[PDFBOX-770] Greek text extraction

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.2.0, 1.2.1, 1.3.1
• Fix Version/s: 1.5.0
• Component/s: Text extraction
• Environment: Ubuntu 10.04
• Assignee: Andreas Lehmkühler
• Reporter: Manos Karampasis
• Created: 04/Jul/10 17:40
• Updated: 08/Mar/11 16:37
• Resolved: 08/Mar/11 16:36

Description

Greek text extraction error
Ι have a greek pdf but
a) after extraction the greek letter π is extracted as pi

for expamle
original text in pdf
"φυσικών προσώπων"

extracted text
"φυσικών piροσώpiων"

b) the greek letter μ is displayed as µ
there is no difference in display except that is different encoding and when searching for μ cannot find it (you find only the uppercase Μ)
if you copy μ as displayed search for that is working fine

e.g. the word is displayed as "κλίµακας" but it is different from the typed word κλίμακα due to the letter μ

due to this problem solr is not indexing documents correctly

is there any configuration I can make?