[GUI Plugin] FanFicFare

[theit8514]

Is there a known issue with unicode and Windows? I am running into a problem with post_process_cmd on Windows with titles that contain unicode characters.

For example, the title on https://archiveofourown.org/works/300125 is En Flambé, but when it comes out of post_process_cmd it appears as En Flambé. Windows runs into problems passing this value around, otherwise I wouldn't care too much and just change it in Calibre.

Is there a setting to change to allow this to work properly? I tried changing the website_encodings around without any success.

Thanks!

[JimmXinu]

Originally Posted by theit8514

Is there a known issue with unicode and Windows? I am running into a problem with post_process_cmd on Windows with titles that contain unicode characters.

Unicode issues get very messy very quickly. What works on one OS frequently breaks on another. And the settings for each individual user (and sometimes individual window) can make a difference.

The post_process_safepattern personal.ini setting is specifically for helping work around those issues by replacing anything that matches in any metadata entries with '_'. The example given is defaults.ini is:

post_process_safepattern:(^\.|/\.|[^a-zA-Z0-9_\. \[\]\(\)&'-]+)

This solution is rather English-language-centric and brute force, but I don't generally have OSes configured in other languages available to test with.

[JimmXinu]

New Test Version Posted

2018-02-22
- Make Get URLs from Page work better with TtH is_adult.

[Tharos]

If I see it correctly, FFF use the ID url:https://www.fanfiction.net/s/12307781/1/ in the metadata. Is it possible to use a different ID? For example calibre:https://www.fanfiction.net/s/12307781/1/? or fff:https://www.fanfiction.net/s/12307781/1/?

I use as ID for example already mal: wiki: isbn: aS: reddit:

[JimmXinu]

Originally Posted by Tharos

If I see it correctly, FFF use the ID url:https://www.fanfiction.net/s/12307781/1/ in the metadata. Is it possible to use a different ID? For example calibre:https://www.fanfiction.net/s/12307781/1/? or fff:https://www.fanfiction.net/s/12307781/1/?

I use as ID for example already mal: wiki: isbn: aS: reddit:

(Emphasis added.)

The FFF code as it is written today is hard coded to use identifier url:. (It also looks for uri: because calibre used to prefer that years ago.)

In theory, I suppose FFF could be made to configurably use a different identifier tag, but I don't really see the point to it--the story URL is the definitive URL for where that story came from.

[Tharos]

A really hard point does not exist. Maybe a small example of my IDs says more:

no:https://www.novelupdates.com/series/a-story-name
gt:http://gravitytales.com/novel/a-story-name
url:http://lightnovelgate.com/novel/a-story-name
db:http://lndb.info/light_novel/a-story-name
rln:https://www.readlightnovel.org/a-story-name

In the end, it was just an idea to make it a little easier for me.

[davidfor]

Originally Posted by Tharos

A really hard point does not exist. Maybe a small example of my IDs says more:

no:https://www.novelupdates.com/series/a-story-name
gt:http://gravitytales.com/novel/a-story-name
url:http://lightnovelgate.com/novel/a-story-name
db:http://lndb.info/light_novel/a-story-name
rln:https://www.readlightnovel.org/a-story-name

In the end, it was just an idea to make it a little easier for me.

I like your suggestion and I understand why JimmXinu isn't likely to adopt it. But, I do use some custom ids. These are defined using the custom identifier rules (Preferences, Book Details and the Create rules button). For one of the above, I would add:

Key: no
Name: Novel Updates
Template: https://www.novelupdates.com/series/{id}

With that, you get "Novel Updates" in the details pane rather than the domain.

For FFF downloaded stories, these get added via a personal plugin that fiddles with the book and formats it the way I like them. I will sometimes remove the URL if I want to make sure I don't accidentally update a book.

[edeniz]

A quick question:

What could be the reason for why ships2chars fails? I mean, I know that if the characters field (or any field for that matter) does not have at least one entry, it's unreasonable to expect it, so that's not what I mean. In the example cases I have noticed, the characters field *does* have an entry, but ships2chars fails. In fact, I have tested it with stories where previously ships2chars *had* worked without trouble, so that's why I know I must have messed it up somewhere along the line.

I tried it with literally only this in replace_metadata.txt:

Spoiler:

 ships2chars=>[ ]*/[ ]*=>/
 ships2chars=>[ ]*&[ ]*=>/
 ships2chars=>([^/]+)/([^/]+)/([^/]+)/(.+)=>\1\,\2\,\3\,\4
 ships2chars=>([^/]+)/([^/]+)/(.+)=>\1\,\2\,\3
 ships2chars=>([^/]+)/(.+)=>\1\,\2
 ships=>([ ]*) And ([ ]*)=>\1/\2

So it can't be that the replaced metadata is the source, right? What am I missing? What did I unwittingly mess up? Because at one point it worked without problems. So where in the ini should I look for the error?

Example url to try it out:
https://archiveofourown.org/works/12793047

And this one is the link to a story where it had split the ships into chars without problems way back when: http://archiveofourown.org/works/2123949


Below is the relevant .ini attached.


Thank you kindly for any and all help.

[JimmXinu]

@edeniz:

Your problem is that you have two add_to_extra_valid_entries entries in the [defaults] section:

[defaults]
add_to_extra_valid_entries:,ships2chars
...
add_to_extra_valid_entries:,rawgenre,rawcategory,rawcharacters,rawships

The INI parser only uses the last entry in a section for the same key word. So ships2chars is ignored. (The add_to_ mechanism is for adding to other sections.)

Combine those two into one line and it should be good again:

[defaults]
add_to_extra_valid_entries:,ships2chars,rawgenre,rawcategory,rawcharacters,rawships

[edeniz]

Originally Posted by JimmXinu

@edeniz:

Your problem is that you have two add_to_extra_valid_entries entries in the [defaults] section:

[defaults]
add_to_extra_valid_entries:,ships2chars
...
add_to_extra_valid_entries:,rawgenre,rawcategory,rawcharacters,rawships

The INI parser only uses the last entry in a section for the same key word. So ships2chars is ignored. (The add_to_ mechanism is for adding to other sections.)

Combine those two into one line and it should be good again:

[defaults]
add_to_extra_valid_entries:,ships2chars,rawgenre,rawcategory,rawcharacters,rawships

Thank you very kindly! I knew I messed up somewhere when I was trying to make the ini easier to navigate for me, but simply couldn't see where, much to my frustration.

[Firedancer885]

Hi Jimm,
I just got the following error. Is this a temporary problem, or something changed in AO3?
<urlopen error [Errno -2] A név vagy a szolgáltatás nem ismert>
https://archiveofourown.org/works/13769718

[JimmXinu]

Originally Posted by Firedancer885

Hi Jimm,
I just got the following error. Is this a temporary problem, or something changed in AO3?
<urlopen error [Errno -2] A név vagy a szolgáltatás nem ismert>
https://archiveofourown.org/works/13769718

Assuming translate.google.com is correct that the error message reads "The name or service is not known" and given that I was able to download it, it's probably a temporary problem, possibly with your DNS.

[Firedancer885]

Thanks

[scarlettruin]

Hi,
When I go to update any of my fics, it's not working. It shows an error message with just 'URLError:'.

calibre, version 3.14.0
ERROR: Unhandled exception: <b>URLError</b>:<urlopen error [Errno 11001] getaddrinfo failed>

calibre 3.14 embedded-python: True is64bit: False
Windows-10-10.0.16299-SP0 Windows ('32bit', 'WindowsPE')
32bit process running on 64bit windows
('Windows', '10', '10.0.16299')
Python 2.7.12+
Windows: ('10', '10.0.16299', 'SP0', u'Multiprocessor Free')
Interface language: en_GB
Successfully initialized third party plugins: DeDRM (6, 5, 2) && View Manager (1, 4, 3) && SmartEject (2, 0, 7) && NormComment (0, 0, 2) && Modify ePub (1, 3, 13) && Count Pages (1, 8, 0) && Search The Internet (1, 7, 6) && Goodreads (1, 1, 13) && Generate Cover (1, 5, 21) && FanFicFare (2, 22, 0) && Author Book Count (2, 1, 5) && Read MP3 AudioBook metadata (1, 0, 79) && Prettify Cover 0.9.12 && Find Duplicates (1, 6, 3) && EpubMerge (2, 1, 1) && Quality Check (1, 9, 11) && Save Composite Custom Columns (1, 1, 0) && EpubSplit (2, 3, 0)
Traceback (most recent call last):
File "calibre_plugins.fanficfare_plugin.fff_plugin" , line 810, in update_anthology
File "calibre_plugins.fanficfare_plugin.fff_plugin" , line 568, in get_urls_from_page
File "calibre_plugins.fanficfare_plugin.fanficfare.getu rls", line 59, in get_urls_from_page
File "calibre_plugins.fanficfare_plugin.fanficfare.conf igurable", line 1221, in _fetchUrl
File "calibre_plugins.fanficfare_plugin.fanficfare.conf igurable", line 1210, in _fetchUrlOpened
File "calibre_plugins.fanficfare_plugin.fanficfare.conf igurable", line 1111, in _fetchUrlOpened
URLError: <urlopen error [Errno 11001] getaddrinfo failed>

Any ideas please?

[JimmXinu]

FFF isn't able to resolve the host name. Could be a problem with your network, your proxy settings (or proxy), or a firewall/security program blocking calibre from making outbound connections.

If you can still see the story page on the same computer using your browser (and don't use a proxy), a firewall program is the most likely cause.