4E PARSER errors

[becbunzen101]

Hello, i'm french and i don't speack english very well, sorry..

I have a problem with 4E Parser. After using compendium mode (i'm a membrer of D&D insider) and crape player handbook 1. I have some errors with Powers, npcs and rituals. others categories are ok . When i make a parse with the player handbook 2, all is OK.

I post here a screenshoot of a error with rituals. I don't understand why i have errors and how correct it. if you can help me ...

The screen of the error on 4E parser :
yfrog.com/5eerrorxtj

Bec

[Griogre]

You should post on the 4E Developer group at FUM for help with the parser: https://www.fouruglymonsters.com/com...FG+Development

You may need to sign up for the website.

[ShadoWWW]

Well, there is an XML error in the XML file (perhaps client.xml?) on the line 53758 in temporary folder. The error is the <p> tag is even there where it shouldn't be. Perhaps there are more errors in the file.

You'll have to find somebody with XML skills and with the XML validator editor, who can correct errors in the file. Then just zip all files in the temporary folder, rename the zip file to *.mod file and copy the mod file to MODULES folder in Fantasy Grounds II folder.

Or you can wait for a new version of 4E PARSER and maybe this bug will be corrected there.

(Sorry for my English, I'm Czech.)

[becbunzen101]

thanx you !

Bec

[tdewitt274]

I reported this error in 4.0.14. As stated above, the extra </p> (find and replace "</p></p>" with "</p>" in Notepad).

This will take care of some of the problems, but there's a couple more. Take a look at the link and it should help.

[becbunzen101]

Thanks you Tdewitt274 ! Nice link !

Bec

[BruntFCA]

Using the parser is a bit of a nightmare. It can't read the xml file generated by itself.

I'm also getting trouble with parsing PHB1. I've limited to only the powers for the first run, still the file is about 22,000 lines long. There are XML errors in the client.xml it generates. The parser program itself highlights these errors, though luckily it *completes* it parse rather than bombs out, this means at least you get a semi useful xml file.

You can't edit giant text files like this in notepad, luckily there is a tool to help sort this out. Download this open source XML editor,

https://www.philo.de/xmledit/

It actually has a built in error detector thats better than what the parser produces anyhow. Some of the errors in "powers" include the use of the \ sign in the xml names. You need to delete all references to this, but only in the xml names where it highlights is (the program takes you to the line number). You will also then need to edit the corresponding "closing" entry since it now won't match now that you've taken the \ symbols out of its partner.

Don't take this as a critisism of the parser, the guy prolly did it for himself, and just decided to release it out of goodwill; he's under no obligation to support it. Also concider that if Wizards change anything the compendium format, this could "break" the parser. This is what I expect may be causing some of these errors.

[Zeus]

To all those experiencing problems scraping and parsing earlier modules like the PHB, keep the faith.

The syntax and formatting of some of WotC DDI content is chaotic at times and therefore there is bound to be the odd bit of content that causes issues.

Tenian (the amazing author of the 4EParser) is aware of the erroneous Compendium scrape output generated for some of the earlier modules like the PHB. Its a monumental and laborious task of finding and correcting these types of errors so try to be patient.

If you follow his dev blog over at 4E Blogspot you'll see he's working on getting this fixed (if possible) for a future release.

[Tenian]

For some of you late comers, it seems you are unaware of the history of the parser and the correct way to report errors:

History:
The parser was originally designed to function by cutting and pasting data from the PDF source material that WOTC used to sell (all books prior to Arcane Power were at one point in time sold in PDF format). This data was available long before the compendium even existsed.

The compendium extraction feature (scrape) is relatively new, it received it's first public testing sometime around the last Gen Con (when the Compendium was free for weekend).

The first book I actually processed using the compendium extraction was Primal Power. Prior to that, the extraction routine was not stable enough to produce reliable results or had significant gaps in what it could process. This means any books published prior to primal power were never fully tested using the extract/parse method. They were in fact processed by copy/paste from PDF or by OCR (which is not fun FYI).

I only recently had the time to go through all the older books (prior to primal power) and test the extraction/parse process. This is not a fun process, FYI. The Adventurer's Vault and Adventurer's Vault 2 take 45 minutes or more to scrape with the debugger attached. And I had to do this 11 times to get the errors corrected. There were a number of errors in the old material. I believe this is mostly due to the old data being in different formats or the odd typo. As of two weeks ago, I had all of the books (I don't do Dungeon, Dragon, RPGA) processing without the need to make any manual edits.

Two stage process:
As for why the extraction process doesn't automatically create a module, that's pretty simple. There's a lot of information that the parser supports that can not be extracted from the compendium. The parser supports Diseases and Familiars for example...the compendium does not provide access to these. And again the extraction was a late addition to the process.

Error Reporting
Any errors with the parser should be reported over on www.fouruglymonsters.com in the D&D 4th Edition FG Development group. This is where I actually look for errors. Posting them in the various groups here is not the correct place and they are unlikely to get fixed if they are not reported in the correct place.

Random Stuff
The XML files the parser creates are checked for validation. It actually runs an XML validation routine on them. If it reports errors you should look at the output XML file.

The best tool for viewing the files is actually IE or Firefox. Both of them are fairly good at locating the errors

You can in fact open the output XML files in Notepad. Notepad will easily handle text files under 10M. Unless you are doing something silly like mashing multiple books into a single file nothing the parser produces will be anywhere close to this size.

[ShadoWWW]

I would like to ask if new parser would be released this week. It there any ETA?