- corrected caching error

- added lorem ipsum pdf test docs
- introduced a new test for detecting if some unwanted caching is occuring
- muting "Unsupported" warning
This commit is contained in:
ari committed 2019-01-23 15:56:25 -05:00
1 parent ac2c462561
commit 98f18c3e32
7 files changed
+43 -8

No files matched your search

+6 -6
View File
@@ -14,38 +14,38 @@ log = get_logger('commands')
# No Support
# @PDFCommands.add('W','W_a', 'sh', 'ET', 'BX', 'EX' ,'BI', 'ID', 'gs')
# def do_no_support(stack: PDFStateStack) -> PDFStateStack:
# # log.warning("Unsupported command")
# # log.debug("Unsupported command")
# return stack
# marked content operators
@PDFCommands.add('MP')
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)
return stack
@PDFCommands.add('DP')
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)
return stack
@PDFCommands.add('BMC')
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)
return stack
@PDFCommands.add('BDC')
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)
return stack
@PDFCommands.add('EMC')
def do_EMC(stack: PDFStateStack) -> PDFStateStack:
# self.device.end_tag(tag)
log.warning("Unsupported command - end_tag")
log.debug("Unsupported command - end_tag")
return stack
@PDFCommands.add('q')
@@ -90,9 +90,9 @@ class LTComponent(LTItem):
##
class LTContainer(LTComponent):
def __init__(self, bbox: Bbox, objs: List[LTItem] = []):
def __init__(self, bbox: Bbox, objs: List[LTItem] = None):
LTComponent.__init__(self, bbox)
self._objs: List[LTItem] = objs
self._objs: List[LTItem] = [] if objs is None else objs
return
def __iter__(self):
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()