Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

tweaked the auto-detection to handle cases where the vast majority of the lines are formatted as block or print

Browse Source
This commit is contained in:
ldolse 2011-01-09 18:38:52 +08:00
parent 696d925232
commit 0f109d699f
1 changed files with 13 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
src/calibre/ebooks/txt/processor.py
View File

@ -133,15 +133,21 @@ def detect_paragraph_type(txt):
hardbreaks = docanalysis.line_histogram(.55) hardbreaks = docanalysis.line_histogram(.55)
if hardbreaks: if hardbreaks:
# Check for print # Determine print percentage
tab_line_count = len(re.findall('(?mu)^(\t|\s{2,}).+$', txt)) tab_line_count = len(re.findall('(?mu)^(\t|\s{2,}).+$', txt))
if tab_line_count / float(txt_line_count) >= .15: print_percent = tab_line_count / float(txt_line_count)
return 'print'
# Determine block percentage
# Check for block
empty_line_count = len(re.findall('(?mu)^\s*$', txt)) empty_line_count = len(re.findall('(?mu)^\s*$', txt))
if empty_line_count / float(txt_line_count) >= .15: block_percent = empty_line_count / float(txt_line_count)
return 'block'
# Compare the two types - the type with the larger number of instances wins
# in cases where only one or the other represents the vast majority of the document neither wins
if print_percent >= block_percent:
if .15 <= print_percent <= .75:
return 'print'
elif .15 <= block_percent <= .75:
return 'block'
# Assume unformatted text with hardbreaks if nothing else matches # Assume unformatted text with hardbreaks if nothing else matches
return 'unformatted' return 'unformatted'