Kozea/WeasyPrint
1
1
Fork 0
mirror of https://github.com/Kozea/WeasyPrint.git synced 2024-10-05 00:21:15 +03:00
Code Issues Projects Releases Wiki Activity

Implemented CSS3 overflow-wrap property

Browse Source 
Implemented the property as described by the W3C draft.  There is still a
bug with shaping characters which does not keep their shapes when wrapped
on new lines.  Implemented as overflow-wrap CSS property, but maybe it
should be refactored to -weasy-overflow-wrap.  Browser implementation is
inconsistent, so there was no clear answer.
This commit is contained in:
Frédérick Deslandes 2013-09-30 14:16:03 -04:00
parent 57b3dd5d07
commit f409ceeb25
5 changed files with 150 additions and 37 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
weasyprint/css/properties.py
View File

@ -119,6 +119,7 @@ INITIAL_VALUES = {
# CSS3 User Interface: http://www.w3.org/TR/css3-ui/#box-sizing
'box_sizing': 'content-box',
'overflow_wrap': 'normal',
# CSS3 Color: http://www.w3.org/TR/css3-color/#transparency
'opacity': 1,
@ -210,6 +211,7 @@ INHERITED = set("""
list_style_position
list_style_type
orphans
overflow_wrap
quotes
text_align
text_decoration

6
weasyprint/css/validation.py
View File

@ -1035,6 +1035,12 @@ def white_space(keyword):
"""``white-space`` property validation."""
return keyword in ('normal', 'pre', 'nowrap', 'pre-wrap', 'pre-line')
@validator()
@single_keyword
def overflow_wrap(keyword):
"""``overflow-wrap`` property validation."""
return keyword in ('normal', 'break-word')
@validator(unprefixed=True)
@single_keyword

10
weasyprint/tests/test_css_validation.py
View File

@ -559,6 +559,16 @@ def test_linear_gradient():
invalid('bottom left, blue')
@assert_no_logs
def test_overflow_wrap():
assert expand_to_dict('overflow-wrap: normal') == {
'overflow_wrap': 'normal'}
assert expand_to_dict('overflow-wrap: break-word') == {
'overflow_wrap': 'break-word'}
assert_invalid('overflow-wrap: none')
assert_invalid('overflow-wrap: normal, break-word')
@assert_no_logs
def test_radial_gradient():
red = (1, 0, 0, 1)

46
weasyprint/tests/test_layout.py
View File

@ -4705,6 +4705,52 @@ def test_hyphenate_limit_chars():
assert line_count('auto 2') == 2
@assert_no_logs
def test_overflow_wrap():
def get_lines(wrap, text):
page, = parse('''
<style>
body {width: 30px; overflow: hidden; }
span {overflow-wrap: %s; white-space: normal; }
</style>
<body style="-weasy-hyphens: auto;" lang="en">
<span>%s
''' % (wrap, text))
html, = page.children
body, = html.children
body_lines = [];
for line in body.children:
box, = line.children
textBox, = box.children
body_lines.append(textBox.text)
return body_lines
# break-word
lines = get_lines('break-word', 'aaaaaaaa')
assert len(lines) == 3
full_text = ''.join(line for line in lines)
assert full_text == 'aaaaaaaa'
# normal
lines = get_lines('normal', 'aaaaaaaa')
assert len(lines) == 1
full_text = ''.join(line for line in lines)
assert full_text == 'aaaaaaaa'
# break-word after hyphenation
lines = get_lines('break-word', 'hyphenation')
assert len(lines) == 5
full_text = ''.join(line for line in lines)
assert full_text == "hy\u2010phenation"
# break word after normal white-space wrap and hyphenation
lines = get_lines('break-word', 'I am a splitted word. I am an hyphenated word.')
assert len(lines) == 18
full_text = ''.join(line for line in lines)
assert full_text == "Iamasplittedword.Iamanhy\u2010phenatedword."
@assert_no_logs
def test_white_space():
"""Test the white-space property."""

123
weasyprint/text.py
View File

@ -58,6 +58,12 @@ ffi.cdef('''
PANGO_STRETCH_EXTRA_EXPANDED,
PANGO_STRETCH_ULTRA_EXPANDED
} PangoStretch;
typedef enum {
PANGO_WRAP_WORD,
PANGO_WRAP_CHAR,
PANGO_WRAP_WORD_CHAR
} PangoWrapMode;
typedef unsigned int guint;
typedef int gint;
@ -98,6 +104,8 @@ ffi.cdef('''
PangoLayout *layout, const char *text, int length);
void pango_layout_set_font_description (
PangoLayout *layout, const PangoFontDescription *desc);
void pango_layout_set_wrap (
PangoLayout *layout, PangoWrapMode wrap);
PangoFontDescription * pango_font_description_new (void);
@ -227,6 +235,12 @@ PANGO_STRETCH = {
'ultra-expanded': pango.PANGO_STRETCH_ULTRA_EXPANDED,
}
PANGO_WRAP_MODE = {
'WRAP_WORD' : pango.PANGO_WRAP_WORD,
'WRAP_CHAR' : pango.PANGO_WRAP_CHAR,
'WRAP_WORD_CHAR' : pango.PANGO_WRAP_WORD_CHAR
}
def utf8_slice(string, slice_):
return string.encode('utf-8')[slice_].decode('utf-8')
@ -318,6 +332,9 @@ class Layout(object):
def get_font_metrics(self):
context = pango.pango_layout_get_context(self.layout)
return FontMetrics(context, self.font)
def set_wrap(self, wrap_mode):
pango.pango_layout_set_wrap(self.layout, wrap_mode)
class FontMetrics(object):
@ -470,46 +487,78 @@ def split_first_line(text, style, hinting, max_width, line_width):
hyphens = style.hyphens
lang = style.lang
total, left, right = style.hyphenate_limit_chars
if hyphens in ('none', 'manual') or lang not in pyphen.LANGUAGES:
# No automatic hyphenation
return first_line_metrics(first_line, text, layout, resume_at)
elif len(next_word) < total:
# Next word is too small
return first_line_metrics(first_line, text, layout, resume_at)
hyphenated = False
# Automatic hyphenation possible and next word is long enough
if hyphens not in ('none', 'manual') and lang in pyphen.LANGUAGES and len(next_word) >= total:
first_line_width, _height = get_size(first_line)
space = max_width - first_line_width
if style.hyphenate_limit_zone.unit == '%':
limit_zone = max_width * style.hyphenate_limit_zone.value / 100.
else:
limit_zone = style.hyphenate_limit_zone.value
if space > limit_zone or space < 0:
# The next word does not fit, try hyphenation
dictionary_key = (lang, left, right, total)
dictionary = PYPHEN_DICTIONARY_CACHE.get(dictionary_key)
if dictionary is None:
dictionary = pyphen.Pyphen(lang=lang, left=left, right=right)
PYPHEN_DICTIONARY_CACHE[dictionary_key] = dictionary
for first_word_part, _ in dictionary.iterate(next_word):
new_first_line = (
first_part + first_word_part + style.hyphenate_character)
temp_layout = create_layout(
new_first_line, style, hinting, max_width)
temp_lines = temp_layout.iter_lines()
temp_first_line = next(temp_lines, None)
temp_second_line = next(temp_lines, None)
if (temp_second_line is None and space >= 0) or space < 0:
hyphenated = True
# TODO: find why there's no need to .encode
resume_at = len(first_part + first_word_part)
layout = temp_layout
first_line = temp_first_line
second_line = temp_second_line
temp_first_line_width, _height = get_size(temp_first_line)
if temp_first_line_width <= max_width:
break
# Step 5: Try to break word if it's too long for the line
overflow_wrap = style.overflow_wrap
first_line_width, _height = get_size(first_line)
space = max_width - first_line_width
if style.hyphenate_limit_zone.unit == '%':
limit_zone = max_width * style.hyphenate_limit_zone.value / 100.
else:
limit_zone = style.hyphenate_limit_zone.value
# If we can break words and the first line is too long
if overflow_wrap == 'break-word' and space < 0:
if hyphenated:
# Is it really OK to remove hyphenation for word-break ?
new_first_line = (new_first_line.rstrip(new_first_line[-(len(style.hyphenate_character)):]))
if second_line is not None:
second_line_index = second_line.start_index
second_part = utf8_slice(text, slice(second_line_index, None))
new_first_line += second_part
hyphenated = False
# TODO: Modify code to preserve W3C condition:
# "Shaping characters are still shaped as if the word were not broken"
# The way new lines are processed in this function (one by one with no memory of the last)
# prevents shaping characters (arabic, for instance) from keeping their shape when
# wrapped on the next line with pango layout. Maybe insert Unicode shaping characters
# in text ?
temp_layout = create_layout(new_first_line, style, hinting, max_width)
temp_layout.set_wrap(PANGO_WRAP_MODE['WRAP_WORD_CHAR'])
temp_lines = temp_layout.iter_lines()
temp_first_line = next(temp_lines, None)
temp_second_line = next(temp_lines, None)
temp_second_line_index = len(new_first_line) if temp_second_line is None else temp_second_line.start_index
resume_at = temp_second_line_index
first_part = utf8_slice(text, slice(temp_second_line_index))
layout = create_layout(first_part, style, hinting, max_width)
lines = layout.iter_lines()
first_line = next(lines, None)
hyphenated = False
if space > limit_zone or space < 0:
# The next word does not fit, try hyphenation
dictionary_key = (lang, left, right, total)
dictionary = PYPHEN_DICTIONARY_CACHE.get(dictionary_key)
if dictionary is None:
dictionary = pyphen.Pyphen(lang=lang, left=left, right=right)
PYPHEN_DICTIONARY_CACHE[dictionary_key] = dictionary
for first_word_part, _ in dictionary.iterate(next_word):
new_first_line = (
first_part + first_word_part + style.hyphenate_character)
temp_layout = create_layout(
new_first_line, style, hinting, max_width)
temp_lines = temp_layout.iter_lines()
temp_first_line = next(temp_lines, None)
temp_second_line = next(temp_lines, None)
if (temp_second_line is None and space >= 0) or space < 0:
hyphenated = True
# TODO: find why there's no need to .encode
resume_at = len(first_part + first_word_part)
layout = temp_layout
first_line = temp_first_line
second_line = temp_second_line
temp_first_line_width, _height = get_size(temp_first_line)
if temp_first_line_width <= max_width:
break
return first_line_metrics(first_line, text, layout, resume_at, hyphenated)