Two features/tweaks

[GoogleCodeExporter]

Parse the comments for html escape characters such as   and &.

Hide the metadata response when querying ComicVine

Original issue reported on code.google.com by cwald...@gmail.com on 21 Dec 2010 at 4:01

[GoogleCodeExporter]

I can add code to remove the html escape characters.  Some of the metadata I 
can hide and some seems to be a timing issue.  I pipe the JSON object into zip 
and sometimes it gets displayed and sometimes it doesn't.  Not sure what 
determines when it shows up.  Most of the other ComicVine displays were in 
there for debug purposes and I haven't gotten around to cleaning them up.

Original comment by andre.me...@gmail.com on 21 Dec 2010 at 6:19

• Added labels: Type-Enhancement
• Removed labels: Type-Defect

[GoogleCodeExporter]

I've changed my html escape code.  Do you have any examples where the metadata 
includes those html escape characters so I can verify this? 
 Also I've suppressed the metadata display

Original comment by andre.me...@gmail.com on 23 Dec 2010 at 4:46

[GoogleCodeExporter]

Here's the metadata for Batman 697. They almost always happen in the 
description. I don't think I've seen it in any other field. it may be an issue 
that originates in how Comicvine.com stores the info.

Archive:  Batman 697.cbz
{"ComicBookInfo/1.0":{"genre":"Superhero","volume":1940,"publicationMonth":5,"ti
tle":"Life After Death, Part 6: 
Liberator","language":"English","publicationYear":2010,"credits":[{"person":"Jar
ed K. Fletcher","role":"letterer"},{"person":"Ian 
Hannin","role":"colorer"},{"person":"Janelle 
Siegel","role":"editor"},{"person":"Mike 
Marts","role":"editor"},{"person":"Sandu 
Florea","role":"inker"},{"person":"Tony 
Daniel","role":"cover"}],"rating":1,"publisher":"DC 
Comics","series":"Batman","comments":"Who wears the Black Mask? The true 
identity of Gotham City's new crimelord is finally revealed and that identity 
will suprise everyone - especially Batman! And after the final showdown between 
Black Mask and The Dark Knight, will things ever be the same for Kittyhawk and 
the Reaper?","issue":"697","country":"United 
States"},"appID":"ComicBookLover/999","lastModified":"2010-12-22 19:02:01 
+0000"}

Original comment by cwald...@gmail.com on 23 Dec 2010 at 5:35

[GoogleCodeExporter]

ok.  I've now got code in place to replace any &xxxx; pattern with a space.  If 
there are other html escape codes that are widely used, I may consider adding 
other ways of handling them.  It looks like   is the most common.

Original comment by andre.me...@gmail.com on 23 Dec 2010 at 6:49

[GoogleCodeExporter]

Another one seems to be &. I haven't seen any others (so far).

Original comment by cwald...@gmail.com on 23 Dec 2010 at 6:53

[GoogleCodeExporter]

Is the Genre written to the zip metadata? I saw it in the data from comicvine, 
but don't see it reflected in any of the imported files.

Original comment by cwald...@gmail.com on 23 Dec 2010 at 9:05

[GoogleCodeExporter]

Should there maybe be a way to flag already scanned files, and an option to 
rescan those. This way if I add a large number of issues to a title I don't 
have to rescan all the files again. But maybe I know the original metadata was 
wrong I could change an option in the script to force re-scan those files. This 
would probably just use a flag in the zip comments, but I don't know if CBL 
would choke on that.

Original comment by cwald...@gmail.com on 24 Dec 2010 at 7:06

[GoogleCodeExporter]

Better handling of misnumbered issues. If the issue number in the meta-data 
doesn't match what's available on comic vine then say that in the log or an 
error message. At the moment the script simply stops (which I've come to deduce 
is a misnumbered issue).

Original comment by cwald...@gmail.com on 24 Dec 2010 at 8:12

[GoogleCodeExporter]

I don't see genre in the comicvine data.  if you see it there, let me know 
where because I'd love to add it.  I believe the Grand Comic Database 
(www.comics.org) has that field but they don't have an api...  If they did, I'd 
definitely look into making an option of which database to query.

I'm not sure how I'd handle a rescan option.  If I add a "scanned and loaded" 
flag to the curent ComicBookInfo tags, it would probably fail validation.  I 
could add it to an app specific section, but there's a posting in the CBL 
forums saying that CBL wipes those out so that would be a bad idea. 

Misnumbered issues is a tricky one because sometimes it's an error on the users 
end and sometimes it's bad data on ComicVine.  Is the current logging not 
recording issues like this?  I can fix that part.  I'm undecided about how the 
logging should handle this.  The purist in me thinks the error list should just 
be files that didn't work.  But the nit-picky side of me thinks it should be 
logged with as much info as possible.  Maybe I should add a log-level option.  
Level 1 is just files that fail. Level 2 is files and brief description.

Original comment by andre.me...@gmail.com on 25 Dec 2010 at 2:31

[GoogleCodeExporter]

My mistake about the genre. It's when the original zip comments were being 
written back that I was seeing it (I'd already entered it).

I was worried about CBL choking on any additional metadata. Rats. I'm keeping 
the seriesId.txt file in each folder, maybe cache the info there. Eh, it'd be 
convenient - but I'm not going to loose sleep over it.

At the moment the script simply dies when it hits a misnumbered issue. Maybe a 
prompt to verify the number and ask for an alternative?

Where is the log created. I've looked for it in ~/Library/Logs.

Original comment by cwald...@gmail.com on 25 Dec 2010 at 3:02

[GoogleCodeExporter]

the log file should be created in the folder you are in when you run the 
script.  If you change the logfilename variable to be a fully qualified path, 
it should put the file there.

Right now I only have it logging when you are running in non-interactive mode 
so it's not too helpful if you are in interactive mode.  This is something else 
I should change... just have to think it through before I implement it!

Original comment by andre.me...@gmail.com on 25 Dec 2010 at 3:08

[GoogleCodeExporter]

I'm not seeing the log file being created, either in the default location of 
the app, or in the path I specify.

Original comment by cwald...@gmail.com on 29 Dec 2010 at 8:58

[GoogleCodeExporter]

Could you give me an example of an issue that is failing to get meta data and 
not being logged?

Original comment by andre.me...@gmail.com on 30 Dec 2010 at 6:19

[GoogleCodeExporter]

Any issue. I don't see the log file being created on failures or on normal 
operation. Both with the default code and with a modified code pointing to my 
home directory or my log folder. Nada.

Original comment by cwald...@gmail.com on 30 Dec 2010 at 7:29

[GoogleCodeExporter]

I've been unable to duplicate this.  I'm afraid I may just be forcing failures 
in the ways the code handles so I'm avoiding the problem code.  Can you give me 
specific steps to reproduce this problem?

Original comment by andre.me...@gmail.com on 3 Jan 2011 at 8:07

[GoogleCodeExporter]

i double-checked the script and now what the problem is. Your choice - either 
"ID10T" or "PEBKAC". I have interactive mode turned on, so it's not logging. I 
didn't notice that option was a condition for logging to work. My apologies.

Original comment by cwald...@gmail.com on 3 Jan 2011 at 8:19

[GoogleCodeExporter]

the latest version has a logLevel variable that should work even if you are in 
interactive mode.

Original comment by andre.me...@gmail.com on 3 Jan 2011 at 8:22

[GoogleCodeExporter]

Closing this ticket out.  In the future, please enter one ticket per feature.   
It's easier for me to track that way.

Original comment by andre.me...@gmail.com on 6 Jan 2011 at 10:51

• Changed state: Fixed