03-18-2009, 01:34 PM | #61 |
Wizard
Posts: 4,552
Karma: 950151
Join Date: Nov 2008
Device: Sony PRS-950, iphone/ipad (Marvin/iBooks/QuickReader)
|
Hopefully that means you now have things up and running? Do you like the results?
|
03-23-2009, 10:23 AM | #62 |
Connoisseur
Posts: 72
Karma: 10
Join Date: Dec 2008
Device: iPhone, Note3
|
Has anyone had a problem with some links working, when others do not?
My website recently got shifted to a new host. When this happened, some of my links no longer worked. The stanza catalog shows a question mark for the cover, and when I try to download a book, I get a notice that the epub contained no data. I can browse the catalog using Firefox, and I checked the links. If I click on Books by Series, then Series: J, then Jane Madison, and then 2: Sorcery And The Single Girl, I get a 404 error. Here is the link that the catalog generates: Code:
http://mywebsite/my_ebook_directory/Mindy%20Klasky/Sorcery%20And%20The%20Single%20Girl%20%28952%29/Sorcery%20And%20The%20Single%20Girl%20-%20Mindy%20Klasky.epub Code:
http://mywebsite/my_ebook_directory/ Code:
http://mywebsite/my_ebook_directory/Mindy%20Klasky/Sorcery%20and%20the%20Single%20Girl%20(952)/Sorcery%20And%20The%20Single%20Girl%20-%20Mindy%20Klasky.epub Compare this to one that works: Jim Butcher's Summer Knight. Here is the catalog link: Code:
http://mywebsite/my_ebook_directory/Jim%20Butcher/Summer%20Knight%20%28183%29/Summer%20Knight%20-%20Jim%20Butcher.epub Code:
http://mywebsite/my_ebook_directory/Jim%20Butcher/Summer%20Knight%20(183)/Summer%20Knight%20-%20Jim%20Butcher.epub What am I missing here? I've verified the books (and cover images) actually exist using my FTP client. I can see it by browsing the directory. But the catalog entry leads to a 404 page. I've got a support ticket in with my webhost, but I'm curious if anyone else has seen this problem. They did mention that they generally discourage spaces in file names, since that could create unusual errors, but since it worked on my last host, they're still looking into it. |
Advert | |
|
03-23-2009, 10:42 AM | #63 |
Wizard
Posts: 4,552
Karma: 950151
Join Date: Nov 2008
Device: Sony PRS-950, iphone/ipad (Marvin/iBooks/QuickReader)
|
There is one difference between the first two URL's - the first one has "And" in the URL while the second one has "and". This will matter if (as is commonly the case) the file names are case significant on the web server.
I have found this can happen if using Calibre on Windows and when updating the metadata the title or author are updated with the only difference being case. Since Windows file systems are case insignificant the file rename is effectively ignored by Windows and you end up with the filename and the calibre metadata disagreeing on the case. The Calibre2Web script always uses the case from the Metadata when generating the catalog files. As this problem does not occur when creating new files, the workaround is to edit the title (or author if that is where the discrepancy occurs) to some other value (I normally just add an "X" on the end) and then after committing that change changing it back. This will make the metadata and filenames agree on case. The point you mention about spaces (or other special characters) should not matter as they should always be escaped using %xx syntax as shown in your examples. |
03-23-2009, 11:04 AM | #64 |
Connoisseur
Posts: 72
Karma: 10
Join Date: Dec 2008
Device: iPhone, Note3
|
Very good catch! I thought I examined the two URLs closer than that.
That makes perfect sense, and I'm doing an FTP sync now. |
03-23-2009, 11:36 AM | #65 |
Connoisseur
Posts: 72
Karma: 10
Join Date: Dec 2008
Device: iPhone, Note3
|
It worked. Although I had to manually change the directory name on the server.
Thanks! |
Advert | |
|
04-10-2009, 12:36 PM | #66 |
Member
Posts: 22
Karma: 10
Join Date: Apr 2009
Device: iphone
|
I rarely do become a member to online forums but in this case I just had too.
I want to say thanks itimpi. Your scrip does exactly what I wanted. When you have 1000+ ebooks it's a little long to pull the calibre webserver feed and finding the ones you want. Took me about 4 hours to create my own webserver and to configure everything for it (for a first timer not to bad *grin*) but once everything was setup.. Again itimpi... THANKS! |
05-23-2009, 03:03 AM | #67 |
Evangelist
Posts: 405
Karma: 692
Join Date: Sep 2006
Device: Samsung Galaxy Note 3 | Kindle Paperwhite | iPad Mini
|
Is there a reason that some catalogs fully download and others do not or why some ebooks have the download ability and others do not?
|
05-23-2009, 04:45 AM | #68 | ||
Wizard
Posts: 4,552
Karma: 950151
Join Date: Nov 2008
Device: Sony PRS-950, iphone/ipad (Marvin/iBooks/QuickReader)
|
Quote:
Make sure that you are using the latest copy of the script. Some of the recent updates were to fix issues around correctly escaping unusual characters at the XML level which could otherwise stop catalogs loading. Quote:
If there is no download link - then that will be because the script currently checks to see if there is an epub file, and only generates the download link if there is one. It is relatively trivial to change the script to either filter out listing books for which there is no epub book, or to allow for other file types. I personally like to see all entries from the Calibre library even if I do not yet have books in the correct electronic format. If there is a download link, and it then fails to start the download it can mean that there is a case discrepancy between the file mentioned in the catalog XML and the actual filename on the web server. Note this only matters if you are running the web server on a system where case in filenames is significant. Correcting the case in such cases will fix the issue. The script always uses the case defined in the calibre metadata.db file, so ideally you should change any files which have the wrong case in their name to match the calibre database. Let me know if this is what is happening for you, as it would be possible to enhance the script to look at the actual files on the hard disk, and use the case from there instead - I will think about that. |
||
05-23-2009, 10:08 AM | #69 | |
Member
Posts: 22
Karma: 10
Join Date: Apr 2009
Device: iphone
|
Quote:
If this is for the iPhone it might be due to the fact that you have not converted to epub format. After doing so try rerunning the script again. |
|
05-23-2009, 10:43 AM | #70 |
Evangelist
Posts: 405
Karma: 692
Join Date: Sep 2006
Device: Samsung Galaxy Note 3 | Kindle Paperwhite | iPad Mini
|
Thanks for the reply. I'll try to recreate what I'm talking about. Let's say I go to Authors and click on the Letter S. The catalog loads but I get "Catalog Error-Error reading catalog data. Please try again later." The catalog's loading is abbreviated in that not all the S authors names are loaded. Then, if I load an individual author's name, from time to time, I'll get a "catalog empty" & the "Catalog Error" message. I know that there are two epub files in say "Sarah Morgan's" folder but the books aren't appearing in Stanza.
As to the other issue of a book not being in epub form but still being in the library, thanks for that explanation. Makes perfect sense and I like looking at the books too. Serves as a not so subtle reminder to convert some books. |
05-23-2009, 12:18 PM | #71 | |
hopeless n00b
Posts: 5,111
Karma: 19597086
Join Date: Jan 2009
Location: in the middle of nowhere
Device: PW4, PW3, Libra H2O, iPad 10.5, iPad 11, iPad 12.9
|
Quote:
|
|
05-23-2009, 01:38 PM | #72 | ||
Wizard
Posts: 4,552
Karma: 950151
Join Date: Nov 2008
Device: Sony PRS-950, iphone/ipad (Marvin/iBooks/QuickReader)
|
Quote:
If you are prepared to upload the Author-S.xml file I can see if it appears to be correct. Another check you can try is to try and load it into Internet Explorer. If there is an error at the XML level then that should show it. Quote:
What are you using as your web server? Knowing that might give an inidcation as to whether you are likely to be hitting the issue I mentioned with the case of files differing at the Calibre level and the File System level. |
||
05-23-2009, 04:26 PM | #73 | |
Evangelist
Posts: 405
Karma: 692
Join Date: Sep 2006
Device: Samsung Galaxy Note 3 | Kindle Paperwhite | iPad Mini
|
Quote:
In another file, (Author-D), the XML is corrupted or cuts off after an apostrophe such as Jacquie D'Allessandro. In the Sarah Morgan file, I couldn't figure out the problem. I've uploaded the file here. Thanks for helping me out guys. You are teaching me how to fish, a little. |
|
05-23-2009, 05:45 PM | #74 |
Wizard
Posts: 4,552
Karma: 950151
Join Date: Nov 2008
Device: Sony PRS-950, iphone/ipad (Marvin/iBooks/QuickReader)
|
That file you uploaded does not load correctly into Internet Explorer - it says there is an invalid character. This suggests that something should have been escaped at the XML or URL level that that has not been. I will see if I can work out exactly which character(s) have caused the problem and when I have done so upload a version of the Calibre2Web script that fixes the problem.
|
05-23-2009, 05:54 PM | #75 |
Wizard
Posts: 4,552
Karma: 950151
Join Date: Nov 2008
Device: Sony PRS-950, iphone/ipad (Marvin/iBooks/QuickReader)
|
That file you uploaded does not load correctly into Internet Explorer - it says there is an invalid character. This suggests that something should have been escaped at the XML or URL level that that has not been. I will see if I can work out exactly which character(s) have caused the problem and when I have done so upload a version of the Calibre2Web script that fixes the problem. At first glance it looks as though the script is not correctly handling non-ASCII characters in a URL. Such URL's can be generated at the moment if you have a non-ASCII character in an author or Title
Last edited by itimpi; 05-23-2009 at 05:59 PM. |