XML feeds without libs; Check system tools; Provide software version

2023-06-06 15:52:31 +02:00 · 2023-06-06 15:52:31 +02:00 · 8e892e096d
parent 1e07f92452
commit 8e892e096d
6 changed files with 160 additions and 51 deletions
--- a/App/Source/Build.py
+++ b/App/Source/Build.py
@ -179,7 +179,7 @@ def BuildMain(Args, FeedEntries):
 	#	os.chdir(Args.InputDir)
 	#	print(f"[I] Current directory: {Args.InputDir}")

-	SiteName = Flags['SiteName'] = OptChoose('', Args.SiteName, ReadConf(SiteConf, 'Site', 'Name'))
+	SiteName = Flags['SiteName'] = DefConfOptChoose('SiteName', Args.SiteName, ReadConf(SiteConf, 'Site', 'Name'))
 	if SiteName:
 		logging.info(f"Compiling: {SiteName}")

@ -246,6 +246,12 @@ def BuildMain(Args, FeedEntries):
 	SiteDomain = Flags['SiteDomain'] = SiteDomain.removesuffix('/')
 	Locale = LoadLocale(SiteLang)

+	if not InSystemPath('pug'):
+		logging.warning("⚠ `pug` not found in system PATH. If you have any .pug pages to be compiled, the program will fail.")
+	if Flags['GemtextOutput'] and not InSystemPath('html2gmi'):
+		logging.warning("⚠ `html2gmi` not found in system PATH. Gemtext generation will be disabled.")
+		Flags['GemtextOutput'] = False
+
 	if DiffBuild:
 		logging.info("Build mode: Differential")
 		LimitFiles = GetModifiedFiles(OutDir)
@ -347,13 +353,13 @@ if __name__ == '__main__':

 	ConfigLogging(Args.Logging)

-	try:
-		import lxml
-		from Modules.Feed import *
-		FeedEntries = Args.FeedEntries if Args.FeedEntries else 'Default'
-	except:
-		logging.warning("⚠ Can't load the XML libraries. XML Feeds Generation is Disabled. Make sure the 'lxml' library is installed.")
-		FeedEntries = 0
+	#try:
+	#	import lxml
+	from Modules.Feed import *
+	FeedEntries = Args.FeedEntries if Args.FeedEntries else 'Default'
+	#except:
+	#	logging.warning("⚠ Can't load the XML libraries. XML Feeds Generation is Disabled. Make sure the 'lxml' library is installed.")
+	#	FeedEntries = 0

 	BuildMain(Args=Args, FeedEntries=FeedEntries)
 	logging.info(f"✅ Done! ({round(time.time()-StartTime, 3)}s)")
--- a/App/Source/Modules/Config.py
+++ b/App/Source/Modules/Config.py
@ -15,6 +15,7 @@ DefConf = {
 	"Threads": 0,
 	"DiffBuild": False,
 	"OutDir": "public",
+	"SiteName": "Untitled Site",
 	"SiteLang": "en",
 	"SiteTemplate": "Default.html",
 	"ActivityPubTypeFilter": "Post",
@ -22,7 +23,7 @@ DefConf = {
 	"CategoriesUncategorized": "Uncategorized",
 	"FeedCategoryFilter": "Blog",
 	"FeedEntries": 10,
-	"JournalRedirect": False
+	"JournalRedirect": False,
 }

 def LoadConfFile(File):
--- a/App/Source/Modules/Elements.py
+++ b/App/Source/Modules/Elements.py
@ -29,18 +29,19 @@ CategoryPageTemplate = """\

 <div><staticoso:Category:{Name}></div>
 """
-RedirectPageTemplate = """\
+RedirectPageTemplate = f"""\
 <!DOCTYPE html>
 <html>
 <head>
 	<meta charset="UTF-8"/>
 	<meta name="viewport" content="width=device-width, initial-scale=1.0"/>
-	<title>{TitlePrefix}Redirect</title>
-	<link rel="canonical" href="{SiteDomain}/{DestURL}"/>
-	<meta http-equiv="refresh" content="0; url='{DestURL}'"/>
+	<title>{{TitlePrefix}}Redirect</title>
+	<link rel="canonical" href="{{SiteDomain}}/{{DestURL}}"/>
+	<meta http-equiv="refresh" content="0; url='{{DestURL}}'"/>
+	<meta name="generator" content="{staticosoNameVersion()}"/>
 </head>
 <body>
-	<p><a href="{DestURL}">{StrClick}</a> {StrRedirect}.</p>
+	<p><a href="{{DestURL}}">{{StrClick}}</a> {{StrRedirect}}.</p>
 </body>
 </html>
 """
--- a/App/Source/Modules/Feed.py
+++ b/App/Source/Modules/Feed.py
@ -7,56 +7,153 @@
 |   Copyright (C) 2022-2023, OctoSpacc |
 | ================================== """

-# TODO: Either switch feed generation lib, or rewrite the 'lxml' module, so that no modules have to be compiled and the program is 100% portable
-
-from Libs.feedgen.feed import FeedGenerator
+from datetime import datetime
 from Modules.Utils import *

-def MakeFeed(Flags:dict, Pages:list, FullSite=False):
+FeedGenerator = None
+try:
+	import lxml
+	from Libs.feedgen.feed import FeedGenerator
+except:
+	logging.warning("⚠ Can't load the native XML libraries. XML Feeds Generation will use the interpreted module.")
+
+def MakeFeed(Flags:dict, Pages:list, FullSite:bool=False):
 	f = NameSpace(Flags)
 	CategoryFilter = Flags['FeedCategoryFilter']
 	MaxEntries = Flags['FeedEntries']

-	Feed = FeedGenerator()
-	Link = Flags['SiteDomain'] if Flags['SiteDomain'] else ' '
-	Feed.id(Link)
-	Feed.title(Flags['SiteName'] if Flags['SiteName'] else 'Untitled Site')
-	Feed.link(href=Link, rel='alternate')
-	Feed.description(Flags['SiteTagline'] if Flags['SiteTagline'] else ' ')
-	if Flags['SiteDomain']:
-		Feed.logo(Flags['SiteDomain'] + '/favicon.png')
-	Feed.language(Flags['SiteLang'])
+	if FeedGenerator:
+		Feed = FeedGenerator()
+		Link = Flags['SiteDomain'] if Flags['SiteDomain'] else ' '
+		Feed.id(Link)
+		Feed.link(href=Link, rel='alternate')
+		Feed.title(Flags['SiteName'])
+		Feed.description(Flags['SiteTagline'] if Flags['SiteTagline'] else ' ')
+		if Flags['SiteDomain']:
+			Feed.logo(f'{Flags["SiteDomain"]}/favicon.png')
+		Feed.language(Flags['SiteLang'])
+	else:
+		FeedData = {
+			'Link': Flags['SiteDomain'],
+			'Title': Flags['SiteName'],
+			'Description': Flags['SiteTagline'],
+			'Language': Flags['SiteLang'],
+			'Entries': [],
+		}

 	DoPages = []
 	for e in Pages:
-		if FullSite or (not FullSite and MaxEntries != 0 and e[3]['Type'] == 'Post' and e[3]['Feed'] == 'True'): # No entry limit if site feed
+		# No entry limit if site feed
+		if FullSite or (not FullSite and MaxEntries != 0 and e[3]['Type'] == 'Post' and e[3]['Feed'] == 'True'):
 			DoPages += [e]
 			MaxEntries -= 1
 	DoPages.reverse()

 	for File, Content, Titles, Meta, ContentHTML, SlimHTML, Description, Image in DoPages:
 		if FullSite or (not FullSite and Meta['Type'] == 'Post' and (not CategoryFilter or (CategoryFilter and (CategoryFilter in Meta['Categories'] or CategoryFilter == '*')))):
-			Entry = Feed.add_entry()
-			FileName = File.split('/')[-1]
-			File = f"{StripExt(File)}.html"
-			Content = ReadFile(f"{Flags['OutDir']}/{File}")
-			Link = Flags['SiteDomain'] + '/' + File if Flags['SiteDomain'] else ' '
+			File = f'{StripExt(File)}.html'
+			Link = Flags['SiteDomain'] + '/' + File if Flags['SiteDomain'] else File
+			Title = Meta['Title'] if Meta['Title'] else Titles[0].lstrip('#') if Titles else File.split('/')[-1]
+			Title = Title.lstrip().rstrip()
 			CreatedOn = GetFullDate(Meta['CreatedOn'])
 			EditedOn = GetFullDate(Meta['EditedOn'])
-			Entry.id(Link)
-			Title = Meta['Title'] if Meta['Title'] else Titles[0].lstrip('#') if Titles else FileName
-			Entry.title(Title.lstrip().rstrip())
-			Entry.description(Description)
-			Entry.link(href=Link, rel='alternate')
-			if not FullSite: # Avoid making an enormous site feed file...
-				Entry.content(ContentHTML, type='html')
-			if CreatedOn:
-				Entry.pubDate(CreatedOn)
-			EditedOn = EditedOn if EditedOn else CreatedOn if CreatedOn and not EditedOn else '1970-01-01T00:00+00:00'
-			Entry.updated(EditedOn)
+			if FullSite: # Avoid making an enormous site feed file...
+				ContentHTML = ''
+			if FeedGenerator:
+				Entry = Feed.add_entry()
+				Entry.id(Link)
+				Entry.link(href=Link, rel='alternate')
+				Entry.title(Title)
+				Entry.description(Description)
+				if ContentHTML: 
+					Entry.content(ContentHTML, type='html')
+				if CreatedOn:
+					Entry.pubDate(CreatedOn)
+				EditedOn = EditedOn if EditedOn else CreatedOn if CreatedOn and not EditedOn else '1970-01-01T00:00+00:00'
+				Entry.updated(EditedOn)
+			else:
+				FeedData['Entries'] += [{
+					'Link': Link,
+					'Title': Title,
+					'Description': Description,
+					'Content': ContentHTML,
+					'PublishedOn': CreatedOn,
+					'UpdatedOn': EditedOn,
+				}]

 	if not os.path.exists(f"{Flags['OutDir']}/feed"):
 		os.mkdir(f"{Flags['OutDir']}/feed")
 	FeedType = 'site.' if FullSite else ''
-	Feed.atom_file(f"{Flags['OutDir']}/feed/{FeedType}atom.xml", pretty=(not Flags['MinifyOutput']))
-	Feed.rss_file(f"{Flags['OutDir']}/feed/{FeedType}rss.xml", pretty=(not Flags['MinifyOutput']))
+	if FeedGenerator:
+		Feed.atom_file(f"{Flags['OutDir']}/feed/{FeedType}atom.xml", pretty=(not Flags['MinifyOutput']))
+		Feed.rss_file(f"{Flags['OutDir']}/feed/{FeedType}rss.xml", pretty=(not Flags['MinifyOutput']))
+	else:
+		Feeds = PyFeedGenerator(FeedData)
+		for Format in ('atom', 'rss'):
+			WriteFile(f"{Flags['OutDir']}/feed/{FeedType}{Format}.xml", Feeds[Format])
+
+def PyFeedGenerator(Data:dict, Format:bool=None):
+	XmlEntries = {'atom': '', 'rss': ''}
+	XmlExtra, AtomExtra, RssExtra = '', '', ''
+	XmlHeader = '<?xml version="1.0" encoding="UTF-8"?>'
+	XmlLang = f'xml:lang="{Data["Language"]}"'
+	XmlTitle = f'<title>{Data["Title"]}</title>'
+	XmlExtra += XmlTitle
+	if Data['Description']:
+		AtomExtra += f'<subtitle>{Data["Description"]}</subtitle>'
+		RssExtra += f'<description>{Data["Description"]}</description>'
+	if Data['Link']:
+		IconUrl = f'{Data["Link"]}/favicon.png'
+		AtomExtra += f'<id>{Data["Link"]}</id><link href="{Data["Link"]}"/><logo>{IconUrl}</logo>'
+		RssExtra += f'<link>{Data["Link"]}</link><image>{XmlTitle}<url>{IconUrl}</url><link>{Data["Link"]}</link></image>'
+	Entries = Data['Entries'] if Data['Entries'] else ()
+	for Entry in Data['Entries']:
+		XmlEntryExtra, AtomEntryExtra, RssEntryExtra = '', '', ''
+		XmlEntryExtra += f'<title>{Entry["Title"]}</title>'
+		if Entry['Description']:
+			RssEntryExtra += f'<description>{Entry["Description"]}</description>'
+		if Entry['Content']:
+			AtomEntryExtra += f'<content type="html">{Entry["Content"]}</content>'
+			RssEntryExtra += f'<content:encoded>{Entry["Content"]}</content:encoded>'
+		if Entry['PublishedOn']:
+			AtomEntryExtra += f'<published>{Entry["PublishedOn"]}</published>'
+			RssEntryExtra += f'<pubDate>{Entry["PublishedOn"]}</pubDate>'
+		if Entry['UpdatedOn']:
+			AtomEntryExtra += f'<updated>{Entry["UpdatedOn"]}</updated>'
+		XmlEntries['atom'] += f'''
+<entry>
+	<id>{Entry['Link']}</id>
+	<link href="{Entry['Link']}"/>
+	{XmlEntryExtra}
+	{AtomEntryExtra}
+</entry>
+		'''
+		XmlEntries['rss'] += f'''
+<item>
+	<guid>{Entry['Link']}</guid>
+	<link>{Entry['Link']}</link>
+	{XmlEntryExtra}
+	{RssEntryExtra}
+</item>
+		'''
+	Feeds = {'atom': f'''{XmlHeader}
+<feed xmlns="http://www.w3.org/2005/Atom" {XmlLang}>
+	{XmlExtra}
+	{AtomExtra}
+	<updated>{datetime.now()}</updated>
+	<generator uri="https://gitlab.com/octtspacc/staticoso" version="{staticosoNameVersion().split(" ")[1]}">staticoso</generator>
+	{XmlEntries['atom']}
+</feed>''', 'rss': f'''{XmlHeader}
+<rss xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/" version="2.0" {XmlLang}>
+	<channel>
+		{XmlExtra}
+		{RssExtra}
+		<language>{Data["Language"]}</language>
+		<lastBuildDate>{datetime.now()}</lastBuildDate>
+		<generator>{staticosoNameVersion()}</generator>
+		{XmlEntries['rss']}
+	</channel>
+</rss>'''}
+	if Format:
+		Feeds = Feeds[Format.lower()]
+	return Feeds
--- a/App/Source/Modules/Utils.py
+++ b/App/Source/Modules/Utils.py
@ -7,8 +7,7 @@
 |   Copyright (C) 2022-2023, OctoSpacc |
 | ================================== """

-import json
-import os
+import json, os, subprocess
 from datetime import datetime
 from multiprocessing import Pool, cpu_count
 from pathlib import Path
@ -22,6 +21,14 @@ def SureList(e):
 def staticosoBaseDir():
 	return f"{os.path.dirname(os.path.abspath(__file__))}/../../"

+def staticosoNameVersion():
+	Version = subprocess.run(('sh', '-c', 'git log | head -n 1'), stdout=subprocess.PIPE).stdout.decode().strip()
+	Version = ' v.' + Version.split(' ')[1][:8] if Version else ''
+	return f'staticoso{Version}'
+
+def InSystemPath(Exec:str):
+	return subprocess.run(('sh', '-c', f'which {Exec}'), stdout=subprocess.PIPE).stdout.decode().strip()
+
 def ReadFile(p:str, m:str='r'):
 	try:
 		with open(p, m) as f:
--- a/App/TODO
+++ b/App/TODO
@ -2,7 +2,6 @@
 - .html input pages bug: // metadata lines not being removed from final file after parsing
 - Multi-line metadata flags
 - Category-based feeds
- Meta tag generator showing software version (git commit hash)
 - Customize date format
 - Misskey for ActivityPub
 - Section marking in pages? (for use with external translators)
@ -25,7 +24,6 @@
 - Support for HTML comment lines (<!-- -->) in any format
 - Support for Wikitext, rST, AsciiDoc (?)
 - Posts in draft state (will not be compiled) (?)
- Check if external tools (pug-cli, html2gmi) are installed
 - Static code syntax highlighing
 - Override internal HTML snippets (meta lines, page lists, redirects, ...) with config file in Templates/NAME.ini
 - Specify input folder(s)
@ -39,7 +37,6 @@
 - Change FolderRoots arg name to CustomPaths
 - Accept Macros as CLI arguments + Deprecate FolderRoots (Macros make it redundant)
 - Fix ordering menu in Site.ini (not working for inner pages)
- Feed generation optionally without native libraries
 - JSON feeds
 - Full XML sitemap
 - SCSS support