Jump to content

User:PotatoBot/Code/5

From Wikipedia, the free encyclopedia
#!/usr/bin/python
# -*- coding: utf-8  -*-

import wikipedia as w
import codecs, catlib, pagegenerators, time, mysave, re
from datetime import date

# PotatoBot Task 5: Creates redirects from trade names to drug articles

excludes = ('', u'Adiuretin', u'Amicacin', u'Apirelina', u'Bayer Brand of Azlocillin‎', u'Benzchinamide', u'BZQ', u'Brevital sodium', u'Cephaloglycine', u'Cephaoglycin acid', u'Chlorazepate', u'Chlorazepic acid', u'Cialis/Tadalafil', u'Cialis/Taladafil Hcl', u'Citrovorum factor', u'Clorazepic acid', u'Co-Vidarabine', u'Covidarabine', u'Dapropterin', u'Deacetyllanatoside C', u'Dextrin, Caloreen', u'Diovan HCT', u'Dl-Tranylcypromine', u'Dolasteron', u'Fish berry', u'Fondaparinux sodium', u'FR-008-III', u'Fujimycin', u'Gamatran citrate', u'Genzyme)', u'Glycocoll', u'Guanabenz(USAN)', u'Indian berry', u'IRINOTECAN, CPT-11', u'Karnitin', u'Kitasamycin A3', u'Kyselina jantarova', u'Lrbesartan', u'Lyophilized Cytoxan', u'Lysuride', u'Metamfetamine', u'Methenamide', u'Methoxamedrine', u'Metossamina', u'Metoxamina', u'Micardis HCT', u'Naphcillin', u'Navelbine Base', u'Oriental berry', u'Polymyxin E. Sulfate', u'Polymyxin E sulfate', u'Propilniazida', u'Pyridium Plus', u'Quetiapin hemifumarate', u'Sapropterin', u'Secremax, SecreFlo', u'Sodium sulamyd', u'Spongoadenosine', u'SU-11248', u'THIORPHAN', u'Trifluopromazine', u'Turimycin A5', u'Wellcome U3B', u'Xiaflextm',
)

def main():
	# Prepare log
	listout = '\n'
	db = {}
	Rtemplate = {False: '{{R from trade name}}', True: '{{R from alternative name}}'}
		
	# Load DrugBank data
	n, id= 0, ''
	with codecs.open('data/drugcards.txt', 'r', 'utf-8') as f:
		for l in f:
			line = l.strip()
			if line[:15] == '#BEGIN_DRUGCARD':
				n = int(line[18:23])
				list = [[], n]
				generic = ''
			elif line[:13] == '#END_DRUGCARD':
				if generic == '':
					w.output('  \03{red}ERROR IN DRUGCARD %d: no generic name\03{default}' % n)
					listout += '# %s: Error in Drugcard %d (no generic name found)\n' % n
				elif n != list[1]:
					w.output('  \03{red}ERROR IN DRUGCARD %d: IDs do not match\03{default}' % n)
					listout += '# %s: Error in Drugcard %d (IDs of start and end tag do not match)\n' % n
				elif generic in db:
					w.output('  \03{red}ERROR IN DRUGCARDS %d, %d: generic name found twice\03{default}' % (n, db[generic][1]))
					listout += '# %s: Error in Drugcards %d, %d (generic name found twice)\n' % (n, db[generic][1])
				elif len(list[0]) > 8:
					db[generic] = list
				n = 0
			elif line[:2] == '# ':
				id = line[2:-1]
			elif line != '' and n != 0:
				if id == 'Generic_Name':
					generic = line
				elif id == 'Brand_Names' and line != 'Not Available' and 'hydrochloride' not in line.lower() and len(line) < 25:
					p = line.find(' (')
					if p == -1: p = line.find(' [')
					if p == -1: p = line.find(' Roche')
					if p > -1: line = line[:p].strip()
					if line not in excludes:
						list[0] += [line]

	w.output('DrugBank entries loaded: %d' % len(db))
	
	# Create redirects
	for drug in db:
		w.output('* %s (#%d)' % (drug, db[drug][1]))
		drugpage = mysave.resolveredir(w.Page(w.getSite(), drug))
		if not drugpage.exists():
			drugpage = mysave.resolveredir(w.Page(w.getSite(), drug.capitalize()))
		notfound = []
		for tradename in db[drug][0]:
			if tradename.lower().find(drug.lower() + ' ') != 0:
				tradepage = w.Page(w.getSite(), tradename)
				if drugpage.title()[-1] == 'e': drug_e = drugpage.title()[:-1]
				else: drug_e = drugpage.title()
				alt = tradename in (drug_e, drug_e + 'a', drug_e + 'e', drug_e + 'o', drug_e + 'um')
				if tradepage.exists() and tradepage.isRedirectPage():
					tradetext = tradepage.get(get_redirect=True)
					brandTsearch = re.search(r'\{\{\s*[Rr] from brand name\s*\}\}', tradetext)
					alterTsearch = re.search(r'\{\{\s*[Rr] from alternative name\s*\}\}', tradetext)
					tradeTsearch = re.search(r'\{\{\s*[Rr] from trade name\s*\}\}', tradetext)
					if tradepage.getRedirectTarget() != drugpage:
						w.output('  \03{yellow}%s doesn\'t redirect to the right page (%s)?\03{default}' \
							% (tradepage.title(), drugpage.title()))
						listout += '# %s: redirects to %s instead of %s\n'\
							% (tradepage.aslink(), tradepage.getRedirectTarget().aslink(), drugpage.aslink())
					elif brandTsearch:
						listout += mysave.savepage(tradepage, tradetext[:brandTsearch.start()] + Rtemplate[alt] + tradetext[brandTsearch.end():], 
							'Replace {{R from brand name}} with ' + Rtemplate[alt], minor = True)
					elif alterTsearch and not alt:
						listout += mysave.savepage(tradepage, tradetext[:alterTsearch.start()] + Rtemplate[alt] + tradetext[alterTsearch.end():], 
							'Replace ' + Rtemplate[not alt] + ' with ' + Rtemplate[alt], minor = True)
					elif tradeTsearch and alt:
						listout += mysave.savepage(tradepage, tradetext[:tradeTsearch.start()] + Rtemplate[alt] + tradetext[tradeTsearch.end():], 
							'Replace ' + Rtemplate[alt] + ' with ' + Rtemplate[not alt], minor = True)
					elif not tradeTsearch and not alterTsearch:
						minusCats = w.removeCategoryLinks(tradetext, w.getSite())
						listout += mysave.savepage(tradepage, minusCats + ' ' + Rtemplate[alt] + tradetext[len(minusCats):], \
							'Add ' + Rtemplate[alt], minor = True)
				elif drugpage.exists():
					listout += mysave.makeredir(tradepage, drugpage, Rtemplate[alt])
				else:
					notfound += [tradepage.aslink()]
		if notfound != []:
			listout += '# %s: target %s not found\n' % (', '.join(notfound), drugpage.aslink())

	# Output log
	listout += '\nTrade names from DrugBank completely included.'
	w.output('')
	logpage = w.Page(w.getSite(), 'User:PotatoBot/Lists/Trade names log')
	mysave.savepage(logpage, logpage.get() + listout, 'Creating trade names log')

if __name__ == "__main__":
	try:
		main()
	finally:
		w.stopme()

mysave.py

[edit]
#!/usr/bin/python
# -*- coding: utf-8  -*-

import pywikibot as w
import re

# Code for saving redirects and other pages

def savepage(page, text, BRFANo, summary = '', minor = False):
	"""Save text to a page and log exceptions."""
	if summary != '':
		w.setAction(summary + '. See [[Wikipedia:Bots/Requests for approval/PotatoBot ' + BRFANo + '|approval]]. Report errors and suggestions at [[User talk:PotatoBot]].')
	try:
		if not '#' in page.title():
			page.put(text, minorEdit = minor)
			w.output('  \03{green}saving %s -> \03{gray}%s\03{default}' % (page.title(), text))
			return ''
		else:
			w.output('  \03{red}cannot save %s because it is a section\03{default}' % page.title())
			return '# %s: this is a secion title' % page.title(aslink=True)
	except w.LockedPage:
		w.output('  \03{red}cannot save %s because it is locked\03{default}' % page.title())
		return '# %s: page was locked\n' % page.title(aslink=True)
	except w.EditConflict:
		w.output('  \03{red}cannot save %s because of edit conflict\03{default}' % page.title())
		return '# %s: edit conflict occurred\n' % page.title(aslink=True)
	except w.SpamfilterError, error:
		w.output('  \03{red}cannot save %s because of spam blacklist entry %s\03{default}' % (page.title(), error.url))
		return '# %s: spam blacklist entry\n' % page.title(aslink=True)
	except:
		w.output('  \03{red}unknown error on saving %s\03{default}' % page.title())
		return '# %s: unknown error occurred\n' % page.title(aslink=True)

def resolveredir(page):
	"""Return target if input is a redirect, else return input."""
	try:
		if page.isRedirectPage():
			try:
				w.output('  \03{gray}resolving redir %s to %s\03{default}'\
					% (page.title(), page.getRedirectTarget().title()))
				return page.getRedirectTarget()
			except:
				w.output('  \03{yellow}target %s is a broken redir\03{default}' % page.title())
				return w.Page(w.getSite(), page.title() + ' (broken redirect)')
		else:
			return page
	except:
		w.output('  \03{yellow}target %s is a bad link\03{default}' % page.title())
		return w.Page(w.getSite(), page.title() + ' (bad link)') # workaround for wikipedia.py breaking wikiasite: links

def makeredir(redirpage, page, BRFANo, templates = ''):
	"""Create a redirect and log existing page that isn't a redirect to the desired article."""
	page = resolveredir(page)
	if redirpage.exists():
		comment = ''
		if redirpage.isDisambig():
			comment = ' (disambiguation)'
			dab = redirpage
		if redirpage.isRedirectPage():
			try:
				if redirpage.getRedirectTarget().title() == page.title() or \
						redirpage.getRedirectTarget().sectionFreeTitle() == page.title():
					# Already a redir to the desired article
					return ''
				elif redirpage.getRedirectTarget().isDisambig():
					comment = ' (redirect to disambiguation)'
					dab = redirpage.getRedirectTarget()
				else:
					comment = ' (redirect)'
			except:
				comment = ' (broken redir)'
		if 'disambiguation' in comment and page in [resolveredir(p) for p in dab.linkedPages()]:
			w.output('  link to %s already on dab page %s' % (page.title(), redirpage.title()))
			return ''
		elif redirpage.title() != page.title():
			w.output('  \03{yellow}redir to %s failed, page %s already exists\03{default}' % (page.title(), redirpage.title()))
			return '# %s: redirecting to %s failed, page already exists%s\n' % (redirpage.title(aslink=True), page.title(aslink=True), comment)
		else:
			return ''
	# Else create redirect, or write page name to list if an error occurs
	else:
		return savepage(redirpage, '#REDIRECT %s %s' % (page.title(aslink=True), templates), BRFANo, 'Redirect to ' + page.title(aslink=True))

def findATCs(page, includeVet = True):
	"""Look for ATC codes in infoboxes."""
	ATCvet, prefix, suffix, supp = False, '', '', ''
	ATCvetpos, prefixpos, suffixpos, supppos = -1, -1, -1, -1
	templatenames = ('Drugbox', 'Chembox Identifiers')
	templates = page.templatesWithParams()
	for tuple in templates:
		if tuple[0] in templatenames:
			idx = templatenames.index(tuple[0])
			templatepos = templates.index(tuple)
			for param in tuple[1]:
				value = param.partition('=')
				if value[0].strip() == 'ATCvet':
					ATCvet = value[2].strip() == 'yes' and includeVet
					ATCvetpos = tuple[1].index(param)
				elif value[0].strip() == ('ATC_prefix', 'ATCCode_prefix')[idx] and value[2].strip().lower != 'none':
					prefix = value[2].strip()
					prefixpos = tuple[1].index(param)
				elif value[0].strip() == ('ATC_suffix', 'ATCCode_suffix')[idx]:
					suffix = value[2].strip()
					suffixpos = tuple[1].index(param)
				elif value[0].strip() == ('ATC_supplemental', 'ATC_Supplemental')[idx]:
					supp = value[2].strip()
					supppos = tuple[1].index(param)
	codes = (prefix != '') * [(ATCvet*'Q' + prefix + suffix)]
	for tupleSupp in page.templatesWithParams(supp):
		if tupleSupp[0] in ['ATC', 'ATCvet']:
			codes.append((tupleSupp[0] == 'ATCvet')*'Q' + tupleSupp[1][0] + tupleSupp[1][1])
	return (codes, ATCvetpos, prefixpos, suffixpos, supppos)

def addTemplateParam(page, newtemplates, BRFANo, summary = 'Updating template', minor = False):
	text = page.get()
	oldtemplates = page.templatesWithParams()
	pointer = 0
	for i in range(len(oldtemplates)):
		search1 = re.compile(r'\{\{\s*(%s|%s)%s\s*\|' % (oldtemplates[i][0][0].upper(), oldtemplates[i][0][0].lower(),\
			oldtemplates[i][0].replace(' ', '( |_)'))).search(text, pointer)
		if search1:
			pointer = end() - 1
		if newtemplates[i] != oldtemplates[i]:
			if newtemplates[i][0].strip() == oldtemplates[i][0].strip():
				for j in range(len(oldtemplates[i][1])):
					oldparam = oldtemplates[i][1][j].partition('=')
					newparam = newtemplates[i][1][j].partition('=')
					# Todo: unnamed params #
					if newparam[0] == oldparam[0]:
						span = re.compile(r'\|\s*%s\s*=\s*([^|}\s]*)\s*(}|\|)' % oldparam[0]).\
							search(text, pointer).span(1)
						pointer = span(1)
						if newparam[2].strip() != oldparam[2].strip():
							text = text[:span(0)] + newparam[2] + text[span(1):]
					else:
						text = text[:] + newtemplates[i][1][j] + text[:]
						pointer = len(text[:] + newtemplates[i][1][j])
			else:
				w.output('\03{yellow}template list does not match page %s: %s vs. %s\03{default}' % \
					(page.title(), newtemplates[i][0].strip(), oldtemplates[i][0].strip()))
				return '# %s: template list did not match templates on page' % page.title(aslink=True)
	if text != page.get():
		return savepage(page, text, BRFANo, summary, minor)
	else:
		return ''

def fmtdate(date):
	"""Format date in English w style."""
	return '%d %s %d' % (date.day, ('', 'January', 'February', 'March', 'April', 'May', 'June', 'July', 'August',\
		'September', 'October', 'November', 'December')[date.month], date.year)