Skip to content

Commit

Permalink
Add new data
Browse files Browse the repository at this point in the history
  • Loading branch information
warioddly committed May 14, 2024
1 parent 89b2225 commit 5881298
Show file tree
Hide file tree
Showing 8 changed files with 33 additions and 11 deletions.
28 changes: 24 additions & 4 deletions parser/bin/parser.dart
Original file line number Diff line number Diff line change
Expand Up @@ -49,19 +49,39 @@ void main(List<String> arguments) async {


Future<void> _scrapy() async {

print('[+] Scrapy started.');

final scrapy = Scrapy();

await scrapy.parse(['', 'https://www.sanjyra.net/man/1']);
void bundler() {
Bundler()
..generate(jsonEncode(scrapy.people.map((e) => e.toJson()).toList()), 'all_data.json')
..generate(jsonEncode(scrapy.people.map((e) => e.toCytoscapeJson()).toList()), 'all_data_site.json');
}


ProcessSignal.sigint.watch().listen((event) {
bundler();
print('\n[+] Exiting...');
exit(0);
});

scrapy.visited.addAll({
'https://www.sanjyra.net/man/1',
'https://www.sanjyra.net/man/2',
'https://www.sanjyra.net/man/32252'
});

await scrapy.parse(['', 'https://www.sanjyra.net/man/44275']);

print('\n[+] Done! Parsed ${scrapy.people.length} people.');

Bundler()
..generate(jsonEncode(scrapy.people.map((e) => e.toJson())), 'all_data.json')
..generate(jsonEncode(scrapy.people.map((e) => e.toCytoscapeJson())), 'all_data_site.json');
bundler();

print('[+] Scrapy finished.');
exit(0);

}


Expand Down
8 changes: 4 additions & 4 deletions parser/lib/scrapy.dart
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@ import 'package:parser/people_model.dart';

class Scrapy {

final _visited = <String>{};
final visited = <String>{};
final people = <People>[];
final bundler = Bundler();

Expand Down Expand Up @@ -69,7 +69,7 @@ class Scrapy {

var url = child.children.first.attributes['href'];

if (url != null && !_visited.contains(url)) {
if (url != null && !visited.contains(url)) {
urls.add(url);
}

Expand All @@ -83,11 +83,11 @@ class Scrapy {


bool _crawl(String url) {
if (_visited.contains(url)) {
if (visited.contains(url)) {
print('[-] Already crawled: $url');
return false;
}
_visited.add(url);
visited.add(url);
print('[+] Crawling $url');
return true;
}
Expand Down
1 change: 0 additions & 1 deletion parser/output/data.json

This file was deleted.

1 change: 1 addition & 0 deletions parser/output/куу уул/all_data.json

Large diffs are not rendered by default.

1 change: 1 addition & 0 deletions parser/output/куу уул/all_data_site.json

Large diffs are not rendered by default.

1 change: 1 addition & 0 deletions parser/output/куу уул/dynamic_all_data.json

Large diffs are not rendered by default.

2 changes: 1 addition & 1 deletion src/assets/data/data.json

Large diffs are not rendered by default.

2 changes: 1 addition & 1 deletion src/components/tree/TreeInteractiveViewer.tsx
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@ export function TreeInteractiveViewer() {

fetch('/api/get-nodes', {
method: 'POST',
body: JSON.stringify({ limit: 2500 }),
body: JSON.stringify({ limit: 4500 }),
}).then((res) => res.json()).then((data: TreeNodeDefinition[]) => {
setGraph(data);
setLoading(false);
Expand Down

0 comments on commit 5881298

Please sign in to comment.