Created
July 5, 2017 12:16
-
-
Save Glavo/0fb35b447a14f50c4fb18d83ed584236 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
package org.glavo.oj.ojs | |
import org.glavo.oj.OJ | |
import org.glavo.oj.Problem | |
import org.jsoup.* | |
import org.jsoup.nodes.Document | |
import org.jsoup.nodes.Element | |
import org.jsoup.select.Elements | |
import java.io.ObjectInputStream | |
import java.io.ObjectOutputStream | |
import java.io.PrintWriter | |
import java.net.URL | |
import java.nio.file.Files | |
import java.nio.file.Path | |
import java.nio.file.Paths | |
import java.util.regex.Pattern | |
import java.util.stream.Stream | |
import kotlin.streams.toList | |
/** | |
* Created by Glavo on 17-7-2. | |
* | |
* @author Glavo | |
* @since 0.1.0 | |
*/ | |
fun parsePage(document: Document): Problem { | |
val foo = document.select("div.lg-content-left div.am-g *") | |
var now = "" | |
val buffer = Elements() | |
for (e in foo) { | |
if (e.nodeName() == "h2") { | |
buffer.clear() | |
} | |
else buffer += e | |
} | |
TODO() | |
} | |
class Luogu : Stream<Problem> by stream, OJ { | |
companion object { | |
val tem: Path by lazy { | |
if (!Files.exists(path / "luogu")) Files.createDirectories(path / "luogu") | |
path / "luogu" | |
} | |
internal val pages: List<String> by lazy { | |
if (Files.exists(tem / "pages") && !Files.isDirectory(tem / "pages")) { | |
Files.newBufferedReader(tem / "pages").use { | |
@Suppress("UNCHECKED_CAST") | |
it.lines().toList() | |
} | |
} else { | |
var pages: List<String> = listOf() | |
var now = "https://www.luogu.org/problem/lists?name=&orderitem=&order=&tag=&page=1" | |
try { | |
while (true) { | |
val html: Document = now.toDocument() | |
pages += html | |
.select("div.lg-content-table-left div[class=am-g lg-table-row lg-table-bg0] a:not(a[target=_blank])") | |
.map { it.attr("href") } | |
.map { "https://www.luogu.org$it" } ?: listOf() | |
val l: MutableList<Element> = html.select("a[rel=next]") | |
if (l.isEmpty()) break | |
else now = "https://www.luogu.org${l[0].attr("href")}" | |
} | |
} catch (e: Exception) { | |
} | |
Files.createFile(tem / "pages") | |
PrintWriter(Files.newBufferedWriter(tem / "pages")).use { stream -> | |
pages.forEach(stream::println) | |
} | |
pages | |
} | |
} | |
private val p = Pattern.compile("""P([0-9]+) +(.+)""") | |
val stream: Stream<Problem> by lazy { | |
pages.stream().map { page -> | |
val title = page | |
val match = p.matcher(title.trim()) | |
if (!match.find()) return@map null | |
val index = match.group(0).toInt() | |
val name = match.group(1) | |
} | |
TODO("") | |
} | |
} | |
} | |
fun String.toDocument(): Document { | |
var exception: Exception? = null | |
for (i in 1..5) { | |
try { | |
return Jsoup.parse(URL(this), 10000) | |
} catch (e: Exception) { | |
exception = e | |
} | |
} | |
throw exception!! | |
} | |
inline fun <T> timed(f: () -> T): Long { | |
val start = System.nanoTime() | |
f() | |
return (System.nanoTime() - start) | |
} | |
val path: Path = Paths.get(System.getProperty("user.home"), ".crawler") | |
get() = | |
if (Files.isDirectory(field)) field | |
else { | |
Files.createDirectories(field) | |
field | |
} | |
operator fun Path.div(other: String): Path = this.resolve(other)!! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment