Tweak footer e layout categorie, fix metapagine, aggiornamento traduttore

Author: octospacc

Scripts/Translate/Main.py

@@ -1,15 +1,17 @@
 #!/usr/bin/env python3
+ModificationMetadataKey = "lastmod"
+KeepOriginalMetadata = ["draft", "date", "lastmod"]
 DestinationLanguages = ["it", "en", "es", "fr"] # "de", "eo"
 IncludePaths = ["/"]
-ExcludePaths = ["/categories", "/note/2024-09-19-Raspberry-Output-Audio-Both.md", "/miscellanea/Devlogs.md"] # "/miscellanea/PicoBlog.md"
+ExcludePaths = ["/note/2024-09-19-Raspberry-Output-Audio-Both.md", "/miscellanea/Devlogs.md"] # "/miscellanea/PicoBlog.md"
 
 import subprocess
 from os import getcwd, listdir
 from os.path import dirname, realpath, isfile, isdir
 from pathlib import Path
 from translate_shell.translate import translate
 
-# TODO somehow handle overriding frontmatter data for some translation (title, slug, ...)
+# TODO somehow handle overriding frontmatter data for some translation (title, slug, ...) (via in other files or commented metadata lines?)
 # TODO handle deleted files? (it should probably be done in another sh script, not here)
 
 def printf(*objects):
@@ -26,11 +28,11 @@ def make_destination_path(document_path, destination_language):
 		+ '/'.join(document_path.split('/')[1:]))
 
 def is_translation_uptodate(source_path, destination_path):
-	original_lines = split_text_with_frontmatter(read_original_document(source_path))[1].splitlines()
-	translated_lines = split_text_with_frontmatter(open(destination_path, 'r').read())[1].splitlines()
+	original_lines = split_with_frontmatter(read_original_document(source_path))[1].splitlines()
+	translated_lines = split_with_frontmatter(open(destination_path, 'r').read())[1].splitlines()
 	for [index, original_line] in enumerate(original_lines):
 		line_key = original_line.split('=')[0]
-		if line_key.strip().lower() == "lastmod":
+		if line_key.strip().lower() == ModificationMetadataKey:
 			if original_line != translated_lines[index]:
 				return False
 			break
@@ -40,7 +42,7 @@ def is_translation_uptodate(source_path, destination_path):
 def needs_translation(source_path, destination_language=None):
 	for exclude_path in ExcludePaths:
 		document_path = ('/' + '/'.join(source_path.split('/')[1:]))
-		if (document_path == exclude_path) or document_path.startswith(exclude_path + '/'):
+		if (document_path == exclude_path) or document_path.startswith(exclude_path.rstrip('/') + '/'):
 			return False
 	if not read_original_document(source_path).strip():
 		return False
@@ -63,7 +65,7 @@ def find_documents(folder_path):
 					documents[document].append(destination_language)
 	return documents
 
-def split_text_with_frontmatter(document_text):
+def split_with_frontmatter(document_text):
 	text_header = document_text.strip().splitlines()[0].strip()
 	if text_header in ["---", "+++"]:
 		text_tokens = document_text.split(text_header)
@@ -76,7 +78,7 @@ def fix_frontmatter(translated_text, reference_text):
 		if translated_line.strip() and (translated_line.lstrip() == translated_line):
 			reference_line = reference_lines[index]
 			line_key = reference_line.split('=')[0]
-			if line_key.strip().lower() in ["draft", "date", "lastmod"]:
+			if line_key.strip().lower() in KeepOriginalMetadata:
 				translated_line = reference_line
 			else:
 				line_value = '='.join(translated_line.split('=')[1:])
@@ -86,11 +88,50 @@ def fix_frontmatter(translated_text, reference_text):
 		result += (translated_line + '\n')
 	return result
 
+# <https://stackoverflow.com/a/18815890>
+def ascii_to_number(text:str) -> int:
+	return int(''.join(format(ord(i), 'b').zfill(8) for i in text), 2)
+
+# <https://stackoverflow.com/a/699891>, <https://stackoverflow.com/a/40559005>
+def number_to_ascii(number:int) -> str:
+	binary = format(int(number), '016b')
+	binary = binary.zfill(len(binary) + (8 - (len(binary) % 8)))
+	return ''.join(chr(int(binary[(i * 8):((i * 8) + 8)], 2)) for i in range(len(binary) // 8))
+
+# TODO add checks for number-strings to ensure they aren't already in the literal text
+# TODO handle code blocks and .notranslate HTML elements
+# TODO fix strange bugs
+def wrap_for_translation(original_text):
+	#return original_text
+	original_text = (original_text
+		.replace("{{%", "{{@%").replace("%}}", "%@}}")
+		.replace("{{<", "{{@<").replace(">}}", ">@}}"))
+	original_tokens = original_text.split("{{@")
+	#[(("{{@" if i else '') + c) for [i, c] in enumerate(original_text.split("{{@"))]
+	for i in range(1, len(original_tokens)):
+		token_tokens = original_tokens[i].split("@}}")
+		token_tokens[0] = (f"{TranslationMagic}__" + str(ascii_to_number("{{@" + token_tokens[0] + "@}}")) + "__").replace("9", "9_")
+		original_tokens[i] = ''.join(token_tokens)
+	#print(unwrap_from_translation(''.join(original_tokens)))
+	#exit(1)
+	return ''.join(original_tokens)
+
+def unwrap_from_translation(translated_text):
+	#return translated_text
+	translated_tokens = translated_text.split(f"{TranslationMagic}__")
+	for i in range(1, len(translated_tokens)):
+		token_tokens = translated_tokens[i].split("__")
+		token_tokens[0] = number_to_ascii(token_tokens[0].replace(' ', '').replace('_', ''))
+		translated_tokens[i] = (token_tokens[0] + "__".join(token_tokens[1:]))
+	return (''.join(translated_tokens)
+		.replace("{{@%", "{{%").replace("%@}}", "%}}")
+		.replace("{{@<", "{{<").replace(">@}}", ">}}"))
+
 def translate_document(document_path, documents):
 	printf(f"* {document_path} ->")
 	for destination_language in documents[document_path]:
 		source_language = get_source_language(document_path)
-		original_text = read_original_document(document_path)
+		original_text = wrap_for_translation(read_original_document(document_path))
 		printf('', destination_language)
 		try:
 			is_python_translator = True
@@ -101,10 +142,13 @@ def translate_document(document_path, documents):
 			printf('❌', exception)
 			try:
 				is_python_translator = False
+				temporary_path = ("./tmp/" + document_path)
+				Path('/'.join(temporary_path.split('/')[:-1])).mkdir(parents=True, exist_ok=True)
+				open(temporary_path, 'w').write(original_text)
 				translated = subprocess.run(
 					("bash", "../Scripts/Lib/translate-shell.bash", "-brief", "-no-autocorrect",
 						"-t", destination_language, "-s", source_language,
-						("file://" + "../content/" + document_path)),
+						("file://" + temporary_path)),
 					stdout=subprocess.PIPE,
 					stderr=subprocess.PIPE)
 				if translated.stderr:
@@ -116,7 +160,7 @@ def translate_document(document_path, documents):
 		translated_text = (translated.results[0].paraphrase
 			if is_python_translator else translated.stdout.decode())
 		translated_preamble = ("\n\n{{< noticeAutomaticTranslation " + source_language + " >}}\n\n")
-		if (translated_tokens := split_text_with_frontmatter(translated_text)):
+		if (translated_tokens := split_with_frontmatter(translated_text)):
 			translated_tokens[1] = fix_frontmatter(translated_tokens[1], original_text.split(translated_tokens[0])[1])
 			if translated_tokens[3].strip():
 				translated_tokens.insert(3, translated_preamble)
@@ -125,7 +169,7 @@ def translate_document(document_path, documents):
 			translated_text = (translated_preamble + translated_text)
 		destination_path = make_destination_path(document_path, destination_language)
 		Path('/'.join(destination_path.split('/')[:-1])).mkdir(parents=True, exist_ok=True)
-		open(destination_path, 'w').write(translated_text)
+		open(destination_path, 'w').write(unwrap_from_translation(translated_text))
 	printf('\n')
 
 def main():
@@ -139,6 +183,8 @@ def main():
 def read_from_scripts(relative_path:str):
 	return open((dirname(realpath(__file__)) + '/../' + relative_path), 'r').read()
 
+TranslationMagic = ("__" + str(ascii_to_number("sitoctt")))
+
 if __name__ == "__main__":
 	globals_text = read_from_scripts('Lib/Globals.sh')
 	exec(globals_text.split('#' + globals_text.splitlines()[0].split('#!')[1] + '!')[0])

assets/ButtonBadges.toml

@@ -74,6 +74,12 @@ file = "Follow-me-on-mastodon.png"
 href = "https://mastodon.uno/@octo"
 rel = "me"
 
+[[20-me]]
+alt = "Visita la OctoSpacc Hub"
+file = "Sites/octospacc-hub-1.png"
+href = "https://hub.octt.eu.org"
+rel = "me"
+
 [[20-me]]
 alt = "Visita il mio Fritto Misto di OctoSpacc"
 file = "Sites/fritto-misto-di-octospacc-1.png"

content/it/_index.md

@@ -1,34 +1,16 @@
 +++
 Title = "il ✨sitoctt✨ — Home"
 Description = "Octt sono io... e questo è letteralmente il mio sito (uwu)."
-Lastmod = 2024-08-27
+Lastmod = 2024-09-21
 +++
 
 Una volta qui era tutta campagn— ehm, volevo dire... era tutta una lista di parole ammassate alla male e peggio. Non solo non piaceva più nemmeno a me, ma non è proprio buona creanza nell'anno del Signore 2024 (e oltre), quindi...
 
 Ora c'è questa lista di tutti i miei ultimissimi articoli, e pagine sfuse che ho aggiornato di recente, mentre la vecchia home è a ["Sul sitoctt"]({{< relref "/miscellanea/Sul-sitoctt/" >}}) finché non avrò riorganizzato tutto. Benvenut<code class="notranslate" data-lang="it"><!--
 --><span class="BlinkA">a</span><!--
---><span class="dn">/</span><!--
+--><span class="Blink_">/</span><!--
 --><span class="BlinkO">o</span></code> nel ✨sitoctt✨!
 
-(Lavori in corso per ancora qualche giorno, quindi forse link rotti e pagine sformate, mi dispiace!!!)
+<!-- (Lavori in corso per ancora qualche giorno, quindi forse link rotti e pagine sformate, mi dispiace!!!) -->
 
-<style>
-/* Animazioni per le desinenze */
-.BlinkA {
-	Animation: BlinkerA 0.25s Step-Start Infinite;
-}
-@Keyframes BlinkerA {
-	0% {Position: Absolute; Visibility: Hidden;}
-	50% {Position: Static; Visibility: Visible;}
-	100% {Position: Absolute; Visibility: Hidden;}
-}
-.BlinkO {
-	Animation: BlinkerO 0.25s Step-Start Infinite;
-}
-@Keyframes BlinkerO {
-	0% {Position: Static; Visibility: Visible;}
-	50% {Position: Absolute; Visibility: Hidden;}
-	100% {Position: Static; Visibility: Visible;}
-}
-</style>
+<link rel="stylesheet" href="/desinenze-blink.css"/>

content/it/blog/_index.md

@@ -1,5 +1,5 @@
 +++
-Title = "Blog"
+Title = "Blog📚️"
 Aliases = [
   "/Posts/index.html",
   "/Categories/Blog.html",

content/it/categories/MicroBlog/_index.md

@@ -2,9 +2,9 @@
 Title = "📒 Vecchi MicroBlog"
 #Index = True
 #URLs = MicroBlog.html Diarylog.html
-Categories = [ "Blog" ]
+#Categories = [ "Blog" ]
 +++
 
-Come annunciato nell'articolo "[🎇 Il resocontoctt di questo 2023, almeno in termini di posting!](../Posts/2023-12-31-Resocontoctt-2023.html#-Il-MicroBlog-nuovissimo-alla-fa)", il Vecchio MicroBlog è ora deprecato, e non verrà più aggiornato. Quello che segue è l'archivio dei vecchi contenuti, che rimarrà leggibile. I nuovi post verranno quindi pubblicati sul Nuovo MicroBlog unificato, raggiungibile dal menu del sito.
+Come annunciato nell'articolo "[🎇 Il resocontoctt di questo 2023, almeno in termini di posting!]({{< relref "/blog/2023-12-31-Resocontoctt-2023/#il-microblog-nuovissimo-alla-faccia-du-rove" >}})", il Vecchio MicroBlog è ora deprecato, e non verrà più aggiornato. Quello che segue è l'archivio dei vecchi contenuti, che rimarrà leggibile. I nuovi post verranno quindi pubblicati sul Nuovo MicroBlog unificato, raggiungibile dal menu del sito.
 
 _Nota: I contenuti negli archivi possono aver subito redazioni non segnalate._