def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
if not epub.identifier():
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
# ---------- lag en kopi av EPUBen ----------
temp_epubdir_obj = tempfile.TemporaryDirectory()
temp_epubdir = temp_epubdir_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_epubdir)
temp_epub = Epub(self, temp_epubdir)
# ---------- gjør tilpasninger i HTML-fila med XSLT ----------
opf_path = temp_epub.opf_path()
if not opf_path:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne OPF-fila i EPUBen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
opf_path = os.path.join(temp_epubdir, opf_path)
opf_xml = ElementTree.parse(opf_path).getroot()
html_file = opf_xml.xpath(
"/*/*[local-name()='manifest']/*[@id = /*/*[local-name()='spine']/*[1]/@idref]/@href"
)
html_file = html_file[0] if html_file else None
if not html_file:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila i OPFen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
html_dir = os.path.dirname(opf_path)
html_file = os.path.join(html_dir, html_file)
if not os.path.isfile(html_file):
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
temp_html_obj = tempfile.NamedTemporaryFile()
temp_html = temp_html_obj.name
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, PrepareForDocx.uid,
"prepare-for-docx.xsl"),
source=html_file,
target=temp_html)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_html, html_file)
archived_path, stored = self.utils.filesystem.storeBook(
temp_epubdir, epub.identifier())
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier(
) + " ble konvertert 👍😄" + epubTitle
return True
def test_copy_locked_files(self):
print("creating book without any locked files")
book1 = os.path.join(self.dir_in, "book1")
os.makedirs(os.path.join(book1, "images"))
Path(os.path.join(book1, "ncc.html")).touch()
Path(os.path.join(book1, "images/Image.png")).touch()
Path(os.path.join(book1, "images/zmage.png")).touch()
print("creating book with a \"Thumbs.db\" file locked by Windows")
book2 = os.path.join(self.dir_in, "book2")
os.makedirs(os.path.join(book2, "images"))
Path(os.path.join(book2, "ncc.html")).touch()
Path(os.path.join(book2, "images/Image.png")).touch()
Path(os.path.join(book2, "images/Thumbs.db")).touch()
Path(os.path.join(book2, "images/zmage.png")).touch()
print("creating book with a \"locked\" file locked by Windows")
book3 = os.path.join(self.dir_in, "book3")
os.makedirs(os.path.join(book3, "images"))
Path(os.path.join(book3, "ncc.html")).touch()
Path(os.path.join(book3, "images/Image.png")).touch()
Path(os.path.join(book3, "images/locked")).touch()
Path(os.path.join(book3, "images/zmage.png")).touch()
target_book1 = os.path.join(self.dir_out, "book1")
target_book2 = os.path.join(self.dir_out, "book2")
target_book3 = os.path.join(self.dir_out, "book3")
print("copy book1 to target_book1")
Filesystem.copy(self.pipeline.utils.report, book1, target_book1)
dirlist = os.listdir(target_book1)
dirlist.sort()
self.assertEqual(dirlist, ["images", "ncc.html"])
dirlist = os.listdir(os.path.join(target_book1, "images"))
dirlist.sort()
self.assertEqual(dirlist, ["Image.png", "zmage.png"])
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[ERROR]")]) == 0)
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[WARN]")]) == 0)
print("copy book1 to target_book1 once more")
Filesystem.copy(self.pipeline.utils.report, book1, target_book1)
dirlist = os.listdir(target_book1)
dirlist.sort()
self.assertEqual(dirlist, ["images", "ncc.html"])
dirlist = os.listdir(os.path.join(target_book1, "images"))
dirlist.sort()
self.assertEqual(dirlist, ["Image.png", "zmage.png"])
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[ERROR]")]) == 0)
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[WARN]")]) == 0)
print("copy book2 to target_book2")
Filesystem.copy(self.pipeline.utils.report, book2, target_book2)
dirlist = os.listdir(target_book2)
dirlist.sort()
self.assertEqual(dirlist, ["images", "ncc.html"])
dirlist = os.listdir(os.path.join(target_book2, "images"))
dirlist.sort()
self.assertEqual(dirlist, ["Image.png", "zmage.png"])
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[ERROR]")]) == 0)
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[WARN]")]) == 0)
print("copy book2 to target_book2 once more")
Filesystem.copy(self.pipeline.utils.report, book2, target_book2)
dirlist = os.listdir(target_book2)
dirlist.sort()
self.assertEqual(dirlist, ["images", "ncc.html"])
dirlist = os.listdir(os.path.join(target_book2, "images"))
dirlist.sort()
self.assertEqual(dirlist, ["Image.png", "zmage.png"])
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[ERROR]")]) == 0)
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[WARN]")]) == 0)
print("copy book3 to target_book3")
Filesystem.copy(self.pipeline.utils.report, book3, target_book3)
dirlist = os.listdir(target_book3)
dirlist.sort()
self.assertEqual(dirlist, ["images", "ncc.html"])
dirlist = os.listdir(os.path.join(target_book3, "images"))
dirlist.sort()
self.assertEqual(dirlist, ["Image.png", "locked", "zmage.png"])
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[ERROR]")]) == 0)
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[WARN]")]) == 0)
print("copy book3 to target_book3 once more")
Filesystem.copy(self.pipeline.utils.report, book3, target_book3)
dirlist = os.listdir(target_book3)
dirlist.sort()
self.assertEqual(dirlist, ["images", "ncc.html"])
dirlist = os.listdir(os.path.join(target_book3, "images"))
dirlist.sort()
self.assertEqual(dirlist, ["Image.png", "locked", "zmage.png"])
self.assertTrue(
len([m for m in self.pipeline.messages
if m.startswith("[WARN]")]) == 0)
self.assertTrue(
len([
m for m in self.pipeline.messages
if m.startswith("[ERROR]") and "/locked" in m
]) >= 1)
def on_book(self):
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return
if not epub.identifier():
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return
temp_obj = tempfile.TemporaryDirectory()
temp_dir = temp_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_dir)
self.utils.report.info("Henter metadata fra api.nlb.no")
creative_work_metadata = None
edition_metadata = None
timeout = 0
while creative_work_metadata is None and timeout < 5:
timeout = timeout + 1
creative_work_metadata = Metadata.get_creative_work_from_api(
self.book["name"],
editions_metadata="all",
use_cache_if_possible=True,
creative_work_metadata="all")
edition_metadata = Metadata.get_edition_from_api(self.book["name"])
if creative_work_metadata is not None:
break
if creative_work_metadata is None:
self.utils.report.warning(
"Klarte ikke finne et åndsverk tilknyttet denne utgaven. Prøver igjen senere."
)
return False
library = edition_metadata["library"].lower()
# in case of wrong upper lower cases
if library == "nlb":
library = "NLB"
elif library == "statped":
library = "Statped"
elif library == "kabb":
library = "KABB"
if library.lower() != "statped":
self.utils.report.error("Ikke en Statped bok. Avbryter")
self.utils.report.should_email = False
return False
# Filesystem.copy(self.utils.report, self.book["source"], temp_dir)
self.utils.report.info("Kopierer til EPUB master-arkiv.")
archived_path, stored = self.utils.filesystem.storeBook(
temp_dir, epub.identifier())
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier(
) + " er valid 👍😄" + epubTitle
self.utils.filesystem.deleteSource()
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
if not epub.identifier() or not epub.identifier().isnumeric():
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
self.utils.report.info("Lager en kopi av EPUBen")
temp_epubdir_obj = tempfile.TemporaryDirectory()
temp_epubdir = temp_epubdir_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_epubdir)
if not os.path.exists(os.path.join(self.dir_out, epub.identifier())):
os.makedirs(os.path.join(self.dir_out, epub.identifier()))
time_created = time.strftime("%Y-%m-%dT%H:%M:%S")
dictfiles = {}
changelog = "changelog.txt"
deleted = "deleted.yml"
files = "files.yml"
extra_files = [changelog, deleted, files, "restore_files.py"]
changes_made = False
new_epub = False
# Overview of deleted files and changelog history
deleted_path = os.path.join(self.dir_out, epub.identifier(), deleted)
changelog_path = os.path.join(self.dir_out, epub.identifier(),
changelog)
deleted_doc = {}
if os.path.isfile(deleted_path):
with open(deleted_path, 'r') as f:
deleted_doc = yaml.load(f, Loader=yaml.FullLoader) or {}
# Dictfiles contains the most recent version of each file, saved to files.yml
for (path, subdir_list,
file_list) in walk(os.path.join(self.dir_out, epub.identifier())):
for file_name in file_list:
if file_name in extra_files:
continue
file_path = os.path.join(path, file_name)
relative_path = file_path.replace(
os.path.join(self.dir_out, epub.identifier()), "")
relative_path = relative_path.strip("/")
short_path = self.short_path_by_one(relative_path)
new_dict = {short_path: relative_path}
if short_path not in dictfiles:
dictfiles.update(new_dict)
elif dictfiles[short_path] < relative_path:
dictfiles.update(new_dict)
new_file_list = []
changelog_string = ""
file_added_again = False
# Changelog.txt contains the history of changes to this nlbpub with timestamps
for temp_path, temp_subdir_list, temp_file_list in walk(temp_epubdir):
for temp_file in temp_file_list:
full_temp_file_path = os.path.join(temp_path, temp_file)
temp_file = full_temp_file_path.replace(temp_epubdir, "")
temp_file = temp_file.strip("/")
new_file_list.append(temp_file)
if temp_file in dictfiles and filecmp.cmp(
full_temp_file_path,
os.path.join(self.dir_out, epub.identifier(),
dictfiles[temp_file])):
os.remove(full_temp_file_path)
elif temp_file in dictfiles and not filecmp.cmp(
full_temp_file_path,
os.path.join(self.dir_out, epub.identifier(),
dictfiles[temp_file])):
changes_made = True
new_location = {
temp_file: os.path.join(time_created, temp_file)
}
dictfiles.update(new_location)
self.utils.report.info("Fil endret: " + temp_file)
changelog_string += ("\n{}: Fil endret: {}".format(
time_created, temp_file))
elif temp_file not in dictfiles:
if dictfiles == {}:
new_epub = True
changes_made = True
new_file = {
temp_file: os.path.join(time_created, temp_file)
}
dictfiles.update(new_file)
if not new_epub:
self.utils.report.info("Fil lagt til: " + temp_file)
changelog_string += (
"\n{}: Fil lagt til: {}".format(
time_created, temp_file))
if temp_file in deleted_doc:
changes_made = True
file_added_again = True
deleted_doc.pop(temp_file, None)
self.utils.report.info("Fil lagt til på nytt: " +
temp_file)
changelog_string += (
"\n{}: Fil lagt til på nytt: {}".format(
time_created, temp_file))
dirs = next(walk(temp_epubdir))[1]
for dir in dirs:
self.del_empty_dirs(temp_epubdir, dir)
if file_added_again:
with open(deleted_path, 'w') as deleted_file:
for key in deleted_doc:
deleted_file.write("\n'{}': '{}'".format(
key.replace("'", "''"),
time_created.replace("'", "''")))
# Deleted file history saved to deleted files.yml
with open(deleted_path,
self.append_write(deleted_path)) as deleted_file:
for key in dictfiles:
if key not in new_file_list and key not in deleted_doc and key not in extra_files:
changes_made = True
self.utils.report.info("Fil slettet: " + key)
changelog_string += ("\n{}: Fil slettet: {}".format(
time_created, key))
deleted_file.write("\n'{}': '{}'".format(
key.replace("'", "''"),
time_created.replace("'", "''")))
# Changelog saved to changelog.txt
with open(changelog_path,
self.append_write(changelog_path)) as changelog_file:
changelog_file.write(changelog_string)
deleted_doc = {}
if os.path.isfile(deleted_path):
with open(deleted_path, 'r') as f:
deleted_doc = yaml.load(f, Loader=yaml.FullLoader) or {}
for del_file in deleted_doc:
try:
del dictfiles[del_file]
except Exception:
self.utils.report.debug(traceback.format_exc(),
preformatted=True)
with open(os.path.join(temp_epubdir, files), 'w') as files_doc:
for file in dictfiles:
files_doc.write("\n'{}': '{}'".format(
file.replace("'", "''"),
dictfiles[file].replace("'", "''")))
# Save copy of different files in NLBPUB master. Different versions of files under NLBPUB-tidligere/xxxxxxx/time
# To restore a certain version copy files from the each folder up to the wanted version to a new folder
archived_path, stored = self.utils.filesystem.storeBook(
temp_epubdir, epub.identifier(), subdir=time_created)
self.utils.report.attachment(None, archived_path, "DEBUG")
if changes_made:
if new_epub:
self.utils.report.info(
"Endringer oppdaget for: " + epub.identifier() +
", ny epub ble kopiert til NLBpub tidligere versjoner.")
self.utils.report.title = self.title + ": " + epub.identifier(
) + " 👍😄" + epubTitle + " , ny epub ble kopiert"
else:
self.utils.report.info(
"Endringer oppdaget for: " + epub.identifier() +
", endrede filer ble kopiert til NLBpub tidligere versjoner."
)
self.utils.report.title = self.title + ": " + epub.identifier(
) + " 👍😄" + epubTitle + " , endring registrert"
else:
self.utils.report.info("Ingen endringer oppdaget for " +
epub.identifier())
self.utils.report.title = self.title + ": " + epub.identifier(
) + " 👍😄" + epubTitle + " , ingen endring registrert"
self.utils.report.should_email = False
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
metadata = Metadata.get_metadata_from_book(self.utils.report,
self.book["source"])
metadata["identifier"] = re.sub(r"[^\d]", "", metadata["identifier"])
if not metadata["identifier"]:
self.utils.report.error(
"Klarte ikke å bestemme boknummer for {}".format(
self.book["name"]))
return False
if metadata["identifier"] != self.book["name"]:
self.utils.report.info("Boknummer for {} er: {}".format(
self.book["name"], metadata["identifier"]))
self.utils.report.info("Lager en kopi av DTBoken")
temp_dtbookdir_obj = tempfile.TemporaryDirectory()
temp_dtbookdir = temp_dtbookdir_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_dtbookdir)
# find DTBook XML
dtbook = None
for root, dirs, files in os.walk(temp_dtbookdir):
for f in files:
if f.endswith(".xml"):
xml = ElementTree.parse(os.path.join(root, f)).getroot()
if xml.xpath(
"namespace-uri()"
) == "http://www.daisy.org/z3986/2005/dtbook/":
dtbook = os.path.join(root, f)
break
if dtbook is not None:
break
if not dtbook:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne DTBook")
return False
# rename all files to lower case
for root, dirs, files in os.walk(temp_dtbookdir):
for f in files:
if not f.lower() == f:
self.utils.report.warn(
"renaming to lowercase: {}".format(f))
shutil.move(os.path.join(root, f),
os.path.join(root, f.lower()))
temp_dtbook_file_obj = tempfile.NamedTemporaryFile()
temp_dtbook_file = temp_dtbook_file_obj.name
self.utils.report.info("Rydder opp i nordisk DTBook")
xslt = Xslt(self,
stylesheet=os.path.join(NordicDTBookToEpub.xslt_dir,
NordicDTBookToEpub.uid,
"nordic-cleanup-dtbook.xsl"),
source=dtbook,
target=temp_dtbook_file)
if not xslt.success:
return False
shutil.copy(temp_dtbook_file, dtbook)
self.utils.report.info("Validerer Nordisk DTBook...")
# create context for Pipeline 2 job
dtbook_dir = os.path.dirname(dtbook)
dtbook_context = {}
for root, dirs, files in os.walk(dtbook_dir):
for file in files:
fullpath = os.path.join(root, file)
relpath = os.path.relpath(fullpath, dtbook_dir)
dtbook_context[relpath] = fullpath
with DaisyPipelineJob(
self,
"nordic-dtbook-validate", {
"dtbook": os.path.basename(dtbook),
"no-legacy": "false"
},
pipeline_and_script_version=[
("1.13.6", "1.4.6"),
("1.13.4", "1.4.5"),
("1.12.1", "1.4.2"),
("1.11.1-SNAPSHOT", "1.3.0"),
],
context=dtbook_context) as dp2_job_dtbook_validate:
dtbook_validate_status = None
if dp2_job_dtbook_validate.status == "SUCCESS":
dtbook_validate_status = "SUCCESS"
elif dp2_job_dtbook_validate.status in ["VALIDATION_FAIL", "FAIL"]:
dtbook_validate_status = "WARN"
else:
dtbook_validate_status = "ERROR"
report_file = os.path.join(dp2_job_dtbook_validate.dir_output,
"html-report/report.xhtml")
if dtbook_validate_status == "WARN":
report_doc = ElementTree.parse(report_file)
errors = report_doc.xpath(
'//*[@class="error" or @class="message-error"]')
for error in errors:
error_text = " ".join(
[e.strip() for e in error.xpath('.//text()')]).strip()
error_text = " ".join(error_text.split()).strip() if bool(
error_text) else error_text
if (bool(error_text) and
(error_text.startswith("[tpb124]")
or error_text.startswith("[tpb43]")
or error_text.startswith("[tpb10] Meta dc:Publisher")
or error_text.startswith("[tpb10] Meta dc:Date")
or error_text.startswith("[opf3g]")
or 'element "h1" not allowed here' in error_text
or 'element "h2" not allowed here' in error_text
or 'element "h3" not allowed here' in error_text
or 'element "h4" not allowed here' in error_text
or 'element "h5" not allowed here' in error_text
or 'element "h6" not allowed here' in error_text
or 'token "toc-brief" invalid' in error_text)):
continue # ignorer disse feilmeldingene
if error_text.startswith("Incorrect file signature"):
magic_number = error.xpath(
'*[@class="message-details"]/*[last()]/*[last()]/text()'
)[0]
magic_number = " ".join(magic_number.split()).strip(
) if bool(magic_number) else magic_number
# JFIF already allowed: 0xFF 0xD8 0xFF 0xE0 0x?? 0x?? 0x4A 0x46 0x49 0x46
if magic_number.startswith(
"0xFF 0xD8 0xFF 0xDB"): # Also allow JPEG RAW
continue
elif magic_number[:
19] == "0xFF 0xD8 0xFF 0xE1" and magic_number[
30:] == ("0x45 0x78 0x69 0x66"
): # Also allow EXIF
continue
else:
dtbook_validate_status = "ERROR"
self.utils.report.error(error_text)
else:
dtbook_validate_status = "ERROR"
self.utils.report.error(error_text)
# get conversion report
if os.path.isfile(report_file):
with open(report_file, 'r') as result_report:
self.utils.report.attachment(
result_report.readlines(),
os.path.join(self.utils.report.reportDir(),
"report-dtbook.html"),
dtbook_validate_status)
if dtbook_validate_status == "ERROR":
self.utils.report.error("Klarte ikke å validere boken")
return False
if dtbook_validate_status == "WARN":
self.utils.report.warn(
"DTBoken er ikke valid, men vi fortsetter alikevel.")
self.utils.report.info(
"Konverterer fra Nordisk DTBook til Nordisk HTML...")
temp_htmldir_obj = tempfile.TemporaryDirectory()
temp_htmldir = temp_htmldir_obj.name
temp_htmlfile = None
with DaisyPipelineJob(
self,
"nordic-dtbook-to-html", {
"dtbook": os.path.basename(dtbook),
"fail-on-error": "false",
"no-legacy": "false"
},
pipeline_and_script_version=[
("1.13.6", "1.4.6"),
("1.13.4", "1.4.5"),
("1.12.1", "1.4.2"),
("1.11.1-SNAPSHOT", "1.3.0"),
],
context=dtbook_context) as dp2_job_dtbook_to_html:
convert_status = "SUCCESS" if dp2_job_dtbook_to_html.status == "SUCCESS" else "ERROR"
convert_report_file = os.path.join(
dp2_job_dtbook_to_html.dir_output, "html-report/report.xhtml")
if convert_status != "SUCCESS":
self.utils.report.error(
"Klarte ikke å konvertere boken fra DTBook til HTML")
# get conversion report
if os.path.isfile(convert_report_file):
with open(convert_report_file, 'r') as result_report:
self.utils.report.attachment(
result_report.readlines(),
os.path.join(self.utils.report.reportDir(),
"report-dtbook-to-html.html"),
convert_status)
return False
dp2_html_dir = os.path.join(dp2_job_dtbook_to_html.dir_output,
"output-dir")
if not os.path.isdir(dp2_html_dir):
self.utils.report.error(
"Finner ikke 'output-dir' for den konverterte boken: {}".
format(dp2_html_dir))
return False
Filesystem.copy(self.utils.report, dp2_html_dir, temp_htmldir)
temp_htmlfile = os.path.join(temp_htmldir,
metadata["identifier"] + ".xhtml")
if not os.path.isfile(temp_htmlfile):
self.utils.report.error(
"Finner ikke den konverterte boken: {}".format(temp_htmlfile))
self.utils.report.info(
"Kanskje filnavnet er forskjellig fra IDen?")
return False
self.utils.report.info("Rydder opp i nordisk HTML")
temp_html_xslt_output_obj = tempfile.NamedTemporaryFile()
temp_html_xslt_output = temp_html_xslt_output_obj.name
xslt = Xslt(self,
stylesheet=os.path.join(NordicDTBookToEpub.xslt_dir,
NordicDTBookToEpub.uid,
"nordic-cleanup-html.xsl"),
source=temp_htmlfile,
target=temp_html_xslt_output)
if not xslt.success:
return False
shutil.copy(temp_html_xslt_output, temp_htmlfile)
self.utils.report.info(
"Konverterer fra Nordisk HTML til Nordisk EPUB3...")
# create context for Pipeline 2 job
html_dir = os.path.dirname(temp_htmlfile)
html_context = {}
for root, dirs, files in os.walk(html_dir):
for file in files:
fullpath = os.path.join(root, file)
relpath = os.path.relpath(fullpath, html_dir)
html_context[relpath] = fullpath
temp_epub_file_obj = tempfile.NamedTemporaryFile()
temp_epub_file = temp_epub_file_obj.name
with DaisyPipelineJob(self,
"nordic-html-to-epub3", {
"html": os.path.basename(temp_htmlfile),
"fail-on-error": "false"
},
pipeline_and_script_version=[
("1.13.6", "1.4.6"),
("1.13.4", "1.4.5"),
("1.12.1", "1.4.2"),
("1.11.1-SNAPSHOT", "1.3.0"),
],
context=html_context) as dp2_job_html_to_epub:
convert_status = "SUCCESS" if dp2_job_html_to_epub.status == "SUCCESS" else "ERROR"
convert_report_file = os.path.join(dp2_job_html_to_epub.dir_output,
"html-report/report.xhtml")
if convert_status != "SUCCESS":
self.utils.report.error("Klarte ikke å konvertere boken")
# get conversion report
if os.path.isfile(convert_report_file):
with open(convert_report_file, 'r') as result_report:
self.utils.report.attachment(
result_report.readlines(),
os.path.join(self.utils.report.reportDir(),
"report-html-to-epub3.html"),
convert_status)
return False
dp2_epub_file = os.path.join(dp2_job_html_to_epub.dir_output,
"output-dir",
metadata["identifier"] + ".epub")
if not os.path.isfile(dp2_epub_file):
self.utils.report.error(
"Finner ikke den konverterte boken: {}".format(
dp2_epub_file))
self.utils.report.info(
"Kanskje filnavnet er forskjellig fra IDen?")
return False
self.utils.report.info("Validerer Nordisk EPUB 3...")
epub_file = dp2_epub_file.asFile()
with DaisyPipelineJob(self,
"nordic-epub3-validate",
{"epub": os.path.basename(epub_file)},
pipeline_and_script_version=[
("1.13.6", "1.4.6"),
("1.13.4", "1.4.5"),
("1.12.1", "1.4.2"),
("1.11.1-SNAPSHOT", "1.3.0"),
],
context={
os.path.basename(epub_file): epub_file
}) as dp2_job_epub_validate:
epub_validate_status = "SUCCESS" if dp2_job_epub_validate.status == "SUCCESS" else "ERROR"
report_file = os.path.join(dp2_job_epub_validate.dir_output,
"html-report/report.xhtml")
if epub_validate_status == "ERROR":
# attach intermediary file from conversion
with open(temp_htmlfile, 'r') as intermediary_htmlfile:
self.utils.report.attachment(
intermediary_htmlfile.readlines(),
os.path.join(self.utils.report.reportDir(),
"intermediary-html.html"), "DEBUG")
epub_validate_status = "WARN"
report_doc = ElementTree.parse(report_file)
errors = report_doc.xpath(
'//*[@class="error" or @class="message-error"]')
for error in errors:
error_text = " ".join([
e.strip() for e in error.xpath('.//text()')
]).strip()
error_text = " ".join(error_text.split()).strip(
) if bool(error_text) else error_text
if (bool(error_text) and
(error_text.startswith("[nordic280]")
or "PKG-021: Corrupted image file encountered."
in error_text)):
continue # ignorer disse feilmeldingene
else:
self.utils.report.warn(
"Not ignoring: {}".format(error_text))
if error_text.startswith("Incorrect file signature"):
magic_number = error.xpath(
'*[@class="message-details"]/*[last()]/*[last()]/text()'
)[0]
magic_number = " ".join(magic_number.split(
)).strip() if bool(magic_number) else magic_number
# JFIF already allowed: 0xFF 0xD8 0xFF 0xE0 0x?? 0x?? 0x4A 0x46 0x49 0x46
if magic_number.startswith(
"0xFF 0xD8 0xFF 0xDB"
): # Also allow JPEG RAW
continue
elif magic_number[:
19] == "0xFF 0xD8 0xFF 0xE1" and magic_number[
30:] == (
"0x45 0x78 0x69 0x66"
): # Also allow EXIF
continue
else:
epub_validate_status = "ERROR"
self.utils.report.error(error_text)
else:
epub_validate_status = "ERROR"
self.utils.report.error(error_text)
# get conversion report
if os.path.isfile(report_file):
with open(report_file, 'r') as result_report:
self.utils.report.attachment(
result_report.readlines(),
os.path.join(self.utils.report.reportDir(),
"report-epub3.html"),
epub_validate_status)
if epub_validate_status == "ERROR":
self.utils.report.error(
"Klarte ikke å validere EPUB 3-versjonen av boken")
return False
Filesystem.copy(self.utils.report, dp2_epub_file, temp_epub_file)
epub = Epub(self.utils.report, temp_epub_file)
if not epub.isepub():
return False
self.utils.report.info(
"Boken ble konvertert. Kopierer til EPUB3-fra-DTBook-arkiv.")
archived_path, stored = self.utils.filesystem.storeBook(
epub.asDir(), metadata["identifier"], overwrite=self.overwrite)
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = "{}: {} ble konvertert 👍😄 ({})".format(
self.title, metadata["identifier"], metadata["title"])
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
return False
if not epub.identifier():
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
return False
if epub.identifier() != self.book["name"].split(".")[0]:
self.utils.report.error(
self.book["name"] +
": Filnavn stemmer ikke overens med dc:identifier: {}".format(
epub.identifier()))
return False
temp_xml_file_obj = tempfile.NamedTemporaryFile()
temp_xml_file = temp_xml_file_obj.name
self.utils.report.info("Lager en kopi av EPUBen")
temp_epubdir_withimages_obj = tempfile.TemporaryDirectory()
temp_epubdir_withimages = temp_epubdir_withimages_obj.name
Filesystem.copy(self.utils.report, self.book["source"],
temp_epubdir_withimages)
self.utils.report.info("Lager en kopi av EPUBen med tomme bildefiler")
temp_epubdir_obj = tempfile.TemporaryDirectory()
temp_epubdir = temp_epubdir_obj.name
Filesystem.copy(self.utils.report, temp_epubdir_withimages,
temp_epubdir)
for root, dirs, files in os.walk(
os.path.join(temp_epubdir, "EPUB", "images")):
for file in files:
fullpath = os.path.join(root, file)
os.remove(fullpath)
Path(fullpath).touch()
temp_epub = Epub(self.utils.report, temp_epubdir)
self.utils.report.info("Rydder opp i nordisk EPUB nav.xhtml")
nav_path = os.path.join(temp_epubdir, temp_epub.nav_path())
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, NordicToNlbpub.uid,
"nordic-cleanup-nav.xsl"),
source=nav_path,
target=temp_xml_file,
parameters={
"cover":
" ".join([item["href"] for item in temp_epub.spine()]),
"base":
os.path.dirname(
os.path.join(temp_epubdir, temp_epub.opf_path())) +
"/"
})
if not xslt.success:
return False
shutil.copy(temp_xml_file, nav_path)
self.utils.report.info("Rydder opp i nordisk EPUB package.opf")
opf_path = os.path.join(temp_epubdir, temp_epub.opf_path())
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, NordicToNlbpub.uid,
"nordic-cleanup-opf.xsl"),
source=opf_path,
target=temp_xml_file)
if not xslt.success:
return False
shutil.copy(temp_xml_file, opf_path)
html_dir_obj = tempfile.TemporaryDirectory()
html_dir = html_dir_obj.name
html_file = os.path.join(html_dir, epub.identifier() + ".xhtml")
self.utils.report.info("Finner ut hvilket bibliotek boka tilhører…")
edition_metadata = Metadata.get_edition_from_api(
epub.identifier(), report=self.utils.report)
library = None
if edition_metadata is not None and edition_metadata[
"library"] is not None:
library = edition_metadata["library"]
else:
library = Metadata.get_library_from_identifier(
epub.identifier(), self.utils.report)
self.utils.report.info(f"Boka tilhører '{library}'")
self.utils.report.info("Zipper oppdatert versjon av EPUBen...")
temp_epub.asFile(rebuild=True)
self.utils.report.info(
"Konverterer fra Nordisk EPUB 3 til Nordisk HTML 5...")
epub_file = temp_epub.asFile()
with DaisyPipelineJob(self,
"nordic-epub3-to-html", {
"epub": os.path.basename(epub_file),
"fail-on-error": "false"
},
pipeline_and_script_version=[
("1.13.6", "1.4.6"),
("1.13.4", "1.4.5"),
("1.12.1", "1.4.2"),
("1.11.1-SNAPSHOT", "1.3.0"),
],
context={os.path.basename(epub_file):
epub_file}) as dp2_job_convert:
convert_status = "SUCCESS" if dp2_job_convert.status == "SUCCESS" else "ERROR"
if convert_status != "SUCCESS":
self.utils.report.error("Klarte ikke å konvertere boken")
return False
dp2_html_dir = os.path.join(dp2_job_convert.dir_output,
"output-dir", epub.identifier())
dp2_html_file = os.path.join(dp2_job_convert.dir_output,
"output-dir", epub.identifier(),
epub.identifier() + ".xhtml")
if not os.path.isdir(dp2_html_dir):
self.utils.report.error(
"Finner ikke den konverterte boken: {}".format(
dp2_html_dir))
return False
if not os.path.isfile(dp2_html_file):
self.utils.report.error(
"Finner ikke den konverterte boken: {}".format(
dp2_html_file))
self.utils.report.info(
"Kanskje filnavnet er forskjellig fra IDen?")
return False
Filesystem.copy(self.utils.report, dp2_html_dir, html_dir)
self.utils.report.info("Rydder opp i nordisk HTML")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, NordicToNlbpub.uid,
"nordic-cleanup.xsl"),
source=html_file,
target=temp_xml_file)
if not xslt.success:
return False
shutil.copy(temp_xml_file, html_file)
self.utils.report.info("Rydder opp i ns0 i page-normal")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, NordicToNlbpub.uid,
"ns0-cleanup.xsl"),
source=html_file,
target=temp_xml_file)
if not xslt.success:
return False
shutil.copy(temp_xml_file, html_file)
self.utils.report.info("Rydder opp i innholdsfortegnelsen")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, NordicToNlbpub.uid,
"fix-toc-span.xsl"),
source=html_file,
target=temp_xml_file)
if not xslt.success:
return False
shutil.copy(temp_xml_file, html_file)
self.utils.report.info(
"Legger til EPUB-filer (OPF, NAV, container.xml, mediatype)...")
nlbpub_tempdir_obj = tempfile.TemporaryDirectory()
nlbpub_tempdir = nlbpub_tempdir_obj.name
nlbpub = Epub.from_html(self, html_dir, nlbpub_tempdir)
if nlbpub is None:
return False
self.utils.report.info(
"Erstatter tomme bildefiler med faktiske bildefiler")
for root, dirs, files in os.walk(
os.path.join(nlbpub_tempdir, "EPUB", "images")):
for file in files:
fullpath = os.path.join(root, file)
relpath = os.path.relpath(fullpath, nlbpub_tempdir)
os.remove(fullpath)
Filesystem.copy(self.utils.report,
os.path.join(temp_epubdir_withimages, relpath),
fullpath)
temp_epub = Epub(self.utils.report, temp_epubdir)
nlbpub.update_prefixes()
self.utils.report.info(
"Boken ble konvertert. Kopierer til NLBPUB-arkiv.")
archived_path, stored = self.utils.filesystem.storeBook(
nlbpub.asDir(), temp_epub.identifier(), overwrite=self.overwrite)
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier(
) + " ble konvertert 👍😄" + epubTitle
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
temp_absdir_obj = tempfile.TemporaryDirectory()
temp_absdir = temp_absdir_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_absdir)
file_exists = {
"abstracts": False,
"back-cover": False,
"test-audio": False
}
if not os.path.isfile(os.path.join(temp_absdir, "ncc.html")):
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎. Er dette en daisy 2.02 lydbok med en ncc.html fil?"
return False
try:
nccdoc = ElementTree.parse(os.path.join(temp_absdir,
"ncc.html")).getroot()
except Exception:
self.utils.report.info(
"Klarte ikke lese ncc fila. Sjekk loggen for detaljer.")
self.utils.report.debug(traceback.format_exc(), preformatted=True)
edition_identifier = ""
audio_title = ""
audio_title = " (" + nccdoc.xpath(
"string(//*[@name='dc:title']/@content)") + ") "
issue_identifier = nccdoc.xpath(
"string(//*[@name='dc:identifier']/@content)")
edition_identifier = issue_identifier[0:6]
if edition_identifier == (""):
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + "Lydbok feilet 😭👎"
return False
try:
smilFile = nccdoc.xpath(
"substring-before(//*[text()='Bokomtale' or text()='Baksidetekst' or text()='Omslagstekst']/@href,'#')"
)
smilFile_Id = nccdoc.xpath(
"substring-after(//*[text()='Bokomtale' or text()='Baksidetekst' or text()='Omslagstekst']/@href,'#')"
)
except Exception:
self.utils.report.debug(traceback.format_exc(), preformatted=True)
self.utils.report.error(
"Det oppstod en feil for" + edition_identifier +
" under lasting av smilfilene. Sjekk loggen for detaljer.")
return False
# Back-cover
if (smilFile != ""):
try:
smildoc = ElementTree.parse(os.path.join(
temp_absdir, smilFile)).getroot()
mp3File = smildoc.xpath("string((//audio/@src)[1])")
mp3File_start = smildoc.xpath(
"substring-before(substring-after(((//par[@id='{0}' or text/@id='{0}']//audio)[1]/@clip-begin),'='),'s')"
.format(smilFile_Id))
mp3File_end = smildoc.xpath(
"substring-before(substring-after(((//par[@id='{0}' or text/@id='{0}']//audio)[last()]/@clip-end),'='),'s')"
.format(smilFile_Id))
if mp3File_start == mp3File_end:
self.utils.report.info(
"Klarte ikke å bestemme start-/slutt-tid for baksidetekst"
)
# Creates audio segment in milliseconds from start to end of the abstract file
mp3 = AudioSegment.from_mp3(os.path.join(temp_absdir, mp3File))
new_mp3 = mp3[float(mp3File_start) * 1000:float(mp3File_end) *
1000]
new_mp3.export(
os.path.join(temp_absdir,
self.parentdirs["back-cover"] + ".mp3"))
self.utils.report.info("Baksidetekst eksportert fra: " +
mp3File)
file_exists["back-cover"] = True
except Exception:
self.utils.report.debug(traceback.format_exc(),
preformatted=True)
self.utils.report.info(
"Klarte ikke hente ut baksidetekst for " +
edition_identifier + " sjekk loggen for detaljer.")
else:
self.utils.report.info("Baksidetekst ikke funnet for " +
edition_identifier)
# creates abstract from ncc --> smil --> mp3
several_smilFiles = []
several_smilFiles_id = []
try:
number_of_smilfiles = int(nccdoc.xpath("count(//@href)"))
for i in range(number_of_smilfiles):
several_smilFiles.append(
nccdoc.xpath(
"substring-before((//@href)[{0}],'#')".format(i + 1)))
several_smilFiles_id.append(
nccdoc.xpath(
"substring-after((//@href)[{0}],'#')".format(i + 1)))
except Exception:
self.utils.report.info(traceback.format_exc(), preformatted=True)
self.utils.report.info("Klarte ikke hente ut .smil filene for " +
edition_identifier + audio_title)
timeout = time.time() + 60 * 2
duration = 0
num = 0
try:
while (duration <= 50 and time.time() < timeout
and int(number_of_smilfiles / 2 + num) < int(
number_of_smilfiles * 0.9)):
smilFile_abstract = several_smilFiles[int(number_of_smilfiles *
0.5 + num)]
smilFile_abstract_id = several_smilFiles_id[int(
number_of_smilfiles * 0.5 + num)]
smildoc_abstract = ElementTree.parse(
os.path.join(temp_absdir, smilFile_abstract)).getroot()
mp3File_abstract_start = float(
smildoc_abstract.xpath(
"substring-before(substring-after(((//par[@id='{0}' or text/@id='{0}']//audio)[1]/@clip-begin),'='),'s')"
.format(smilFile_abstract_id)))
if (smilFile_abstract == several_smilFiles[
int(number_of_smilfiles * 0.5 + num) + 1]):
smilFile_abstract_id = several_smilFiles_id[
int(number_of_smilfiles * 0.5 + num) + 1]
mp3File_abstract_end = float(
smildoc_abstract.xpath(
"substring-before(substring-after(((//par[@id='{0}' or text/@id='{0}']//audio)[last()]/@clip-end),'='),'s')"
.format(smilFile_abstract_id)))
duration = mp3File_abstract_end - mp3File_abstract_start
num = num + 1
mp3File_abstract = smildoc_abstract.xpath(
"string((//audio/@src)[1])")
except Exception:
self.utils.report.info(traceback.format_exc(), preformatted=True)
self.utils.report.info("Lydutdrag fra smilfiler feilet.")
if (duration >= 75):
mp3File_abstract_end = mp3File_abstract_start + 75
# As a last resort, just use an mp3 of sufficient length
if (duration < 20):
try:
for item in os.listdir(temp_absdir):
if (item.endswith(".mp3")):
try_mp3 = AudioSegment.from_mp3(
os.path.join(temp_absdir, item))
if (len(try_mp3) / 1000 > duration):
mp3File_abstract = item
mp3File_abstract_start = 0
mp3File_abstract_end = len(try_mp3) / 1000
duration = mp3File_abstract_end
if (duration > 75):
mp3File_abstract_start = 0.0
mp3File_abstract_end = 75.0
break
except Exception:
self.utils.report.debug(traceback.format_exc(),
preformatted=True)
self.utils.report.info(
"Klarte ikke hente ut lydutdrag basert på mp3 filene i mappa. Sjekk loggen for detaljer."
)
# Export abstract
try:
mp3_abstract = AudioSegment.from_mp3(
os.path.join(temp_absdir, mp3File_abstract))
new_mp3_abstract = mp3_abstract[mp3File_abstract_start *
1000:mp3File_abstract_end * 1000]
final_mp3 = new_mp3_abstract.fade_out(3000)
final_mp3.export(
os.path.join(temp_absdir,
self.parentdirs["abstracts"] + ".mp3"))
self.utils.report.info("Lydutdrag eksportert fra: " +
mp3File_abstract)
file_exists["abstracts"] = True
except Exception:
self.utils.report.info(traceback.format_exc(), preformatted=True)
self.utils.report.error(
"Klarte ikke eksportere excerpt.mp3. Har du ffmpeg kodeken for .mp3 filer?"
)
# Copies abstract and back cover to dir_out
if (os.path.isfile(
os.path.join(temp_absdir,
self.parentdirs["back-cover"] + ".mp3"))
or os.path.isfile(
os.path.join(temp_absdir,
self.parentdirs["abstracts"] + ".mp3"))):
if (file_exists["back-cover"]):
shutil.copy(
os.path.join(temp_absdir,
self.parentdirs["back-cover"] + ".mp3"),
os.path.join(temp_absdir,
self.parentdirs["test-audio"] + ".mp3"))
file_exists["test-audio"] = True
if (self.parentdirs["abstracts"]):
self.utils.report.info(
"Baksidetekst og lydutdrag funnet. Kopierer til {}.mp3"
.format(self.parentdirs["test-audio"]))
else:
self.utils.report.info(
"Baksidetekst funnet. Kopierer til {}.mp3".format(
self.parentdirs["test-audio"]))
elif (self.parentdirs["abstracts"]):
shutil.copy(
os.path.join(temp_absdir,
self.parentdirs["abstracts"] + ".mp3"),
os.path.join(temp_absdir,
self.parentdirs["test-audio"] + ".mp3"))
file_exists["test-audio"] = True
self.utils.report.info("Lydutdrag funnet. Kopierer til " +
self.parentdirs["test-audio"])
for key in self.parentdirs:
if (file_exists[key]):
archived_path, stored = self.utils.filesystem.storeBook(
os.path.join(temp_absdir,
self.parentdirs[key] + ".mp3"),
edition_identifier,
parentdir=self.parentdirs[key],
file_extension="mp3")
if edition_identifier != issue_identifier:
archived_path, stored = self.utils.filesystem.storeBook(
os.path.join(temp_absdir,
self.parentdirs[key] + ".mp3"),
issue_identifier,
parentdir=self.parentdirs[key],
file_extension="mp3")
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + edition_identifier + " lydutdrag ble eksportert 👍😄" + audio_title
else:
self.utils.report.title = (
"Klarte ikke hente ut hverken baksidetekst eller lydutdrag 😭👎. "
) + audio_title
return False
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# check that this is an EPUB (we only insert metadata into EPUBs)
if not epub.isepub():
return False
if not epub.identifier():
self.utils.report.error(
self.book["name"] +
": Klarte ikke ├Ц bestemme boknummer basert p├Ц dc:identifier."
)
return False
if epub.identifier() != self.book["name"].split(".")[0]:
self.utils.report.error(
self.book["name"] +
": Filnavn stemmer ikke overens med dc:identifier: {}".format(
epub.identifier()))
return False
should_produce, metadata_valid = Metadata.should_produce(
epub.identifier(),
self.publication_format,
report=self.utils.report)
if not metadata_valid:
self.utils.report.info(
"{} har feil i metadata for {}. Avbryter.".format(
epub.identifier(), self.publication_format))
self.utils.report.title = "{}: {} har feil i metadata for {} ЪўГЪЉј {}".format(
self.title, epub.identifier(), self.publication_format,
epubTitle)
return False
if not should_produce:
self.utils.report.info(
"{} skal ikke produseres som {}. Avbryter.".format(
epub.identifier(), self.publication_format))
self.utils.report.title = "{}: {} Skal ikke produseres som {} Ъци {}".format(
self.title, epub.identifier(), self.publication_format,
epubTitle)
return True
self.utils.report.info("Lager en kopi av EPUBen")
temp_epubdir_obj = tempfile.TemporaryDirectory()
temp_epubdir = temp_epubdir_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_epubdir)
temp_epub = Epub(self.utils.report, temp_epubdir)
is_valid = Metadata.insert_metadata(
self.utils.report,
temp_epub,
publication_format=self.publication_format,
report_metadata_errors=False)
if not is_valid:
self.utils.report.error(
"Bibliofil-metadata var ikke valide. Avbryter.")
return False
self.utils.report.info(
"Boken ble oppdatert med format-spesifikk metadata. Kopierer til {}-arkiv."
.format(self.publication_format))
archived_path, stored = self.utils.filesystem.storeBook(
temp_epub.asDir(), epub.identifier())
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = "{}: {} har f├Цtt {}-spesifikk metadata og er klar til ├Ц produseres ЪЉЇЪўё {}".format(
self.title, epub.identifier(), self.publication_format,
temp_epub.meta("dc:title"))
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
if not epub.identifier():
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
# language must be exctracted from epub or else docx default language (nb) wil be used in the converted file
language = ""
try:
#language = " (" + epub.meta("dc:language") + ") "
language = epub.meta("dc:language")
except Exception:
pass
# ---------- lag en kopi av EPUBen ----------
temp_epubdir_obj = tempfile.TemporaryDirectory()
temp_epubdir = temp_epubdir_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_epubdir)
temp_epub = Epub(self, temp_epubdir)
opf_path = temp_epub.opf_path()
if not opf_path:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne OPF-fila i EPUBen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
opf_path = os.path.join(temp_epubdir, opf_path)
opf_xml = ElementTree.parse(opf_path).getroot()
html_file = opf_xml.xpath(
"/*/*[local-name()='manifest']/*[@id = /*/*[local-name()='spine']/*[1]/@idref]/@href"
)
html_file = html_file[0] if html_file else None
if not html_file:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila i OPFen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
html_file = os.path.join(os.path.dirname(opf_path), html_file)
if not os.path.isfile(html_file):
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
temp_xml_file_obj = tempfile.NamedTemporaryFile()
temp_xml_file = temp_xml_file_obj.name
self.utils.report.info(
"Konverterer fra ASCIIMath til norsk punktnotasjon…")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, NLBpubToDocx.uid,
"nordic-asciimath-epub.xsl"),
source=html_file,
target=temp_xml_file)
if not xslt.success:
return False
shutil.copy(temp_xml_file, html_file)
# ---------- konverter HTML-fila til DOCX ----------
temp_docxdir_obj = tempfile.TemporaryDirectory()
temp_docxdir = temp_docxdir_obj.name
try:
self.utils.report.info("Konverterer fra XHTML til DOCX...")
process = self.utils.filesystem.run([
"/usr/bin/ebook-convert",
html_file,
os.path.join(temp_docxdir,
epub.identifier() + "_calibre.docx"),
"--chapter=/",
"--chapter-mark=none",
"--page-breaks-before=/",
"--no-chapters-in-toc",
"--toc-threshold=0",
"--docx-page-size=a4",
# "--linearize-tables",
"--extra-css=" +
os.path.join(Xslt.xslt_dir, self.uid, 'extra.css'),
# NOTE: microsoft fonts must be installed:
# sudo apt-get install ttf-mscorefonts-installer
"--embed-font-family=Verdana",
"--docx-page-margin-top=42",
"--docx-page-margin-bottom=42",
"--docx-page-margin-left=70",
"--docx-page-margin-right=56",
#"--language="+epub.meta('dc:language'),
("--language=" + language) if language else "",
"--base-font-size=13",
#"--remove-paragraph-spacing",
#"--remove-paragraph-spacing-indent-size=-1",
"--font-size-mapping=13,13,13,13,13,13,13,13"
])
if process.returncode == 0:
self.utils.report.info("Boken ble konvertert.")
# ------------- script from kvile ---------------
document = Document(
os.path.join(temp_docxdir,
epub.identifier() + "_calibre.docx"))
emptyParagraph = False
normalParagraph = "Normal"
normalParagraphNoIndent = "NormalNoIndent"
headingIndent = Cm(1.25)
fontSize = Pt(13)
# ny kode 2021-01-20
#folder = os.path.join(temp_docxdir)
folder = Path(temp_docxdir)
# slutt ny kode
#self.utils.report.info("Folder: "+folder)
def zipdir(src, dst, zip_name):
os.chdir(dst)
ziph = zipfile.ZipFile(zip_name, 'w', zipfile.ZIP_DEFLATED)
for root, dirs, files in os.walk(src):
for file in files:
ziph.write(os.path.join(root, file),
arcname=os.path.join(
root.replace(src, ""), file))
ziph.close()
def writeFile(txt, dst):
tempFile = open(folder / dst, "w+")
tempFile.write(txt)
tempFile.close()
def delete_paragraph(paragraph):
# self.utils.report.info("Delete paragraph: ")
p = paragraph._element
p.getparent().remove(p)
p._p = p._element = None
def delete_element(element):
element.getparent().remove(element)
element._element = None
indent = Cm(0.44)
hangingIndentList = Cm(0.63)
document.styles[normalParagraph].font.size = fontSize
document.styles[
normalParagraph].paragraph_format.first_line_indent = indent
styleNoIndent = document.styles.add_style(
'NormalNoIndent', WD_STYLE_TYPE.PARAGRAPH)
styleNoIndent.base_style = document.styles[normalParagraph]
document.styles[
normalParagraphNoIndent].paragraph_format.first_line_indent = Cm(
0)
# set style to normal for regular paragraphs, set keep_with_next to false, remove multiple empty paragraphs, and remove empty p after page nr or heading
for paragraph in document.paragraphs:
# deleting empty text-elements
emptyTextElementList = document.element.xpath(
"//w:t[. = '']")
for emptyTextElement in emptyTextElementList:
delete_element(emptyTextElement)
paragraph.paragraph_format.keep_with_next = None
if re.match("Para 0[1-9]|[0-9] Block|Para [0-9]",
paragraph.style.name
) and paragraph.style.font.underline != True:
paragraph.style = normalParagraph
if len(paragraph.text) <= 1 or re.match(
r"^--- \d+ til ", paragraph.text
) or paragraph.style.name[
0:
7] == "Heading": # if empty p or page nr or heading
paragraph.text = re.sub(
r"^\s(.*)", r"\1",
paragraph.text) #remove space at beginning av p
# self.utils.report.info("Paragraph.text <= 1 ")
if len(
paragraph.text
) == 0 and emptyParagraph: #if last p also was empty or page nr
# self.utils.report.info("Paragraph.text == 0 ")
delete_paragraph(paragraph)
emptyParagraph = True
else:
emptyParagraph = False
if re.match(r"^\s*STATPED_DUMMYTEXT_LI_OL\s*$",
paragraph.text):
paragraph.text = ""
# no indent after Heading, page-nr, or paragraphs starting with "Bilde: ", paragraphs in only bold (text=^_[^_]*_$) and the paragraph after p in only bold, or on empty p.
removeIndent = False
for paragraph in document.paragraphs:
#remove space at beginning of line after <br/>
spaceAfterBreakList = paragraph._element.xpath(
r'w:r/w:br[@w:clear="none"]/following::w:t[@xml:space="preserve"][1]'
)
if len(spaceAfterBreakList) > 0:
for spaceAfterBreakElement in spaceAfterBreakList:
if re.match(
'^ ', spaceAfterBreakElement.text
) and not (spaceAfterBreakElement.xpath(
r'preceding-sibling::*[1][self::w:t]')):
spaceAfterBreakElement.text = re.sub(
r"^ ", r"", spaceAfterBreakElement.text)
#remove break before paragraph end
breakBeforeParagraphEndList = paragraph._element.xpath(
r'w:r[last()]/w:br[@w:clear="none" and not(following-sibling::*)]'
)
if len(breakBeforeParagraphEndList) > 0:
delete_element(breakBeforeParagraphEndList[0])
t = paragraph.text.strip()
if re.match(
r"^Bilde: |^Forklaring: |^--- \d+ til |^_[^_]*_$|^STATPED_DUMMYTEXT_LIST_UNSTYLED|^STATPED_DUMMYTEXT_P_BEFORE_DL",
t) or ((removeIndent or len(t) == 0)
and paragraph.style.name == "Normal"):
paragraph.style = normalParagraphNoIndent
# Remove dummy-text and set hengemarg
if re.match(
r"^(STATPED_DUMMYTEXT_LIST_UNSTYLED|STATPED_DUMMYTEXT_DL)",
paragraph.text):
paragraph.paragraph_format.left_indent = hangingIndentList #Pt(0)
paragraph.paragraph_format.first_line_indent = -hangingIndentList #Pt(-20)
if re.match(r"^STATPED_DUMMYTEXT", paragraph.text):
paragraph.text = re.sub(
r"^(STATPED_DUMMYTEXT_LIST_UNSTYLED|STATPED_DUMMYTEXT_DL|STATPED_DUMMYTEXT_P_BEFORE_DL)",
"", paragraph.text)
if len(t) == 0 or paragraph.style.name[
0:7] == "Heading" or re.match(
r"^--- \d+ til |^_[^_]*_$", t):
removeIndent = True
else:
removeIndent = False
# remove bold from Headings.
paraStylesWithoutBoldOrUnderline = [
] #list of all para-styles without underline or bold
paraStylesWithoutUnderline = [
] #list of all para-styles without underline
for style in document.styles:
if style.name[0:7] == "Heading":
style.font.bold = None
style.paragraph_format.left_indent = headingIndent #Pt(0)
style.paragraph_format.first_line_indent = -headingIndent #Pt(-20)
style.paragraph_format.space_before = Pt(0)
style.paragraph_format.space_after = Pt(0)
style_element = style._element
spacing = style_element.xpath(r'w:pPr/w:spacing')[0]
spacing.set(qn('w:beforeLines'), "0")
spacing.set(qn('w:afterLines'), "0")
if style.name[0:5] == "Para ":
if style.font.underline != True:
paraStylesWithoutUnderline.append(style.name)
if style.font.bold != True:
paraStylesWithoutBoldOrUnderline.append(
style.name)
# find all para-styles with wanted properties in tables and change style
paraStylesInTables = []
#for paraStyleWithoutBoldOrUnderline in paraStylesWithoutBoldOrUnderline:
for paraStyleWithoutUnderline in paraStylesWithoutUnderline:
for element in document.element.xpath(
"//w:tbl//w:p//w:pStyle[@w:val = '" +
paraStyleWithoutUnderline + "']"):
paraStylesInTables.append(element)
for paraStyleInTables in paraStylesInTables:
paraStyleInTables.attrib[
'{http://schemas.openxmlformats.org/wordprocessingml/2006/main}val'] = normalParagraphNoIndent # or normalParagraph
# uncomment if you want to modify first p in a cell
# firstParaStylesInTables = []
# for paraStyleWithoutBoldOrUnderline in paraStylesWithoutBoldOrUnderline:
# for element in document.element.xpath("//w:tc//w:p[position()=1]//w:pStyle[@w:val = '" + normalParagraph + "']"):
# firstParaStylesInTables.append(element)
# for paraStyleInTables in firstParaStylesInTables:
# paraStyleInTables.attrib['{http://schemas.openxmlformats.org/wordprocessingml/2006/main}val'] = normalParagraphNoIndent
# tables missing required <w:tblGrid>, so throws: docx.oxml.exceptions.InvalidXmlError: required ``<w:tblGrid>`` child element not present
#from docx.table import _Cell, Table
#from docx.oxml.text.paragraph import CT_P
# for row in table.columns:
# try:
# for cell in row.cells:
# firstP = True
# for p in cell.paragraphs:
# if p.style.font.underline != True and re.match(r"^Para | Block",p.style.name):
# if firstP:
# p.style = "NormalNoIndent"
# firstP = False
# else:
# p.style = "Normal"
# except Exception as e:
# pass
document.save(
os.path.join(temp_docxdir,
epub.identifier() + ".docx"))
self.utils.report.info(
"Temp-fil ble lagret: " +
os.path.join(temp_docxdir,
epub.identifier() + ".docx"))
wordFile = os.path.join(temp_docxdir,
epub.identifier() + ".docx")
zipDocument = zipfile.ZipFile((folder / wordFile))
tempFolder = "temp"
zipDocument.extractall(folder / tempFolder)
zipDocument.close()
zippedFile = tempFolder + "/word/numbering.xml"
xmlFile = open((folder / zippedFile), 'r+')
xmlText = xmlFile.read()
xmlText = re.sub(r'w:left="1152"', r'w:left="360"', xmlText)
xmlText = re.sub(r'w:left="1512"', r'w:left="720"', xmlText)
xmlText = re.sub(r'w:left="1872"', r'w:left="1080"', xmlText)
xmlText = re.sub(
r'<w:numFmt w:val="lowerLetter"/><w:lvlText w:val="%([1-9])\."/>',
r'<w:numFmt w:val="lowerLetter"/><w:lvlText w:val="%\1)"/>',
xmlText) # a. as a) in lists
#xmlText = re.sub(r'<w:lvlText w:val="%(1|2)\."/>', r'<w:lvlText w:val="%\1)"/>', xmlText) # a. as a), and 1. as 1) in lists
writeFile(xmlText, zippedFile)
zipdir(str(folder / tempFolder), str(folder),
os.path.join(temp_docxdir,
epub.identifier() + ".docx"))
# ---------- end script from kvile -------
else:
self.utils.report.error(
"En feil oppstod ved konvertering til DOCX for " +
epub.identifier())
self.utils.report.debug(traceback.format_stack())
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
except subprocess.TimeoutExpired:
self.utils.report.error(
"Det tok for lang tid å konvertere " + epub.identifier() +
" til DOCX, og Calibre-prosessen ble derfor stoppet.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
except Exception:
self.utils.report.error(
"En feil oppstod ved konvertering til DOCX for " +
epub.identifier())
self.utils.report.info(traceback.format_exc(), preformatted=True)
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
archived_path, stored = self.utils.filesystem.storeBook(
temp_docxdir, epub.identifier())
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier(
) + " ble konvertert 👍😄" + epubTitle
return True
def on_book(self):
self.utils.report.info("Validerer Daisy 2.02 lydbok")
if self.dp1_home == "" or self.validator_script == "":
if not self.init_environment():
self.utils.report.error(
"Pipeline1 ble ikke funnet. Avbryter..")
return False
folder = self.book["name"]
if self.book["name"].isnumeric() is False:
self.utils.report.warn(
f"{folder} er ikke et tall, prosesserer ikke denne boka. Mulig det er en multivolum bok."
)
self.utils.report.should_email = False
return False
if os.path.isdir(os.path.join(self.dir_out, folder)):
self.utils.report.error(
f"{folder} finnes allerede på share, avbryter.")
return False
if self.nlbsamba_out == "":
self.nlbsamba_out = Config.get("nlbsamba.dir")
if self.nlbsamba_out is None:
self.nlbsamba_out = ""
temp_obj = tempfile.TemporaryDirectory()
temp_dir = temp_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_dir)
if not os.path.isfile(os.path.join(temp_dir, "ncc.html")):
self.utils.report.error("Finner ikke ncc fila")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎. Er dette en daisy 2.02 lydbok med en ncc.html fil?"
return False
try:
ncc_tree = ElementTree.parse(os.path.join(temp_dir, "ncc.html"))
ncc_encoding = ncc_tree.docinfo.encoding.lower()
nccdoc = ncc_tree.getroot()
except Exception:
self.utils.report.info(
"Klarte ikke lese ncc fila. Sjekk loggen for detaljer.")
self.utils.report.debug(traceback.format_exc(), preformatted=True)
return False
edition_identifier = ""
audio_title = ""
audio_title = nccdoc.xpath("string(//*[@name='dc:title']/@content)")
edition_identifier = nccdoc.xpath(
"string(//*[@name='dc:identifier']/@content)")
if ncc_encoding != 'utf-8':
self.utils.report.error(
self.book["name"] +
": Encodingen til filen er ikke utf-8, (f{ncc_encoding}) avbryter."
)
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
str_edition_identifier = str(edition_identifier)
str_book_name = str(self.book["name"])
if edition_identifier == (
"") or str_edition_identifier != str_book_name:
self.utils.report.error(
self.book["name"] +
f": Klarte ikke å bestemme boknummer basert på dc:identifier. dc:identifier: {str_edition_identifier} mappenavn: {str_book_name}"
)
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
self.utils.report.info("Henter metadata fra api.nlb.no")
creative_work_metadata = None
edition_metadata = None
timeout = 0
while creative_work_metadata is None and timeout < 5:
timeout = timeout + 1
creative_work_metadata = Metadata.get_creative_work_from_api(
edition_identifier,
editions_metadata="all",
use_cache_if_possible=True,
creative_work_metadata="all")
edition_metadata = Metadata.get_edition_from_api(
edition_identifier)
if creative_work_metadata is not None:
break
if creative_work_metadata is None:
self.utils.report.warning(
"Klarte ikke finne et åndsverk tilknyttet denne utgaven. Prøver igjen senere."
)
return False
library = edition_metadata["library"].lower()
# in case of wrong upper lower cases
if library == "nlb":
library = "NLB"
elif library == "statped":
library = "Statped"
elif library == "kabb":
library = "KABB"
periodical = False
if creative_work_metadata[
"newspaper"] is True or creative_work_metadata[
"magazine"] is True:
periodical = True
if len(edition_identifier) != 12:
self.utils.report.error(
f"Boka {edition_identifier} er en avis eller et magasin, men utgavenummeret har ikke 12 siffer"
)
return False
else:
if len(edition_identifier) != 6:
self.utils.report.error(
f"Boka {edition_identifier} har ikke 6 siffer")
return False
root_directory = Path(temp_dir)
max_size = 702545920 - 20971520
size = sum(f.stat().st_size for f in root_directory.glob('**/*')
if f.is_file())
multi_volume = False
if size >= max_size:
self.utils.report.info(
f"{edition_identifier} er på størrelse {size}, sjekker om det er en multivolum bok."
)
multi_volume = True
else:
self.utils.report.info(
f"{edition_identifier} er på størrelse {size} bytes")
multi_volume_dirs = []
if multi_volume:
files_dir = os.listdir(self.dir_in)
for file in files_dir:
if file.startswith(self.book["name"]) and file[-1].isdigit(
) and file[-2] == "_":
self.utils.report.info(
f"{file} er en del av multi volum boka {edition_identifier}"
)
multi_volume_dirs.append(file)
multi_volume_directory = Path(
os.path.join(self.dir_in, file))
multi_volume_size = size = sum(
f.stat().st_size
for f in multi_volume_directory.glob('**/*')
if f.is_file())
if multi_volume_size >= max_size:
self.utils.report.info(
f" Multi volum mappen {file} er på størrelse {multi_volume_size}, dette er for stort"
)
self.utils.report.title = self.title + ": " + self.book[
"name"] + " Lydbok feilet 😭👎"
return False
else:
multi_volume_files = os.listdir(multi_volume_directory)
self.utils.report.info(
f"Validerer filer til multi volum {file}...")
if self.check_files(edition_identifier,
multi_volume_files, library,
multi_volume_directory,
multi_volume) is False:
return False
if len(multi_volume_dirs) <= 0:
self.utils.report.error(
f"{edition_identifier} bør være en multivolum bok, men har ikke flere multivolum mapper. Avbryter."
)
self.utils.report.title = self.title + ": " + self.book[
"name"] + "Lydbok feilet 😭👎"
return False
files_book = os.listdir(temp_dir)
if "default.css" in files_book and library != "Statped":
self.utils.report.info("Erstatter default.css med en tom fil")
open(os.path.join(temp_dir, "default.css"), 'w').close()
self.utils.report.info("Validerer filer...")
if self.check_files(edition_identifier, files_book, library, temp_dir,
False) is False:
return False
dc_creator = nccdoc.xpath("string(//*[@name='dc:creator']/@content)")
if not len(dc_creator) >= 1:
self.utils.report.error(
f"{edition_identifier} finner ikke dc:creator, dette må boka ha"
)
return False
dc_narrator = nccdoc.xpath(
"string(//*[@name='ncc:narrator']/@content)")
if not len(dc_narrator) >= 1:
self.utils.report.error(
f"{edition_identifier} finner ikke ncc:narrator, dette må boka ha"
)
return False
multimedia_types = [
"audioOnly", "audioNcc", "audioPartText", "audioFullText",
"textPartAudio", "textNcc"
]
ncc_multimedia_type = nccdoc.xpath(
"string(//*[@name='ncc:multimediaType']/@content)")
if ncc_multimedia_type not in multimedia_types:
self.utils.report.error(
f"{edition_identifier} har ikke en valid ncc:multimediaType, dette må boka ha. Multimediatype er {ncc_multimedia_type}"
)
return False
first_head_class = nccdoc.xpath(
"string(//*[local-name()='h1'][1]/@class)")
second_head = nccdoc.xpath("string(//*[local-name()='h1'][2])").lower()
accepted_second_head = [
"lydbokavtalen", "audiobook agreement", "the audiobook agreement",
"tigar announcement", "nlb"
]
if first_head_class != "title":
self.utils.report.error(
f"{edition_identifier} første heading {first_head_class} er ikke title"
)
return False
if second_head not in accepted_second_head and library == "NLB" and creative_work_metadata[
"newspaper"] is False and not (
creative_work_metadata["magazine"] is True
and library == "KABB"):
self.utils.report.error(
f"{edition_identifier} andre heading {second_head} er ikke Lydbokavtalen, Audiobook agreement, eller Tigar announcement"
)
return False
if library != "Statped":
status = self.validate_book(os.path.join(temp_dir, "ncc.html"))
if status == "ERROR" or status is False:
self.utils.report.error(
"Pipeline validator: Boka er ikke valid. Se rapport.")
return False
self.utils.report.info("Pipeline validator: Boka er valid")
if multi_volume:
for folder in multi_volume_dirs:
self.utils.report.debug(f"Flytter multivolum fil {folder}")
archived_path_multi, stored = self.utils.filesystem.storeBook(
os.path.join(self.dir_in, folder), folder)
self.utils.report.attachment(None, archived_path_multi,
"DEBUG")
if self.nlbsamba_out != "":
archived_path_samba_multi, stored_samba_multi = self.utils.filesystem.storeBook(
os.path.join(self.dir_in, folder),
folder,
dir_out=self.nlbsamba_out)
self.utils.report.attachment(None,
archived_path_samba_multi,
"DEBUG")
shutil.rmtree(os.path.join(self.dir_in, folder))
if library == "Statped":
css_format = "Statped"
elif edition_metadata["includesText"] is True:
css_format = "daisy202"
else:
css_format = "daisy202-ncc"
self.utils.report.info(f"Inserting CSS: {css_format}")
if library != "Statped":
self.utils.filesystem.insert_css(
os.path.join(temp_dir, "default.css"), library, css_format)
files_temp = os.listdir(temp_dir)
archived_path, stored = self.utils.filesystem.storeBook(
temp_dir, edition_identifier)
if self.nlbsamba_out != "":
archived_path_samba, stored_samba = self.utils.filesystem.storeBook(
temp_dir, edition_identifier, dir_out=self.nlbsamba_out)
self.utils.report.attachment(None, archived_path_samba, "DEBUG")
files_out = os.listdir(os.path.join(self.dir_out, edition_identifier))
if self.nlbsamba_out != "":
if len(files_temp) == len(
os.listdir(
os.path.join(self.nlbsamba_out, edition_identifier))):
with open(
os.path.join(self.nlbsamba_out, edition_identifier,
'.donedaisy'), 'w') as file:
self.utils.report.debug(".donedaisy created")
else:
self.utils.report.error(
f"MANGLER FILER i {self.nlbsamba_out}, sjekk utmappa")
return False
if len(files_temp) == len(files_out):
with open(
os.path.join(self.dir_out, edition_identifier,
'.donedaisy'), 'w') as file:
self.utils.report.debug(".donedaisy created")
else:
self.utils.report.error(
f"MANGLER FILER i {self.dir_out}, sjekk utmappa")
return False
self.utils.report.info("Boka er godkjent og overført")
if periodical:
available_title = ""
if creative_work_metadata["newspaper"] is False:
available_title = audio_title
Bibliofil.book_available("DAISY 2.02",
edition_identifier,
title=available_title)
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + edition_identifier + " er valid 👍😄" + audio_title
self.utils.filesystem.deleteSource()
return True
def on_book(self):
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return
if not epub.identifier():
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return
self.utils.report.info("Lager en kopi av EPUBen med tomme bildefiler")
temp_noimages_epubdir_obj = tempfile.TemporaryDirectory()
temp_noimages_epubdir = temp_noimages_epubdir_obj.name
Filesystem.copy(self.utils.report, epub.asDir(), temp_noimages_epubdir)
if os.path.isdir(os.path.join(temp_noimages_epubdir, "EPUB",
"images")):
temp_xml_obj = tempfile.NamedTemporaryFile()
temp_xml = temp_xml_obj.name
opf_image_references = []
html_image_references = {}
for root, dirs, files in os.walk(
os.path.join(temp_noimages_epubdir, "EPUB")):
for file in files:
if file.endswith(".opf"):
opf_file = os.path.join(root, file)
self.utils.report.info(
"Fjerner alle bildereferanser fra OPFen, og erstatter med en referanse til dummy.jpg..."
)
opf_xml_document = ElementTree.parse(opf_file)
opf_xml = opf_xml_document.getroot()
image_items = opf_xml.xpath(
"//*[local-name()='item' and starts-with(@media-type, 'image/')]"
)
replaced = False
for image_item in image_items:
if image_item.attrib[
"href"] not in opf_image_references:
opf_image_references.append(
image_item.attrib["href"])
if image_item.get("href") == "images/cover.jpg":
pass # don't change the reference to cover.jpg
elif not replaced:
image_item.attrib["href"] = "images/dummy.jpg"
replaced = True
else:
image_item.getparent().remove(image_item)
opf_xml_document.write(opf_file,
method='XML',
xml_declaration=True,
encoding='UTF-8',
pretty_print=False)
if file.endswith(".xhtml"):
html_file = os.path.join(root, file)
html_xml_document = ElementTree.parse(html_file)
html_xml = html_xml_document.getroot()
image_references = html_xml.xpath(
"//@href | //@src | //@altimg")
for reference in image_references:
path = reference.split("#")[0]
if path.startswith("images/"):
if path not in html_image_references:
html_image_references[path] = []
html_image_references[path].append(file)
self.utils.report.info(
"Erstatter alle bildereferanser med images/dummy.jpg..."
)
self.utils.report.debug("dummy-jpg.xsl")
self.utils.report.debug(" source = " + html_file)
self.utils.report.debug(" target = " + temp_xml)
xslt = Xslt(self,
stylesheet=os.path.join(
Xslt.xslt_dir, IncomingNordic.uid,
"dummy-jpg.xsl"),
source=html_file,
target=temp_xml)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_xml, html_file)
# validate for the presence of image files here, since epubcheck won't be able to do it anymore after we change the EPUB
image_files_present = []
for root, dirs, files in os.walk(
os.path.join(temp_noimages_epubdir, "EPUB", "images")):
for file in files:
fullpath = os.path.join(root, file)
relpath = os.path.relpath(
fullpath, os.path.join(temp_noimages_epubdir, "EPUB"))
image_files_present.append(relpath)
image_error = False
for file in image_files_present:
if file not in opf_image_references:
self.utils.report.error(
"Bildefilen er ikke deklarert i OPFen: " + file)
image_error = True
for file in opf_image_references:
if file not in image_files_present:
self.utils.report.error(
"Bildefilen er deklarert i OPFen, men finnes ikke: " +
file)
image_error = True
for file in html_image_references:
if file not in opf_image_references:
self.utils.report.error(
"Bildefilen er deklarert i HTMLen, men finnes ikke: " +
file + " (deklarert i: " +
", ".join(html_image_references[file]) + ")")
image_error = True
if image_error:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
for root, dirs, files in os.walk(
os.path.join(temp_noimages_epubdir, "EPUB", "images")):
for file in files:
if file == "cover.jpg":
continue # don't delete the cover file
fullpath = os.path.join(root, file)
os.remove(fullpath)
shutil.copy(
os.path.join(Xslt.xslt_dir, IncomingNordic.uid,
"reference-files", "demobilde.jpg"),
os.path.join(temp_noimages_epubdir, "EPUB", "images",
"dummy.jpg"))
temp_noimages_epub = Epub(self.utils.report, temp_noimages_epubdir)
self.utils.report.info(
"Validerer EPUB med epubcheck og nordiske retningslinjer...")
epub_noimages_file = temp_noimages_epub.asFile()
with DaisyPipelineJob(self,
"nordic-epub3-validate",
{"epub": os.path.basename(epub_noimages_file)},
priority="high",
pipeline_and_script_version=[
("1.13.6", "1.4.6"),
("1.13.4", "1.4.5"),
("1.12.1", "1.4.2"),
("1.11.1-SNAPSHOT", "1.3.0"),
],
context={
os.path.basename(epub_noimages_file):
epub_noimages_file
}) as dp2_job:
# get validation report
report_file = os.path.join(dp2_job.dir_output,
"html-report/report.xhtml")
if os.path.isfile(report_file):
with open(report_file, 'r') as result_report:
self.utils.report.attachment(
result_report.readlines(),
os.path.join(self.utils.report.reportDir(),
"report.html"),
"SUCCESS" if dp2_job.status == "SUCCESS" else "ERROR")
if dp2_job.status != "SUCCESS":
self.utils.report.error("Klarte ikke å validere boken")
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return
self.utils.report.debug("Making a copy of the EPUB to work on…")
epub_fixed, epub_fixed_obj = epub.copy()
epub_unzipped = epub_fixed.asDir()
nav_path = os.path.join(epub_unzipped, epub_fixed.nav_path())
mathML_validation_result = True
mathml_error_count = 0
mathml_errors_not_shown = 0
mathml_report_errors_max = 10
for root, dirs, files in os.walk(epub_unzipped):
for f in files:
file = os.path.join(root, f)
if not file.endswith(".xhtml") or file is nav_path:
continue
self.utils.report.info("Checking MathML in " + file)
mathml_validation = Mathml_validator(
self,
source=file,
report_errors_max=mathml_report_errors_max)
if not mathml_validation.success:
mathml_error_count += mathml_validation.error_count
mathml_errors_not_shown += max(
(mathml_validation.error_count -
mathml_report_errors_max), 0)
if mathml_error_count > mathml_report_errors_max:
mathml_report_errors_max = 0 # don't put any more errors for the other HTML documents in the main report
mathML_validation_result = False
if mathml_errors_not_shown > 0:
self.utils.report.error(
"{} additional MathML errors not shown in the main report. Check the log for details."
.format(mathml_errors_not_shown))
if mathML_validation_result is False:
return False
self.utils.report.debug(
"Making sure that the EPUB has the correct file and directory permissions…"
)
epub_fixed.fix_permissions()
try:
self.utils.report.info("Genererer ACE-rapport...")
ace_dir = os.path.join(self.utils.report.reportDir(),
"accessibility-report")
process = self.utils.filesystem.run(
[IncomingNordic.ace_cli, "-o", ace_dir,
epub_fixed.asFile()])
if process.returncode == 0:
self.utils.report.info("ACE-rapporten ble generert.")
else:
self.utils.report.warn(
"En feil oppstod ved produksjon av ACE-rapporten for " +
epub.identifier())
self.utils.report.debug(traceback.format_stack())
# attach report
ace_status = None
with open(os.path.join(ace_dir, "report.json")) as json_report:
ace_status = json.load(
json_report)["earl:result"]["earl:outcome"]
if ace_status == "pass":
ace_status = "SUCCESS"
else:
ace_status = "WARN"
self.utils.report.attachment(None,
os.path.join(ace_dir, "report.html"),
ace_status)
except subprocess.TimeoutExpired:
self.utils.report.warn(
"Det tok for lang tid å lage ACE-rapporten for " +
epub.identifier() + ", og prosessen ble derfor stoppet.")
except Exception:
self.utils.report.warn(
"En feil oppstod ved produksjon av ACE-rapporten for " +
epub.identifier())
self.utils.report.debug(traceback.format_exc(), preformatted=True)
self.utils.report.info(
"Boken er valid. Kopierer til EPUB master-arkiv.")
archived_path, stored = self.utils.filesystem.storeBook(
epub_fixed.asDir(), epub.identifier())
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier(
) + " er valid 👍😄" + epubTitle
self.utils.filesystem.deleteSource()
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
self.utils.report.info("Locating HTML file")
epub = Epub(self.utils.report, self.book["source"])
if not epub.isepub():
return False
assert epub.isepub(), "The input must be an EPUB"
spine = epub.spine()
if not len(spine) == 1:
self.utils.report.warn(
"There must only be one item in the EPUB spine")
return False
html_file = os.path.join(self.book["source"],
os.path.dirname(epub.opf_path()),
spine[0]["href"])
identifier = epub.identifier()
self.utils.report.info("lag en kopi av boka")
temp_resultdir_obj = tempfile.TemporaryDirectory()
temp_resultdir = temp_resultdir_obj.name
Filesystem.copy(self.utils.report, os.path.dirname(html_file),
temp_resultdir)
temp_result = os.path.join(temp_resultdir, identifier + ".xml")
self.utils.report.info("sletter EPUB-spesifikke filer")
for root, dirs, files in os.walk(temp_resultdir):
for file in files:
if Path(file).suffix.lower() in [
".xhtml", ".html", ".smil", ".mp3", ".wav", ".opf"
]:
os.remove(os.path.join(root, file))
shutil.copy(html_file, temp_result)
temp_xslt_output_obj = tempfile.NamedTemporaryFile()
temp_xslt_output = temp_xslt_output_obj.name
# MATHML to stem
self.utils.report.info("Erstatter evt. MathML i boka...")
mathml_validation = Mathml_validator(self, source=temp_result)
if not mathml_validation.success:
return False
mathML_result = Mathml_to_text(self,
source=temp_result,
target=temp_result)
if not mathML_result.success:
return False
self.utils.report.info("Fikser Webarch-oppmerking")
self.utils.report.debug("webarch-fixup.xsl")
self.utils.report.debug(" source = " + temp_result)
self.utils.report.debug(" target = " + temp_xslt_output)
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir,
NlbpubToNarrationEpub.uid,
"webarch-fixup.xsl"),
source=temp_result,
target=temp_xslt_output)
if not xslt.success:
return False
shutil.copy(temp_xslt_output, temp_result)
self.utils.report.info("Setter inn lydbokavtalen...")
self.utils.report.debug("bokinfo-tts-dtbook.xsl")
self.utils.report.debug(" source = " + temp_result)
self.utils.report.debug(" target = " + temp_xslt_output)
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir,
NlbpubToTtsDtbook.uid,
"bokinfo-tts-dtbook.xsl"),
source=temp_result,
target=temp_xslt_output)
if not xslt.success:
return False
shutil.copy(temp_xslt_output, temp_result)
creative_work_metadata = None
timeout = 0
while creative_work_metadata is None and timeout < 5:
timeout = timeout + 1
creative_work_metadata = Metadata.get_creative_work_from_api(
identifier,
editions_metadata="all",
use_cache_if_possible=True,
creative_work_metadata="all")
if creative_work_metadata is not None:
if creative_work_metadata["magazine"] is True:
self.utils.report.info(
"Fjerner sidetall fordi det er et tidsskrift...")
self.utils.report.debug("remove-pagenum.xsl")
self.utils.report.debug(" source = " + temp_result)
self.utils.report.debug(" target = " + temp_xslt_output)
xslt = Xslt(self,
stylesheet=os.path.join(
Xslt.xslt_dir, NlbpubToTtsDtbook.uid,
"remove-pagenum.xsl"),
source=temp_result,
target=temp_xslt_output)
if not xslt.success:
return False
shutil.copy(temp_xslt_output, temp_result)
break
if creative_work_metadata is None:
self.utils.report.warning(
"Klarte ikke finne et åndsverk tilknyttet denne utgaven. Konverterer likevel."
)
library = epub.meta("schema:library")
library = library.upper() if library else library
logo = os.path.join(Xslt.xslt_dir, NlbpubToTtsDtbook.uid,
"{}_logo.png".format(library))
if os.path.isfile(logo):
# epub_dir = os.path.join(temp_resultdir, "EPUB")
image_dir = os.path.join(temp_resultdir, "images")
if not os.path.isdir(image_dir):
os.mkdir(image_dir)
shutil.copy(logo, image_dir)
self.utils.report.info("Konverterer fra XHTML5 til DTBook...")
self.utils.report.debug("html-to-dtbook.xsl")
self.utils.report.debug(" source = " + temp_result)
self.utils.report.debug(" target = " + temp_xslt_output)
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir,
NlbpubToTtsDtbook.uid,
"html-to-dtbook.xsl"),
source=temp_result,
target=temp_xslt_output)
if not xslt.success:
return False
shutil.copy(temp_xslt_output, temp_result)
self.utils.report.info("Gjør tilpasninger i DTBook")
self.utils.report.debug("dtbook-cleanup.xsl")
self.utils.report.debug(" source = " + temp_result)
self.utils.report.debug(" target = " + temp_xslt_output)
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir,
NlbpubToTtsDtbook.uid,
"dtbook-cleanup.xsl"),
source=temp_result,
target=temp_xslt_output)
if not xslt.success:
return False
shutil.copy(temp_xslt_output, temp_result)
# Fjern denne transformasjonen hvis det oppstår kritiske proplemer med håndteringen av komplekst innhold
self.utils.report.info(
"Legger inn ekstra informasjon om komplekst innhold")
self.utils.report.debug("optimaliser-komplekst-innhold.xsl")
self.utils.report.debug(" source = " + temp_result)
self.utils.report.debug(" target = " + temp_xslt_output)
xslt = Xslt(self,
stylesheet=os.path.join(
Xslt.xslt_dir, NlbpubToTtsDtbook.uid,
"optimaliser-komplekst-innhold.xsl"),
source=temp_result,
target=temp_xslt_output)
if not xslt.success:
return False
shutil.copy(temp_xslt_output, temp_result)
self.utils.report.info("Validerer DTBook...")
# NOTE: This RelaxNG schema assumes that we're using DTBook 2005-3 and MathML 3.0
dtbook_relax = Relaxng(
self,
relaxng=os.path.join(
Xslt.xslt_dir, NlbpubToTtsDtbook.uid,
"dtbook-schema/rng/dtbook-2005-3.mathml-3.integration.rng"),
source=temp_result)
dtbook_sch = Schematron(self,
schematron=os.path.join(
Xslt.xslt_dir, NlbpubToTtsDtbook.uid,
"dtbook-schema/sch/dtbook.mathml.sch"),
source=temp_result)
if not dtbook_relax.success:
self.utils.report.error("Validering av DTBook feilet (RelaxNG)")
if not dtbook_sch.success:
self.utils.report.error("Validering av DTBook feilet (Schematron)")
if not dtbook_relax.success or not dtbook_sch.success:
tempfile_stored = os.path.join(self.utils.report.reportDir(),
os.path.basename(temp_result))
shutil.copy(temp_result, tempfile_stored)
self.utils.report.info(
f"Validering av DTBook feilet, lagrer temp fil for feilsøking: {tempfile_stored}"
)
self.utils.report.attachment(None, tempfile_stored, "DEBUG")
return False
self.utils.report.info(
"Boken ble konvertert. Kopierer til DTBook-arkiv.")
archived_path, stored = self.utils.filesystem.storeBook(
temp_resultdir, identifier)
self.utils.report.attachment(None, archived_path, "DEBUG")
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return
if not epub.identifier():
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return
# ---------- lag en kopi av EPUBen ----------
temp_epubdir_obj = tempfile.TemporaryDirectory()
temp_epubdir = temp_epubdir_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_epubdir)
temp_epub = Epub(self.utils.report, temp_epubdir)
# ---------- gjør tilpasninger i HTML-fila med XSLT ----------
opf_path = temp_epub.opf_path()
if not opf_path:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne OPF-fila i EPUBen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return
opf_path = os.path.join(temp_epubdir, opf_path)
opf_xml = ElementTree.parse(opf_path).getroot()
html_file = opf_xml.xpath(
"/*/*[local-name()='manifest']/*[@id = /*/*[local-name()='spine']/*[1]/@idref]/@href"
)
html_file = html_file[0] if html_file else None
if not html_file:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila i OPFen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return
html_file = os.path.join(os.path.dirname(opf_path), html_file)
if not os.path.isfile(html_file):
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return
temp_xml_obj = tempfile.NamedTemporaryFile()
temp_xml = temp_xml_obj.name
self.utils.report.info("Flater ut NLBPUB")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, NlbpubToEpub.uid,
"nlbpub-flatten.xsl"),
source=html_file,
target=temp_xml)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return
shutil.copy(temp_xml, html_file)
self.utils.report.info("Deler opp NLBPUB i flere HTML-filer")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, NlbpubToEpub.uid,
"nlbpub-split.xsl"),
source=html_file,
target=temp_xml,
parameters={"output-dir": os.path.dirname(html_file)})
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return
os.remove(html_file)
spine_hrefs = []
for href in sorted(os.listdir(os.path.dirname(html_file))):
if href.endswith(".xhtml") and href not in [
"nav.xhtml", os.path.basename(html_file)
]:
spine_hrefs.append(href)
self.utils.report.info("Oppdaterer OPF-fil")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, NlbpubToEpub.uid,
"update-opf.xsl"),
source=opf_path,
target=temp_xml,
parameters={"spine-hrefs": ",".join(spine_hrefs)})
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return
shutil.copy(temp_xml, opf_path)
nav_path = os.path.join(temp_epubdir, temp_epub.nav_path())
self.utils.report.info("Lager nytt navigasjonsdokument")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, NlbpubToEpub.uid,
"generate-nav.xsl"),
source=opf_path,
target=nav_path)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return
self.utils.report.info("Legger til properties i OPF etter behov")
temp_epub.update_opf_properties()
if Epubcheck.isavailable():
epubcheck = Epubcheck(self, opf_path)
if not epubcheck.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return
else:
self.utils.report.warn(
"Epubcheck not available, EPUB will not be validated!")
self.utils.report.info(
"Boken ble konvertert. Kopierer til e-bok-arkiv.")
archived_path, stored = self.utils.filesystem.storeBook(
temp_epubdir, temp_epub.identifier())
self.utils.report.attachment(None, archived_path, "DEBUG")
Bibliofil.book_available(NlbpubToEpub.publication_format,
temp_epub.identifier())
self.utils.report.title = self.title + ": " + epub.identifier(
) + " ble konvertert 👍😄" + epubTitle
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
if not epub.identifier():
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
# ---------- lag en kopi av EPUBen ----------
temp_epubdir_obj = tempfile.TemporaryDirectory()
temp_epubdir = temp_epubdir_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_epubdir)
temp_epub = Epub(self.utils.report, temp_epubdir)
# ---------- gjør tilpasninger i HTML-fila med XSLT ----------
opf_path = temp_epub.opf_path()
if not opf_path:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne OPF-fila i EPUBen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
opf_path = os.path.join(temp_epubdir, opf_path)
opf_xml = ElementTree.parse(opf_path).getroot()
html_file = opf_xml.xpath(
"/*/*[local-name()='manifest']/*[@id = /*/*[local-name()='spine']/*[1]/@idref]/@href"
)
html_file = html_file[0] if html_file else None
if not html_file:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila i OPFen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
html_dir = os.path.dirname(opf_path)
html_file = os.path.join(html_dir, html_file)
if not os.path.isfile(html_file):
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
temp_xml_obj = tempfile.NamedTemporaryFile()
temp_xml = temp_xml_obj.name
# MATHML to stem
self.utils.report.info("Erstatter evt. MathML i boka...")
mathml_validation = Mathml_validator(self, source=html_file)
if not mathml_validation.success:
self.utils.report.error(
"NLBPUB contains MathML errors, aborting...")
return False
mathML_result = Mathml_to_text(self,
source=html_file,
target=html_file)
if not mathML_result.success:
return False
self.utils.report.info(
"Lager skjulte overskrifter der det er nødvendig")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, PrepareForEbook.uid,
"create-hidden-headlines.xsl"),
source=html_file,
target=temp_xml,
parameters={
"cover-headlines": "from-type",
"frontmatter-headlines": "from-type",
"bodymatter-headlines": "from-text",
"backmatter-headlines": "from-type"
})
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_xml, html_file)
self.utils.report.info("Tilpasser innhold for e-bok...")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, PrepareForEbook.uid,
"prepare-for-ebook.xsl"),
source=html_file,
target=temp_xml)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_xml, html_file)
# Use library-specific logo and stylesheet if available
library = temp_epub.meta("schema:library")
library = library.upper() if library else library
logo = os.path.join(Xslt.xslt_dir, PrepareForEbook.uid,
"{}_logo.png".format(library))
if os.path.isfile(logo):
shutil.copy(logo, os.path.join(html_dir, os.path.basename(logo)))
PrepareForEbook.update_css()
stylesheet = PrepareForEbook.css_tempfile_obj.name
if library is not None and library.lower() == "statped":
stylesheet = PrepareForEbook.css_tempfile_statped_obj.name
shutil.copy(stylesheet, os.path.join(html_dir, "ebok.css"))
self.utils.report.info("Legger til logoen i OPF-manifestet")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, PrepareForEbook.uid,
"add-to-opf-manifest.xsl"),
source=opf_path,
target=temp_xml,
parameters={
"href": os.path.basename(logo),
"media-type": "image/png"
})
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_xml, opf_path)
self.utils.report.info("Legger til CSS-fila i OPF-manifestet")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, PrepareForEbook.uid,
"add-to-opf-manifest.xsl"),
source=opf_path,
target=temp_xml,
parameters={
"href": "ebok.css",
"media-type": "text/css"
})
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_xml, opf_path)
# add cover if missing
opf_xml = ElementTree.parse(opf_path).getroot()
cover_id = opf_xml.xpath(
"/*/*[local-name()='manifest']/*[contains(concat(' ', @properties, ' '), ' cover-image ')]/@id"
) # from properties
if not cover_id:
cover_id = opf_xml.xpath(
"/*/*[local-name()='manifest']/*[@name='cover']/@content"
) # from metadata
if not cover_id:
cover_id = opf_xml.xpath(
"/*/*[local-name()='manifest']/*[starts-with(@media-type, 'image/') and contains(@href, 'cover')]/@id"
) # from filename
cover_id = cover_id[0] if cover_id else None
if not cover_id:
# cover not found in the book, let's try NLBs API
# NOTE: identifier at this point is the e-book identifier
edition_url = "{}/editions/{}?creative-work-metadata=none&edition-metadata=all".format(
Config.get("nlb_api_url"), epub.identifier())
response = requests.get(edition_url)
self.utils.report.debug(
"looking for cover image in: {}".format(edition_url))
if response.status_code == 200:
response_json = response.json()
if "data" not in response_json:
self.utils.report.debug("response as JSON:")
self.utils.report.debug(str(response_json))
raise Exception(
"No 'data' in response: {}".format(edition_url))
data = response_json["data"]
cover_url = data["coverUrlLarge"]
if cover_url is not None and cover_url.startswith("http"):
response = requests.get(cover_url)
if response.status_code == 200:
_, extension = os.path.splitext(cover_url)
target_href = "cover" + extension
target_dir = os.path.dirname(opf_path)
with open(os.path.join(target_dir, target_href),
"wb") as target_file:
target_file.write(response.content)
self.utils.report.info(
"Legger til bildet av bokomslaget i OPF-manifestet"
)
media_type = None
if extension.lower() in [
".png"
]: # check for png, just in case. Should always be jpg though.
media_type = "image/png"
else:
media_type = "image/jpeg"
xslt = Xslt(self,
stylesheet=os.path.join(
Xslt.xslt_dir, PrepareForEbook.uid,
"add-to-opf-manifest.xsl"),
source=opf_path,
target=temp_xml,
parameters={
"href": target_href,
"media-type": media_type
})
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_xml, opf_path)
opf_xml = ElementTree.parse(opf_path).getroot()
cover_id = opf_xml.xpath(
"/*/*[local-name()='manifest']/*[@href = '{}']/@id"
.format(target_href)) # from filename
cover_id = cover_id[0] if cover_id else None
if cover_id is None or len(cover_id) == 0:
self.utils.report.warn(
"Klarte ikke å finne bilde av bokomslaget for {}".format(
epub.identifier()))
self.utils.report.info("Legger til properties i OPF etter behov")
temp_epub.update_opf_properties()
# validate with epubcheck
if Epubcheck.isavailable():
epubcheck = Epubcheck(self, opf_path)
if not epubcheck.success:
tempfile_stored_opf = os.path.join(
self.utils.report.reportDir(), os.path.basename(opf_path))
shutil.copy(opf_path, tempfile_stored_opf)
tempfile_stored = os.path.join(self.utils.report.reportDir(),
os.path.basename(html_file))
shutil.copy(html_file, tempfile_stored)
self.utils.report.info(
f"Validering av DTBook feilet, lagrer temp fil for feilsøking: {tempfile_stored}"
)
self.utils.report.attachment(None, tempfile_stored, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return
else:
self.utils.report.warn(
"Epubcheck er ikke tilgjengelig, EPUB blir ikke validert!")
# ---------- lagre filsett ----------
self.utils.report.info(
"Boken ble konvertert. Kopierer til HTML-arkiv.")
archived_path, stored = self.utils.filesystem.storeBook(
temp_epubdir, epub.identifier())
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier(
) + " ble konvertert 👍😄" + epubTitle
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
self.utils.report.info("Lager en kopi av filsettet")
temp_htmldir_obj = tempfile.TemporaryDirectory()
temp_htmldir = temp_htmldir_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_htmldir)
self.utils.report.info("Finner HTML-fila")
html_file = None
for root, dirs, files in os.walk(temp_htmldir):
for f in files:
if f.endswith("html"):
html_file = os.path.join(root, f)
if not html_file or not os.path.isfile(html_file):
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne en HTML-fil.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet "
return False
html_xml = ElementTree.parse(html_file).getroot()
identifier = html_xml.xpath(
"/*/*[local-name()='head']/*[@name='dc:identifier']")
metadata = Metadata.get_metadata_from_book(self.utils.report,
temp_htmldir)
line_spacing = "single"
duplex = "true"
for e in html_xml.xpath(
"/*/*[local-name()='head']/*[@name='dc:format.linespacing']"):
if "double" == e.attrib["content"]:
line_spacing = "double"
for e in html_xml.xpath(
"/*/*[local-name()='head']/*[@name='dc:format.printing']"):
if "single-sided" == e.attrib["content"]:
duplex = "false"
self.utils.report.info("Linjeavstand: {}".format(
"åpen" if line_spacing == "double" else "enkel"))
self.utils.report.info("Trykk: {}".format("enkeltsidig" if duplex ==
"false" else "dobbeltsidig"))
bookTitle = ""
bookTitle = " (" + html_xml.xpath(
"string(/*/*[local-name()='head']/*[local-name()='title']/text())"
) + ") "
identifier = identifier[0].attrib[
"content"] if identifier and "content" in identifier[
0].attrib else None
if not identifier:
self.utils.report.error(
self.book["name"] +
": Klarte ikke å finne boknummer i HTML-fil.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet "
return False
epub_identifier = html_xml.xpath(
"/*/*[local-name()='head']/*[@name='nlbprod:identifier.epub']")
epub_identifier = epub_identifier[0].attrib[
"content"] if epub_identifier and "content" in epub_identifier[
0].attrib else None
# ---------- konverter til PEF ----------
# create context for Pipeline 2 job
html_dir = os.path.dirname(html_file)
html_context = {}
for root, dirs, files in os.walk(html_dir):
for file in files:
kind = mimetypes.guess_type(file)[0]
if kind is not None and kind.split("/")[0] in [
"image", "video", "audio"
]:
continue # ignore media files
fullpath = os.path.join(root, file)
relpath = os.path.relpath(fullpath, html_dir)
html_context[relpath] = fullpath
script_id = "nlb:html-to-pef"
pipeline_and_script_version = [
("1.11.1-SNAPSHOT", "1.10.0-SNAPSHOT"),
]
braille_arguments = {
"source": os.path.basename(html_file),
"braille-standard": "(dots:6)(grade:0)",
"line-spacing": line_spacing,
"duplex": duplex,
}
# for custom Statped options using NLBs PIP (remove `and False` or replace with `or True` to test)
if metadata["library"].lower() == "statped" and False:
# see: https://github.com/nlbdev/pipeline/blob/nlb/nlb/book-to-pef/src/main/resources/xml/html-to-pef.xpl#L146-L167
#
# (1) 'http://www.nlb.no/pipeline/modules/braille/pre-processing.xsl',
# (2) 'http://www.daisy.org/pipeline/modules/braille/xml-to-pef/generate-toc.xsl',
# (3) if ($default-table-class = '') then resolve-uri('add-table-classes.xsl') else (),
# (4) if ($insert-boilerplate = 'true') then 'http://www.nlb.no/pipeline/modules/braille/insert-boilerplate.xsl' else (),
# (5) if ($apply-default-stylesheet = 'true') then 'http://www.nlb.no/pipeline/modules/braille/default.scss' else (),
# (6) if ($stylesheet) then tokenize($stylesheet,',') else ()),' ')"/>
braille_arguments["insert-boilerplate"] = "false" # disable (4)
braille_arguments[
"apply-default-stylesheet"] = "false" # disable (5)
# (1-3) will still be included. Specifying (6) let's us include replacements for (4) and (5)
braille_arguments["stylesheet"] = ",".join([
"https://raw.githubusercontent.com/StatpedEPUB/nlb-scss/master/src/xslt/insert-boilerplate.xsl",
"https://raw.githubusercontent.com/StatpedEPUB/nlb-scss/master/src/scss/braille.scss"
])
# for custom Statped options using DAISYs PIP (remove `and False` or replace with `or True` to test)
if metadata["library"].lower() == "statped" and True:
# use DAISYs version of PIP instead
script_id = "html-to-pef"
pipeline_and_script_version = [
("1.14.6", None),
("1.14.5", None),
("1.14.4", "4.2.0"),
("1.14.4-SNAPSHOT", "4.1.1"),
("1.14.3", "4.1.1"),
("1.14.2", "4.1.0"),
("1.13.6", "1.4.6"),
("1.13.4", "1.4.5"),
("1.12.1", "1.4.2"),
("1.11.1-SNAPSHOT", "1.3.0"),
]
braille_arguments = {
"html":
os.path.basename(html_file),
"transform":
"(formatter:dotify)(translator:liblouis)(dots:6)(grade:0)",
"stylesheet":
" ".join([
# 1. better volume breaking, and also removes title page and print toc, moves the colophon and copyright page to the end of the book
# "https://raw.githubusercontent.com/nlbdev/pipeline/nlb/nlb/book-to-pef/src/main/resources/xml/pre-processing.xsl",
"https://raw.githubusercontent.com/StatpedEPUB/nlb-scss/master/src/xslt/pre-processing.xsl",
#"https://raw.githubusercontent.com/daisy/pipeline/master/modules/braille/xml-to-pef/src/main/resources/xml/xslt/generate-toc.xsl",
# 3. NLB: Add table classes based on the dimensions of the table, for better handling of tables
"https://raw.githubusercontent.com/nlbdev/pipeline/nlb/nlb/book-to-pef/src/main/resources/xml/add-table-classes.xsl",
# 4. NLB: Generate a new title page and about page in the frontmatter
# "https://raw.githubusercontent.com/nlbdev/pipeline/nlb/nlb/book-to-pef/src/main/resources/xml/insert-boilerplate.xsl",
"https://raw.githubusercontent.com/StatpedEPUB/nlb-scss/master/src/xslt/insert-boilerplate.xsl",
# 5. Statped-specific SCSS
"https://raw.githubusercontent.com/StatpedEPUB/nlb-scss/master/src/scss/braille.scss",
]),
"page-width":
'38',
"page-height":
'29',
"toc-depth":
'2',
"maximum-number-of-sheets":
'50',
"include-production-notes":
'true',
"hyphenation":
'false',
"allow-volume-break-inside-leaf-section-factor":
'10',
"prefer-volume-break-before-higher-level-factor":
'1',
"stylesheet-parameters":
"(skip-margin-top-of-page:true)",
}
pef_tempdir_object = tempfile.TemporaryDirectory()
self.utils.report.info("Konverterer fra HTML til PEF...")
found_pipeline_version = None
found_script_version = None
with DaisyPipelineJob(
self,
script_id,
braille_arguments,
pipeline_and_script_version=pipeline_and_script_version,
context=html_context) as dp2_job:
found_pipeline_version = dp2_job.found_pipeline_version
found_script_version = dp2_job.found_script_version
# get conversion report
if os.path.isdir(
os.path.join(dp2_job.dir_output, "preview-output-dir")):
Filesystem.copy(
self.utils.report,
os.path.join(dp2_job.dir_output, "preview-output-dir"),
os.path.join(self.utils.report.reportDir(), "preview"))
self.utils.report.attachment(
None,
os.path.join(self.utils.report.reportDir(),
"preview" + "/" + identifier + ".pef.html"),
"SUCCESS" if dp2_job.status == "SUCCESS" else "ERROR")
if dp2_job.status != "SUCCESS":
self.utils.report.info("Klarte ikke å konvertere boken")
self.utils.report.title = self.title + ": " + identifier + " feilet 😭👎" + bookTitle
return False
dp2_pef_dir = os.path.join(dp2_job.dir_output, "pef-output-dir")
dp2_new_pef_dir = os.path.join(dp2_job.dir_output, "output-dir")
if not os.path.exists(dp2_pef_dir) and os.path.exists(
dp2_new_pef_dir):
dp2_pef_dir = dp2_new_pef_dir
if not os.path.isdir(dp2_pef_dir):
self.utils.report.info("Finner ikke den konverterte boken.")
self.utils.report.title = self.title + ": " + identifier + " feilet 😭👎" + bookTitle
return False
Filesystem.copy(self.utils.report, dp2_pef_dir,
pef_tempdir_object.name)
self.utils.report.info("Boken ble konvertert.")
self.utils.report.info("Kopierer metadata fra HTML til PEF...")
try:
pef_file = None
for root, dirs, files in os.walk(pef_tempdir_object.name):
for f in files:
if f.endswith(".pef"):
pef_file = os.path.join(root, f)
if not pef_file or not os.path.isfile(pef_file):
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne en PEF-fil.")
else:
additional_metadata = []
additional_metadata.append(
("daisy-pipeline-engine-version", "nlbprod",
"http://www.nlb.no/production", None,
found_pipeline_version))
additional_metadata.append(
("daisy-pipeline-script-id", "nlbprod",
"http://www.nlb.no/production", None, script_id))
additional_metadata.append(
("daisy-pipeline-script-version", "nlbprod",
"http://www.nlb.no/production", None,
found_script_version))
for argument in braille_arguments:
if argument in ["source", "html"]:
continue # skip HTML file path
values = braille_arguments[argument]
values = values if isinstance(values, list) else [values]
for value in values:
additional_metadata.append(
("daisy-pipeline-argument", "nlbprod",
"http://www.nlb.no/production", argument, value))
transfer_metadata_from_html_to_pef(html_file, pef_file,
additional_metadata)
except Exception:
self.utils.report.warning(traceback.format_exc(),
preformatted=True)
self.utils.report.error(
"An error occured while trying to insert metadata about the conversion"
)
self.utils.report.info("Kopierer til PEF-arkiv.")
archived_path, stored = self.utils.filesystem.storeBook(
pef_tempdir_object.name, identifier)
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + identifier + " ble konvertert 👍😄" + bookTitle
return True
def on_book(self):
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
self.utils.report.title = self.title + ": " + self.book["name"] + " feilet 😭👎" + epubTitle
return
if not epub.identifier():
self.utils.report.error(self.book["name"] + ": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book["name"] + " feilet 😭👎" + epubTitle
return
self.utils.report.should_email = self.should_email_default
self.utils.report.should_message_slack = self.should_message_slack
self.utils.report.info("Lager kopi av EPUB...")
nordic_epubdir_obj = tempfile.TemporaryDirectory()
nordic_epubdir = nordic_epubdir_obj.name
Filesystem.copy(self.pipeline.utils.report, epub.asDir(), nordic_epubdir)
nordic_epub = Epub(self.utils.report, nordic_epubdir)
html_file = os.path.join(nordic_epubdir, "EPUB", nordic_epub.identifier() + ".xhtml")
nav_file = os.path.join(nordic_epubdir, "EPUB", "nav" + ".xhtml")
package_file = os.path.join(nordic_epubdir, "EPUB", "package" + ".opf")
nlbpub_files = [html_file, nav_file, package_file]
for file in nlbpub_files:
if not os.path.isfile(file):
self.utils.report.error(file + " Not found. This is not a valid NLBPUB")
self.utils.report.info("Validerer NLBPUB")
schematron_files = ["nordic2015-1.sch", "nordic2015-1.nav-references.sch", "nordic2015-1.opf.sch"]
rng_files = "nordic-html5.rng"
html_sch = Schematron(self, schematron=os.path.join(Xslt.xslt_dir, "incoming-NLBPUB", schematron_files[0]), source=html_file)
nav_sch = Schematron(self, schematron=os.path.join(Xslt.xslt_dir, "incoming-NLBPUB", schematron_files[1]), source=nav_file)
opf_sch = Schematron(self, schematron=os.path.join(Xslt.xslt_dir, "incoming-NLBPUB", schematron_files[2]), source=package_file)
warning_sch = Schematron(self,
schematron=os.path.join(Xslt.xslt_dir, "incoming-NLBPUB", "nlbpub-check-need-for-manual-intervention.sch"),
source=html_file)
schematron_list = [html_sch, nav_sch, opf_sch]
html_relax = Relaxng(self, relaxng=os.path.join(Xslt.xslt_dir, "incoming-NLBPUB", rng_files), source=html_file)
for i in range(0, len(schematron_list)):
if not schematron_list[i].success:
self.utils.report.error("Validering av NLBPUB feilet etter schematron: " + schematron_files[i])
return False
if not html_relax.success:
self.utils.report.error("Validering av NLBPUB feilet etter RELAXNG: " + rng_files)
return False
self.utils.report.info("Boken er valid.")
if not self.skip_warning:
#warning_sch = Schematron(self, schematron=os.path.join(Xslt.xslt_dir, "incoming-NLBPUB", "nlbpub-check-need-for-manual-intervention.sch"), source=html_file)
if warning_sch.success is False:
if self.uid == "NLBPUB-incoming-warning":
archived_path, stored = self.utils.filesystem.storeBook(nordic_epubdir, epub.identifier())
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier() + " er valid, men må sjekkes manuelt 👍😄" + epubTitle
self.utils.report.should_email = True
self.utils.report.should_message_slack = True
return True
else:
self.utils.report.should_email = False
self.utils.report.should_message_slack = False
self.utils.report.title = self.title + ": " + epub.identifier() + " er valid, men må sjekkes manuelt 👍😄" + epubTitle
return True
else:
if self.uid == "NLBPUB-incoming-validator":
archived_path, stored = self.utils.filesystem.storeBook(nordic_epubdir, epub.identifier())
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier() + " er valid 👍😄" + epubTitle
self.utils.filesystem.deleteSource()
return True
else:
self.utils.report.info(epub.identifier() + " er valid og har ingen advarsler.")
return True
archived_path, stored = self.utils.filesystem.storeBook(nordic_epubdir, epub.identifier())
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier() + " er valid 👍😄" + epubTitle
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return
if not epub.identifier():
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
# ---------- lag en kopi av EPUBen ----------
temp_epubdir_obj = tempfile.TemporaryDirectory()
temp_epubdir = temp_epubdir_obj.name
Filesystem.copy(self.utils.report, self.book["source"], temp_epubdir)
temp_epub = Epub(self.utils.report, temp_epubdir)
# ---------- gjør tilpasninger i HTML-fila med XSLT ----------
opf_path = temp_epub.opf_path()
if not opf_path:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne OPF-fila i EPUBen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
opf_path = os.path.join(temp_epubdir, opf_path)
opf_xml = ElementTree.parse(opf_path).getroot()
html_file = opf_xml.xpath(
"/*/*[local-name()='manifest']/*[@id = /*/*[local-name()='spine']/*[1]/@idref]/@href"
)
html_file = html_file[0] if html_file else None
if not html_file:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila i OPFen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
html_file = os.path.join(os.path.dirname(opf_path), html_file)
if not os.path.isfile(html_file):
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
temp_html_obj = tempfile.NamedTemporaryFile()
temp_html = temp_html_obj.name
self.utils.report.info("Tilpasser innhold for punktskrift...")
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir,
PrepareForBraille.uid,
"prepare-for-braille.xsl"),
source=html_file,
target=temp_html)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_html, html_file)
# ---------- hent nytt boknummer fra /html/head/meta[@name='dc:identifier'] og bruk som filnavn ----------
html_xml = ElementTree.parse(temp_html).getroot()
result_identifier = html_xml.xpath(
"/*/*[local-name()='head']/*[@name='dc:identifier']")
result_identifier = result_identifier[0].attrib[
"content"] if result_identifier and "content" in result_identifier[
0].attrib else None
if not result_identifier:
self.utils.report.error(
self.book["name"] +
": Klarte ikke å finne boknummer i ny HTML-fil.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
shutil.copy(html_file, temp_html)
os.remove(html_file)
html_file = os.path.join(
os.path.dirname(html_file), result_identifier +
".html") # Bruk html istedenfor xhtml når det ikke er en EPUB
shutil.copy(temp_html, html_file)
# TODO: sett inn HTML5 doctype: <!DOCTYPE html>
# ---------- slett EPUB-spesifikke filer ----------
items = opf_xml.xpath("/*/*[local-name()='manifest']/*")
for item in items:
delete = False
if "properties" in item.attrib and "nav" in re.split(
r'\s+', item.attrib["properties"]):
delete = True
if "media-type" in item.attrib:
if item.attrib["media-type"].startswith("audio/"):
delete = True
elif item.attrib["media-type"] == "application/smil+xml":
delete = True
if not delete or "href" not in item.attrib:
continue
fullpath = os.path.join(os.path.dirname(opf_path),
item.attrib["href"])
os.remove(fullpath)
os.remove(opf_path)
# ---------- lagre HTML-filsett ----------
html_dir = os.path.dirname(opf_path)
self.utils.report.info(
"Boken ble konvertert. Kopierer til arkiv for punkt-klare HTML-filer."
)
archived_path, stored = self.utils.filesystem.storeBook(
html_dir, self.book["name"])
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " ble konvertert 👍😄" + epubTitle
return True
def on_book(self):
self.utils.report.attachment(None, self.book["source"], "DEBUG")
epub = Epub(self.utils.report, self.book["source"])
epubTitle = ""
try:
epubTitle = " (" + epub.meta("dc:title") + ") "
except Exception:
pass
# sjekk at dette er en EPUB
if not epub.isepub():
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
if not epub.identifier():
self.utils.report.error(
self.book["name"] +
": Klarte ikke å bestemme boknummer basert på dc:identifier.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎"
return False
# ---------- lag en kopi av EPUBen ----------
narration_epubdir_obj = tempfile.TemporaryDirectory()
narration_epubdir = narration_epubdir_obj.name
Filesystem.copy(self.utils.report, self.book["source"],
narration_epubdir)
nlbpub = Epub(self.utils.report, narration_epubdir)
# ---------- gjør tilpasninger i HTML-fila med XSLT ----------
opf_path = nlbpub.opf_path()
if not opf_path:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne OPF-fila i EPUBen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
opf_path = os.path.join(narration_epubdir, opf_path)
xml = ElementTree.parse(opf_path).getroot()
html_file = xml.xpath(
"/*/*[local-name()='manifest']/*[@id = /*/*[local-name()='spine']/*[1]/@idref]/@href"
)
html_file = html_file[0] if html_file else None
if not html_file:
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila i OPFen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
html_file = os.path.join(os.path.dirname(opf_path), html_file)
if not os.path.isfile(html_file):
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
temp_html_obj = tempfile.NamedTemporaryFile()
temp_html = temp_html_obj.name
self.utils.report.info(
"Fjerner elementer som ikke skal være med i lydboka...")
self.utils.report.debug("ta-vekk-innhold.xsl")
self.utils.report.debug(" source = " + html_file)
self.utils.report.debug(" target = " + temp_html)
xslt = Xslt(self,
stylesheet=os.path.join(NlbpubToNarrationEpub.xslt_dir,
NlbpubToNarrationEpub.uid,
"ta-vekk-innhold.xsl"),
source=html_file,
target=temp_html)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_html, html_file)
self.utils.report.info("Fikser Webarch-oppmerking")
self.utils.report.debug("webarch-fixup.xsl")
self.utils.report.debug(" source = " + html_file)
self.utils.report.debug(" target = " + temp_html)
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir,
NlbpubToNarrationEpub.uid,
"webarch-fixup.xsl"),
source=html_file,
target=temp_html)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_html, html_file)
self.utils.report.info("Fikser dikt-oppmerking")
self.utils.report.debug("unwrap-poem-chapters.xsl")
self.utils.report.debug(" source = " + html_file)
self.utils.report.debug(" target = " + temp_html)
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir,
NlbpubToNarrationEpub.uid,
"unwrap-poem-chapters.xsl"),
source=html_file,
target=temp_html)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_html, html_file)
self.utils.report.info("Lager usynlige overskrifter der det trengs...")
self.utils.report.debug("create-hidden-headlines.xsl")
self.utils.report.debug(" source = " + html_file)
self.utils.report.debug(" target = " + temp_html)
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, PrepareForEbook.uid,
"create-hidden-headlines.xsl"),
source=html_file,
target=temp_html)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_html, html_file)
self.utils.report.info("Tilpasser innhold for innlesing...")
self.utils.report.debug("prepare-for-narration.xsl")
self.utils.report.debug(" source = " + html_file)
self.utils.report.debug(" target = " + temp_html)
xslt = Xslt(self,
stylesheet=os.path.join(NlbpubToNarrationEpub.xslt_dir,
NlbpubToNarrationEpub.uid,
"prepare-for-narration.xsl"),
source=html_file,
target=temp_html)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_html, html_file)
self.utils.report.info("Lager synkroniseringspunkter...")
self.utils.report.debug("lag-synkroniseringspunkter.xsl")
self.utils.report.debug(" source = " + html_file)
self.utils.report.debug(" target = " + temp_html)
xslt = Xslt(self,
stylesheet=os.path.join(NlbpubToNarrationEpub.xslt_dir,
NlbpubToNarrationEpub.uid,
"lag-synkroniseringspunkter.xsl"),
source=html_file,
target=temp_html)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_html, html_file)
self.utils.report.info("Gjør HTMLen litt penere...")
self.utils.report.debug("pretty-print.xsl")
self.utils.report.debug(" source = " + html_file)
self.utils.report.debug(" target = " + temp_html)
xslt = Xslt(self,
stylesheet=os.path.join(Xslt.xslt_dir, Epub.uid,
"pretty-print.xsl"),
source=html_file,
target=temp_html)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_html, html_file)
# ---------- erstatt metadata i OPF med metadata fra HTML ----------
temp_opf_obj = tempfile.NamedTemporaryFile()
temp_opf = temp_opf_obj.name
xslt = Epub.html_to_opf(self, opf_path, temp_opf)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
shutil.copy(temp_opf, opf_path)
# ---------- hent nytt filnavn fra OPF (det endrer seg basert på boknummer) ----------
try:
xml = ElementTree.parse(opf_path).getroot()
new_html_file = xml.xpath(
"/*/*[local-name()='manifest']/*[@id = /*/*[local-name()='spine']/*[1]/@idref]/@href"
)
new_html_file = os.path.join(
os.path.dirname(opf_path),
new_html_file[0]) if new_html_file else None
except Exception:
self.utils.report.info(traceback.format_exc(), preformatted=True)
self.utils.report.error(self.book["name"] +
": Klarte ikke å finne HTML-fila i OPFen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
if html_file != new_html_file:
shutil.copy(html_file, new_html_file)
os.remove(html_file)
html_file = new_html_file
# ---------- lag nav.xhtml på nytt ----------
nav_path = nlbpub.nav_path()
if not nav_path:
self.utils.report.error(
self.book["name"] +
": Klarte ikke å finne navigasjonsfila i OPFen.")
self.utils.report.title = self.title + ": " + self.book[
"name"] + " feilet 😭👎" + epubTitle
return False
nav_path = os.path.join(narration_epubdir, nav_path)
xslt = Epub.html_to_nav(self, html_file, nav_path)
if not xslt.success:
self.utils.report.title = self.title + ": " + epub.identifier(
) + " feilet 😭👎" + epubTitle
return False
# ---------- legg til logo ----------
library = nlbpub.meta("schema:library")
library = library.upper() if library else library
logo = os.path.join(Xslt.xslt_dir, PrepareForEbook.uid,
"{}_logo.png".format(library))
if os.path.isfile(logo) and library == "STATPED":
shutil.copy(
logo,
os.path.join(os.path.dirname(html_file),
os.path.basename(logo)))
# ---------- save EPUB ----------
self.utils.report.info(
"Boken ble konvertert. Kopierer til innlesingsklart EPUB-arkiv.")
archived_path, stored = self.utils.filesystem.storeBook(
nlbpub.asFile(),
nlbpub.identifier(),
file_extension="epub",
move=True)
self.utils.report.attachment(None, archived_path, "DEBUG")
self.utils.report.title = self.title + ": " + epub.identifier(
) + " ble konvertert 👍😄" + epubTitle
return True