I've got some weird filenames

[hostmaster]

I've got some weird filenames seems filenames are not decoded properly. I'm not sure but these are PDF files mostly

=== CUT BEGIN === -rwx------ 1 khomyi khomyi 0 Jan 1 1970 Introduction to Probability and Statistics Using R.pdf ?????????? ? ? ? ? ? IYOCGwP_book1.pdfk rel='http://schemas.google.com/acl/2007#accessCon? -rwx------ 1 khomyi khomyi 0 Jan 1 1970 Jacob Perkins - Python Text Processing with NTLK 2.0 Cookbook - 2010.pdf

=== CUT END ===

[jcline]

Is it an actual PDF you uploaded to Google drive, or is it one that was shared with you or uploaded to Google docs? What about the ones that are correct?

It does appear as though it is reading in the wrong thing from the XML, at a guess... Not sure how, though.

[hostmaster]

This are files which I upload to Google Docs and some of them have been uploaded to Google Drive via Windows desktop client. I also found a couple of broken MP4 files which have been uploaded to Google Docs example below === CUT BEGIN === -rwx------ 1 khomyi khomyi 0 Jan 1 1970 2012-01-15-325.MP4 Authority -rwx------ 1 khomyi khomyi 0 Jan 1 1970 2012-01-15-336.MP4 -rwx------ 1 khomyi khomyi 0 Jan 1 1970 2012-01-15-337.MP4tViewed -rwx------ 1 khomyi khomyi 0 Jan 1 1970 2012-01-15-350.MP4esourceId === CUT END === Original file names are ended with MP4.

Seems like this is not PDF specific issue.

[jcline]

If you could try it on a commit on or after 2a361a3eeb752d20d3e28c6215c29f7617635953, which is the most recent at this time, I would appreciate it. I fixed a few things that could be causing it to read past the end of the string.

[hostmaster]

Unfortunately I do not see any changes

[jcline]

I recently changed large portions of the string handling. Any changes? If not I have another idea that may work.

[hostmaster]

Seems like everything is fine now. Thanks !