lua-htmlparser/tst/init.lua at master · lineCode/lua-htmlparser · GitHub

History

328 lines (306 loc) · 7.55 KB

Raw

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

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

-- vim: ft=lua ts=2 sw=2

-- Omit next line in actual module clients; it's only to support development of the module itself

package.path = "../src/?.lua;" .. package.path

pcall(require, "luacov")

print("------------------------------------")

print("Lua version: " .. (jit and jit.version or _VERSION))

print("------------------------------------")

print("")

local HAS_RUNNER = not not lunit

local lunitx

if not HAS_RUNNER then

lunitx = require"lunitx"

else

lunitx = require"lunit"

end

local TEST_CASE = lunitx.TEST_CASE

local LUA_VER = _VERSION

local unpack, pow, bit32 = unpack, math.pow, bit32

local _ENV = TEST_CASE"html"

local htmlparser = require("htmlparser")

function test_void()

local tree = htmlparser.parse([[

<p>

<br>

<br/>

<br >

<br />

</p>

<br>

<br/>

<br >

<br />

]])

assert_equal(5, #tree.nodes, "top level")

for _,n in ipairs(tree.nodes) do

if n.name == "p" then

assert_equal(4, #n.nodes, "deeper level")

else

assert_equal("br", n.name, "name")

assert_equal("", n:getcontent(), "content")

end

function test_id()

local tree = htmlparser.parse([[

</n>

]])

assert_equal(1, #tree.nodes, "top level")

assert_equal("n", tree("#4711")[1].name, "#4711")

assert_equal("m", tree("#1174")[1].name, "#1174")

end

function test_class()

local tree = htmlparser.parse([[

</n>

]])

assert_equal(3, #tree.nodes, "top level")

assert_equal(1, #tree(".one"), ".one")

assert_equal(2, #tree(".two"), ".two")

assert_equal(2, #tree(".three"), ".three")

assert_equal(1, #tree(".two.three"), ".two.three")

assert_equal(0, #tree(".four"), ".four")

end

function test_attr()

local tree = htmlparser.parse([[

<n a1 a2= a3='' a4=""

a5='a"5"' a6="a'6'" a7='#.[] :()' a8='|*+-=?$^%&/'

a9=a9

a10></n>

]])

assert_equal(1, #tree.nodes, "top level")

assert(tree("[a1]")[1], "a1")

assert(tree("[a2]")[1], "a2")

assert(tree("[a3]")[1], "a3")

assert(tree("[a4]")[1], "a4")

assert(tree("[a5]")[1], "a5")

assert(tree("[a6]")[1], "a6")

assert(tree("[a7]")[1], "a7")

assert(tree("[a8]")[1], "a8")

assert(tree("[a9]")[1], "a9")

assert(tree("[a10]")[1], "a10")

end

function test_attr_equal()

local tree = htmlparser.parse([[

<n a1 a2= a3='' a4=""

a5='a"5"' a6="a'6'" a7='#.[]:()' a8='|*+-=?$^%&/'

a9=a9

a10></n>

]])

assert_equal(1, #tree.nodes, "top level")

assert(tree("[a1='']")[1], "a1=''")

assert(tree("[a2='']")[1], "a2=''")

assert(tree("[a3='']")[1], "a3=''")

assert(tree("[a4='']")[1], "a4=''")

assert(tree("[a5='a\"5\"']")[1], "a5='a\"5\"'")

assert(tree("[a6=\"a'6'\"]")[1], "a6=\"a'6'\"")

assert(tree("[a7='#.[]:()']")[1], "a7='#.[]:()'")

assert(tree("[a8='|*+-=?$^%&/']")[1], "a8='|*+-=?$^%&/'")

assert(tree("[a9='a9']")[1], "a9='a9'")

assert(tree("[a10='']")[1], "a10=''")

assert(tree("[a10=]")[1], "a10=")

end

function test_attr_notequal()

local tree = htmlparser.parse([[

<n></n>

]])

assert_equal(4, #tree.nodes, "top level")

assert_equal(3, #tree("[a1!='a1']"), "a1!='a1'")

assert_equal(4, #tree("[a1!='b1']"), "a1!='b1'")

assert_equal(3, #tree("[a1!='']"), "a1!=''")

assert_equal(3, #tree("[a1!=]"), "a1!=")

end

function test_attr_prefix_start_end()

local tree = htmlparser.parse([[

<n></n>

]])

assert_equal(5, #tree.nodes, "top level")

assert_equal(3, #tree("[a1|='en']"), "a1|='en'")

assert_equal(4, #tree("[a1^='en']"), "a1^='en'")

assert_equal(2, #tree("[a1$='en']"), "a1$='en'")

end

function test_attr_word()

local tree = htmlparser.parse([[

<n></n>

]])

assert_equal(4, #tree.nodes, "top level")

assert_equal(1, #tree("[a1~='two']"), "a1~='two'")

assert_equal(2, #tree("[a1~='three']"), "a1~='three'")

assert_equal(1, #tree("[a1~='four']"), "a1~='four'")

end

function test_attr_contains()

local tree = htmlparser.parse([[

<n></n>

]])

assert_equal(6, #tree.nodes, "top level")

assert_equal(2, #tree("[a1*='one']"), "a1*='one'")

assert_equal(2, #tree("[a1*='t']"), "a1*='t'")

assert_equal(1, #tree("[a1*='f']"), "a1*='f'")

assert_equal(5, #tree("[a1*='']"), "a1*=''")

assert_equal(5, #tree("[a1*=]"), "a1*=")

end

function test_descendants()

local tree = htmlparser.parse([[

<parent>1

<child>2

</child>

</arbitrary>

</parent>

<parent>2

<child>6

</child>

</arbitrary>

</parent>

</arbitrary>

]])

assert_equal(8, #tree("parent child"), 'parent child')

end

function test_children()

local tree = htmlparser.parse([[

<parent>1

<child>2

</child>

</arbitrary>

</parent>

<parent>2

<child>4

</child>

</arbitrary>

</parent>

</arbitrary>

]])

assert_equal(4, #tree("parent > child"), 'parent > child')

end

function test_not()

local tree = htmlparser.parse([[

</n>

]])

assert_equal(2, #tree.nodes, "top level")

assert_equal(1, #tree(":not([a1=1])"), ":not([a1=1])")

assert_equal(1, #tree(":not([a2])"), ":not([a2])")

assert_equal(1, #tree(":not(n)"), ":not(n)")

assert_equal(2, #tree(":not(m)"), ":not(m)")

end

function test_combine()

local tree = htmlparser.parse([[

</n>

</e>

]])

assert_equal(2, #tree.nodes, "top level")

assert_equal(2, #tree("e.c:not([a|='1']) > n[b*='2']"), "e.c:not([a|='1']) > n[b*='2']")

assert_equal(3, #tree("e.c:not([a|='1']) n[b*='2']"), "e.c:not([a|='1']) n[b*='2']")

assert_equal(1, #tree("#123 .c[b]"), "#123 .c[b]")

end

function test_order()

local tree = htmlparser.parse([[

<1>

<2>

<3>

<4>

</4>

</3>

</2>

</1>

]])

assert_equal(1, #tree.nodes, "top level")

local n = tree("n")

assert_equal(10, #n, "n")

for i,v in pairs(n) do

assert_equal(i, tonumber(v:getcontent()), "n order")

end

local notn = tree(":not(n)")

assert_equal(4, #notn, "notn")

for i,v in pairs(notn) do

assert_equal(i, tonumber(v.name), "notn order")

end

function test_tagnames_with_hyphens()

local tree = htmlparser.parse([[

<tag-name id="9999">

</tag-name>

]])

assert_equal(1, #tree.nodes, "top level")

assert_equal("tag-name", tree("#9999")[1].name, "#9999")

assert_equal("m", tree("#10000")[1].name, "#10000")

end

function test_loop_limit()

local tree = htmlparser.parse([[

<p>

<a id="unclosed>Element"> with unclosed attribute</a>

</p>

]]) -- issue#42

assert(1==1)

end

lua-htmlparser/tst/init.lua at master · lineCode/lua-htmlparser · GitHub

Expand file tree

Breadcrumbs

init.lua

Latest commit

History

Breadcrumbs

init.lua

File metadata and controls