-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathparser.jl
215 lines (170 loc) · 6.09 KB
/
parser.jl
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
207
208
209
210
211
212
213
214
215
#=
Copyright (C) 2023 Yiding Song
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
=#
using EzXML
using ProgressBars
include("wikigraph.jl")
function hmean(freqs)
num = 0
rec = 0
for f in freqs
num += 1
rec += 1 / f
end
return ceil(num / rec)
end
function strength(
text::AbstractString,
phrase::AbstractString
)
return 1
#=
phrase = lowercase(phrase)
words = [i.match for i in eachmatch(r"[a-z']+", phrase)]
numwords = length(words)
if numwords == 0
return length(findall(phrase, text))
end
return hmean([length(findall(w, text)) for w in words]);
=#
end
function mineLinks(
text::AbstractString,
lowertext::AbstractString,
wiki_namespaces::Vector{String}
)
linkTargets = AbstractString[]
linkWeights = Integer[]
len = length(text)
inlink = false
last = ""
cache = ""
counter = 1
for i in eachindex(text)
if counter > (len - 3)
break
end
c = text[i:i]
if inlink
if c in ("|", "#") || last*c == "]]"
title = normalise(cache)
if !(title in linkTargets)
push!(linkTargets, title)
push!(linkWeights, strength(lowertext, cache))
end
inlink = false
last = ""
cache = ""
elseif last == "]"
cache *= "]" * c
elseif c != "]"
cache *= c
end
if cache in wiki_namespaces
inlink = false
last = ""
cache = ""
end
elseif last*c == "[["
inlink = true
end
last = c
counter += 1
end
return linkTargets .=> linkWeights
#= Above equivalent to:
[ linkTargets[0] => linkWeights[0],
linkTargets[1] => linkWeights[1],
...,
linkTargets[n] => linkWeights[n] ]
=#
end
function mineXML(
xmlPath::AbstractString,
wgDir::AbstractString,
titlesPath::AbstractString,
logPath::AbstractString,
numPages::Integer
)
if wgIntegrity(wgDir)
println("Loading Wikigraph from last checkpoint")
@time wg = loadwg(wgDir, titlesPath)
else
println("Initialising empty Wikigraph from titles")
@time wg = Wikigraph(titlesPath)
end
println("\nTotal page count: $(wg.pm.totalpages)")
println("\nReading XML document")
@time doc = readxml(xmlPath)
xmlns = namespace(doc.root)
wiki_namespaces = String[]
for i in findall("/x:mediawiki/x:siteinfo/x:namespaces/x:namespace", doc.root, ["x" => xmlns])
if i.content != ""
push!(wiki_namespaces, normalise(i.content) * ":")
push!(wiki_namespaces, ":" * normalise(i.content) * ":")
end
end
println("\nWikipedia namepsaces: $(wiki_namespaces)\n")
open(logPath, "a") do logs
counter = 0
numlinks = []
iter = ProgressBar(eachelement(doc.root))
for ele in iter
if ele.name != "page"
continue
end
children = elements(ele)
@assert (children[1].name == "title") "Title not first child element"
title = normalise(children[1].content)
if !haskey(wg.pm.title2id, title) || # Not an article
wg.pm.redirs[wg.pm.title2id[title]] != wg.pm.title2id[title] || # Is redirected
length(wg.links[wg.pm.title2id[title]]) != 0 # Already mined
# println("========== SKIPPED $(title) ==========\n")
elseif children[end-1].name == "redirect"
redir = normalise(children[end - 1]["title"])
if haskey(wg.pm.title2id, redir)
tie(wg.pm, wg.pm.title2id[title], wg.pm.title2id[redir])
end
# println("========== REDIRECTED $(title) ==========\n")
else
@assert (children[end].name == "revision") "Revision not last child element"
revision_children = elements(children[end])
@assert (revision_children[end-1].name == "text")
"Text not second last revision child element"
text = revision_children[end-1].content
lowertext = lowercase(text)
lks = mineLinks(text, lowertext, wiki_namespaces)
title_id = wg.pm.title2id[title]
for (t, w) in lks
try
link(wg, title_id, wg.pm.title2id[t], w)
catch err
write(logs, "$(title)\t$(t)\n")
# println("Title error! ", err)
end
end
push!(numlinks, length(wg.links[wg.pm.title2id[title]]))
# println("========== COMPLETED $(title) ==========\n")
end
counter += 1
set_description(iter, "$(counter)/$(numPages)")
end
printlog(s) = (write(logs, s); println(s))
printlog("\nAverage number of links: $(trunc(sum(numlinks)/length(numlinks), digits=3))\n")
printlog("Pages with links: $(length(numlinks))\n")
printlog("Number of pages (not counting redirects): $(wg.pm.numpages)")
end
println("\nSaving Wikigraph")
@time savewg(wgDir, wg)
return wg
end