From 7c16beb6538ccf024b552d475a26b9345bf550ec Mon Sep 17 00:00:00 2001 From: mjfernez Date: Mon, 18 Oct 2021 19:53:03 -0400 Subject: Fixes is_hidden_path, RSS. Adds txt support, RSS This commit fixes the is_hidden_path function to work for subdirectories and also remove entries from the RSS file view_functions.py has been refactored use the siteconfig file instead of the app to remove unecessary imports and avoid circularly imports This also moves logic for the default site directories from server.py to siteconfig.py so it's accessible before the app runs. This has the benefit of cleaning up the server file, but the drawback of adding the os import to the siteconfig file. Settings have been moved around for (hopefully) easier reading --- rss_generator.py | 47 ++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 38 insertions(+), 9 deletions(-) (limited to 'rss_generator.py') diff --git a/rss_generator.py b/rss_generator.py index 4397a45..f248003 100644 --- a/rss_generator.py +++ b/rss_generator.py @@ -1,12 +1,24 @@ import os from time import strftime, strptime, ctime from siteconfig import siteconfig +from view_functions import is_hidden_path class RSS_Item: + """ + RSS_Item - a (very) basic implementation of an object in an RSS + feed using only essential parameters as specified in: + https://www.rssboard.org/rss-specification#hrelementsOfLtitemgt + + Item data is generated from a given file path + """ PARAGRAPHS = siteconfig.rss_channel_config['DESCRIPTION_LENGTH'] class NotAFile(Exception): + """ + Throws an exception if an RSS_Item is made out of a + directory or invalid file + """ def __init__(self, path: str): self.path = path self.message = f"{path} not a file" @@ -18,6 +30,7 @@ class RSS_Item: self.FULL_PATH = path self.TITLE = path.rsplit('.', 1)[0].split('/')[-1] + self.FILE_TYPE = path.rsplit('.', 1)[1] self.DESCRIPTION = self.parse_file() self.LAST_UPDATE = self.file_last_modified() self.URI = self.get_uri() @@ -35,7 +48,8 @@ class RSS_Item: """ parse_file - reads the file at FULL_PATH and saves the content from when the first

tag is hit up to and including the - closing

tag. Expects an HTML style file + closing

tag. Other files are interpreted as text files + and, just reads the first 3 paragraphs (two new lines in a row) """ with open(self.FULL_PATH) as f: in_body = False @@ -45,13 +59,19 @@ class RSS_Item: if paragraphs >= self.PARAGRAPHS: break line = line.strip() - if line.startswith("

"): - in_body = True - if in_body: + if self.FILE_TYPE in ['html', 'html!']: + if line.startswith("

"): + in_body = True + if in_body: + description += line + if line.endswith("

"): + in_body = False + paragraphs += 1 + else: description += line - if line.endswith("

"): - in_body = False - paragraphs += 1 + # remember, we stripped the line + if line == '': + paragraphs += 1 return ''.join(description) @@ -59,16 +79,25 @@ class RSS_Item: return ctime(os.stat(self.FULL_PATH).st_ctime) def get_uri(self): + # return everything after "./templates/" return '/'.join(self.FULL_PATH.split('/')[2:]) def get_rss_channel(): + """ + get_rss_channel - list all files from the BASE_DIR, and if allowed, + add them as RSS_Items to populate feed.xml. Called by feed.xml view + """ items = [] + extensions = siteconfig.rss_channel_config['RSS_FILE_EXT'] for root, dirs, files in os.walk(siteconfig.BASE_DIR): for f in files: + # remember, path will be like "./templates/site/..." path = os.path.join(root, f) if ( - path.endswith(".html") or f.endswith(".html!") - ) and path not in siteconfig.RSS_OMIT: + path.split(".")[-1] in extensions + and path not in siteconfig.RSS_OMIT + and not is_hidden_path(path.split('.', 1)[1]) + ): items.append(RSS_Item(path)) return items -- cgit v1.2.3