find if there are any coauthors for the work
coauthors = self.fetch_coauthors(new_story)
for coauthor in coauthors:
self._convert_author_join(new_story, coauthor)
current = print_progress(current, total, "stories converted")
return self.sql.execute_and_fetchall(self.working_open_doors,
"SELECT * FROM stories")
def convert_stories(self):
"""
Convert eFiction stories to the Open Doors format.
:return: The Open Doors stories table as a dict.
"""
self.logger.info("Converting stories...")
old_stories, current, total = self.sql.read_table_with_total(
self.working_original, "stories")
for old_story in old_stories:
new_story = {
'id': old_story['sid'],
'title': (old_story['title'] or '').strip(),
'summary': normalize(old_story['summary']),
'notes': (old_story['storynotes'] or '').strip(),
'date': str(old_story['date']),
'updated': str(old_story['updated']),
}
self.logger.debug(
f"Converting story metadata for '{new_story['title']}'")
query = f"""
INSERT INTO stories (id, title, summary, notes, date, updated)
VALUES {new_story['id'], new_story['title'], new_story['summary'],
new_story['notes'], new_story['date'], new_story['updated']};
"""
self.sql.execute(self.working_open_doors, query)
self.logger.debug(f" tags...")
tags = self._convert_story_tags(old_story)
self._convert_tags_join(new_story, tags)
self.logger.debug(f" authors...")
self._convert_author_join(new_story, old_story['uid'])
coauthors = []
if old_story[
'coauthors'] is not None and old_story['coauthors'] != "":
for authorid in old_story['coauthors'].split(","):
coauthors.append(authorid.strip())
for coauthor in coauthors:
self._convert_author_join(new_story, coauthor)
current = print_progress(current, total, "stories converted")
return self.sql.execute_and_fetchall(self.working_open_doors,
"SELECT * FROM stories")