USFM Document import - parsing error

[eb1]

Adapt It Mobile Version: 1.0 beta 11

Severity: 1 - data corruption

Description: I received a USFM document yesterday that reads RTL. There seems to be a parsing error in AIM such that the markers are not getting flagged as markers, and instead end up as regular source. Even the chapter breaks are not identified, which is causing the entire document to be read in as one chapter. Since AIM reads one chapter in at a time for editing, it meant that the whole file was being read in -- and everything was really slow.

I'm not sure if this condition is a general issue for our parser on RTL documents, or if it's specific to this one document. Investigating...

[eb1]

Fixed with previous checkin:

The problem wasn't RTL scripts; instead, AIM was not checking for upper case file extensions, so the parser was loading the file as if it were plain text. With the fix, any file extension we support should be parsed correctly (regardless of the extension case).