bss/incorporeal-cms
1
0
Fork 0
Code Issues 5 Pull Requests Releases 54 Wiki Activity

remove os.chdir usage, rely on absolute and relative paths more

Browse Source 
os.chdir was getting confusing and hurting the log output, and
potentially the cause of a couple bugs left to fix, so this removes it,
but it means we need to pass around the pages/ absolute path into the
markdown parser, because it relies on knowing both the absolute path
now (to open files), and also the path relative to the pages dir in
order to know where to stop reading parent files/how to generate proper
URL-like references to other files.

probably this should be refactored at some point to inherit the pages/
path from the SSG somehow, rather than passing it through a bunch of
methods, but this seems to work for now

fixes #22

Signed-off-by: Brian S. Stephan <bss@incorporeal.org>
This commit is contained in:
Brian S. Stephan
2025-03-21 10:35:56 -05:00
parent e75d5c48d2
commit 8c75947088
7 changed files with 91 additions and 67 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
incorporealcms/feed.py
View File

@@ -24,6 +24,9 @@ logger = logging.getLogger(__name__)
def generate_feed(feed_type: str, instance_dir: str, dest_dir: str) -> None:
"""Generate the Atom or RSS feed as requested.
Feed entries should be symlinks to .md files in the pages/ directory, so that they
are also linkable and can be browsed outside of the feed.
Args:
feed_type: 'atom' or 'rss' feed
instance_dir: the directory for the instance, containing both the feed dir and pages
@@ -37,16 +40,18 @@ def generate_feed(feed_type: str, instance_dir: str, dest_dir: str) -> None:
fg.link(href=f'https://{Config.DOMAIN_NAME}', rel='alternate')
fg.subtitle(f"Blog posts and other interesting materials from {Config.TITLE_SUFFIX}")
# feed symlinks should all be within the core content subdirectory
pages_dir = os.path.join(instance_dir, 'pages')
# get recent feeds
feed_path = os.path.join(instance_dir, 'feed')
feed_entry_paths = [os.path.join(dirpath, filename) for dirpath, _, filenames in os.walk(feed_path)
for filename in filenames if os.path.islink(os.path.join(dirpath, filename))]
for feed_entry_path in sorted(feed_entry_paths):
# get the actual file to parse it
os.chdir(os.path.abspath(os.path.join(instance_dir, 'pages')))
resolved_path = os.path.relpath(os.path.realpath(feed_entry_path), os.path.join(instance_dir, 'pages'))
resolved_path = os.path.relpath(os.path.realpath(feed_entry_path), pages_dir)
try:
content, md, page_name, page_title, mtime = parse_md(resolved_path)
content, md, page_name, page_title, mtime = parse_md(os.path.join(pages_dir, resolved_path), pages_dir)
link = f'https://{Config.DOMAIN_NAME}{instance_resource_path_to_request_path(resolved_path)}'
except (OSError, ValueError, TypeError):
logger.exception("error loading/rendering markdown!")

48
incorporealcms/markdown.py
View File

@@ -44,18 +44,21 @@ def instance_resource_path_to_request_path(path):
return '/' + re.sub(r'.md$', '', re.sub(r'index.md$', '', path))
def parse_md(path: str):
def parse_md(path: str, pages_root: str):
"""Given a file to parse, return file content and other derived data along with the md object.
Args:
path: the path to the file to render
pages_root: the absolute path to the pages/ dir, which the path should be within. necessary for
proper resolution of resolving parent pages (which needs to know when to stop)
"""
try:
logger.debug("opening path '%s'", path)
with open(path, 'r') as input_file:
absolute_path = os.path.join(pages_root, path)
logger.debug("opening path '%s'", absolute_path)
with open(absolute_path, 'r') as input_file:
mtime = datetime.datetime.fromtimestamp(os.path.getmtime(input_file.name), tz=datetime.timezone.utc)
entry = input_file.read()
logger.debug("path '%s' read", path)
logger.debug("path '%s' read", absolute_path)
md = init_md()
content = Markup(md.convert(entry)) # nosec B704
except (OSError, FileNotFoundError):
@@ -67,17 +70,25 @@ def parse_md(path: str):
logger.debug("file metadata: %s", md.Meta)
page_name = get_meta_str(md, 'title') if md.Meta.get('title') else instance_resource_path_to_request_path(path)
rel_path = os.path.relpath(path, pages_root)
page_name = get_meta_str(md, 'title') if md.Meta.get('title') else instance_resource_path_to_request_path(rel_path)
page_title = f'{page_name} - {Config.TITLE_SUFFIX}' if page_name else Config.TITLE_SUFFIX
logger.debug("title (potentially derived): %s", page_title)
return content, md, page_name, page_title, mtime
def handle_markdown_file_path(path: str) -> str:
"""Given a location on disk, attempt to open it and render the markdown within."""
content, md, page_name, page_title, mtime = parse_md(path)
parent_navs = generate_parent_navs(path)
def handle_markdown_file_path(path: str, pages_root: str) -> str:
"""Given a location on disk, attempt to open it and render the markdown within.
Args:
path: the path to the file to parse and produce metadata for
pages_root: the absolute path to the pages/ dir, which the path should be within. necessary for
proper resolution of resolving parent pages (which needs to know when to stop)
"""
content, md, page_name, page_title, mtime = parse_md(path, pages_root)
relative_path = os.path.relpath(path, pages_root)
parent_navs = generate_parent_navs(relative_path, pages_root)
extra_footer = get_meta_str(md, 'footer') if md.Meta.get('footer') else None
template_name = get_meta_str(md, 'template') if md.Meta.get('template') else 'base.html'
@@ -92,14 +103,21 @@ def handle_markdown_file_path(path: str) -> str:
description=get_meta_str(md, 'description'),
image=Config.BASE_HOST + get_meta_str(md, 'image'),
content=content,
base_url=Config.BASE_HOST + instance_resource_path_to_request_path(path),
base_url=Config.BASE_HOST + instance_resource_path_to_request_path(relative_path),
navs=parent_navs,
mtime=mtime.strftime('%Y-%m-%d %H:%M:%S %Z'),
extra_footer=extra_footer)
def generate_parent_navs(path):
"""Create a series of paths/links to navigate up from the given resource path."""
def generate_parent_navs(path, pages_root: str):
"""Create a series of paths/links to navigate up from the given resource path.
Args:
path: the path to parse and generate parent metadata nav links for
pages_root: the absolute path to the pages/ dir, which the path should be within. path is relative,
but opening parents requires the full path
"""
logger.debug("path to generate navs for: %s", path)
if path == 'index.md':
# bail and return the domain name as a terminal case
return [(Config.DOMAIN_NAME, '/')]
@@ -124,14 +142,14 @@ def generate_parent_navs(path):
# read the resource
try:
with open(path, 'r') as entry_file:
with open(os.path.join(pages_root, path), 'r') as entry_file:
entry = entry_file.read()
_ = Markup(md.convert(entry)) # nosec B704
page_name = (" ".join(md.Meta.get('title')) if md.Meta.get('title')
else request_path_to_breadcrumb_display(request_path))
return generate_parent_navs(parent_resource_path) + [(page_name, request_path)]
return generate_parent_navs(parent_resource_path, pages_root) + [(page_name, request_path)]
except FileNotFoundError:
return generate_parent_navs(parent_resource_path) + [(request_path, request_path)]
return generate_parent_navs(parent_resource_path, pages_root) + [(request_path, request_path)]
def request_path_to_breadcrumb_display(path):

19
incorporealcms/ssg.py
View File

@@ -86,7 +86,6 @@ class StaticSiteGenerator(object):
convert_markdown: whether or not to convert Markdown files (or simply copy them)
"""
cprint(f"copying files from '{source_dir}' to '{dest_dir}'", 'green')
os.chdir(source_dir)
for base_dir, subdirs, files in os.walk(source_dir):
logger.debug("starting to build against %s || %s || %s", base_dir, subdirs, files)
# remove the absolute path of the directory from the base_dir
@@ -113,9 +112,12 @@ class StaticSiteGenerator(object):
dest_dir: the output directory to place the subdir in
"""
dst = os.path.join(dest_dir, base_dir, subdir)
if os.path.islink(os.path.join(base_dir, subdir)):
absolute_dir = os.path.join(source_dir, base_dir, subdir)
logger.debug("checking if %s is a symlink or not", absolute_dir)
if os.path.islink(absolute_dir):
logger.debug("symlink; raw destination is %s", os.path.realpath(absolute_dir))
# keep the link relative to the output directory
src = self.symlink_to_relative_dest(source_dir, os.path.join(base_dir, subdir))
src = self.symlink_to_relative_dest(source_dir, absolute_dir)
print(f"creating directory symlink '{dst}' -> '{src}'")
os.symlink(src, dst, target_is_directory=True)
else:
@@ -137,9 +139,12 @@ class StaticSiteGenerator(object):
dest_dir: the output directory to place the subdir in
"""
dst = os.path.join(dest_dir, base_dir, file_)
if os.path.islink(os.path.join(base_dir, file_)):
absolute_file = os.path.join(source_dir, base_dir, file_)
logger.debug("checking if %s is a symlink or not", absolute_file)
if os.path.islink(absolute_file):
logger.debug("symlink; raw destination is %s", os.path.realpath(absolute_file))
# keep the link relative to the output directory
src = self.symlink_to_relative_dest(source_dir, os.path.join(base_dir, file_))
src = self.symlink_to_relative_dest(source_dir, absolute_file)
print(f"creating symlink '{dst}' -> '{src}'")
os.symlink(src, dst, target_is_directory=False)
if src.endswith('.md') and convert_markdown:
@@ -151,7 +156,7 @@ class StaticSiteGenerator(object):
os.symlink(second_src, second_dst, target_is_directory=False)
else:
src = os.path.join(base_dir, file_)
src = os.path.join(source_dir, base_dir, file_)
print(f"copying file '{src}' -> '{dst}'")
shutil.copy2(src, dst)
@@ -160,7 +165,7 @@ class StaticSiteGenerator(object):
rendered_file = dst.removesuffix('.md') + '.html'
print(f"rendering file '{src}' -> '{rendered_file}'")
try:
content = handle_markdown_file_path(src)
content = handle_markdown_file_path(src, source_dir)
except UnicodeDecodeError:
# perhaps this isn't a markdown file at all for some reason; we
# copied it above so stick with tha