POC doc88.com Downloader

This is a POC downloader of documents from doc88.com. It saves pages of a given document as PNGs or JPEGs. It doesn't have any dependencies — it's a bit of JavaScript that you paste into Developer Tools' Console. It was tested in Chrome and Firefox.

Then, having pages saved as images, a searchable PDF can be reconstructed from them.

Step 1: Save pages of a document as images

Option A: Bookmark

Create a browser bookmark, pasting content of this file (exactly as it is) in its URL field.

From now on, clicking the bookmark on a document page will capture all pages as JPEGs, bundle them in a ZIP archive and download it.

Important

Don't interact with the browser during the process.
Be patient, especially with large documents containing hundreds of pages.
You can assess the progress of the process in doc88's page selector (e.g. "17 / 42").
Check that all desired pages were captured correctly.

Option B: Manual (finer control over the process)

Navigate to the desired document in your browser.
Make sure browser's zoom level is set to 100% — based on some tests it seems that zoom levels lower than 100% can result in lower quality of captured pages.
Open Developer Tools (e.g. press Ctrl+Shift+I).
Switch to JavaScript Console.
Paste this JavaScript in Console and confirm with Enter.
Type the following in Console and hit Enter:
```
downloadPages()
```
This will capture and bundle all the pages in a ZIP file.
Pages will be automatically preloaded and captured one by one.

See options section below for options.

Important

Don't interact with the browser during the process.
Be patient, especially with large documents containing hundreds of pages.
Wait until it ends, printing Finished downloading pages in the Console.
Check that all desired pages were captured correctly.

Options

downloadPages function takes an optional options object:

downloadPages({fromPage: 2, toPage: 10, quality: 0.8, imageNamePrefix: 'temp_'})

Possible options are:

fromPage – first page in range to be downloaded; number; default is 1
toPage – last page in range to be downloaded; number; default is total number of pages in the document
format – downloaded image format; string; either 'jpg' or 'png'; default is 'jpg'
quality – quality of images; applicable when format is 'jpg'; number between 0 and 1; default is 0.9
imageNamePrefix – prefix for names of downloaded images; string; default is 'page' (resulting in downloaded file names e.g.: page001.jpg, page002.jpg, etc. assuming format is 'jpg')
archive – type of archive to put the captured images in; string; either 'zip' or 'none'; default is 'zip'; 'none' will result in each image downloaded as a separate file

Note

In case of Chrome, if you set archive to none, the first time you download pages you may see a popup stating that "This site is attempting to download multiple files". You have to allow it, as with this option each page will be downloaded as a separate file.

Step 2: Converting images back to a PDF

Under Linux you can easily convert downloaded images back to a PDF.

Install ImageMagick package:
```
sudo apt-get install imagemagick
```
If you want the PDF to be OCRed (recognize the text in it and make it searchable), install the OCRmyPDF package:
```
sudo apt-get install ocrmypdf
```
Use the convert-images-to-pdf.sh script to convert downloaded images back to a PDF, e.g.:
```
./convert-images-to-pdf.sh image-directory output.pdf
```
Run it with -h argument for help.

Troubleshooting

If you see errors from ImageMagick with the message "attempt to perform an operation not allowed by the security policy 'PDF'", see this StackOverflow question and answers for a likely quick fix.

Developing

Run build-bookmark.sh to update minified bookmark code in bookmark.min.js.

apankowski/doc88-downloader