meds-processor, part 1 of 4, snippet 9

HzzoHtmlScraper.cs

ISet<HzzoMedsDownloadDto> ParseHtmlDocuments(IDocument[] docs) =>
	docs.Aggregate(
		new HashSet<HzzoMedsDownloadDto>(),
		(docList, doc) => new HashSet<HzzoMedsDownloadDto>(docList.Concat(ParseHtmlDocument(doc)))
	);

static ISet<HzzoMedsDownloadDto> ParseMedsLiElements(IEnumerable<IElement> elems) =>
  elems.Aggregate(new HashSet<HzzoMedsDownloadDto>(), (medsList, li) =>
  {
    var href = li.QuerySelector("a").GetAttribute("href");

    // NOTE: this domain is not available, links don't work :-(
    if (!href.Contains("cdn.hzzo.hr"))
    {
      var downloadDto = new HzzoMedsDownloadDto(
        href,
        li.TextContent.TrimEnd().Split(' ').LastOrDefault(),
        DOWNLOAD_DIR
      );

      // NOTE: that's it folks, docs from 2013 and older are messed up
      // and can't be approached with this generic parser in this app
      // A more sophisticated parser (more if/else loops...) would be needed
      if (downloadDto.ValidFrom > filterDtParsable2013)
        medsList.Add(downloadDto);
    }

    return medsList;
  });

static ISet<HzzoMedsDownloadDto> ParseHtmlDocument(IDocument doc) =>
  ParseMedsLiElements(SelectLiElements(doc));

static IEnumerable<IElement> SelectLiElements(IDocument doc) =>
  doc.QuerySelectorAll("section#main > ul li").Where(_predicateForListLiMeds);

static Func<IElement, bool> _predicateForListLiMeds =
  x =>
  // primary list:
  x.TextContent.Contains("Osnovna lista lijekova") ||
  // supplementary list:
  x.TextContent.Contains("Dopunska lista lijekova");