Describe the bug
Search query has quotes around it, which can prevent results from appearing if the query has more than the title alone.

To Reproduce
Unpacked to C:\hpx
setup with postgre, import galleries, enable plugins, start trying to match metadata
most folders are limited in information, most only have the artist/title

Screenshots

Environment:

• OS: Windows 10 1803
• Firefox 66
• HPX Version - Client 0.1.1, Server 0.10.2, Database 0.7.0, postgreSQL11

File Metadata parser fails for info.json files generated from HDoujin Downloader when inside zip files. Same info.json when extracted parses with no issues whatsoever.

Here is the plugin.log:

Sep-09 00:16:49--INFO pluginctx.file-metadata.main: Attempting with DataType.eze
Sep-09 00:16:49--WARNING pluginctx.file-metadata.extractors.common: An error occured while trying to parse file into a dict
Sep-09 00:16:49--INFO pluginctx.file-metadata.main: Skipping DataType.eze
Sep-09 00:16:49--INFO pluginctx.file-metadata.main: Attempting with DataType.hdoujin
Sep-09 00:16:49--WARNING pluginctx.file-metadata.extractors.common: An error occured while trying to parse file into a dict
Sep-09 00:16:49--INFO pluginctx.file-metadata.main: Skipping DataType.hdoujin

Let me know if you need an exmaple, but really this is happening with all my files.

Hi i using "koromo copy"(hitomi downloader) program.
this program can save gallery metadata for made json file.
but hpx not support this json file.
I have so many files that it is hard to query with ehenti metadata plugin
can you add plugin with "koromo copy"s json file?
i can give json files if you want. thx

koromo copy github: https://github.com/dc-koromo/koromo-copy/issues
i attach info.txt(metadata file)
Info.txt

Would it be possible to add support for external json files saved using https://github.com/dnsev-h/ehentai-archive-info, with the same name as the archive?

Example:
[GRINP (Neko Toufu)] Onii-chan wa Oshimai! Tokubetsu Hen [Digital].cbz
[GRINP (Neko Toufu)] Onii-chan wa Oshimai! Tokubetsu Hen [Digital].info.json

I'm importing zip files with an info.txt file for the metadata. I have the File Metadata plugin enabled but it only fetches some of the metadata. Those being the name, artist, parody, language, pages, tags and links. The ones it doesn't add are type, series, description and publisher. The main one being series, since it just adds the name as the series.
Note: I didn't fill all options but left them blank (example)

To Reproduce:
Enable File Metadata plugin.
Scan folder with zip files (example zip with white as pages).
Submit.
Check in inbox.

Expected behavior:
I would expect all the metadata to get imported.

Logs:
It recognizes it as an hdoujin metadata file (in both activity.log and plugin.log) and doesn't give me any errors:
--INFO [plugin].context.file-metadata.main: Attempting with DataType.hdoujin
--INFO [plugin].context.file-metadata.main: DataType.hdoujin matched!

Environment:

• OS: Linux
• Brave 1.23.71, Firefox 84.0.1
• HPX Version 0.13.3

Ignore.

When i tried querying metadata for my doujins ( 2600 or so ), after around 500 i was struck with a ban hammer on exhentai. I had every delay on default as that happened.

That got me thinking about possible resolutions to this problem. I'm listing them in no particular order and you can add them all or only some of them.

1. Add a warning notification when adding many, lets say above 200-300 doujins to metadata queue that you can be struck with a ban hammer.
2. Add logic to the metadata queue so it notices when you get a warning page on exhentai, I forget about specifics but it says something along these lines "you are doing this too much, if you continue you will be temp banned." and dissapears after few minutes. When that would be detected queue would initiate cooldown of for example 60 minutes and then resume until another warning.
3. Same as above but hardcode limit of 200-300 doujins between 60 minutes cooldowns.

Edit:
First ban was 24h after around 500 doujins
Second ban was 3 days after around 100 doujins.
fun stuff.

You should get in touch with Tenboro and work out safe values for this.

While Chaika and NH work fine E-H does not work for me

To Reproduce
Login using EHentai Login
Clicking download gallery using the Firefox extension gives the following in the log
[---INFO---]--2019-Oct-30 21:57:22-: Finished handling item
[---INFO---]--2019-Oct-30 21:57:22-: Failed to import item: BadZipFile("File is not a zip file",)
[---INFO---]--2019-Oct-30 21:57:21-: Waiting on post-processing from "exhentai"...
[---INFO---]--2019-Oct-30 21:57:15-: Downloading "2" requests for https://exhentai.org/g/1509162/87cf84cdae/
[---INFO---]--2019-Oct-30 21:57:15-: Received 2 requests from "exhentai"
[---INFO---]--2019-Oct-30 21:57:10-: Querying resources from "exhentai"...
[---INFO---]--2019-Oct-30 21:57:09-: Handling url https://exhentai.org/g/1509162/87cf84cdae/
[---INFO---]--2019-Oct-30 21:57:09-: Adding https://exhentai.org/g/1509162/87cf84cdae/

Expected behavior
To download from EH

Logs
error.log

Environment:
Windows 10
Firefox 72
HPX 0.12.3

Hello, firstly I'm really love your work here and very excited to know that you will coming back soon.
So recently I've been using a downloader call Hitomi-downloader https://github.com/KurtBestor/Hitomi-Downloader-issues I think it's very convinient to download archives from exhentai also the downloader already had tags/artist/group... categories itself for each archive.
But it's a shame that the downloader is not plugin-supported by HappypandaX so, I would like to request for one please.

I recently encountered a lot of incorrectly tagged libraries which was caused by doujins with really generic names (e.g. "Cheeky") resulting in the plugin picking the wrong chaika gallery for metadata.
Since HPX already recognized and tagged the artist of the doujin correctly during the import (due to the file name), would it be possible to make an optional setting which incorporates the artist tag (if present) in the chaika search?

(There hasn't been much activity on this plugin in a while so please let me know if this request is low priority/not worth working on because in that case I'd try to look into this myself.)

• fetch metadata
• download galleries

Can we include the "CHARACTERS" tag created in the info.txt file as a tag?

Currently, the information about the Characters exist in the info.txt file, but are not being included as a tag, due to not being part of the "TAGS" field, but the separate "CHARACTERS" field. I'm sure this can be arranged somehow. My understanding of python is extremely limited, so the modification required to make this work is beyond my skill.

I've got some 4000 touhou doujins in the Database, but after import i realized i can not search by character. I'm thinking that by simply adding the "CHARACTERS" field to the tags this would greatly help my cause. Ideally HPX would have a seperate Characters field like e-hentai does, but seeing how it doesn't it feels appropriate to merge the TAGS and CHARACTERS field.

I've included a sample info.txt to show what it looks like.

info.txt

The plugin loads correctly, but upon adding a gallery the following errors are produced

plugin.log: https://pastebin.com/wVLM9gSB
plugin_debug.log: https://pastebin.com/WLf5JjwN

Using Server version 0.5.0 and the latest plugin version.

Hi. I'd been frustrated by this plugin's inability to prefer english language search results. When I read the readme for the plugin it mentioned a "preferred_language" option that I didn't even know about. Since it's supposed to default to 'english', it should have just been working anyway but it wasn't.

I'm not a programmer but I'm pretty sure I was able to trace the issue to line 223 in EHentai Metadata/main.py:

found_urls = [(str(x.a.string), x.a['href']) for x in results] # title, url

I don't understand this line, but I think that (str(x.a.string) is supposed to extract the gallery title from each search result and for it me is consistently returning "None". When the plugin later goes to look for "english" in that string is then always false.

Since I don't actually know python enough to fix it, I've just added + " language:english" to the search string (i_title) and that'll suit my needs. 😆

Describe the bug
The EHentai Metadata plugin failed to initialize

To Reproduce
Not sure

Plugin Logs
Jun-06 23:41:04--INFO pluginctx.ehentai-metadata: Installing plugin
Jun-06 23:41:04--INFO pluginctx.ehentai-metadata: Marking plugin for initialization
Jun-06 23:41:06--INFO pluginctx.ehentai-metadata: Initiating plugin
Jun-06 23:41:06--ERROR pluginctx.ehentai-metadata: An unhandled exception 'SyntaxError' was raised during plugin initialization
Traceback (most recent call last):
File "happypanda\core\plugins.py", line 789, in init
File "", line 3, in
File "C:\Users\Gudrith\Documents\HappyPanda X\data\plugins\1572a650-5352-4d62-802f-8dfb983f9231.hpxplugin\main.py", line 470
gcircles.append(CircleData(name=ca pitalize_text(circlename)))
^
SyntaxError: invalid syntax
Jun-06 23:41:06--INFO pluginctx.ehentai-metadata: Unloading plugin
Jun-06 23:41:06--INFO pluginctx.ehentai-metadata: Disabling plugin: Failed to initialise plugin
Jun-06 23:41:06--INFO pluginctx.ehentai-metadata: Marking plugin for initialization
Jun-06 23:41:06--INFO pluginctx.ehentai-metadata: Plugin failed
Jun-06 23:41:06--INFO pluginctx.ehentai-metadata: Marking plugin for initialization

Screenshots

Environment:

• OS: Windows 10
• Chrome
• HPX Version 0.10

The current EHentai Login plugin only collects ipb_member_id and ipb_pass_hash as credentials. This leads to login failure usually on my device and this problem can be solved by adding igneous and sk to the cookie. I tried this out by modifying the python script. Please collect igneous information for ExHentai login.

When downloading gallery from exhentai using the plugins, the log says "DownloadError("The resource could not be found in the cache directory cache\downloads")". Is there any steps that can be taken to initailize that directory?

Describe the bug
Queries on EHentai for metadata don't work. It always returns 0 results.
I'm logged in EHentai and ExHentai as seen on the log.

Example of gallery with the problem:
https://ibb.co/JxXwM0j

EHentai link:
https://exhentai.org/g/1387488/a7df1b2944/

The ZIP file has the same name as the gallery.
I tried changing filename_search to false with the same result.

config.yaml:
https://pastebin.com/kmxSg07k

I tried more than 30 different galleries (downloaded from EHentai) with the same results.

If I add the gallery URL manually in the "External links" in HPX it works and apply the metadata.

To Reproduce
Steps to reproduce the behavior:

1. Query for metadata on a gallery with no "External links"
2. Nothing changed

Expected behavior
Metadata should be found and applied.

Logs
plugin.log:
https://pastebin.com/RU7KySdi

Environment:

• OS: Win10
• Browser and browser version Chrome 73
• HPX Version 0.9.0 (also presents on 0.8.1)

Generated.zip
Real.zip

I've included two zip files, Generated.zip provides a sample that I edited to fill up all the possible metadata fields and a few screenshots of how it actually looks on the program cause the actual xml file would look messy.

Real.zip contains a few un-edited .xml files that were produced directly from HDoujinDownloader.

E-Hentai plugins need updating.

• E-Hentai Login one needs exhentai to be removed.
• E-Hentai Metadata needs to be updated to remove exhentai as well as to always send cookies/session which will allow certain users to still search for forbidden content on E-Hentai frontend.
• E-Hentai Metadata needs to be updated for new site design.

See https://github.com/CirnoT/happypandax-plugins/blob/master/plugins/EHentai%20Metadata/main.py as well as https://github.com/CirnoT/happypandax-plugins/blob/master/plugins/EHentai%20Login/main.py for example of how such updates could be done. Do note however that E-Hentai Metadata one was updated only for thumbnails view as this is what I use.

• fetch metadata

Describe the issue
"No matching download handler found for url https://nhentai.net/g/..." whenever I try to add an url for download.

To Reproduce
Steps to reproduce the behavior:

Clean install of HPX
Create an admin user
Start server
Open Webclient and login
Go to Downloads
Add a random nhentai gallery url
See error in log window (No matching download handler found for url https://nhentai.net/g/...)

Expected behavior
The hentai gallery to be downloaded without errors.

Environment:

OS: Windows 10
Browser and browser version: Firefox latest
HPX Version: v0.13.3

Additional context
Tested on two separate machines and same issue. The same issue also occurs with e-hentai galleries.

• fetch metadata
• download galleries

• fetch metadata
• download galleries

HI,
I have here: [Zucchini] (C97) [Z.A.P. (Zucchini)] Artoria wa Makerarenai. (Fate／Grand Order) [Chinese] [黎欧x新桥月白日语社汉化] (356060)

And well, instead of Chinese it will recognized as English. No idea why the last number makes the result wrong.

File Metadata 0.2.0b installed
happypandax0.5.2.win

When importing my manga into HPX the "Parody:" entry does not automatically fill in even when the data is there in the info.txt file using the File Metadata plugin.

I use hdoujin downloader to grab my manga and create the appropriate info.txt v2 metadata file titled info.txt. An example of the affected manga is attached. info.txt

I have also noticed that the "Circle:" entry is missing when clicking edit on an existing imported gallery. So I can't add or edit an existing circle if it was found by the plugin. Is this by design?

downloaded via hdoujin, cbz format, metadata in info.txt
the metadata can't be detected.
is it actually only support folder?
thanks before

The plugin needs a delay mechanic implemented. As it is now e-hentai.org is banning due to too much bandwidth use/no delay.

"Your IP address has been temporarily banned for excessive pageloads which indicates that you are using automated mirroring/harvesting software. "

I've been receiving "No matching download handler found for url https://nhentai.net" errors even after installing both plugins. After messing around with the downloader to make it log the response content, I figured it out.

NHentai enabled Cloudflare some time ago and this breaks both NHentai plugins - the requests receive a 503 response with the Cloudflare "please wait" screen. I think this could be improved in two ways:

• the logging output for plugins should be more specific: HTTP error codes and messages would help diagnose issues;
• I'm not familiar enough with Python or Cloudflare to figure out a proper fix, but I figure it should be done at the hpx.command.SingleGETRequest().request level. Some way to trick Cloudflare into believing that it's a real browser doing the interaction?

I've attached both of the edited plugins for anyone who encounters the same problem. This isn't really a fix since the cf_clearance cookie will eventually expire and this forces you to continue editing the plugins. @twiddli , thank you for making this, it's a great piece of software and nothing I've used so far is even close.

nh plugins.zip

While running metadata with only chaika enabled, this error will appear in the console.

Logs:
logs.zip

HPX 13.3
Windows 10