def test_get_files(self):
self.tmp_dir = tempfile.mkdtemp(prefix='Door43_test_file_utils_')
_, tmp_file1 = tempfile.mkstemp(dir=self.tmp_dir)
_, tmp_file2 = tempfile.mkstemp(dir=self.tmp_dir)
tmp_subdir = os.path.join(self.tmp_dir, 'subdir')
os.mkdir(tmp_subdir)
_, tmp_file3 = tempfile.mkstemp(dir=tmp_subdir, suffix=".md")
files = file_utils.get_files(self.tmp_dir,
relative_paths=False,
include_directories=True)
self.assertEqual(len(files), 4)
self.assertTrue(any(self.paths_equal(tmp_file1, d) for d in files))
self.assertTrue(any(self.paths_equal(tmp_file2, d) for d in files))
self.assertTrue(any(self.paths_equal(tmp_subdir, d) for d in files))
self.assertTrue(any(self.paths_equal(tmp_file3, d) for d in files))
files = file_utils.get_files(self.tmp_dir, extensions=['.md'])
self.assertEqual(len(files), 1)
files = file_utils.get_files(self.tmp_dir,
relative_paths=True,
include_directories=True)
self.assertEqual(len(files), 4)
self.assertTrue(
any(
self.paths_equal(os.path.relpath(tmp_file1, self.tmp_dir), d)
for d in files))
self.assertTrue(
any(
self.paths_equal(os.path.relpath(tmp_file2, self.tmp_dir), d)
for d in files))
self.assertTrue(
any(
self.paths_equal(os.path.relpath(tmp_subdir, self.tmp_dir), d)
for d in files))
self.assertTrue(
any(
self.paths_equal(os.path.relpath(tmp_file3, self.tmp_dir), d)
for d in files))
def get_files(self, relative_paths: bool) -> List[str]:
"""
relative_paths can be True or False
Returns a sorted list of .md files to be processed
"""
if self.single_dir:
dir_path = os.path.join(self.source_dir, self.single_dir)
sub_files = sorted(
get_files(directory=dir_path,
relative_paths=relative_paths,
exclude=self.EXCLUDED_FILES,
extensions=['.md']))
files = []
for f in sub_files:
files.append(os.path.join(self.single_dir, f))
else:
files = sorted(
get_files(directory=self.source_dir,
relative_paths=relative_paths,
exclude=self.EXCLUDED_FILES,
extensions=['.md']))
return files
def test_get_files(self):
tmp_dir = tempfile.mkdtemp()
_, tmp_file1 = tempfile.mkstemp(dir=tmp_dir)
_, tmp_file2 = tempfile.mkstemp(dir=tmp_dir)
tmp_subdir = tmp_dir + "/subdir"
os.mkdir(tmp_subdir)
_, tmp_file3 = tempfile.mkstemp(dir=tmp_subdir)
files = file_utils.get_files(tmp_dir,
relative_paths=False,
include_directories=True)
self.assertEqual(len(files), 4)
self.assertTrue(any(self.paths_equal(tmp_file1, d) for d in files))
self.assertTrue(any(self.paths_equal(tmp_file2, d) for d in files))
self.assertTrue(any(self.paths_equal(tmp_subdir, d) for d in files))
self.assertTrue(any(self.paths_equal(tmp_file3, d) for d in files))
files = file_utils.get_files(tmp_dir,
relative_paths=True,
include_directories=True)
self.assertEqual(len(files), 4)
self.assertTrue(
any(
self.paths_equal(os.path.relpath(tmp_file1, tmp_dir), d)
for d in files))
self.assertTrue(
any(
self.paths_equal(os.path.relpath(tmp_file2, tmp_dir), d)
for d in files))
self.assertTrue(
any(
self.paths_equal(os.path.relpath(tmp_subdir, tmp_dir), d)
for d in files))
self.assertTrue(
any(
self.paths_equal(os.path.relpath(tmp_file3, tmp_dir), d)
for d in files))
def update_from_files(self, path):
path = path.rstrip('/')
found_markdown = False
found_usfm = False
found_html = False
found_text_in_numbered_dir = False
if not self.format:
for f in get_files(path):
if f.endswith('usfm'):
found_usfm = True
elif f.endswith('.md'):
found_markdown = True
elif f.endswith('.html'):
found_html = True
elif f.endswith('.txt'):
try:
if int(os.path.basename(os.path.dirname(f))):
found_text_in_numbered_dir = True
except Exception:
pass
if found_usfm:
if not self.format:
self.format = 'usfm'
if not self.resource['id']:
self.resource['id'] = 'bible'
self.resource['name'] = 'Bible'
elif found_markdown:
if not self.format:
self.format = 'markdown'
elif found_html:
if not self.format:
self.format = 'html'
if not self.generator['name']:
for subdir in glob(os.path.join(path, '*')):
if os.path.isdir(subdir):
dir_name = subdir[len(path) + 1:]
try:
if int(dir_name) and len(
glob(os.path.join(subdir, '*.txt'))) > 0:
self.generator['name'] = 'ts'
break
except Exception:
continue
def convert_obs(self) -> None:
self.log.info("Converting OBS markdown files…")
# Find the first directory that has md files.
files = get_files(directory=self.files_dir,
exclude=self.EXCLUDED_FILES)
current_dir = os.path.dirname(os.path.realpath(__file__))
with open(os.path.join(current_dir, 'templates',
'template.html')) as template_file:
html_template = string.Template(template_file.read())
# found_chapters = {}
for filepath in sorted(files):
if filepath.endswith('.md'):
# Convert files that are markdown files
base_name_part = os.path.splitext(
os.path.basename(filepath))[0]
# found_chapters[base_name] = True
try:
md = read_file(filepath)
except Exception as e:
self.log.error(
f"Error reading {base_name_part+'.md'}: {e}")
continue
html = markdown.markdown(md)
html = html_template.safe_substitute(
title=self.repo_subject.replace('_', ' '), content=html)
html_filename = base_name_part + '.html'
output_filepath = os.path.join(self.output_dir, html_filename)
write_file(output_filepath, html)
self.log.info(
f"Converted {os.path.basename(filepath)} to {os.path.basename(html_filename)}."
)
else:
# Directly copy over files that are not markdown files
try:
output_filepath = os.path.join(self.output_dir,
os.path.basename(filepath))
if not os.path.exists(output_filepath):
copyfile(filepath, output_filepath)
except:
pass
self.log.info("Finished processing OBS markdown files.")
def mock_s3_bible_project(self,
test_file_name,
project_key,
multi_part=False):
converted_proj_dir = os.path.join(self.resources_dir,
'converted_projects')
test_file_base = test_file_name.split('.zip')[0]
zip_file = os.path.join(converted_proj_dir, test_file_name)
out_dir = os.path.join(self.temp_dir, test_file_base)
unzip(zip_file, out_dir)
project_dir = os.path.join(out_dir, test_file_base) + os.path.sep
self.project_files = file_utils.get_files(out_dir)
self.project_key = project_key
for filename in self.project_files:
sub_path = filename.split(project_dir)[1].replace(
os.path.sep, '/') # Make sure it is a bucket path
AppSettings.cdn_s3_handler().upload_file(
filename, '{0}/{1}'.format(project_key, sub_path))
if multi_part: # copy files from cdn to door43
base_name = os.path.basename(filename)
if '.html' in base_name:
with open(filename, 'r') as f:
soup = BeautifulSoup(f, 'html.parser')
# add nav tag
new_tag = soup.new_tag('div', id='right-sidebar')
soup.body.append(new_tag)
html = str(soup)
file_utils.write_file(
filename, html.encode('ascii', 'xmlcharrefreplace'))
AppSettings.door43_s3_handler().upload_file(
filename, '{0}/{1}'.format(project_key, base_name))
# u, user, repo = project_key
AppSettings.door43_s3_handler().upload_file(
os.path.join(self.resources_dir, 'templates', 'project-page.html'),
'templates/project-page.html')
def convert(self) -> bool:
"""
Main function to convert info in TSV files into HTML files.
"""
AppSettings.logger.debug(
"Tsv2HtmlConverter processing the TSV files …")
# Find the first directory that has usfm files.
filepaths = get_files(directory=self.files_dir,
exclude=self.EXCLUDED_FILES)
# convert_only_list = self.check_for_exclusive_convert()
convert_only_list = [] # Not totally sure what the above line did
# Process the manifest file
self.manifest_dict = None
for source_filepath in filepaths:
if 'manifest.yaml' in source_filepath:
self.process_manifest(source_filepath)
break
current_dir = os.path.dirname(os.path.realpath(__file__))
with open(os.path.join(current_dir, 'templates',
'template.html')) as template_file:
# Simple HTML template which includes $title and $content fields
template_html = template_file.read()
# Convert tsv files and copy across other files
num_successful_books = num_failed_books = 0
for source_filepath in sorted(filepaths):
base_name = os.path.basename(source_filepath)
if source_filepath.endswith('.tsv'):
if convert_only_list and (
base_name not in convert_only_list
): # see if this is a file we are to convert
continue
# Convert the TSV file
self.log.info(
f"Tsv2HtmlConverter converting TSV file: {base_name} …"
) # Logger also issues DEBUG msg
filebase = os.path.splitext(
os.path.basename(source_filepath))[0]
# Do the actual TSV -> HTML conversion
converted_html = self.buildSingleHtml(source_filepath)
# AppSettings.logger.debug(f"Got converted html: {converted_html[:5000]}{' …' if len(converted_html)>5000 else ''}")
# Now what are we doing with the converted html ???
template_soup = BeautifulSoup(template_html, 'html.parser')
template_soup.head.title.string = self.repo_subject
converted_soup = BeautifulSoup(converted_html, 'html.parser')
content_div = template_soup.find('div', id='content')
content_div.clear()
if converted_soup and converted_soup.body:
content_div.append(converted_soup.body)
content_div.body.unwrap()
num_successful_books += 1
else:
content_div.append('ERROR! NOT CONVERTED!')
self.log.warning(
f"TSV parsing or conversion error for {base_name}")
# AppSettings.logger.debug(f"Got converted html: {converted_html[:600]}{' …' if len(converted_html)>600 else ''}")
if not converted_soup:
AppSettings.logger.debug(f"No converted_soup")
elif not converted_soup.body:
AppSettings.logger.debug(f"No converted_soup.body")
# from bs4.diagnose import diagnose
# diagnose(converted_html)
num_failed_books += 1
html_filename = filebase + '.html'
output_filepath = os.path.join(self.output_dir, html_filename)
#print("template_soup type is", type(template_soup)) # <class 'bs4.BeautifulSoup'>
write_file(output_filepath, str(template_soup))
#print("Got converted x2 html:", str(template_soup)[:500])
self.log.info(
f"Converted {os.path.basename(source_filepath)} to {os.path.basename(html_filename)}."
)
else:
# Directly copy over files that are not TSV files
try:
output_filepath = os.path.join(self.output_dir, base_name)
if not os.path.exists(output_filepath):
copyfile(source_filepath, output_filepath)
except:
pass
if num_failed_books and not num_successful_books:
self.log.error(f"Conversion of all books failed!")
self.log.info("Finished processing TSV files.")
return True
def convert(self):
AppSettings.logger.debug("Processing the Bible USFM files …")
# Find the first directory that has usfm files.
files = get_files(directory=self.files_dir,
exclude=self.EXCLUDED_FILES)
# convert_only_list = self.check_for_exclusive_convert()
convert_only_list = [] # Not totally sure what the above line did
current_dir = os.path.dirname(os.path.realpath(__file__))
with open(os.path.join(current_dir, 'templates',
'template.html')) as template_file:
# Simple HTML template which includes $title and $content fields
template_html = template_file.read()
# Convert usfm files and copy across other files
num_successful_books = num_failed_books = 0
for filename in sorted(files):
if filename.endswith('.usfm'):
base_name = os.path.basename(filename)
if convert_only_list and (
base_name not in convert_only_list
): # see if this is a file we are to convert
continue
# Convert the USFM file
self.log.info(f"Converting Bible USFM file: {base_name} …"
) # Logger also issues DEBUG msg
# Copy just the single file to be converted into a single scratch folder
scratch_dir = tempfile.mkdtemp(
prefix='tX_convert_usfm_scratch_')
delete_scratch_dir_flag = True # Set to False for debugging this code
copyfile(filename,
os.path.join(scratch_dir, os.path.basename(filename)))
filebase = os.path.splitext(os.path.basename(filename))[0]
# Do the actual USFM -> HTML conversion
warning_list = UsfmTransform.buildSingleHtml(
scratch_dir, scratch_dir, filebase)
if warning_list:
for warning_msg in warning_list:
self.log.warning(f"{filebase} - {warning_msg}")
# This code seems to be cleaning up or adjusting the converted HTML file
html_filename = filebase + '.html'
with open(os.path.join(scratch_dir, html_filename),
'rt',
encoding='utf-8') as html_file:
converted_html = html_file.read()
converted_html_length = len(converted_html)
# AppSettings.logger.debug(f"### Usfm2HtmlConverter got converted html of length {converted_html_length:,}")
# AppSettings.logger.debug(f"Got converted html: {converted_html[:500]}{' …' if len(converted_html)>500 else ''}")
if '</p></p></p>' in converted_html:
AppSettings.logger.debug(
f"Usfm2HtmlConverter got multiple consecutive paragraph closures in converted {html_filename}"
)
# Now what are we doing with the converted html ???
template_soup = BeautifulSoup(template_html, 'html.parser')
template_soup.head.title.string = self.repo_subject
converted_soup = BeautifulSoup(converted_html, 'html.parser')
content_div = template_soup.find('div', id='content')
content_div.clear()
if converted_soup and converted_soup.body:
content_div.append(converted_soup.body)
content_div.body.unwrap()
num_successful_books += 1
else:
content_div.append("ERROR! NOT CONVERTED!")
self.log.warning(
f"USFM parsing or conversion error for {base_name}")
AppSettings.logger.debug(
f"Got converted html: {converted_html[:600]}{' …' if len(converted_html)>600 else ''}"
)
if not converted_soup:
AppSettings.logger.debug(f"No converted_soup")
elif not converted_soup.body:
AppSettings.logger.debug(f"No converted_soup.body")
# from bs4.diagnose import diagnose
# diagnose(converted_html)
num_failed_books += 1
output_filepath = os.path.join(self.output_dir, html_filename)
#print("template_soup type is", type(template_soup)) # <class 'bs4.BeautifulSoup'>
template_soup_string = str(template_soup)
write_file(output_filepath, template_soup_string)
template_soup_string_length = len(template_soup_string)
# AppSettings.logger.debug(f"### Usfm2HtmlConverter wrote souped-up html of length {template_soup_string_length:,} from {converted_html_length:,}")
if '</p></p></p>' in template_soup_string:
AppSettings.logger.warning(
f"Usfm2HtmlConverter got multiple consecutive paragraph closures in {html_filename}"
)
if template_soup_string_length < converted_html_length * 0.67: # What is the 33% or so that's lost ???
AppSettings.logger.debug(
f"### Usfm2HtmlConverter wrote souped-up html of length {template_soup_string_length:,} from {converted_html_length:,} = {template_soup_string_length*100.0/converted_html_length}%"
)
self.log.warning(
f"Usfm2HtmlConverter possibly lost converted html for {html_filename}"
)
AppSettings.logger.info(
f"Usfm2HtmlConverter {html_filename} was {converted_html_length:,} now {template_soup_string_length:,}"
)
# AppSettings.logger.debug(f"Usfm2HtmlConverter {html_filename} was: {converted_html}")
write_file(
os.path.join(scratch_dir,
filebase + '.converted.html'),
template_soup_string)
if prefix and debug_mode_flag:
delete_scratch_dir_flag = False
#print("Got converted x2 html:", str(template_soup)[:500])
# self.log.info(f"Converted {os.path.basename(filename)} to {os.path.basename(html_filename)}.")
if delete_scratch_dir_flag:
remove_tree(scratch_dir)
else:
# Directly copy over files that are not USFM files
try:
output_filepath = os.path.join(self.output_dir,
os.path.basename(filename))
if not os.path.exists(output_filepath):
copyfile(filename, output_filepath)
except:
pass
if num_failed_books and not num_successful_books:
self.log.error(f"Conversion of all books failed!")
self.log.info(
f"Finished processing {num_successful_books} Bible USFM files.")
return True
def convert_markdown(self) -> None:
self.log.info("Converting Markdown files…")
# Find the first directory that has md files.
files = get_files(directory=self.files_dir,
exclude=self.EXCLUDED_FILES)
# convert_only_list = self.check_for_exclusive_convert()
convert_only_list = [] # Not totally sure what the above line did
current_dir = os.path.dirname(os.path.realpath(__file__))
with open(os.path.join(current_dir, 'templates',
'template.html')) as template_file:
# Just a very simple template with $title and $content place-holders
html_template = string.Template(template_file.read())
# found_chapters = {}
for filepath in sorted(files):
if filepath.endswith('.md'):
base_name_part = os.path.splitext(
os.path.basename(filepath))[0]
filename = base_name_part + '.md'
if convert_only_list and (
filename not in convert_only_list
): # see if this is a file we are to convert
continue
html_filename = base_name_part + '.html'
AppSettings.logger.debug(
f"Converting '{filename}' to '{html_filename}' …")
# Convert files that are markdown files
try:
md = read_file(filepath)
except Exception as e:
self.log.error(f"Error reading {filename}: {e}")
continue
# if 0: # test code—creates html1
# headers = {"content-type": "application/json"}
# url = "http://bg.door43.org/api/v1/markdown"
# payload = {
# 'Context': "",
# 'Mode': "normal",
# 'Text': md,
# 'Wiki': False
# }
# # url = "http://bg.door43.org/api/v1/markdown/raw"
# AppSettings.logger.debug(f"Making callback to {url} with payload:")
# AppSettings.logger.debug(json.dumps(payload)[:256] + '…')
# try:
# response = requests.post(url, json=payload, headers=headers)
# # response = requests.post(url, data=md, headers=headers)
# except requests.exceptions.ConnectionError as e:
# AppSettings.logger.critical(f"Markdown->HTML connection error: {e}")
# response = None
# if response:
# #AppSettings.logger.info(f"response.status_code = {response.status_code}, response.reason = {response.reason}")
# #AppSettings.logger.debug(f"response.headers = {response.headers}")
# AppSettings.logger.debug(f"response.text = {response.text[:256] + '…'}")
# html1 = response.text
# if response.status_code != 200:
# AppSettings.logger.critical(f"Failed to submit Markdown->HTML job:"
# f" {response.status_code}={response.reason}")
# # callback_status = response.status_code
# # if (callback_status >= 200) and (callback_status < 299):
# # AppSettings.logger.debug("Markdown->HTML callback finished.")
# # else:
# # AppSettings.logger.error(f"Error calling callback code {callback_status}: {response.reason}")
# else: # no response
# AppSettings.logger.error("Submission of job to Markdown->HTML got no response")
if 1: # old/existing code—creates html2
if self.repo_subject in [
'Translation_Academy',
]:
html2 = markdown2.markdown(
md, extras=['markdown-in-html', 'tables'])
if prefix and debug_mode_flag:
write_file(
os.path.join(self.debug_dir,
base_name_part + '.1.html'),
html2)
else:
html2 = markdown.markdown(md)
# if 0:
# if html2 == html1:
# AppSettings.logger.debug("HTML responses are identical.")
# else:
# AppSettings.logger.error(f"HTML responses differ: {len(html1)} and {len(html2)}")
# AppSettings.logger.debug(repr(html1)[:256] + ' …… ' + repr(html1)[-256:])
# AppSettings.logger.debug(repr(html2)[:256] + ' …… ' + repr(html2)[-256:])
# try: html = html1
# except UnboundLocalError: html = html2
# else:
html = html2
# print(f"md len = {len(md):,} html len = {len(html):,} ratio = {len(html)/len(md):.2f}")
if len(html) < len(md): # Seems created html was too short
AppSettings.logger.error(
f"Converter error: {filename} HTML ended up smaller than the original markdown!"
)
self.log.info(f"Markdown was {md[:20]} …… {md[-200:]}")
self.log.info(f"HTML is {html[:20]} ' …… ', {html[-200:]}")
html = html_template.safe_substitute(
title=self.repo_subject.replace('_', ' '), content=html)
if prefix and debug_mode_flag:
write_file(
os.path.join(self.debug_dir,
base_name_part + '.2.html'), html)
html = fix_naked_urls(html)
if prefix and debug_mode_flag:
write_file(
os.path.join(self.debug_dir,
base_name_part + '.3.html'), html)
# Change headers like <h1><a id="verbs"/>Verbs</h1> to <h1 id="verbs">Verbs</h1>
soup = BeautifulSoup(html, 'html.parser')
for tag in soup.findAll('a', {'id': True}):
if tag.parent and tag.parent.name in [
'h1', 'h2', 'h3', 'h4', 'h5', 'h6'
]:
tag.parent['id'] = tag['id']
tag.parent['class'] = tag.parent.get(
'class', []) + ['section-header']
tag.extract()
html = str(soup)
# Write the file
base_name_part = os.path.splitext(
os.path.basename(filepath))[0]
# found_chapters[base_name_part] = True
output_file = os.path.join(self.output_dir, html_filename)
write_file(output_file, html)
self.log.info(f"Converted {filename} to {html_filename}.")
else:
# Directly copy over files that are not markdown files
try:
output_file = os.path.join(self.output_dir,
os.path.basename(filepath))
if not os.path.exists(output_file):
copyfile(filepath, output_file)
except:
pass
self.log.info("Finished processing generic markdown files.")
