User:Andlabs/Sega of Japan Dreamcast Master List

because I've had it now
ok more explanation: Guardiana (where our DC master list appears to comes from) prioritizes JP name over US name so I kept getting caught off-guard when making lists (what tipped me over the edge was seeing HuneX's website list the JP and US versions of games separately) or just has wrong info (wonders never cease); sometimes the Sega pages have wrong serial numbers (happens more often with Saturn) or randomly combine versions or something, and again I want to make sure everything here is correct

Sometimes there will be a translated name on the left spine that differs from the Japanese name; that will be listed second, after a slash. I personally consider this to be a secondary name :/

Legend
Everything (all fields in table + page categories) verified
Everything verified, but box/master list genre mismatch
Problem with verification; cell(s) left white explains what (everything else is fine)
Unverified (either I didn't get to it yet or the page doesn't exist)

1998 1999 2000 2001 2002 2003 2004 2005-2007

1998

Romanized Name	Japanese Name	Publisher	Genre on Master List	Genre on Box	Release Date	Price	Catalogue Page	Homepage	Serial on Master List	Serial on Box	CERO	Overseas Names
July/?	Ｊｕｌｙ	45XLV	ADV	ADV	1998/11/27	5,800円	[1]		T-35401M			No
Pen Pen TriIcelon/(check)	ペンペン トライアイスロン	GE	RCG	RCG	1998/11/27	5,800円	[2]		T-17001M	T-17001M		Same
	バーチャファイター３ｔｂ リピート版	Sega	ACT		1998/11/27	5,800円	[3]
Virtua Fighter 3tb	バーチャファイター３ｔｂ	Sega	ACT		1998/11/27	5,800円	[4]
Godzilla Generations	GODZILLA GENERATIONS	Sega	ACT	ACT	1998/11/27	5,800円	[5]		HDR-0004	HDR-0004		No
?/Incoming Humanity Last Battle	インカミング 人類最終決戦	Imagineer	STG	STG	1998/12/17	5,800円	[6]		T-15001M	T-15001M		Incoming
Sonic Adventure	ＳＯＮＩＣ ＡＤＶＥＮＴＵＲＥ	Sega	ACT	ACT	1998/12/23	5,800円	[7]		HDR-000	HDR-0001		Same
Tetris 4D	ＴＥＴＲＩＳ ４Ｄ（テトリス フォーディー）	BPS	PZL	PZL	1998/12/23	4,800円	[8]		T-20801M	T-20801M		No
Seventh Cross	ＳＥＶＥＮＴＨ ＣＲＯＳＳ	NEC	SRPG	SRPG	1998/12/23	5,800円	[9]		T-38802M			US: Seventh Cross Evolution EU: none

generator

# 27-28 oct 2011
# NOTE: PYTHON 3

import sys
import urllib.request
import io
from lxml import etree

def pageURLGen():
	p_hw  = 10  # Dreamcast
	p_sr  = 640 # start record according to clicking on the last page of results
	p_rpp = 20  # decrement p_sr each time; sega seems to ignore this parameter so we have to deal with it ourselves
	while p_sr >= 0:
		yield "http://sega.jp/cgi-bin/csgame.cgi?rpp={}&sr={}&tt=&sy=&gr=&hw={}&fw=&sort=2".format(p_rpp, p_sr, p_hw)
		p_sr -= p_rpp

# constants
genreIcons = {
	"/shared/images/icons/genre_act.gif": "ACT",
	"/shared/images/icons/genre_ftg.gif": "FTG",
	"/shared/images/icons/genre_stg.gif": "STG",
	"/shared/images/icons/genre_tbl.gif": "TBL",
	"/shared/images/icons/genre_rcg.gif": "RCG",
	"/shared/images/icons/genre_slg.gif": "SLG",
	"/shared/images/icons/genre_spg.gif": "SPG",
	"/shared/images/icons/genre_etc.gif": "ETC",
	"/shared/images/icons/genre_rpg.gif": "RPG",
	"/shared/images/icons/genre_adv.gif": "ADV",
	"/shared/images/icons/genre_pzl.gif": "PZL",
	"/shared/images/icons/genre_arpg.gif": "ARPG",
	"/shared/images/icons/genre_srpg.gif": "SRPG",
	"/shared/images/icons/genre_aadv.gif": "AADV"
}
ceroIcons = {
	"/shared/images/icons/icon_cero-a_xsmall.gif": "a",
	"/shared/images/icons/icon_cero-b_xsmall.gif": "b",
	"/shared/images/icons/icon_cero-c_xsmall.gif": "c",
	"/shared/images/icons/icon_cero-d_xsmall.gif": "d",
	"/shared/images/icons/icon_cero-z_xsmall.gif": "z",
	"?? free": "free",
	"?? 12": "12",
	"?? 15": "15",
	"/shared/images/icons/icon_cero-h_xsmall.gif": "18"
}
noCEROIcon = "/shared/images/icons/icon_cero-_xsmall.gif"
partnerIcon = "/shared/images/icons/partners.gif"
dreamcastIcon = "/shared/images/icons/hard_DC.gif"
networkIcon = "/shared/images/icons/function_network.gif"
# TODO other icons that may or may not be skipped
homepageLinkIcon = "/shared/images/icons/btn_official.gif"
cataloguePageLinkIcon = "/shared/images/icons/btn_product.gif"

def isComment(e):
	return e.tag == etree.Comment

def getImgTag(element):
	for e in element:
		if isComment(e):
			continue
		if e.tag.lower() == "img":
			return e
	raise Exception("expected img element; found none")

def handleDivClassIcon(element):
	imgTag = getImgTag(element)
	src = imgTag.attrib["src"]
	if src == dreamcastIcon: # skip this icon
		return ""
	elif src == partnerIcon: # note that this is third party
		return "\n| publisher=THIRD PARTY"
	elif src == networkIcon:
		return "\n| network={{yes}}"
	elif src in genreIcons:
		return "\n| genrelist={}".format(genreIcons[src]) + \
			"\n| genrebox="
	else:
		raise Exception("unknown icon {}".format(src))

def handleDivClassTitleText(element):
	return "\n| title=" + \
		"\n| jptitle={}".format(element.text) + \
		"\n| overseas="

def handleATag(element):
	url = element.attrib["href"]
	img = getImgTag(element)
	imgsrc = img.attrib["src"]
	if imgsrc == homepageLinkIcon:
		return "\n| homepage={}".format(url)
	elif imgsrc == cataloguePageLinkIcon:
		return "\n| catalogue=http://sega.jp{}".format(url)
	else:
		raise Exception("unknown image link {} -> {}".format(src, url))

def handleDivRightColumn(element):
	s = ""
	for e in element.getiterator():
		if isComment(e):
			continue
		if e.tag.lower() == "a":
			s += handleATag(e)
		elif e.tag.lower() == "div":
			divclass = e.attrib.get("class")
			if divclass == "icon":
				s += handleDivClassIcon(e)
			elif divclass == "titleText":
				s += handleDivClassTitleText(e)
	return s

def handleFirstColumn(element):
	for e in element.getiterator():
		if isComment(e):
			continue
		if e.tag.lower() == "div" and e.attrib.get("class") == "rightColumn":
			return handleDivRightColumn(e)
	return ""

def handleCEROCell(element):
	for e in element.getiterator():
		if isComment(e):
			continue
		if e.tag.lower() == "img":
			if e.attrib.get("src") in ceroIcons:
				return "\n| cero={}".format(ceroIcons[e.attrib["src"]])
			elif e.attrib.get("src") == noCEROIcon:
				return ""
			else:
				raise Exception("unknown CERO icon {}".format(e.attrib["src"]))
	return ""

def handleRow(element):
	if element[0].attrib.get("class").startswith("th1"): # skip table header
		return ""
	s = "{{wipdclist"
	s += handleFirstColumn(element[0])
	s += "\n| release={}".format(element[1].text)
	s += "\n| price={}".format(element[2].text)
	s += handleCEROCell(element[3])
	s += "\n| seriallist="
	s += "\n| serialbox="
	return s + "\n}}"

from lxml import html

#try:
for no in [1]:
	for page in pageURLGen():
		tree = html.parse(page)
		glist = []
		for e in tree.getiterator(): # search through all tags for the one we want
			if isComment(e):
				continue
			if e.tag.lower() == "div" and e.attrib.get("id") == "resultTable":
				for tr in e[0]: # we assume this is a <table>
					game = handleRow(tr)
					if game != "":
						glist += [game]
				break
		glist.reverse() # pages store in reverse chronological order
		for game in glist:
			print(game)
# TODO figure out the cleanest way to get a traceback