AZEN-SGG/Kavita
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Fixed Docker not being able to parse PDF Images (#418)

Browse source 
* Fixed an issue with downloading pdf files

* Bumped sentry version to latest

* Refactored PDF Page to image extraction to a private method. Bumped Sentry version in backend.

* Fixed issue with PDFs not being parsable (image) within Docker due to libgdiplus not being installed.
This commit is contained in:
Joseph Milazzo 2021-07-23 10:35:26 -05:00 • committed by GitHub
parent b0df67cdda
commit 199ab7946c
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
9 changed files with 134 additions and 139 deletions
Download patch file Download diff file Expand all files Collapse all files

80
API/Services/BookService.cs
View file

@ -14,6 +14,7 @@ using API.Interfaces.Services;
using API.Parser;
using Docnet.Core;
using Docnet.Core.Models;
using Docnet.Core.Readers;
using ExCSS;
using HtmlAgilityPack;
using Microsoft.Extensions.Logging;
@ -375,22 +376,7 @@ namespace API.Services
for (var pageNumber = 0; pageNumber < pages; pageNumber++)
{
using var pageReader = docReader.GetPageReader(pageNumber);
var rawBytes = pageReader.GetImage();
var width = pageReader.GetPageWidth();
var height = pageReader.GetPageHeight();
using var doc = new Bitmap(width, height, PixelFormat.Format32bppArgb);
using var bmp = new Bitmap(width, height, PixelFormat.Format32bppArgb);
AddBytesToBitmap(bmp, rawBytes);
for (int y = 0; y < bmp.Height; y++)
{
bmp.SetPixel(bmp.Width - 1, y, bmp.GetPixel(bmp.Width - 2, y));
}
var g = Graphics.FromImage(doc);
g.FillRegion(Brushes.White, new Region(new Rectangle(0, 0, width, height)));
g.DrawImage(bmp, new Point(0, 0));
g.Save();
using var stream = new MemoryStream();
doc.Save(stream, ImageFormat.Jpeg);
using var stream = GetPdfPage(docReader, pageNumber);
File.WriteAllBytes(Path.Combine(targetDirectory, "Page-" + pageNumber + ".png"), stream.ToArray());
}
}
@ -417,15 +403,12 @@ namespace API.Services
if (coverImageContent == null) return Array.Empty<byte>();
if (createThumbnail)
{
using var stream = new MemoryStream(coverImageContent.ReadContent());
if (!createThumbnail) return coverImageContent.ReadContent();
using var thumbnail = Image.ThumbnailStream(stream, MetadataService.ThumbnailWidth);
return thumbnail.WriteToBuffer(".jpg");
}
using var stream = new MemoryStream(coverImageContent.ReadContent());
using var thumbnail = Image.ThumbnailStream(stream, MetadataService.ThumbnailWidth);
return thumbnail.WriteToBuffer(".jpg");
return coverImageContent.ReadContent();
}
catch (Exception ex)
{
@ -442,32 +425,14 @@ namespace API.Services
using var docReader = DocLib.Instance.GetDocReader(fileFilePath, new PageDimensions(1080, 1920));
if (docReader.GetPageCount() == 0) return Array.Empty<byte>();
using var pageReader = docReader.GetPageReader(0);
var rawBytes = pageReader.GetImage();
var width = pageReader.GetPageWidth();
var height = pageReader.GetPageHeight();
using var doc = new Bitmap(width, height, PixelFormat.Format32bppArgb);
using var bmp = new Bitmap(width, height, PixelFormat.Format32bppArgb);
AddBytesToBitmap(bmp, rawBytes);
for (int y = 0; y < bmp.Height; y++)
{
bmp.SetPixel(bmp.Width - 1, y, bmp.GetPixel(bmp.Width - 2, y));
}
var g = Graphics.FromImage(doc);
g.FillRegion(Brushes.White, new Region(new Rectangle(0, 0, width, height)));
g.DrawImage(bmp, new Point(0, 0));
g.Save();
using var stream = new MemoryStream();
doc.Save(stream, ImageFormat.Jpeg);
using var stream = GetPdfPage(docReader, 0);
stream.Seek(0, SeekOrigin.Begin);
if (createThumbnail)
{
using var thumbnail = Image.ThumbnailStream(stream, MetadataService.ThumbnailWidth);
return thumbnail.WriteToBuffer(".png");
}
if (!createThumbnail) return stream.ToArray();
using var thumbnail = Image.ThumbnailStream(stream, MetadataService.ThumbnailWidth);
return thumbnail.WriteToBuffer(".png");
return stream.ToArray();
}
catch (Exception ex)
{
@ -479,6 +444,29 @@ namespace API.Services
return Array.Empty<byte>();
}
private static MemoryStream GetPdfPage(IDocReader docReader, int pageNumber)
{
using var pageReader = docReader.GetPageReader(pageNumber);
var rawBytes = pageReader.GetImage();
var width = pageReader.GetPageWidth();
var height = pageReader.GetPageHeight();
using var doc = new Bitmap(width, height, PixelFormat.Format32bppArgb);
using var bmp = new Bitmap(width, height, PixelFormat.Format32bppArgb);
AddBytesToBitmap(bmp, rawBytes);
for (int y = 0; y < bmp.Height; y++)
{
bmp.SetPixel(bmp.Width - 1, y, bmp.GetPixel(bmp.Width - 2, y));
}
using var g = Graphics.FromImage(doc);
g.FillRegion(Brushes.White, new Region(new Rectangle(0, 0, width, height)));
g.DrawImage(bmp, new Point(0, 0));
g.Save();
var stream = new MemoryStream();
doc.Save(stream, ImageFormat.Jpeg);
return stream;
}
private static string RemoveWhiteSpaceFromStylesheets(string body)
{
body = Regex.Replace(body, @"[a-zA-Z]+#", "#");