Custom recipes (archive, read-only)

[Starson17]

Originally Posted by lenik

calibre 0.6.13/ubuntu 9.10.

calibre 0.6.13 is 35 versions old. You may be able to find help, but running a more current version will provide lots of benefits, even if it doesn't solve your problem.

[kiklop74]

The thing is that linux users always try to use what is available in the distribution repository and the versions over there are always quite older than the actual.

To all linux users I can just say - do not use calibre from distribution repository it is always better to manually install latest version.

[gambarini]

Originally Posted by Starson17

That page sends you to another page:

http://www.ilgiornale.it/interni/romiti_il_corriere_non_volevo_e_stato_bazoli_fare_tutto/18-04-2010/articolo-id=438656-page=0-comments=1

I see a print link on that page at:

http://www.ilgiornale.it/interni/romiti_il_corriere_non_volevo_e_stato_bazoli_fare_tutto/18-04-2010/articolostampa-id=438656-page=1-comments=1

yes, but i am not able to find the print page in the recipe.
I have tried with a replace and with a soup.find but ... nothing... !

[Starson17]

Originally Posted by gambarini

yes, but i am not able to find the print page in the recipe.
I have tried with a replace and with a soup.find but ... nothing... !

I suspect your problem is with the part where I wrote :"That page sends you to another page:" That's why I posted that part. The first thing I would do is see if you are being redirected to the page that has the print link on it. It is quite possible that there is a cookie or a referer issue when Calibre gets page 1 that prevents it from getting page 2 that has the actual print link. There are many ways that a browser will work, but a simple calibre recipe will not. You need to check that out and see if you are getting to the next page. If not, you need to solve that problem first with a more complex recipe.

Edit: Also, if you are getting to the right page, but the print link isn't there, you may want to turn off the script stripping. I've seen that code strip off links I needed.

[Starson17]

Originally Posted by kiklop74

The thing is that linux users always try to use what is available in the distribution repository and the versions over there are always quite older than the actual.

To all linux users I can just say - do not use calibre from distribution repository it is always better to manually install latest version.

Yes. At some point a linux/calibre user may step up and begin maintaining and updating his distribution repository on a more regular basis, but until then, it's up to the individual user to install a reasonably current version.

[gambarini]

i have made some adjustments on the La_Repubblica.recipe
Some feeds now work (feedsportal, sport for example)
add "Roma" and "Torino" local version of newspaper.
What is the correct procedure to post the new recipe?

[kiklop74]

You have two options.

1. Post updated recipe in this thread compressed in zip format
2. Open an issue in calibre trac and attach a file to it

[gambarini]

New Recipe for La_Repubblica, an italian news paper.
new Motori feed, some feedsportal now work, "Roma" e "Torino" local news.

[lenik]

Originally Posted by Starson17

calibre 0.6.13 is 35 versions old. You may be able to find help, but running a more current version will provide lots of benefits, even if it doesn't solve your problem.

was the problem with getting dozens of files for every news feed actually solved in any of these 35 versions?

[DoctorOhh]

Originally Posted by lenik

was the problem with getting dozens of files for every news feed actually solved in any of these 35 versions?

Since I have no clue what you're describing it might have been fixed. As an example I download my local newspaper, the recipe has 8 separate feeds and I end up with one epub document that contains them all.

See the attached for screen shots of how this epub (containing 8 feeds) looks on my device.

[Starson17]

Originally Posted by lenik

was the problem with getting dozens of files for every news feed actually solved in any of these 35 versions?

I've been reading most messages here for the last 6 months and I've never seen this problem. Unless someone here has seen it, there's no way to tell if your problem would be fixed by upgrading. OTOH, you'll be able to get more help for a problem seen on a current version than a problem that occurred 35 versions ago.

[gambarini]

Originally Posted by Starson17

I suspect your problem is with the part where I wrote :"That page sends you to another page:" That's why I posted that part. The first thing I would do is see if you are being redirected to the page that has the print link on it. It is quite possible that there is a cookie or a referer issue when Calibre gets page 1 that prevents it from getting page 2 that has the actual print link. There are many ways that a browser will work, but a simple calibre recipe will not. You need to check that out and see if you are getting to the next page. If not, you need to solve that problem first with a more complex recipe.

Edit: Also, if you are getting to the right page, but the print link isn't there, you may want to turn off the script stripping. I've seen that code strip off links I needed.

in the calibre output i can find the print link;
this is an example

<div class="calibre26">
<a href="/sport/inter_strepitosa_coppa_sneijder_maicon_e_milito_il_bara_e_annichilito_3-1/calcio-inter-champions_league-maicon-ibrahimovic-mourinho-etoo-barcellona-milito-guardiola-semifinale-messi-sneijder/20-04-2010/articolostampa-id=439200-page=1-comments=1"><img src="images/img10.jpg" border="0" class="calibre15"/>*Stampa</a>

so, i think that the print link is in the redirect link of the original url.
Can i use attribute calibre26 to find the href link? No, because this output was generated by calibre...
...
i am a bit confused

[Starson17]

Originally Posted by gambarini

in the calibre output i can find the print link;

If it's in the calibre output, it's in the input, too.

so, i think that the print link is in the redirect link of the original url.

Yes, I agree.

Can i use attribute calibre26 to find the href link? No, because this output was generated by calibre...

Correct, you have to find it in the input.

i am a bit confused

I would simply use the pre or post process html to print the soup for your page. That's one of my common steps during debug. Redirect your output to a text file (my default setup is always like this) and search for the print link. It must be there or it wouldn't get to Calibre.

I just used Firebug to find it.

It's in the first "A" child tag of the second "DIV" child tag of the DIV tag having id=strumenti_utili_bottom. Learn the BeautifulSoup navigation options to find it.

[gambarini]

Originally Posted by Starson17

If it's in the calibre output, it's in the input, too.


Yes, I agree.



Correct, you have to find it in the input.



I would simply use the pre or post process html to print the soup for your page. That's one of my common steps during debug. Redirect your output to a text file (my default setup is always like this) and search for the print link. It must be there or it wouldn't get to Calibre.

I just used Firebug to find it.

It's in the first "A" child tag of the second "DIV" child tag of the DIV tag having id=strumenti_utili_bottom. Learn the BeautifulSoup navigation options to find it.

thanks.
i must investigate the navitation options... I am not yet finded the correct syntax... The print link is too annidated for my actual knowledge!!!

[Starson17]

Originally Posted by gambarini

thanks.
i must investigate the navitation options... I am not yet finded the correct syntax... The print link is too annidated for my actual knowledge!!!

Here is the documentation. I've read that page many times and keep going back. There would be many ways to find the tag you need with BeautifulSoup navigation. You can use a regular expression to match it, or a findNextSibliing, etc. I am certainly not an expert, but I would try to construct a command that finds this tag, then print the tag found, then see if what gets printed is what you expected. If not, adjust, etc. until it prints the right tag.