volumetric-regimes-book/command-line/update.py

import urllib.request
import os
import re
import json
import jinja2

# Notes are here: https://pad.vvvvvvaria.org/volumetric-regimes-in-process.

STATIC_FOLDER_PATH = '.' # without trailing slash
PUBLIC_STATIC_FOLDER_PATH = '.' # without trailing slash
TEMPLATES_DIR = './templates'

# This uses a low quality copy of all the images 
# (using a folder with the name "images-small",
# which stores a copy of all the images generated with:
# $ mogrify -quality 5% -adaptive-resize 25% -remap pattern:gray50 * )
fast = False

def API_request(url, pagename):
	"""
		url = API request url (string)
		data =  { 'query': 
					'pages' : 
						pageid : { 					
							'links' : {
								'?' : '?'
								'title' : 'pagename'
							}
						} 
					}  
				}
	"""
	response = urllib.request.urlopen(url).read()
	data = json.loads(response)

	# Save response as JSON to be able to inspect API call
	json_file = f'{ STATIC_FOLDER_PATH }/{ pagename }.json'
	print('Saving JSON:', json_file)
	with open(json_file, 'w') as out:
		out.write(json.dumps(data, indent=4))
		out.close()

	return data

def download_media(html, images, wiki):
	"""
		html = string (HTML)
		images = list of filenames (str)
	"""
	# check if 'images/' already exists
	if not os.path.exists(f'{ STATIC_FOLDER_PATH }/images'):
		os.makedirs(f'{ STATIC_FOLDER_PATH }/images')

	# tmp list for filename replacements 
	replaced = []
	
	images.sort()
	images.reverse() # reverse to make sure that 01.png does not override Image01.png in the filename replacements later

	# download media files
	for filename in images:
		filename = filename.replace(' ', '_') # safe filenames

		# check if the image is already downloaded
		# if not, then download the file
		if not os.path.isfile(f'{ STATIC_FOLDER_PATH }/images/{ filename }'):

			# first we search for the full filename of the image
			url = f'{ wiki }/api.php?action=query&list=allimages&aifrom={ filename }&format=json'
			response = urllib.request.urlopen(url).read()
			data = json.loads(response)

			# we select the first search result
			# (assuming that this is the image we are looking for)
			image = data['query']['allimages'][0] 

			# then we download the image
			image_url = image['url']
			image_filename = image['name']
			print('Downloading:', image_filename)
			image_response = urllib.request.urlopen(image_url).read()

			# and we save it as a file
			image_path = f'{ STATIC_FOLDER_PATH }/images/{ image_filename }'
			out = open(image_path, 'wb') 
			out.write(image_response)
			out.close()

			import time
			time.sleep(3) # do not overload the server

		# replace src image link (from wiki folder structure to local folder)
		image_path = f'{ PUBLIC_STATIC_FOLDER_PATH }/images/{ filename }' # here the images need to link to the / of the domain, for flask :/// confusing! this breaks the whole idea to still be able to make a local copy of the file
		
		img_path_patterns = [rf'(?<!\.)/images/.*?px-{ filename }', rf'(?<!\.)/images/.*?{ filename }']
		for img_path_pattern in img_path_patterns:
			matches = re.findall(img_path_pattern, html) # for debugging
			if matches:
				for match in matches:
					if match not in replaced:
						# print(f'    { match } --> { image_path }') # for debugging: each image should have the correct match!
						html = html.replace(match, image_path)
						replaced.append(match)

	return html

def add_item_inventory_links(html):
	"""
		html = string (HTML)
	"""
	# THROUGHOUT THE BOOK
	# Find all references in the text to the item index
	matches = re.findall(r'\w.*?Item \d\d\d.*?\w\w\w', html) # Dodgy attempt to find unique patterns for each mentioning of Item ###
	index = {}
	for match in matches:
		item_match = re.search(r'Item \d\d\d', match).group()
		number = item_match.replace('Item ', '').strip()
		text_before = re.search(rf'\w.*?Item { number }', match).group().replace(f'Item { number }', '')
		text_after = re.search(rf'Item { number }.*?\w\w\w', match).group().replace(f'Item { number }', '')
		if not number in index:
			index[number] = []
			count = 1
		else:
			count = index[number][-1] + 1
		index[number].append(count)
		item_id = f'ii-{ number }-{ index[number][-1] }'
		# print(f'match: { number } --> { item_id } --> { match }')
		html = html.replace(match, f'{ text_before }<a id="{ item_id }" href="#Item_Index">Item { number }</a>{ text_after }')

	# IN THE ITEM INDEX
	# Also add a <span> around the index nr to style it
	matches = re.findall(r'<li>\d\d\d', html)
	for match in matches:
		html = html.replace(match, f'<li><span class="item_nr">{ match }</span>')

	print("\n-------------\n")
	print("The following items ('###') appear [#, #, ...] many times in the book:\n")
	sorted_index = dict(sorted(index.items()))
	print(sorted_index)
	print("\n-------------\n")
	
	return html

def tweaking(html):
	"""
		html = string (HTML)
	"""
	html = html.replace('<a href="#X,_y,_z_(4_filmstills)"', '<a href="#x,_y,_z_(4_filmstills)"') # change the anchor link in the TOC to lowercase
	html = html.replace('<a href="#Rehearsal_as_the_%E2%80%98Other%E2%80%99_to_Hypercomputation"', '<a href="#Rehearsal_as_the_‘Other’_to_Hypercomputation"') # change the anchor link in the TOC to lowercase
	html = html.replace('<a href="#We_hardly_encounter_anything_that_didn%E2%80%99t_really_matter"', '<a href="#We_hardly_encounter_anything_that_didn’t_really_matter"') # change the anchor link in the TOC to lowercase
	html = re.sub(r'''<h3><span class="mw-headline" id="References.*?">References</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h3>
<ul>''', '''<h3 class="references"><span class="mw-headline" id="References">References</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h3>
<ul class="references">''', html) # add id="references" to h3 and ul, so the elements can be selected with CSS
	html = html.replace('src="./images/Userinfo.jpg"', 'src="./images/Userinfo.svg"') # This image is not on the wiki
	html = html.replace('src="./images/Topology-typography-1A.png"', 'src="./images/Topology-typography-1A.svg"') # This image is not on the wiki
	html = html.replace('src="./images/Topology-typography-1B.png"', 'src="./images/Topology-typography-1B.svg"') # This image is not on the wiki
	html = html.replace('src="./images/Topology-typography-2A.png"', 'src="./images/Topology-typography-2A.svg"') # This image is not on the wiki
	html = html.replace('src="./images/Topology-typography-2B.png"', 'src="./images/Topology-typography-2B.svg"') # This image is not on the wiki
	html = html.replace('trans*feminis', 'trans✶feminis') # changing stars
	html = html.replace('Trans*feminis', 'Trans✶feminis') # changing stars
	html = html.replace('star (*)', 'star (✶)') # changing stars
	html = html.replace('Our trans*feminist lens is sharpened by queer and anti-colonial sensibilities, and oriented towards (but not limited to) trans*generational, trans*media, trans*disciplinary, trans*geopolitical, trans*expertise, and trans*genealogical forms of study.', 'Our trans✶feminist lens is sharpened by queer and anti-colonial sensibilities, and oriented towards (but not limited to) trans✶generational, trans✶media, trans✶disciplinary, trans✶geopolitical, trans✶expertise, and trans✶genealogical forms of study.') # changing stars
	html = html.replace('<h2><span class="mw-headline" id="Invasive_imagination_and_its_agential_cuts">Invasive imagination and its agential cuts</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>', '<h2><span class="mw-headline" id="Invasive_imagination_and_its_agential_cuts">Invasive imagination <br>and its agential cuts</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>') 
	html = html.replace('<h2><span class="mw-headline" id="Volumetric_Regimes:_Material_cultures_of_quantified_presence">Volumetric Regimes: Material cultures of quantified presence</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>', '<h2><span class="mw-headline" id="Volumetric_Regimes:_Material_cultures_of_quantified_presence">Volumetric Regimes:<br>Material cultures of<br>quantified presence</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>') 
	html = html.replace('<h2><span id="Somatopologies_(materials_for_a_movie_in_the_making)"></span><span class="mw-headline" id="Somatopologies_.28materials_for_a_movie_in_the_making.29">Somatopologies (materials for a movie in the making)</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>', '<h2><span id="Somatopologies_(materials_for_a_movie_in_the_making)"></span><span class="mw-headline" id="Somatopologies_.28materials_for_a_movie_in_the_making.29">Somatopologies (materials<br> for a movie in the making)</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>') 
	html = html.replace('<h1><span class="mw-headline" id="Signs_of_clandestine_disorder:_The_continuous_aftermath_of_3D-computationalism"><a href="#Clandestine_disorder" title="Clandestine disorder">Signs of clandestine disorder: The continuous aftermath of 3D-computationalism</a></span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h1>', '<h1><span class="mw-headline" id="Signs_of_clandestine_disorder:_The_continuous_aftermath_of_3D-computationalism"><a href="#Clandestine_disorder" title="Clandestine disorder">Signs of clandestine disorder:<br>The continuous<br>aftermath of 3D-<br>computationalism</a></span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h1>') 
	html = html.replace('<h2><span class="mw-headline" id="The_Industrial_Continuum_of_3D">The Industrial Continuum of 3D</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>', '<h2><span class="mw-headline" id="The_Industrial_Continuum_of_3D">The Industrial Continuum<br>of 3D</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>') 
	html = html.replace('src="./images/Continuum_brighton.png"', 'src="./images/Continuum_brighton.svg"') # This image is not on the wiki
	html = html.replace('<h1><span class="mw-headline" id="Depths_and_Densities:_Accidented_and_dissonant_spacetimes"><a href="#Depths_and_densities" title="Depths and densities">Depths and Densities: Accidented and dissonant spacetimes</a></span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h1>', '<h1><span class="mw-headline" id="Depths_and_Densities:_Accidented_and_dissonant_spacetimes"><a href="#Depths_and_densities" title="Depths and densities">Depths and Densities:<br>Accidented<br>and dissonant<br>spacetimes</a></span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h1>') 
	html = html.replace('<h2><span class="mw-headline" id="Open_Boundary_Conditions:_a_grid_for_intensive_study">Open Boundary Conditions: a grid for intensive study</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>', '<h2><span class="mw-headline" id="Open_Boundary_Conditions:_a_grid_for_intensive_study">Open Boundary Conditions:<br>a grid for intensive study</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>') 
	html = html.replace('<h2><span class="mw-headline" id="Depths_and_Densities:_A_Bugged_Report">Depths and Densities: A Bugged Report</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>', '<h2><span class="mw-headline" id="Depths_and_Densities:_A_Bugged_Report">Depths and Densities:<br>A Bugged Report</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>') 
	html = html.replace('T*fRP', 'T✶fRP') 
	html = html.replace('trans*', 'trans✶') 
	html = html.replace('Trans*', 'trans✶') 
	html = html.replace('(*)', '(✶)') 
	html = html.replace('✶', '<span class="star">✶</span>') 
	html = html.replace('<p><a href="#File', '<p class="image"><a href="#File') # give <p>'s that contain an non-thumb image a .image class
	html = html.replace(' – ', '<span class="endash"> – </span>') # control the white spaces around an endash
	html = html.replace(' — ', '<span class="endash"> — </span>') # control the white spaces around an endash
	
	pattern1 = r'''<h2><span class="mw-headline" id=".*">.*</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>
<p><b>.*?</b>
</p>''' # title + author 
	pattern2 = r'''<h2><span class="mw-headline" id=".*?">.*?</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>
<h2 style="display:none;"><span class="mw-headline" id=".*?">.*?</span></h2>
<p><b>.*?</b>
</p>''' # exceptions: custom running headers
	pattern3 = r'''<h2><span class="mw-headline" id=".*?">.*?</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>''' # only title
	pattern4 = r'''<h2><span id="x,_y,_z_\(4_filmstills\)"></span><span class="mw-headline" id="x.2C_y.2C_z_.284_filmstills.29">x, y, z \(4 filmstills\)</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>
<p><b>Jara Rocha, Femke Snelting</b>
</p>''' 
	pattern5 = r'''<h2><span id="Somatopologies_\(materials_for_a_movie_in_the_making\)"></span><span class="mw-headline" id="Somatopologies_.28materials_for_a_movie_in_the_making.29">Somatopologies \(materials<br> for a movie in the making\)</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>
<p><b>Possible Bodies \(Jara Rocha, Femke Snelting\)</b>
</p>''' 
	pattern6 = r'''<h2><span id="Rehearsal_as_the_\‘Other\’_to_Hypercomputation"></span><span class="mw-headline" id="Rehearsal_as_the_\.E2\.80\.98Other\.E2\.80\.99_to_Hypercomputation">Rehearsal as the \‘Other\’ to Hypercomputation</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>
<p><b>Maria Dada</b>
</p>''' 
	pattern7 = r'''<h2><span id="We_hardly_encounter_anything_that_didn’t_really_matter"></span><span class="mw-headline" id="We_hardly_encounter_anything_that_didn\.E2\.80\.99t_really_matter">We hardly encounter anything that didn’t really matter</span><span class="mw-editsection"><span class="mw-editsection-bracket"></span></span></h2>
<p><b>Phil Langley in conversation with Possible Bodies</b>
</p>'''

	results = re.findall(rf'{pattern1}|{pattern2}|{pattern3}|{pattern4}|{pattern5}|{pattern6}|{pattern7}', html)
	for match in results:
		html = html.replace(match, f'<div class="title-wrapper">{ match }</div>')

	return html

def clean_up(html):
	"""
		html = string (HTML)
	"""
	html = re.sub(r'\[.*edit.*\]', '', html) # remove the [edit]
	html = re.sub(r'href="/index.php\?title=', 'href="#', html) # remove the internal wiki links
	html = re.sub(r'&#91;(?=\d)', '', html) # remove left footnote bracket [
	html = re.sub(r'(?<=\d)&#93;', '', html) # remove right footnote bracket ]
	return html

def fast_loader(html):
	"""
		html = string (HTML)
	"""
	if fast == True:
		html = html.replace('/images/', '/images-small/')
		print('--- rendered in FAST mode ---')

	return html

def parse_page(pagename, wiki):
	"""
		pagename = string
		html = string (HTML)
	"""
	parse = f'{ wiki }/api.php?action=parse&page={ pagename }&pst=True&format=json'
	data = API_request(parse, pagename)
	# print(json.dumps(data, indent=4))
	if 'parse' in data:
		html = data['parse']['text']['*']
		images = data['parse']['images']
		html = download_media(html, images, wiki)
		html = clean_up(html)
		html = add_item_inventory_links(html)
		html = tweaking(html)
		html = fast_loader(html)
	else: 
		html = None

	return html

def save(html, pagename):
	"""
		html = string (HTML)
		pagename = string
	"""
	if __name__ == "__main__":
		# command-line

		# save final page that will be used with PagedJS
		template_file = open(f'{ STATIC_FOLDER_PATH }/{ TEMPLATES_DIR }/template.html').read()
		template = jinja2.Template(template_file)
		doc = template.render(publication_unfolded=html, title=pagename)
		
		html_file = f'{ STATIC_FOLDER_PATH }/{ pagename }.html'
		print('Saving HTML:', html_file)
		with open(html_file, 'w') as out:
			out.write(doc)
			out.close()

		# save extra html page for debugging (CLI only)
		template_file = open(f'{ STATIC_FOLDER_PATH }/{ TEMPLATES_DIR }/template.inspect.html').read()
		template = jinja2.Template(template_file)
		doc = template.render(publication_unfolded=html, title=pagename)

		html_file = f'{ STATIC_FOLDER_PATH }/{ pagename }.inspect.html'
		print('Saving HTML:', html_file)
		with open(html_file, 'w') as out:
			out.write(doc)
			out.close()

	else:
		# Flask application 

		with open(f'{ STATIC_FOLDER_PATH }/Unfolded.html', 'w') as out:
			out.write(html) # save the html to a file (without <head>)

def update_material_now(pagename, wiki):
	"""
		pagename = string
		publication_unfolded = string (HTML)
	"""
	publication_unfolded = parse_page(pagename, wiki)

	return publication_unfolded

# ---

if __name__ == "__main__":

	wiki = 'https://volumetricregimes.xyz' # remove tail slash '/'
	pagename = 'Unfolded'
	
	publication_unfolded = update_material_now(pagename, wiki) # download the latest version of the page
	save(publication_unfolded, pagename) # save the page to file