4 Commits
Author SHA1 Message Date
ari bab104885f - fixed caching error when running the interpreter multiple times on items with xobjects 2019-01-23 15:56:46 -05:00
ari 98f18c3e32 - corrected caching error
- added lorem ipsum pdf test docs
- introduced a new test for detecting if some unwanted caching is occuring
- muting "Unsupported" warning
2019-01-23 15:56:25 -05:00
ari ac2c462561 - improved html converter 2019-01-22 13:16:46 -05:00
ari 19712f1f0f cleaner html converter 2019-01-22 13:16:17 -05:00
13 changed files with 90 additions and 119 deletions

No files matched your search

+1 -1
View File
@@ -1 +1 @@
v1.3.4 v1.3.6
+1 -1
View File
@@ -97,7 +97,7 @@ OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
""" """
__version__ = (1, 3, 4) __version__ = (1, 3, 6)
if __name__ == '__main__': if __name__ == '__main__':
print( print(
+2 -2
View File
@@ -27,9 +27,9 @@ def convert_file(
): ):
if output_file is None: if output_file is None:
output_file = os.path.join( output_file = os.path.join(
os.path.dirname(input_file.name), os.path.dirname(input_file),
os.path.basename( os.path.basename(
re.sub(r"\.\w+$", "."+out_type, input_file.name) re.sub(r"\.\w+$", "."+out_type, input_file)
) )
) )
if re.search(re.escape(out_type) + r'$', output_file ) is None: if re.search(re.escape(out_type) + r'$', output_file ) is None:
+38 -101
View File
@@ -5,6 +5,8 @@ from ..interpreter import LTImage, LTTextBlock, LTCharBlock, LTChar, LTCurve, LT
from ..interpreter.commands import LTItem from ..interpreter.commands import LTItem
from ..interpreter.commands.state import CurvePath, PDFColor from ..interpreter.commands.state import CurvePath, PDFColor
INF = (1<<31) - 1
def convert_to_html( def convert_to_html(
input_file_path: str, input_file_path: str,
output_file_path: str, output_file_path: str,
@@ -79,33 +81,33 @@ def render_page(html: HTMLMaker, ltpage: PageInterpreter):
}): }):
html.write(item.get_text()) html.write(item.get_text())
elif isinstance(item, LTTextBlock): elif isinstance(item, LTTextBlock):
for char in item: with html.elm('span', {'class': 'text-block'}):
render(char) for char in item:
render(char)
elif isinstance(item, LTCharBlock): elif isinstance(item, LTCharBlock):
with html.elm('span', {'class': 'char-block'}): with html.elm('span', { 'class': 'char-block' }, {
with html.elm('p', { 'class': 'text-block' }, { 'position': 'absolute',
'position': 'absolute', 'left': f'{item.x0}',
'left': f'{item.x0}', 'bottom': f'{item.y0}',
'bottom': f'{item.y0}', 'font-size': f'{item.size}px',
'font-size': f'{item.size}px', 'font-family': item.fontname,
'font-family': item.fontname, 'font-weight': item.font.font_weight,
'font-weight': item.font.font_weight, 'color': get_color(item.color, default="black"),
'color': get_color(item.color, default="black"), 'width': str(item.width) + "px",
'width': str(item.width) + "px", 'min-width': str(item.width) + "px",
'min-width': str(item.width) + "px", 'max-width': str(item.width) + "px",
'max-width': str(item.width) + "px", 'height': str(item.height) + "px",
'height': str(item.height) + "px", 'min-height': str(item.height) + "px",
'min-height': str(item.height) + "px", 'max-height': str(item.height) + "px",
'max-height': str(item.height) + "px", 'transform': f'skew({item.font.italic_angle}deg, 0deg)',
'transform': f'skew({item.font.italic_angle}deg, 0deg)', 'text-align': 'center',
'text-align': 'center', 'display': 'flex',
'display': 'flex', 'flex-flow': 'row nowrap',
'flex-flow': 'row nowrap', 'justify-content': 'space-between',
'justify-content': 'space-between', # 'padding-left': item.font.leading
# 'padding-left': item.font.leading }):
}): for char in item:
for char in item: render(char)
render(char)
with html.elm('div', { "class": 'page', "id": f"page-{ltpage.page_num}" }, { with html.elm('div', { "class": 'page', "id": f"page-{ltpage.page_num}" }, {
'width': f'{ltpage.width}px', 'width': f'{ltpage.width}px',
@@ -147,7 +149,7 @@ def render_curve(html: HTMLMaker, ltpage: PageInterpreter, item: LTCurve):
'right': f'{item.x1}px', 'right': f'{item.x1}px',
'width': f'{item.width}px', 'width': f'{item.width}px',
'height': f'{item.height}px', 'height': f'{item.height}px',
}) as svg: }):
html.singleton('path', { html.singleton('path', {
"d": path, "d": path,
**path_css **path_css
@@ -156,11 +158,6 @@ def render_curve(html: HTMLMaker, ltpage: PageInterpreter, item: LTCurve):
}) })
def create_svg_path(item: LTCurve): def create_svg_path(item: LTCurve):
css = {
'position': 'absolute',
'left': f'{item.x0}px',
'bottom': f'{item.y0}px',
}
path = "" path = ""
svg_attr = { svg_attr = {
'class': 'curve', 'class': 'curve',
@@ -215,73 +212,13 @@ def create_svg_path(item: LTCurve):
return path, svg_attr return path, svg_attr
def __render_curve(html: HTMLMaker, ltpage: PageInterpreter, item: LTCurve): # def render_textblock(html: HTMLMaker, ltpage: PageInterpreter, item: LTTextBlock):
css = { # (x0, y0, x1, y1) = (INF, INF, -INF, -INF)
'position': 'absolute',
'left': f'{item.x0}px',
'bottom': f'{item.y0}px',
'top': f'{item.y1}px',
'right': f'{item.x1}px',
'width': f'{item.width}px',
'height': f'{item.height}px',
}
path = ""
svg_attr = {
'class': 'curve',
'z-index': 2
}
path_css = {
} # for c_block in item:
# x0 = min(x0, c_block.x0)
# Compute ViewBox # y0 = min(y0, c_block.y0)
p_vals = { "x": [], "y": [] } # x1 = max(x1, c_block.x1)
# y1 = max(y1, c_block.y1)
for p in item.paths:
for pt in p.points: # with html.elm('span')
p_vals['x'].append(pt.x)
p_vals['y'].append(pt.y)
mx_x = max([*p_vals['x'], 1])
mn_x = min([*p_vals['x'], 0])
mx_y = max([*p_vals['y'], 1])
mn_y = min([*p_vals['y'], 0])
if mx_x == mn_x:
mn_x -= 1
if mx_y == mn_y:
mn_y -= 1
dx = 1/( mx_x - mn_x )
dy = (item.height/item.width)/( mx_y - mn_y )
ROUND_VAL = 5
svg_attr.update({
# "viewBox": f"{min(p_vals['x'])} {min(p_vals['y'])} {max(p_vals['x'])} {max(p_vals['y'])}",
"viewBox": f"{round(mn_x*dx, ROUND_VAL)} {round(mn_y*dy, ROUND_VAL)} {round(mx_x*dx , ROUND_VAL)} {round(mx_y*dy, ROUND_VAL)}",
})
# Compute shape
for p in item.paths:
if p.method == CurvePath.METHOD.MOVE_TO:
path += f'M{" ".join( f"{round(dx*pt.x, ROUND_VAL)} {round(dy*pt.y, ROUND_VAL)}" for pt in p.points )} '
elif p.method == CurvePath.METHOD.LINE_TO:
path += f'L{" ".join( f"{round(dx*pt.x, ROUND_VAL)} {round(dy*pt.y, ROUND_VAL)}" for pt in p.points )} '
elif p.method == CurvePath.METHOD.CLOSE_PATH:
path += 'Z'
elif p.method == CurvePath.METHOD.CURVE_BOTH_TO:
path += f'C{" ".join( f"{round(dx*pt.x, ROUND_VAL)} {round(dy*pt.y, ROUND_VAL)}" for pt in p.points )} '
stroke_col = get_color(item.stroke)
fill_col = get_color(item.fill)
if fill_col:
path_css['fill'] = fill_col
if stroke_col:
path_css['stroke'] = stroke_col
with html.elm('svg', svg_attr, css):
html.singleton('path', {
"d": path,
**path_css
}, {
**path_css
})
+6 -6
View File
@@ -14,38 +14,38 @@ log = get_logger('commands')
# No Support # No Support
# @PDFCommands.add('W','W_a', 'sh', 'ET', 'BX', 'EX' ,'BI', 'ID', 'gs') # @PDFCommands.add('W','W_a', 'sh', 'ET', 'BX', 'EX' ,'BI', 'ID', 'gs')
# def do_no_support(stack: PDFStateStack) -> PDFStateStack: # def do_no_support(stack: PDFStateStack) -> PDFStateStack:
# # log.warning("Unsupported command") # # log.debug("Unsupported command")
# return stack # return stack
# marked content operators # marked content operators
@PDFCommands.add('MP') @PDFCommands.add('MP')
def do_MP(stack: PDFStateStack, tag) -> PDFStateStack: def do_MP(stack: PDFStateStack, tag) -> PDFStateStack:
log.warning(f"Unsupported command - do_tag {[tag]}") log.debug(f"Unsupported command - do_tag {[tag]}")
# self.device.do_tag(tag) # self.device.do_tag(tag)
return stack return stack
@PDFCommands.add('DP') @PDFCommands.add('DP')
def do_DP(stack: PDFStateStack, tag, props=None) -> PDFStateStack: def do_DP(stack: PDFStateStack, tag, props=None) -> PDFStateStack:
log.warning(f"Unsupported command - do_tag {[tag.name, type(tag), props]}") log.debug(f"Unsupported command - do_tag {[tag.name, type(tag), props]}")
# self.device.do_tag(tag) # self.device.do_tag(tag)
return stack return stack
@PDFCommands.add('BMC') @PDFCommands.add('BMC')
def do_BMC(stack: PDFStateStack, tag) -> PDFStateStack: def do_BMC(stack: PDFStateStack, tag) -> PDFStateStack:
log.warning(f"Unsupported command - begin_tag {[tag.name, type(tag)]}") log.debug(f"Unsupported command - begin_tag {[tag.name, type(tag)]}")
# self.device.begin_tag(tag) # self.device.begin_tag(tag)
return stack return stack
@PDFCommands.add('BDC') @PDFCommands.add('BDC')
def do_BDC(stack: PDFStateStack, tag, props=None) -> PDFStateStack: def do_BDC(stack: PDFStateStack, tag, props=None) -> PDFStateStack:
log.warning(f"Unsupported command - begin_tag {[tag.name, type(tag), props]}") log.debug(f"Unsupported command - begin_tag {[tag.name, type(tag), props]}")
# self.device.begin_tag(tag) # self.device.begin_tag(tag)
return stack return stack
@PDFCommands.add('EMC') @PDFCommands.add('EMC')
def do_EMC(stack: PDFStateStack) -> PDFStateStack: def do_EMC(stack: PDFStateStack) -> PDFStateStack:
# self.device.end_tag(tag) # self.device.end_tag(tag)
log.warning("Unsupported command - end_tag") log.debug("Unsupported command - end_tag")
return stack return stack
@PDFCommands.add('q') @PDFCommands.add('q')
@@ -90,9 +90,9 @@ class LTComponent(LTItem):
## ##
class LTContainer(LTComponent): class LTContainer(LTComponent):
def __init__(self, bbox: Bbox, objs: List[LTItem] = []): def __init__(self, bbox: Bbox, objs: List[LTItem] = None):
LTComponent.__init__(self, bbox) LTComponent.__init__(self, bbox)
self._objs: List[LTItem] = objs self._objs: List[LTItem] = [] if objs is None else objs
return return
def __iter__(self): def __iter__(self):
+1 -1
View File
@@ -10,7 +10,7 @@ import sys
with open('README.md') as readme_file: with open('README.md') as readme_file:
SETUP = { SETUP = {
"name": "pdfmajor", "name": "pdfmajor",
"version": "1.3.4", "version": "1.3.6",
"packages": find_packages(include=['pdfmajor*']), "packages": find_packages(include=['pdfmajor*']),
"install_requires": [ "install_requires": [
'pycryptodome', 'pycryptodome',
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
+35
View File
@@ -0,0 +1,35 @@
import os
from tqdm import tqdm
# from pdfmajor.extractor import extract_items_from_pdf
from pdfmajor.interpreter import LTCharBlock, PDFInterpreter, logging
from pdfmajor.interpreter import PageInterpreter
from pdfmajor.interpreter.commands import LTContainer
CUR_PATH = os.path.dirname(os.path.dirname(__file__))
INPUT_FOLDER = os.path.join(
CUR_PATH, "tests/samples/pdf"
)
FILES = os.listdir(INPUT_FOLDER)
NUM_LOOPS = 10
def count_elms(file_path: str):
count = 0
for page in PDFInterpreter(file_path):
for item in page:
if isinstance(item, LTContainer):
count += len([*item])
else:
count += 1
return count
with tqdm(desc='Cache Test', total=len(FILES)*NUM_LOOPS) as waiter:
for file_name in FILES:
waiter.set_description(f"Cache Test - [{file_name}]")
file_path = os.path.join(INPUT_FOLDER, file_name)
last_count = count_elms(file_path)
for _ in range(NUM_LOOPS):
new_count = count_elms(file_path)
assert new_count == last_count
last_count = new_count
waiter.update()
+4 -5
View File
@@ -52,11 +52,10 @@ def main(args=None):
if len(parsed_args.files) == 1: if len(parsed_args.files) == 1:
if parsed_args.output_file != "-": if parsed_args.output_file != "-":
func_args['output_file'] = open(parsed_args.output_file, "wb") func_args['output_file'] = open(parsed_args.output_file, "wb")
with open(parsed_args.files[0], 'rb') as input_file: convert_file(
convert_file( parsed_args.files[0],
input_file, **func_args
**func_args )
)
else: else:
print("Multiple files not supported yet.") print("Multiple files not supported yet.")
return 1 return 1