How to force calibre to download images?

[fenuks]

Hello. I made recipe for page which uses flash player to display images (cgm.pl). In preprocess_html function I extract image ID from flash tag and replace it with image tag. Problem is that that images aren't downloaded. I meant they aren't downloaded to file. When I open it calibre ebook viewer download all photos to memory. How to force calibre to download those images to file? Thanks for help!

preprocess_html function

Spoiler:

def preprocess_html(self, soup):
        gallery=soup.find('div', attrs={'class':'galleryFlash'})
        if gallery:
            img=gallery.find('embed')
            if img:
                img=img['src'][35:]
                gallery.replaceWith('<img src="http://www.cgm.pl/_vault/_gallery/_photo/'+img+'" />')
            
        return soup

flash tag

Spoiler:

<div class="galleryFlash">
                                        <object classid="clsid:d27cdb6e-ae6d-11cf-96b8-444553540000" codebase="http://fpdownload.macromedia.com/pub/shockwave/cabs/flash/swflash.cab#version=8,0,0,0" width="689" height="442" id="single_photo" align="middle">
                                            <param name="allowScriptAccess" value="sameDomain">
                                            <param name="movie" value="_flash/single_photo.swf?photo_file=21578.jpg">

                                            <param name="quality" value="high">
                                            <param name="bgcolor" value="#000000">
                                            <embed src="_flash/single_photo.swf?photo_file=21578.jpg" quality="high" bgcolor="#000000" width="689" height="442" name="single_photo" align="middle" allowscriptaccess="sameDomain" type="application/x-shockwave-flash" pluginspage="http://www.macromedia.com/go/getflashplayer">
                                        </object>
                                    </div>

[kovidgoyal]

That should work just fine. The news download system looks for images to download after preprocess has run. Look in the log to see why the images are not downloading. Also rather than using replaceWith just set

img.name = 'img'
img['src'] = 'whatever'

[fenuks]

Thanks. That works.

Originally Posted by kovidgoyal

img.name = 'img'
img['src'] = 'whatever'

I looked at log and when I used replaceWith() function calibre didn't know that it must fetch those images. Same thing when I did insert

sometag.insert(pos, '<img src="link" />')

. When I parsed this tag into soup it inserts '\n'. Anyway, first method works. Best regards.

[kiavash]

Thanks for this post.

This is how I achived the same. You can see I used the same trick to make article title better standing out.

Spoiler:

PHP Code:

    def preprocess_html(self, soup):
        # Finds all the jpg links
        for figure in soup.findAll('a', attrs = {'href' : lambda x: x and 'jpg' in x}):
            
            # makes sure that the link points to the absolute web address
            if figure['href'].startswith('/'):
                figure['href'] = self.site + figure['href']
            
            figure.name = 'img' # converts the links to img
            figure['src']= figure['href'] # with the same address as href        
        
        # Makes the title standing out
        title = soup.find('a', attrs = {'class': 'commonSectionTitle'})
        title.name = 'h1'

        return soup 


[kiavash]

Originally Posted by kovidgoyal

That should work just fine. The news download system looks for images to download after preprocess has run. Look in the log to see why the images are not downloading. Also rather than using replaceWith just set

img.name = 'img'
img['src'] = 'whatever'

After fetching all the images using the above code, they all become inline with the text. I would like to put a new line between an image and the text before/after. Tried couple of techniques including Tag(soup,'br /') and tag.insert but all ended up eliminating the image all together in the final file.

I also attached the example epub that shows the behavior I am referring to.

Spoiler:

PHP Code:

    def preprocess_html(self, soup):
        # Includes all the figures inside the final ebook
        # Finds all the jpg links
        for figure in soup.findAll('a', attrs = {'href' : lambda x: x and 'jpg' in x}):
            
            # makes sure that the link points to the absolute web address
            if figure['href'].startswith('/'):
                figure['href'] = self.site + figure['href']
                
            figure.name = 'img' # converts the links to img
            figure['src'] = figure['href'] # with the same address as href
            del figure['href']
            del figure['target']
        return soup 


Any idea?

[kovidgoyal]

img['style'] = 'display:block'

[kiavash]

Perfect! thanks!