/// <summary>
/// Decide if
/// Is the time to get file informations (the current page represent a file content)
/// Or need to continue the process (the current page represent a folder list)
/// </summary>
/// <param name="parser">Contain full DOM Elements (IDocument) and helpers to use it</param>
/// <param name="root">Indicate if the URL is the main/master/first/root</param>
/// <returns></returns>
protected virtual async Task DeterminePageContentAsync(GitHubParser parser, bool root)
{
if (root)
{
// Here, i use the cache! The key is Last commit hash.
// Uses the URL like Key is fastest then last commit hash, because to get last commit, i need to load the first page
// But if new commit was push, the URL key returns outdated result
this.lastCommitHash = parser.GetLastCommitHash();
var cachedFileInformation = this.cache.Get <ConcurrentBag <ItemFileInformationResponse> >(lastCommitHash);
if (cachedFileInformation != default)
{
this.temporaryFiles = cachedFileInformation;
this.usedCache = true;
return;
}
}
var pageType = parser.DiscoverPageType();
switch (pageType)
{
case GitHubParser.GitHubPageType.FileContent:
this.GetSynthesizedFileInformation(parser);
break;
case GitHubParser.GitHubPageType.FolderList:
await this.IterateFolderListAsync(parser);
break;
}
}
public async Task ShouldGetLastCommitHash()
{
// Given
var source = @"
<div class=""Box-header Box-header--blue position-relative"">
<h2 class=""sr-only"">Latest commit</h2>
<div class=""js-details-container Details d-flex rounded-top-1 flex-items-center flex-wrap"" data-issue-and-pr-hovercards-enabled="""">
<div class=""flex-1 d-flex flex-items-center ml-3 min-width-0"">
<div class=""css-truncate css-truncate-overflow text-gray"">
<span class=""commit-author user-mention"">Paulo Justino</span>
<span class=""d-none d-sm-inline"">
<a data-pjax=""true"" title=""Documentation"" class=""link-gray-dark"" href=""/paulojsilva/web-scraping-nolayer/commit/1cf47cd55b5c745bb8dfb734c8f96614c6b30273"">Documentation</a>
</span>
</div>
<span class=""hidden-text-expander ml-2 d-inline-block d-inline-block d-lg-none"">
<button type=""button"" class=""hx_bg-black-fade-15 text-gray-dark ellipsis-expander js-details-target"" aria-expanded=""false"">…</button>
</span>
<div class=""d-flex flex-auto flex-justify-end ml-3 flex-items-baseline"">
<a href=""/paulojsilva/web-scraping-nolayer/commit/1cf47cd55b5c745bb8dfb734c8f96614c6b30273"" class=""f6 link-gray text-mono ml-2 d-none d-lg-inline"" data-pjax="""">1cf47cd</a>
<a href=""/paulojsilva/web-scraping-nolayer/commit/1cf47cd55b5c745bb8dfb734c8f96614c6b30273"" class=""link-gray ml-2"" data-pjax="""">
<relative-time datetime=""2021-02-02T14:54:52Z"" class=""no-wrap"" title=""2 de fev. de 2021 11:54 BRT"">12 hours ago</relative-time>
</a>
</div>
</div>
</div>
</div>";
var document = await context.OpenAsync(req => req.Content(source));
var parser = new GitHubParser(document, "github.com");
// When
var lastCommitHash = parser.GetLastCommitHash();
// Then
lastCommitHash.Should().Be("1cf47cd55b5c745bb8dfb734c8f96614c6b30273");
}