public static IObservable <WikiPage> CreateCategoryDownloader(WikiSite site, string categoryTitle)
{
var page = new WikiPage(site, categoryTitle);
var generator = new CategoryMembersGenerator(page);
var observable = generator.EnumPagesAsync()
.Where((page, b) => !page.IsSpecialPage)
.ToObservable();
return(observable);
}
public async Task RunAsync()
{
await zhWarriorsSite.Initialization;
var gen = new CategoryMembersGenerator(zhWarriorsSite, "没有图片的猫物")
{
PaginationSize = 50,
MemberTypes = CategoryMemberTypes.Page,
};
var enu = gen.EnumPagesAsync(PageQueryOptions.FetchContent);
var fixedPages = @"
焦风
微光毛_(黑莓星的风暴)
".Split(new[] { '\r', '\n' }, StringSplitOptions.RemoveEmptyEntries).Select(t => new WikiPage(zhWarriorsSite, t)).ToList();
await fixedPages.RefreshAsync(PageQueryOptions.FetchContent | PageQueryOptions.ResolveRedirects);
enu = fixedPages.ToAsyncEnumerable();
var counter = 0;
await foreach (var page in enu)
{
counter++;
var query = CPRepository.CreateQuery(@"
SELECT ?link {
?link schema:isPartOf <https://warriors.huijiwiki.com/>;
schema:name @title.
}");
query.SetLiteral("title", page.Title, "zh");
if (CPRepository.ExecuteQuery(query).Any())
{
Logger.LogWarning("Exists {}", page);
continue;
}
Logger.LogInformation("[{}] Processing {}", counter, page);
RETRY:
try
{
await ExportEntityAsync(page);
}
catch (WikiClientException ex)
{
Console.WriteLine(ex);
Console.ReadKey();
await page.RefreshAsync(PageQueryOptions.FetchContent);
goto RETRY;
}
}
}
static async Task Main(string[] args)
{
var client = new WikiClient {
ClientUserAgent = "ClientWikiMaintenance/1.0"
};
var site = new WikiSite(client, Credentials.ApiEndpoint);
await site.Initialization;
await site.LoginAsync(Credentials.UserName, Credentials.Password);
site.ModificationThrottler.ThrottleTime = TimeSpan.FromSeconds(1);
var gen = new CategoryMembersGenerator(site, "猫物")
{
MemberTypes = CategoryMemberTypes.Page,
PaginationSize = 50
};
await WorkAsync(gen.EnumPagesAsync(PageQueryOptions.FetchContent));
}
public async Task RunAsync()
{
await zhWarriorsSite.Initialization;
var gen = new CategoryMembersGenerator(zhWarriorsSite, "猫物")
{
PaginationSize = 50,
MemberTypes = CategoryMemberTypes.Page,
};
var counter = 0;
using (var ie = gen.EnumPagesAsync(PageQueryOptions.FetchContent).GetEnumerator())
{
while (await ie.MoveNext())
{
counter++;
var page = ie.Current;
var query = CPRepository.CreateQuery(@"
SELECT ?link {
?link schema:isPartOf <https://warriors.huijiwiki.com/>;
schema:name @title.
}");
query.SetLiteral("title", page.Title, "zh");
if (CPRepository.ExecuteQuery(query).Any())
{
Logger.LogWarning("Exists {}", page);
continue;
}
Logger.LogInformation("[{}] Processing {}", counter, page);
RETRY:
try
{
await ExportEntityAsync(page);
}
catch (WikiClientException ex)
{
Console.WriteLine(ex);
Console.ReadKey();
await page.RefreshAsync(PageQueryOptions.FetchContent);
goto RETRY;
}
}
}
}
public async Task WikiaCategoryMembersGeneratorTest()
{
var site = await WikiaTestSiteAsync;
var cat = new WikiPage(site, "Category:BlogListingPage");
await cat.RefreshAsync();
Output.WriteLine(cat.ToString());
var generator = new CategoryMembersGenerator(cat)
{
PaginationSize = 50
};
var pages = await generator.EnumPagesAsync().ToList();
TracePages(pages);
AssertTitlesDistinct(pages);
var catInfo = cat.GetPropertyGroup <CategoryInfoPropertyGroup>();
Assert.Equal(catInfo.MembersCount, pages.Count);
}
public async Task WpCategoryMembersGeneratorTest()
{
var site = await WpTest2SiteAsync;
var cat = new WikiPage(site, "Category:Template documentation pages");
await cat.RefreshAsync();
WriteOutput(cat);
var generator = new CategoryMembersGenerator(cat)
{
PaginationSize = 50
};
var pages = await generator.EnumPagesAsync().ToListAsync();
TracePages(pages);
AssertTitlesDistinct(pages);
var catInfo = cat.GetPropertyGroup <CategoryInfoPropertyGroup>();
Assert.Equal(catInfo.MembersCount, pages.Count);
}
static async Task HelloWikiGenerators()
{
// Create a MediaWiki API client.
var wikiClient = new WikiClient();
// Create a MediaWiki Site instance.
var site = new WikiSite(wikiClient, "https://en.wikipedia.org/w/api.php");
await site.Initialization;
// List all pages starting from item "Wiki", without redirect pages.
var allpages = new AllPagesGenerator(site)
{
StartTitle = "Wiki",
RedirectsFilter = PropertyFilterOption.WithoutProperty
};
// Take the first 1000 results
var pages = await allpages.EnumPagesAsync().Take(1000).ToList();
foreach (var p in pages)
{
Console.WriteLine("{0, -30} {1, 8}B {2}", p, p.ContentLength, p.LastTouched);
}
// List the first 10 subcategories in Category:Cats
Console.WriteLine();
Console.WriteLine("Cats");
var catmembers = new CategoryMembersGenerator(site, "Category:Cats")
{
MemberTypes = CategoryMemberTypes.Subcategory
};
pages = await catmembers.EnumPagesAsync().Take(10).ToList();
foreach (var p in pages)
{
Console.WriteLine("{0, -30} {1, 8}B {2}", p, p.ContentLength, p.LastTouched);
}
}
public async Task CollectAsync(string outputPath)
{
var site = await WikiSite.CreateAsync(wikiClient, EndpointUrl);
site.Logger = WikiLogger.Default;
var gen = new CategoryMembersGenerator(site, "Characters")
{
PagingSize = 50,
MemberTypes = CategoryMemberTypes.Page,
};
var counter = 0;
using (var writer = File.CreateText(outputPath))
{
writer.WriteLine("% Powered by Graywing Prolog Generator.");
writer.WriteLine("% Generated from Warriors Wiki on {0:O}.", DateTime.Now);
writer.WriteLine();
writer.WriteLine(":- discontiguous[name/2, male/1, female/1, belongsto/2, child/3, apprentice/2].");
writer.WriteLine();
await gen.EnumPagesAsync(PageQueryOptions.FetchContent)
//.Take(100)
.ForEachAsync(page =>
{
counter++;
Console.WriteLine("{0}: {1}", counter, page.Title);
writer.Write("% ");
writer.WriteLine(page.Title);
foreach (var l in FactsFromPage(page))
{
writer.WriteLine(l);
}
writer.WriteLine();
writer.Flush();
});
}
}