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

Adjust joining lines into paragraphs

Browse Source 
Add more margin processing
This commit is contained in:
MisterAP 2024-10-28 18:36:06 +00:00
parent cec78caa1f
commit 80b60a7389
1 changed files with 27 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

37
src/calibre/ebooks/pdf/reflow.py
View File

@ -72,8 +72,8 @@ RIGHT_FACTOR = 1.8
CENTER_FACTOR = 0.15 CENTER_FACTOR = 0.15
# How near does text right need to be to right margin # How near does text right need to be to right margin
# to be considered right aligned. 0.1 = 10% # to be considered right aligned. 0.05 = 5%
RIGHT_FLOAT_FACTOR = 0.1 RIGHT_FLOAT_FACTOR = 0.05
#### Indents and line spacing #### Indents and line spacing
# How near must pixel values be to appear the same # How near must pixel values be to appear the same
@ -289,6 +289,18 @@ class Text(Element):
has_float = '<span style="float:right">' has_float = '<span style="float:right">'
has_gap = 1 has_gap = 1
#else leave has_gap #else leave has_gap
old_float = re.match('^(.*)(<span style="float:right">.*)</span>\s*$', self.raw)
if old_float:
# There is already a float as parts of a line are near the right.
# Remove the </span> and put it after this part
r1 = old_float.group(1)
r2 = old_float.group(2)
if not r1:
r1 = ''
if not r2:
r2 = ''
self.raw = r1 + r2
has_float = ' ' # Empty, but True
# Insert multiple spaces # Insert multiple spaces
while has_gap > 0: while has_gap > 0:
self.text_as_string += ' ' self.text_as_string += ' '
@ -967,17 +979,23 @@ class Page:
# Can two lines be merged into one paragraph? # Can two lines be merged into one paragraph?
# Some PDFs have a wandering left margin which is consistent on a page # Some PDFs have a wandering left margin which is consistent on a page
# but not within the whole document. Hence use self.stats_left # but not within the whole document. Hence use self.stats_left
# Try to avoid close double quote at end of one and open double quote at start of next # Try to avoid close double quote at end of one and open double quote at start of next.
# #
# The left can wander by a few (SAME_INDENT) pixels.
# "float:left" occurs where there is a multi-line character, so indentation is messed up # "float:left" occurs where there is a multi-line character, so indentation is messed up
lchar = re.match('.*([^ ])\s*$', first_text.text_as_string)
last_char = ' ' # Nothing interesting
if lchar is not None:
last_char = lchar.group(1) # Final non-space char
same_left = bool(first_text.last_left-SAME_INDENT <= second_text.left <= first_text.last_left+SAME_INDENT)
if ((second_text.left < left + second_text.average_character_width \ if ((second_text.left < left + second_text.average_character_width \
and (second_text.left == first_text.last_left \ and (same_left \
or (second_text.left < first_text.last_left \ or (second_text.left < first_text.last_left \
and (first_text.indented > 0 or '"float:left"' in first_text.raw)))) \ and (first_text.indented > 0 or '"float:left"' in first_text.raw)))) \
or (second_text.left == first_text.last_left \ or (same_left \
and first_text.indented == 0 \ and first_text.indented == 0 \
and second_text.left >= indent) \ and second_text.left >= indent) \
or (second_text.left == first_text.last_left \ or (same_left \
and first_text.indented == second_text.indented \ and first_text.indented == second_text.indented \
and second_text.indented > 1) \ and second_text.indented > 1) \
or (second_text.left >= first_text.last_left \ or (second_text.left >= first_text.last_left \
@ -987,10 +1005,9 @@ class Page:
and first_text.bottom + stats.line_space + (stats.line_space*LINE_FACTOR) \ and first_text.bottom + stats.line_space + (stats.line_space*LINE_FACTOR) \
>= second_text.bottom \ >= second_text.bottom \
and first_text.final_width > self.width*self.opts.unwrap_factor \ and first_text.final_width > self.width*self.opts.unwrap_factor \
and not (re.match('.*[.!?].$', first_text.text_as_string) is not None \ and not ( (last_char == '\u0022' and second_text.text_as_string[0] == '\u0022') \
and ((first_text.text_as_string[-1] == '\u0022' and second_text.text_as_string[0] == '\u0022') \ or (last_char == '\u2019' and second_text.text_as_string[0] == '\u2018') \
or (first_text.text_as_string[-1] == '\u2019' and second_text.text_as_string[0] == '\u2018') \ or (last_char == '\u201d' and second_text.text_as_string[0] == '\u201c')):
or (first_text.text_as_string[-1] == '\u201d' and second_text.text_as_string[0] == '\u201c'))):
# This has checked for single quotes (9...6), double quotes (99...66), and "..." # This has checked for single quotes (9...6), double quotes (99...66), and "..."
# at end of 1 line then start of next as a check for Don't merge # at end of 1 line then start of next as a check for Don't merge
return True return True