[LanguageTool]: Grammar check

[brent63]

I changed False to True and followed your previous instructions, see below.

File Name Line Column Context Catgory Rule Message
Section0001.xhtml 9 4 ... >>F???<< STYLE PROFANITY Profanity: This word is considered offensive.

[Doitsu]

Originally Posted by brent63

I changed False to True and followed your previous instructions, see below.

Spoiler:

File Name Line Column Context Catgory Rule Message
Section0001.xhtml 9 4 ... >>F???<< STYLE PROFANITY Profanity: This word is considered offensive.

This means that the plugin works. Is your problem that the plugin doesn't display the message in the Sigil validation window or do you want LanguageTool to ignore profanities?

If you don't want LanguageTool to complain about profanities, insert ,PROFANITY at the end of the disabledRules entry (after SENTENCE_WHITESPACE and before the closing quotation mark):

{
"disabledRules": "MORFOLOGIK_RULE_EN_US,MORFOLOGIK_RULE_EN_GB,ENGLISH_WORD_REPEAT_BEGINNING_RULE,WHITESPACE_RULE,COMMA_PARENTHESIS_WHITESPACE,SENTENCE_WHITESPACE,PROFANITY",
"disabledCategories": "REDUNDANCY",
"update_check": false,
"allFiles": false,
"clipboard_copy": false,
"ltPath": "C:/Users/Brent/Language/LanguageTool-4.6/LanguageTool-4.6/languagetool-commandline.jar"
}

(You'll have to scroll to the right to see the change.)

D.

[brent63]

Thank you working now.

[DiapDealer]

Please use common sense when posting the results of a profanity test. It was suggested that you add "the f-word" to test. It should have been a no-brainer that a similar obfuscation should have been used when giving the results.

[lumpynose]

I'm using LanguageTool_v0.4.2.zip for the plugin and LanguageTool-4.6 for the java app. When I run it it blows up with

Command line: java -Xmx1024m -Xms512m -jar C:/blah/blah/blah
Java error message: working on ... java.lang.OutOfMemoryError at blah blah blah

I'd copy and paste the errors here but I can't figure out how to copy what I want in the Validation Results window. I get either the partial results, the stuff in yellow, or the header lines.

I increased the -Xmx and -Xms values and the error stopped.

[Doitsu]

Originally Posted by lumpynose

I'm using LanguageTool_v0.4.2.zip for the plugin and LanguageTool-4.6 for the java app. When I run it it blows up with

Command line: java -Xmx1024m -Xms512m -jar C:/blah/blah/blah
Java error message: working on ... java.lang.OutOfMemoryError at blah blah blah

The default settings (-Xmx1024m -Xms512m) work fine for me with LanguageTool-4.6.

1. How much RAM does your machine have?
2. Did the plugin fail for all epubs or only epubs with very large (X)HTML files?

[lumpynose]

Originally Posted by Doitsu

The default settings (-Xmx1024m -Xms512m) work fine for me with LanguageTool-4.6.

1. How much RAM does your machine have?
2. Did the plugin fail for all epubs or only epubs with very large (X)HTML files?

Huh, weird.

The failure was sort of random. At first I selected all the xhtml files and it failed when it started the 2nd one. Selecting 2 or 3 would also fail after the first one.

I have 64gb of ram. There's nothing big running using ram.

[Doitsu]

Originally Posted by lumpynose

The failure was sort of random. At first I selected all the xhtml files and it failed when it started the 2nd one. Selecting 2 or 3 would also fail after the first one.

If you're working on a Public Domain book, please share a download link. (If it's a commercial book, you can scramble it with the Calibre ScrambleEbook plugin.)

Also, please post the contents of LanguageTool.json.

Originally Posted by lumpynose

I have 64gb of ram. There's nothing big running using ram.

My Windows 10 testing machine only has 8GB of RAM. I.e., it's definitely not a RAM issue.

[KevinH]

Assuming the program in question is using recursive descent/parsing techniques, it may not be a total RAM issue or even RAM available to the JVM, but instead a single thread stack size issue. Each recursive call pushes a new frame on the stack with return info and local variables. If that is the case, increasing the stack size (the Xss) may be needed.

This is just a guess, but given the OP has 64gb of ram and is still running into trouble, this may in fact be the cause.

[lumpynose]

Originally Posted by Doitsu

If you're working on a Public Domain book, please share a download link. (If it's a commercial book, you can scramble it with the Calibre ScrambleEbook plugin.)

Also, please post the contents of LanguageTool.json.


My Windows 10 testing machine only has 8GB of RAM. I.e., it's definitely not a RAM issue.

Ok, thanks. Will do. I agree with KevinH that it's likely a stack issue.

[JSWolf]

Lumpy,are you using a 32-bit or 64-bit Java? That have been some recent issues using 32-bit Java like stack issues with epubcheck. Plus, 32-bit Java has access to less memory and that could be your out of memory problem.

[Leonatus]

Windows 10, 64 bit, Sigil V. 1.2.0, LT V. 0.4.3.
I'm actually trying to check a book in French. When I start the plugin, it opens a window looking for "LT (languagetool-commandline.jar)". As I don't know what to do, I close this window and the check fails with the error message: "LanguageTool path not selected or not found!"
I slightly remember that this behaviour did already appear once, but I can't find the post. Can you help, please, Doitsu?

[Doitsu]

Originally Posted by Leonatus

Windows 10, 64 bit, Sigil V. 1.2.0, LT V. 0.4.3.
I'm actually trying to check a book in French. When I start the plugin, it opens a window looking for "LT (languagetool-commandline.jar)". As I don't know what to do, I close this window and the check fails with the error message: "LanguageTool path not selected or not found!"
I slightly remember that this behaviour did already appear once, but I can't find the post. Can you help, please, Doitsu?

You'll need to download the desktop version of LanguageTool and unzip it.
The unzipped folder will contain languagetool-commandline.jar.

There's also another Sigil grammar checker plugin for French books that you might find helpful: Grammalecte.

[Leonatus]

Originally Posted by Doitsu

You'll need to download the desktop version of LanguageTool and unzip it.

But there has always been a Sigil plugin, and it used to work really well. Is this gone? I scarcely can believe it.

Now I understood the necessity of the desktop version, sorry!

Grammalecte runs very well, thank you!

But when I ran Language tool, there appeared another error:

Status: failed

Working on Section0001_0001.xhtml...
Traceback (most recent call last):
  File "C:\Program Files\Sigil\plugin_launchers\python\launcher.py", line 136, in launch
    self.exitcode = target_script.run(container)
  File "C:\Users\Juerg\AppData\Local\sigil-ebook\sigil\plugins\LanguageTool\plugin.py", line 491, in run
    stdout = result[0].decode('utf-8')
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xe4 in position 94: invalid continuation byte
Error: 'utf-8' codec can't decode byte 0xe4 in position 94: invalid continuation byte

Edit: The error apparently only occurs at the french book.

[Doitsu]

Originally Posted by Leonatus

But when I ran Language tool, there appeared another error:

UnicodeDecodeError: 'utf-8' codec can't decode byte 0xe4 in position 94: invalid continuation byte

This usually only happens if the epub is incorrectly encoded. Your book most likely has no encoding declaration or an UTF8 declaration but is most likely CP1252 encoded. However, valid epubs may only be utf8 or utf16 encoded.

0xe4 (228) in CP1252 is the German a umlaut (ä).

If the French book is a PD book, please send me the download link in a PM. I might be able to fix the code page issue.