aboutsummaryrefslogtreecommitdiff
path: root/backend/server.py
blob: 8e4a90fee528222ac8f5f01d7f5072108f403da8 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
#!/usr/bin/python3

import sys, re, sqlite3, json
import os.path
from http.server import HTTPServer, BaseHTTPRequestHandler
import urllib.parse

hostname = "localhost"
port = 8000
dbFile = "data/data.db"
imgDir = "../public/img/"
SEARCH_SUGG_LIMIT = 5

usageInfo =  f"usage: {sys.argv[0]}\n"
usageInfo += "Starts a server that listens for GET requests to http://" + hostname + ":" + str(port) + ".\n"
usageInfo += "Responds to path+query /data/type1?name=name1 with JSON data.\n"
usageInfo += "\n"
usageInfo += "If type1 is 'node': Responds with map from names to objects representing node name1 and it's children.\n"
usageInfo += "If type1 is 'chain': Like 'node', but gets nodes from name1 up to the root, and their direct children.\n"
usageInfo += "If type1 is 'search': Responds with a tolnode name that has alt-name name1, or null.\n"
if len(sys.argv) > 1:
	print(usageInfo, file=sys.stderr)
	sys.exit(1)

# Connect to db, and load spellfix extension
dbCon = sqlite3.connect(dbFile)
dbCon.enable_load_extension(True)
dbCon.load_extension('./data/spellfix')
# Some functions
def lookupNodes(names):
	nodeObjs = {}
	cur = dbCon.cursor()
	# Get node info
	query = "SELECT name, children, parent, tips, p_support FROM nodes WHERE" \
		" name IN ({})".format(",".join(["?"] * len(names)))
	namesForImgs = []
	firstSubnames = {}
	secondSubnames = {}
	for row in cur.execute(query, names):
		name = row[0]
		nodeObj = {
			"children": json.loads(row[1]),
			"parent": None if row[2] == "" else row[2],
			"tips": row[3],
			"pSupport": True if row[4] == 1 else False,
			"commonName": None,
			"imgName": None,
		}
		# Check for image file
		match = re.fullmatch(r"\[(.+) \+ (.+)]", name)
		if match == None:
			namesForImgs.append(name)
		else:
			name1 = match.group(1)
			name2 = match.group(2)
			namesForImgs.extend([name1, name2])
			firstSubnames[name1] = name
			secondSubnames[name2] = name
		# Add node object
		nodeObjs[name] = nodeObj
	# Get image names
	query = "SELECT name, id FROM eol_ids WHERE" \
		" name IN ({})".format(",".join(["?"] * len(namesForImgs)))
	for [n, id] in cur.execute(query, namesForImgs):
		filename = str(id) + ".jpg"
		if not os.path.exists(imgDir + filename):
			continue
		if n in firstSubnames:
			nodeName = firstSubnames[n]
			nodeObjs[nodeName]["imgName"] = filename
		elif n in secondSubnames:
			nodeName = secondSubnames[n]
			if nodeObjs[nodeName]["imgName"] == None:
				nodeObjs[nodeName]["imgName"] = filename
		else:
			nodeObjs[n]["imgName"] = filename
	# Get preferred-name info
	query = "SELECT name, alt_name FROM names WHERE pref_alt = 1 AND" \
		" name IN ({})".format(",".join(["?"] * len(names)))
	for row in cur.execute(query, names):
		[name, altName] = row
		nodeObjs[name]["commonName"] = altName
	#
	return nodeObjs
def getNodeImg(name):
	cur = dbCon.cursor()
	row = cur.execute("SELECT name, id FROM eol_ids WHERE name = ?", (name,)).fetchone()
	if row != None:
		eolId = row[1]
		filename = str(eolId) + ".jpg"
		if os.path.exists(imgDir + filename):
			return filename
	return None
def lookupName(name):
	cur = dbCon.cursor()
	results = []
	hasMore = False
	#for row in cur.execute(
	#	"SELECT DISTINCT name, alt_name FROM names WHERE alt_name LIKE ? LIMIT ?",
	#	(name, SEARCH_SUGG_LIMIT)):
	#	results.append({"name": row[0], "altName": row[1]})
	#for row in cur.execute(
	#	"SELECT DISTINCT names.name, names.alt_name, nodes.tips FROM" \
	#		" names INNER JOIN nodes ON names.name = nodes.name " \
	#		" WHERE alt_name LIKE ? ORDER BY nodes.tips DESC LIMIT ?",
	#	(name + "%", SEARCH_SUGG_LIMIT)):
	#	results.append({"name": row[0], "altName": row[1]})
	for row in cur.execute(
		"SELECT word, alt_name, name FROM" \
			" spellfix_alt_names INNER JOIN names ON alt_name = word" \
			" WHERE word MATCH ? LIMIT ?",
		(name, SEARCH_SUGG_LIMIT)):
		results.append({"name": row[2], "altName": row[0]})
	if len(results) > SEARCH_SUGG_LIMIT:
		hasMore = True
		del results[-1]
	return [results, hasMore]
def lookupNodeInfo(name):
	cur = dbCon.cursor()
	# Get node desc
	row = cur.execute("SELECT desc, redirected from descs WHERE descs.name = ?", (name,)).fetchone()
	desc = {"text": row[0], "fromRedirect": row[1] == 1} if row != None else None
	# Get img info
	imgInfoQuery = "SELECT eol_id, source_url, license, copyright_owner FROM" \
		" images INNER JOIN eol_ids ON images.eol_id = eol_ids.id WHERE eol_ids.name = ?"
	row = cur.execute(imgInfoQuery, (name,)).fetchone()
	imgInfo = None
	if row != None:
		imgInfo = {"eolId": row[0], "sourceUrl": row[1], "license": row[2], "copyrightOwner": row[3]}
	#
	return {"desc": desc, "imgInfo": imgInfo}

class DbServer(BaseHTTPRequestHandler):
	def do_GET(self):
		# Parse URL
		urlParts = urllib.parse.urlparse(self.path)
		path = urllib.parse.unquote(urlParts.path)
		queryDict = urllib.parse.parse_qs(urlParts.query)
		# Check first element of path
		match = re.match(r"/([^/]+)/(.+)", path)
		if match != None and match.group(1) == "data" and "name" in queryDict:
			reqType = match.group(2)
			name = queryDict["name"][0]
			# Check query string
			if reqType == "node":
				nodeObjs = lookupNodes([name])
				if len(nodeObjs) > 0:
					nodeObj = nodeObjs[name]
					childNodeObjs = lookupNodes(nodeObj["children"])
					childNodeObjs[name] = nodeObj
					self.respondJson(childNodeObjs)
					return
			elif reqType == "chain":
				results = {}
				ranOnce = False
				while True:
					# Get node
					nodeObjs = lookupNodes([name])
					if len(nodeObjs) == 0:
						if not ranOnce:
							self.respondJson(results)
							return
						print("ERROR: Parent-chain node {} not found".format(name), file=sys.stderr)
						break
					nodeObj = nodeObjs[name]
					results[name] = nodeObj
					# Conditionally add children
					if not ranOnce:
						ranOnce = True
					else:
						childNamesToAdd = []
						for childName in nodeObj["children"]:
							if childName not in results:
								childNamesToAdd.append(childName)
						childNodeObjs = lookupNodes(childNamesToAdd)
						results.update(childNodeObjs)
					# Check if root
					if nodeObj["parent"] == None:
						self.respondJson(results)
						return
					else:
						name = nodeObj["parent"]
			elif reqType == "search":
				self.respondJson(lookupName(name))
				return
			elif reqType == "info":
				self.respondJson(lookupNodeInfo(name))
				return
		self.send_response(404)
		self.end_headers()
		self.end_headers()
	def respondJson(self, val):
		self.send_response(200)
		self.send_header("Content-type", "application/json")
		self.end_headers()
		self.wfile.write(json.dumps(val).encode("utf-8"))

server = HTTPServer((hostname, port), DbServer)
print("Server started at http://{}:{}".format(hostname, port))
try:
	server.serve_forever()
except KeyboardInterrupt:
	pass
server.server_close()
dbCon.close()
print("Server stopped")