diff --git a/pdfmajor/converters/html.py b/pdfmajor/converters/html.py new file mode 100644 index 0000000..bcb24a3 --- /dev/null +++ b/pdfmajor/converters/html.py @@ -0,0 +1,15 @@ +from ..extractor import extract_items_from_pdf + +def convert_to_html( + input_file_path: str, + output_file: io.TextIOWrapper = None, + image_folder_path: str = None, + codec: str = 'utf-8', + maxpages: int = 0, + password: str = None, + caching: bool = True, + check_extractable: bool = True, + pagenos: List[int] = None, + debug_level: int = logging.WARNING, +): + \ No newline at end of file diff --git a/pdfmajor/extractor/PDFExtractor.py b/pdfmajor/extractor/PDFExtractor.py deleted file mode 100644 index 209629b..0000000 --- a/pdfmajor/extractor/PDFExtractor.py +++ /dev/null @@ -1,12 +0,0 @@ -from ..utils import get_logger - -log = get_logger(__name__) - -from ..layouts import PDFLayoutAnalyzer, PDFPage, LTPage - -class PDFExctractor(PDFLayoutAnalyzer): - def end_page(self, page: PDFPage): - assert not self._stack, str(len(self._stack)) - assert isinstance(self.cur_item, LTPage), str(type(self.cur_item)) - self.pageno += 1 - return self.cur_item diff --git a/pdfmajor/extractor/__init__.py b/pdfmajor/extractor/__init__.py index 50e423d..bf7e5a9 100644 --- a/pdfmajor/extractor/__init__.py +++ b/pdfmajor/extractor/__init__.py @@ -6,17 +6,13 @@ from io import BytesIO, StringIO, TextIOWrapper from tqdm import tqdm -from ..interpreter.PDFResourceManager import PDFResourceManager from ..interpreter.PDFPage import PDFPage -from ..interpreter.PDFPageInterpreter import PDFPageInterpreter -from ..interpreter_v2 import PDFPageInterpreter as PDFPageInterpreterV2 +from ..interpreter_v2 import PDFPageInterpreter from ..imagewriter import ImageWriter from ..utils import set_log_level, get_logger, logging -from .PDFExtractor import PDFExctractor log = get_logger(__name__) - def extract_items_from_pdf( input_file_path: str, maxpages: int = 0, @@ -29,42 +25,10 @@ def extract_items_from_pdf( set_log_level(debug_level) log.debug("Creating resources....") - rsrcmgr = PDFResourceManager(caching=caching) - device = PDFExctractor( - rsrcmgr=rsrcmgr, - pageno=0 - ) - interpreter = PDFPageInterpreter(rsrcmgr, device) + interpreter = PDFPageInterpreter() log.debug("Reading pages....") with open(input_file_path, 'rb') as input_file: pages = PDFPage.get_pages(input_file, pagenos=pagenos, maxpages=maxpages, password=password, caching=caching, check_extractable=check_extractable) - - page_waiter = tqdm(enumerate(pages), disable=debug_level > logging.INFO) - for _, page in page_waiter: - page_waiter.set_description("Processing pdf...") - yield interpreter.process_page(page) - page_waiter.close() - device.close() - - -def v2_extract_items_from_pdf( - input_file_path: str, - maxpages: int = 0, - password: str = None, - caching: bool = True, - check_extractable: bool = True, - pagenos: List[int] = None, - debug_level: int = logging.WARNING, -): - set_log_level(debug_level) - - log.debug("Creating resources....") - rsrcmgr = PDFResourceManager(caching=caching) - interpreter = PDFPageInterpreterV2(rsrcmgr) - log.debug("Reading pages....") - with open(input_file_path, 'rb') as input_file: - pages = PDFPage.get_pages(input_file, pagenos=pagenos, maxpages=maxpages, password=password, caching=caching, check_extractable=check_extractable) - page_waiter = tqdm(enumerate(pages), disable=debug_level > logging.INFO) for _, page in page_waiter: page_waiter.set_description("Processing pdf...") diff --git a/pdfmajor/interpreter/PDFPage.py b/pdfmajor/interpreter/PDFPage.py index 73757a1..cb47b52 100644 --- a/pdfmajor/interpreter/PDFPage.py +++ b/pdfmajor/interpreter/PDFPage.py @@ -91,12 +91,12 @@ class PDFPage(object): tree_type = tree.get('type') if tree_type is LITERAL_PAGES and 'Kids' in tree: - # log.info('Pages: Kids=%r', tree['Kids']) + log.debug('Pages: Kids=%r', tree['Kids']) for c in list_value(tree['Kids']): for x in search(c, tree): yield x elif tree_type is LITERAL_PAGE: - log.info('Page: %r', tree) + log.debug('Page: %r', tree) yield (objid, tree) pages = False diff --git a/pdfmajor/interpreter_v2/PageInterpreter.py b/pdfmajor/interpreter_v2/PageInterpreter.py new file mode 100644 index 0000000..82339fc --- /dev/null +++ b/pdfmajor/interpreter_v2/PageInterpreter.py @@ -0,0 +1,78 @@ +from pdfmajor.interpreter.PDFPage import PDFPage +from pdfmajor.interpreter.PDFStream import list_value +from pdfmajor.interpreter.PDFContentParser import PDFContentParser +from pdfmajor.interpreter.PSStackParser import PSKeyword +from pdfmajor.interpreter.PSStackParser import keyword_name + +from .commands import PDFStateStack, PDFCommands +from .commands import LTXObject +from .utils import init_resources +from .XObjectInterpreter import XObjectInterpreter + +class PageInterpreter: + class PDFInterpreterError(Exception): pass + + def __init__(self, page: PDFPage, page_num: int, font_cache: dict = {}): + (x0, y0, x1, y1) = page.mediabox + if page.rotate == 90: + ctm = [0, -1, 1, 0, -y0, x1] + elif page.rotate == 180: + ctm = [-1, 0, 0, -1, x1, y1] + elif page.rotate == 270: + ctm = [0, 1, -1, 0, y1, -x0] + else: + ctm = [1, 0, 0, 1, -x0, -y0] + + # Locals + self.font_cache = font_cache + self.page = page + self.page_num = page_num + + # Init State + self.state: PDFStateStack = PDFStateStack() + self.state.t_matrix = ctm + self.state.resources = page.resources + # set some global states. + self.state.graphics.ncolor.color_space = self.state.graphics.scolor.color_space = None + if self.state.colorspace_map: + col_space = next(iter(self.state.colorspace_map.values())) + self.state.graphics.ncolor.color_space = self.state.graphics.scolor.color_space = col_space + # Init Resources + init_resources(self.state, self.font_cache) + + def __iter__(self): + parser = PDFContentParser(list_value(self.page.contents)) + for obj in parser: + if isinstance(obj, PSKeyword): + name = keyword_name(obj) + method = name.replace('*', '_a').replace('"', '_w').replace("'", '_q') + if method in PDFCommands.commands.keys(): + func = PDFCommands.commands.get(method) + nargs = func.__code__.co_argcount-1 + if nargs: + args = self.state.pop(nargs) + if len(args) == nargs: + func(self.state, *args) + else: + raise self.PDFInterpreterError(f"Invalid Number of Args provided {method}") + else: + # log.debug('exec: %s', name) + func(self.state) + else: + raise self.PDFInterpreterError('Unknown operator: %r' % name) + else: + self.state.argstack.append(obj) + + for complete_item in self.state.complete_layout_items: + if isinstance(complete_item, LTXObject): + interpeter = XObjectInterpreter( + streams=[complete_item.stream], + resources=complete_item.resources, + ctm=complete_item.t_matrix + ) + for item in interpeter: + complete_item.add(item) + yield complete_item + self.state.complete_layout_items = [] + def __repr__(self) -> str: + return f"" \ No newline at end of file diff --git a/pdfmajor/interpreter_v2/XObjectInterpreter.py b/pdfmajor/interpreter_v2/XObjectInterpreter.py new file mode 100644 index 0000000..afad9fa --- /dev/null +++ b/pdfmajor/interpreter_v2/XObjectInterpreter.py @@ -0,0 +1,71 @@ +from pdfmajor.utils import MATRIX_IDENTITY +from pdfmajor.interpreter.PSStackParser import literal_name +from pdfmajor.interpreter.PDFParser import PDFStream +from pdfmajor.interpreter.PSStackParser import PSEOF +from pdfmajor.interpreter.PSStackParser import PSKeyword +from pdfmajor.interpreter.PSStackParser import keyword_name +from pdfmajor.interpreter.PSStackParser import PSStackParser +from pdfmajor.interpreter.PDFResourceManager import PDFResourceManager +from pdfmajor.interpreter.constants import LITERAL_PDF, LITERAL_TEXT, LITERAL_FONT + +from pdfmajor.interpreter.PDFStream import resolve1, list_value, dict_value +from pdfmajor.interpreter.PDFStream.PDFObjRef import PDFObjRef +from pdfmajor.interpreter.PDFContentParser import PDFContentParser + +from .commands.state import PDFStateStack, PDFColorSpace, PREDEFINED_COLORSPACE +from .commands import PDFCommands, LTXObject +from .utils import init_resources + +class XObjectInterpreter: + + class PDFInterpreterError(Exception): pass + + def __init__(self, streams, resources, ctm): + self.streams = streams + self.font_cache = {} + + self.state = PDFStateStack() + self.state.t_matrix = ctm + self.state.resources = resources + + # set some global states. + self.state.graphics.ncolor.color_space = self.state.graphics.scolor.color_space = None + if self.state.colorspace_map: + col_space = next(iter(self.state.colorspace_map.values())) + self.state.graphics.ncolor.color_space = self.state.graphics.scolor.color_space = col_space + init_resources(self.state, self.font_cache) + + def __iter__(self): + parser = PDFContentParser(self.streams) + for obj in parser: + if isinstance(obj, PSKeyword): + name = keyword_name(obj) + method = name.replace('*', '_a').replace('"', '_w').replace("'", '_q') + if method in PDFCommands.commands.keys(): + func = PDFCommands.commands.get(method) + nargs = func.__code__.co_argcount-1 + if nargs: + args = self.state.pop(nargs) + if len(args) == nargs: + func(self.state, *args) + else: + raise self.PDFInterpreterError(f"Invalid Number of Args provided {method}") + else: + # log.debug('exec: %s', name) + func(self.state) + else: + raise self.PDFInterpreterError('Unknown operator: %r' % name) + else: + self.state.argstack.append(obj) + + for complete_item in self.state.complete_layout_items: + if isinstance(complete_item, LTXObject): + interpeter = self.__class__( + streams=[complete_item.stream], + resources=complete_item.resources, + ctm=complete_item.t_matrix + ) + for item in interpeter: + complete_item.add(item) + yield complete_item + self.state.complete_layout_items = [] \ No newline at end of file diff --git a/pdfmajor/interpreter_v2/__init__.py b/pdfmajor/interpreter_v2/__init__.py index 5bdd975..a123a2d 100644 --- a/pdfmajor/interpreter_v2/__init__.py +++ b/pdfmajor/interpreter_v2/__init__.py @@ -1,124 +1,72 @@ -from pdfmajor.utils import MATRIX_IDENTITY -from pdfmajor.interpreter.PSStackParser import literal_name -from pdfmajor.interpreter.PSStackParser import PSEOF -from pdfmajor.interpreter.PSStackParser import PSKeyword -from pdfmajor.interpreter.PSStackParser import keyword_name -from pdfmajor.interpreter.PSStackParser import PSStackParser -from pdfmajor.interpreter.PDFParser import PDFStream +from typing import List -from pdfmajor.interpreter.PDFStream import resolve1, list_value, dict_value -from pdfmajor.interpreter.PDFStream.PDFObjRef import PDFObjRef -from pdfmajor.interpreter.PDFContentParser import PDFContentParser +from tqdm import tqdm +from pdfmajor.interpreter.PDFPage import PDFPage +from pdfmajor.utils import set_log_level, get_logger, logging from .commands.state import PDFStateStack, PDFColorSpace, PREDEFINED_COLORSPACE from .commands import PDFCommands -from .commands.state.PDFItem import PDFXObject, PDFImage, PDFItem, PDFShape, PDFText +from .commands import LTCharBlock, LTChar +from .commands import LTCurve, LTLine, LTHorizontalLine, LTVerticalLine, LTRect +from .commands import LTImage +from .commands import LTXObject +from .PageInterpreter import PageInterpreter -def get_colorspace(spec): - if isinstance(spec, list): - name = literal_name(spec[0]) - else: - name = literal_name(spec) - if name == 'ICCBased' and isinstance(spec, list) and 2 <= len(spec): - return PDFColorSpace(name, PDFStream.validated_stream(spec[1])['N']) - elif name == 'DeviceN' and isinstance(spec, list) and 2 <= len(spec): - return PDFColorSpace(name, len(list_value(spec[1]))) - else: - return PREDEFINED_COLORSPACE.get(name) +log = get_logger(__name__) -class PDFPageInterpreter: +class PDFInterpreter: + class AccessError(Exception): pass - class PDFInterpreterError(Exception): pass - - def __init__(self, rsrcmgr): - self.rsrcmgr = rsrcmgr - self.state: PDFStateStack = PDFStateStack() + def __init__(self, + input_file_path: str, + maxpages: int = 0, + caching: bool = True, + check_extractable: bool = True, + password: str = None, + pagenos: List[int] = None, + preload: bool = False, + debug_level: int = logging.WARNING, + ): + self.input_file_path = input_file_path + self.maxpages = maxpages + self.caching = caching + self.check_extractable = check_extractable + self.password = password + self.pagenos = pagenos + self.debug_level = debug_level + self.__pages = [] + if preload: + for _ in self.__load_pages(): + pass + + def __load_pages(self): + set_log_level(self.debug_level) + log.info("Opening file...") + with open(self.input_file_path, 'rb') as input_file: + font_cache = {} + log.info("Parsing file...") + pages = PDFPage.get_pages( + input_file, + pagenos=self.pagenos, + maxpages=self.maxpages, + password=self.password, + caching=self.caching, + check_extractable=self.check_extractable + ) + for page_num, page in tqdm(enumerate(pages), desc="Reading pages...", disable=self.debug_level > logging.INFO): + self.__pages.append(PageInterpreter(page, page_num, font_cache)) + yield self.__pages[-1] + log.info(f"Done Reading {len(self.__pages)} pages.") + set_log_level(logging.WARNING) + if len(self.__pages) == 0: + raise self.AccessError("No pages found in pdf-file") - # init_resources(resources): - # Prepare the fonts and XObjects listed in the Resource attribute. - def init_resources(self, resources): - self.state.resources = resources - for (k, v) in iter(dict_value(self.state.resources).items()): - # log.debug('Resource: %r: %r', k, v) - if k == 'Font': - for (fontid, spec) in iter(dict_value(v).items()): - objid = None - if isinstance(spec, PDFObjRef): - objid = spec.objid - spec = dict_value(spec) - self.state.fontmap[fontid] = self.rsrcmgr.get_font(objid, spec) - elif k == 'ColorSpace': - for (csid, spec) in iter(dict_value(v).items()): - self.state.colorspace_map[csid] = get_colorspace(resolve1(spec)) - elif k == 'ProcSet': - self.rsrcmgr.get_procset(list_value(v)) - elif k == 'XObject': - for (xobjid, xobjstrm) in iter(dict_value(v).items()): - self.state.xobjmap[xobjid] = xobjstrm - - - # init_state(ctm) - # Initialize the text and graphic states for rendering a page. - def init_state(self, ctm): - self.state = PDFStateStack() - self.state.t_matrix = ctm - # set some global states. - self.state.graphics.ncolor.color_space = self.state.graphics.scolor.color_space = None - if self.state.colorspace_map: - col_space = next(iter(self.state.colorspace_map.values())) - self.state.graphics.ncolor.color_space = self.state.graphics.scolor.color_space = col_space - - # render_contents(resources, streams, ctm) - # Render the content streams. - # This method may be called recursively. - def render_contents(self, resources, streams, ctm=MATRIX_IDENTITY): - self.init_state(ctm) - self.init_resources(resources) - return self.execute(list_value(streams)) - - def process_page(self, page): - (x0, y0, x1, y1) = page.mediabox - if page.rotate == 90: - ctm = [0, -1, 1, 0, -y0, x1] - elif page.rotate == 180: - ctm = [-1, 0, 0, -1, x1, y1] - elif page.rotate == 270: - ctm = [0, 1, -1, 0, y1, -x0] + def __iter__(self): + if len(self.__pages) > 0: + set_log_level(self.debug_level) + for page in tqdm(self.__pages, desc="Reading pages from store...", disable=self.debug_level > logging.INFO): + yield page + set_log_level(logging.WARNING) else: - ctm = [1, 0, 0, 1, -x0, -y0] - return self.render_contents(page.resources, page.contents, ctm=ctm) - - def execute(self, streams): - parser = PDFContentParser(streams) - for obj in parser: - if isinstance(obj, PSKeyword): - name = keyword_name(obj) - method = name.replace('*', '_a').replace('"', '_w').replace("'", '_q') - if method in PDFCommands.commands.keys(): - func = PDFCommands.commands.get(method) - nargs = func.__code__.co_argcount-1 - if nargs: - args = self.state.pop(nargs) - if len(args) == nargs: - func(self.state, *args) - else: - raise self.PDFInterpreterError(f"Invalid Number of Args provided {method}") - else: - # log.debug('exec: %s', name) - func(self.state) - else: - raise self.PDFInterpreterError('Unknown operator: %r' % name) - else: - self.state.argstack.append(obj) - - for complete_item in self.state.complete_items: - if isinstance(complete_item, PDFXObject): - interpeter = self.__class__(self.rsrcmgr) - for item in interpeter.render_contents( - streams=[complete_item.stream], - resources=complete_item.resources, - ctm=complete_item.ctm - ): - complete_item.children.append(item) - yield complete_item - self.state.complete_items = [] \ No newline at end of file + for page in self.__load_pages(): + yield page \ No newline at end of file diff --git a/pdfmajor/interpreter_v2/commands/PDFCommands.py b/pdfmajor/interpreter_v2/commands/PDFCommands.py index d7801ed..039c50c 100644 --- a/pdfmajor/interpreter_v2/commands/PDFCommands.py +++ b/pdfmajor/interpreter_v2/commands/PDFCommands.py @@ -4,8 +4,8 @@ from pdfmajor.utils import MATRIX_IDENTITY, mult_matrix from pdfmajor.interpreter.PSStackParser import literal_name from pdfmajor.interpreter.PDFStream import PDFStream, list_value, dict_value from pdfmajor.interpreter.constants import LITERAL_FORM, LITERAL_IMAGE -from ..state import PDFStateStack, PDFColorSpace -from ..state.Curves import CurveMethod, CurvePath, CurvePoint +from .state import PDFStateStack, PDFColorSpace +from .state.Curves import CurveMethod, CurvePath, CurvePoint class PDFCommands: commands = {} diff --git a/pdfmajor/interpreter_v2/commands/__init__.py b/pdfmajor/interpreter_v2/commands/__init__.py index 23e16ae..c132cff 100644 --- a/pdfmajor/interpreter_v2/commands/__init__.py +++ b/pdfmajor/interpreter_v2/commands/__init__.py @@ -1,4 +1,9 @@ from .commands import PDFCommands from .state import PDFStateStack, PDFGraphicState, PDFTextState from .state.PDFGraphicState import PDFColor, PDFColorSpace, PREDEFINED_COLORSPACE -from .state.PDFTextState import PDFFont \ No newline at end of file +from .state import PDFFont, get_font +from .state import LTCharBlock, LTChar +from .state import LTCurve, LTLine, LTHorizontalLine, LTVerticalLine, LTRect +from .state import LTImage +from .state import LTXObject +from .state import LTItem, LTComponent, LTContainer \ No newline at end of file diff --git a/pdfmajor/interpreter_v2/commands/commands.py b/pdfmajor/interpreter_v2/commands/commands.py index 3c3ade6..48188a5 100644 --- a/pdfmajor/interpreter_v2/commands/commands.py +++ b/pdfmajor/interpreter_v2/commands/commands.py @@ -1,4 +1,4 @@ -from pdfmajor.utils import mult_matrix, MATRIX_IDENTITY +from pdfmajor.utils import mult_matrix, MATRIX_IDENTITY, apply_matrix_pt from pdfmajor.interpreter.PSStackParser import literal_name from pdfmajor.interpreter.PDFStream import PDFStream, list_value, dict_value from pdfmajor.interpreter.constants import LITERAL_FORM, LITERAL_IMAGE @@ -6,7 +6,8 @@ from pdfmajor.interpreter.constants import LITERAL_FORM, LITERAL_IMAGE from .PDFCommands import PDFCommands from .state import PDFStateStack, PDFGraphicState from .state.Curves import CurveMethod, CurvePath, CurvePoint -from .state.PDFItem import PDFImage, PDFShape, PDFText, PDFXObject +# from .state.PDFItem import PDFImage, PDFShape, PDFText, PDFXObject +from .state import make_char_block, make_curve, make_image, make_xobject # No Support @PDFCommands.add('W','W_a', 'sh', 'ET', 'BX', 'EX' ,'BI', 'ID', 'gs') @@ -15,17 +16,17 @@ def do_no_support(stack: PDFStateStack) -> PDFStateStack: # marked content operators @PDFCommands.add('MP', 'DP') -def do_MP(stack: PDFStateStack, tag) -> PDFStateStack: +def do_MP(stack: PDFStateStack) -> PDFStateStack: # self.device.do_tag(tag) return stack @PDFCommands.add('BMC', 'BDC') -def do_BMC(stack: PDFStateStack, tag) -> PDFStateStack: +def do_BMC(stack: PDFStateStack) -> PDFStateStack: # self.device.begin_tag(tag) return stack @PDFCommands.add('EMC') -def do_EMC(stack: PDFStateStack, tag) -> PDFStateStack: +def do_EMC(stack: PDFStateStack) -> PDFStateStack: # self.device.end_tag(tag) return stack @@ -166,7 +167,8 @@ def do_re(stack: PDFStateStack, x, y, w, h) -> PDFStateStack: # stroke, fill @PDFCommands.add('S', 'f', 'F', 'B') def do_complete_path(stack: PDFStateStack) -> PDFStateStack: - stack.complete_items.append(PDFShape( + stack.complete_layout_items.append(make_curve( + stack.t_matrix, stack.graphics.copy(), False, stack.curvestacks @@ -177,7 +179,8 @@ def do_complete_path(stack: PDFStateStack) -> PDFStateStack: # sroke, fill-even-odd @PDFCommands.add('f_a', 'B_a') def do_complete_path_evenodd(stack: PDFStateStack) -> PDFStateStack: - stack.complete_items.append(PDFShape( + stack.complete_layout_items.append(make_curve( + stack.t_matrix, stack.graphics.copy(), True, stack.curvestacks @@ -191,7 +194,8 @@ def do_close_complete(stack: PDFStateStack) -> PDFStateStack: stack.curvestacks.append(CurvePath( CurvePath.METHOD.CLOSE_PATH )) - stack.complete_items.append(PDFShape( + stack.complete_layout_items.append(make_curve( + stack.t_matrix, stack.graphics.copy(), False, stack.curvestacks @@ -205,7 +209,8 @@ def do_close_complete_evenodd(stack: PDFStateStack) -> PDFStateStack: stack.curvestacks.append(CurvePath( CurvePath.METHOD.CLOSE_PATH )) - stack.complete_items.append(PDFShape( + stack.complete_layout_items.append(make_curve( + stack.t_matrix, stack.graphics.copy(), True, stack.curvestacks @@ -411,11 +416,11 @@ def do_T_a(stack: PDFStateStack) -> PDFStateStack: def do_TJ(stack: PDFStateStack, seq: bytearray) -> PDFStateStack: if stack.text.font is None: raise PDFCommands.InvalidOperation("No Font Specified") - stack.complete_items.append(PDFText.make( + stack.complete_layout_items.append(make_char_block( seq, stack.t_matrix, stack.text.copy(), - stack.graphics.copy() + stack.graphics.ncolor.copy() )) return stack @@ -439,7 +444,9 @@ def do__w(stack: PDFStateStack, wordspace, charspace, b) -> PDFStateStack: @PDFCommands.add('EI') def do_EI(stack: PDFStateStack, obj) -> PDFStateStack: if 'W' in obj and 'H' in obj: - stack.complete_items.append(PDFImage(obj)) + stack.complete_layout_items.append(make_image( + obj, stack.ctm + )) return stack # invoke an XObject @@ -460,11 +467,17 @@ def do_Do(stack: PDFStateStack, xobjid) -> PDFStateStack: # instead of having their own Resources entry. xobjres = xobj.get('Resources') resources = dict_value(xobjres) if xobjres else stack.resources.copy() - stack.complete_items.append(PDFXObject( - xobj, bbox, resources, mult_matrix(stack.t_matrix, matrix) + stack.complete_layout_items.append(make_xobject( + obj=xobj, + bbox=bbox, + ctm=stack.t_matrix, + matrix=matrix, + resources=resources )) elif subtype is LITERAL_IMAGE and 'Width' in xobj and 'Height' in xobj: - stack.complete_items.append(PDFImage(obj)) + stack.complete_layout_items.append(make_image( + obj, stack.ctm + )) else: # unsupported xobject type. pass diff --git a/pdfmajor/interpreter_v2/commands/state/PDFGraphicState/__init__.py b/pdfmajor/interpreter_v2/commands/state/PDFGraphicState/__init__.py index 94de01f..f3fc3a1 100644 --- a/pdfmajor/interpreter_v2/commands/state/PDFGraphicState/__init__.py +++ b/pdfmajor/interpreter_v2/commands/state/PDFGraphicState/__init__.py @@ -1,4 +1,5 @@ from .PDFColor import PDFColor, PDFColorSpace +from .PDFColorSpace import PREDEFINED_COLORSPACE class PDFGraphicState(object): class InvalidOperation(Exception): pass diff --git a/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont.py b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont.py deleted file mode 100644 index 7369eb7..0000000 --- a/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont.py +++ /dev/null @@ -1,755 +0,0 @@ - -import sys -import struct -from io import BytesIO -from abc import abstractmethod - -from pdfmajor.interpreter.cmapdb import CMapDB -from pdfmajor.interpreter.cmapdb import CMapParser -from pdfmajor.interpreter.cmapdb import FileUnicodeMap -from pdfmajor.interpreter.cmapdb import CMap -from pdfmajor.interpreter.EncodingDB import EncodingDB, name2unicode -from pdfmajor.interpreter.PSStackParser import PSStackParser -from pdfmajor.interpreter.PSStackParser import PSEOF -from pdfmajor.interpreter.PSStackParser import LIT -from pdfmajor.interpreter.PSStackParser import KWD -from pdfmajor.interpreter.PSStackParser import PSLiteral -from pdfmajor.interpreter.PSStackParser import literal_name -from pdfmajor.interpreter.PDFStream import PDFException -from pdfmajor.interpreter.PDFStream import resolve1 -from pdfmajor.interpreter.PDFStream import int_value -from pdfmajor.interpreter.PDFStream import num_value -from pdfmajor.interpreter.PDFStream import list_value -from pdfmajor.interpreter.PDFStream import dict_value -from pdfmajor.interpreter.PDFStream import PDFStream -from pdfmajor.interpreter.constants import COMMON_FONT_ATTRIBUTES -from pdfmajor.utils import settings, FONT_METRICS, apply_matrix_norm, nunpack, choplist, isnumber - -def get_widths(seq): - widths = {} - r = [] - for v in seq: - if isinstance(v, list): - if r: - char1 = r[-1] - for (i, w) in enumerate(v): - widths[char1+i] = w - r = [] - elif isnumber(v): - r.append(v) - if len(r) == 3: - (char1, char2, w) = r - for i in range(char1, char2+1): - widths[i] = w - r = [] - return widths -#assert get_widths([1]) == {} -#assert get_widths([1,2,3]) == {1:3, 2:3} -#assert get_widths([1,[2,3],6,[7,8]]) == {1:2,2:3, 6:7,7:8} - -def get_widths2(seq): - widths = {} - r = [] - for v in seq: - if isinstance(v, list): - if r: - char1 = r[-1] - for (i, (w, vx, vy)) in enumerate(choplist(3, v)): - widths[char1+i] = (w, (vx, vy)) - r = [] - elif isnumber(v): - r.append(v) - if len(r) == 5: - (char1, char2, w, vx, vy) = r # pylint: disable=E0632 - for i in range(char1, char2+1): - widths[i] = (w, (vx, vy)) - r = [] - return widths -#assert get_widths2([1]) == {} -#assert get_widths2([1,2,3,4,5]) == {1:(3, (4,5)), 2:(3, (4,5))} -#assert get_widths2([1,[2,3,4,5],6,[7,8,9]]) == {1:(2, (3,4)), 6:(7, (8,9))} - - -## FontMetricsDB -## -class FontMetricsDB(object): - - @classmethod - def get_metrics(klass, fontname): - return FONT_METRICS[fontname] - - -## Type1FontHeaderParser -## -class Type1FontHeaderParser(PSStackParser): - - KEYWORD_BEGIN = KWD(b'begin') - KEYWORD_END = KWD(b'end') - KEYWORD_DEF = KWD(b'def') - KEYWORD_PUT = KWD(b'put') - KEYWORD_DICT = KWD(b'dict') - KEYWORD_ARRAY = KWD(b'array') - KEYWORD_READONLY = KWD(b'readonly') - KEYWORD_FOR = KWD(b'for') - KEYWORD_FOR = KWD(b'for') - - def __init__(self, data): - PSStackParser.__init__(self, data) - self._cid2unicode = {} - return - - def get_encoding(self): - while 1: - try: - (cid, name) = self.nextobject() - except PSEOF: - break - try: - self._cid2unicode[cid] = name2unicode(name) - except KeyError: - pass - return self._cid2unicode - - def do_keyword(self, pos, token): - if token is self.KEYWORD_PUT: - ((_, key), (_, value)) = self.pop(2) # pylint: disable=E0632 - if (isinstance(key, int) and - isinstance(value, PSLiteral)): - self.add_results((key, literal_name(value))) - return - - -NIBBLES = ('0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '.', 'e', 'e-', None, '-') - - -## CFFFont -## (Format specified in Adobe Technical Note: #5176 -## "The Compact Font Format Specification") -## -def getdict(data): - d = {} - fp = BytesIO(data) - stack = [] - while 1: - c = fp.read(1) - if not c: - break - b0 = ord(c) - if b0 <= 21: - d[b0] = stack - stack = [] - continue - if b0 == 30: - s = '' - loop = True - while loop: - b = ord(fp.read(1)) - for n in (b >> 4, b & 15): - if n == 15: - loop = False - else: - s += NIBBLES[n] - value = float(s) - elif 32 <= b0 and b0 <= 246: - value = b0-139 - else: - b1 = ord(fp.read(1)) - if 247 <= b0 and b0 <= 250: - value = ((b0-247) << 8)+b1+108 - elif 251 <= b0 and b0 <= 254: - value = -((b0-251) << 8)-b1-108 - else: - b2 = ord(fp.read(1)) - if 128 <= b1: - b1 -= 256 - if b0 == 28: - value = b1 << 8 | b2 - else: - value = b1 << 24 | b2 << 16 | struct.unpack('>H', fp.read(2))[0] - stack.append(value) - return d - - -class CFFFont(object): - - STANDARD_STRINGS = ( - '.notdef', 'space', 'exclam', 'quotedbl', 'numbersign', - 'dollar', 'percent', 'ampersand', 'quoteright', 'parenleft', - 'parenright', 'asterisk', 'plus', 'comma', 'hyphen', 'period', - 'slash', 'zero', 'one', 'two', 'three', 'four', 'five', 'six', - 'seven', 'eight', 'nine', 'colon', 'semicolon', 'less', 'equal', - 'greater', 'question', 'at', 'A', 'B', 'C', 'D', 'E', 'F', 'G', - 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', - 'U', 'V', 'W', 'X', 'Y', 'Z', 'bracketleft', 'backslash', - 'bracketright', 'asciicircum', 'underscore', 'quoteleft', 'a', - 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', - 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', - 'braceleft', 'bar', 'braceright', 'asciitilde', 'exclamdown', - 'cent', 'sterling', 'fraction', 'yen', 'florin', 'section', - 'currency', 'quotesingle', 'quotedblleft', 'guillemotleft', - 'guilsinglleft', 'guilsinglright', 'fi', 'fl', 'endash', - 'dagger', 'daggerdbl', 'periodcentered', 'paragraph', 'bullet', - 'quotesinglbase', 'quotedblbase', 'quotedblright', - 'guillemotright', 'ellipsis', 'perthousand', 'questiondown', - 'grave', 'acute', 'circumflex', 'tilde', 'macron', 'breve', - 'dotaccent', 'dieresis', 'ring', 'cedilla', 'hungarumlaut', - 'ogonek', 'caron', 'emdash', 'AE', 'ordfeminine', 'Lslash', - 'Oslash', 'OE', 'ordmasculine', 'ae', 'dotlessi', 'lslash', - 'oslash', 'oe', 'germandbls', 'onesuperior', 'logicalnot', 'mu', - 'trademark', 'Eth', 'onehalf', 'plusminus', 'Thorn', - 'onequarter', 'divide', 'brokenbar', 'degree', 'thorn', - 'threequarters', 'twosuperior', 'registered', 'minus', 'eth', - 'multiply', 'threesuperior', 'copyright', 'Aacute', - 'Acircumflex', 'Adieresis', 'Agrave', 'Aring', 'Atilde', - 'Ccedilla', 'Eacute', 'Ecircumflex', 'Edieresis', 'Egrave', - 'Iacute', 'Icircumflex', 'Idieresis', 'Igrave', 'Ntilde', - 'Oacute', 'Ocircumflex', 'Odieresis', 'Ograve', 'Otilde', - 'Scaron', 'Uacute', 'Ucircumflex', 'Udieresis', 'Ugrave', - 'Yacute', 'Ydieresis', 'Zcaron', 'aacute', 'acircumflex', - 'adieresis', 'agrave', 'aring', 'atilde', 'ccedilla', 'eacute', - 'ecircumflex', 'edieresis', 'egrave', 'iacute', 'icircumflex', - 'idieresis', 'igrave', 'ntilde', 'oacute', 'ocircumflex', - 'odieresis', 'ograve', 'otilde', 'scaron', 'uacute', - 'ucircumflex', 'udieresis', 'ugrave', 'yacute', 'ydieresis', - 'zcaron', 'exclamsmall', 'Hungarumlautsmall', 'dollaroldstyle', - 'dollarsuperior', 'ampersandsmall', 'Acutesmall', - 'parenleftsuperior', 'parenrightsuperior', 'twodotenleader', - 'onedotenleader', 'zerooldstyle', 'oneoldstyle', 'twooldstyle', - 'threeoldstyle', 'fouroldstyle', 'fiveoldstyle', 'sixoldstyle', - 'sevenoldstyle', 'eightoldstyle', 'nineoldstyle', - 'commasuperior', 'threequartersemdash', 'periodsuperior', - 'questionsmall', 'asuperior', 'bsuperior', 'centsuperior', - 'dsuperior', 'esuperior', 'isuperior', 'lsuperior', 'msuperior', - 'nsuperior', 'osuperior', 'rsuperior', 'ssuperior', 'tsuperior', - 'ff', 'ffi', 'ffl', 'parenleftinferior', 'parenrightinferior', - 'Circumflexsmall', 'hyphensuperior', 'Gravesmall', 'Asmall', - 'Bsmall', 'Csmall', 'Dsmall', 'Esmall', 'Fsmall', 'Gsmall', - 'Hsmall', 'Ismall', 'Jsmall', 'Ksmall', 'Lsmall', 'Msmall', - 'Nsmall', 'Osmall', 'Psmall', 'Qsmall', 'Rsmall', 'Ssmall', - 'Tsmall', 'Usmall', 'Vsmall', 'Wsmall', 'Xsmall', 'Ysmall', - 'Zsmall', 'colonmonetary', 'onefitted', 'rupiah', 'Tildesmall', - 'exclamdownsmall', 'centoldstyle', 'Lslashsmall', 'Scaronsmall', - 'Zcaronsmall', 'Dieresissmall', 'Brevesmall', 'Caronsmall', - 'Dotaccentsmall', 'Macronsmall', 'figuredash', 'hypheninferior', - 'Ogoneksmall', 'Ringsmall', 'Cedillasmall', 'questiondownsmall', - 'oneeighth', 'threeeighths', 'fiveeighths', 'seveneighths', - 'onethird', 'twothirds', 'zerosuperior', 'foursuperior', - 'fivesuperior', 'sixsuperior', 'sevensuperior', 'eightsuperior', - 'ninesuperior', 'zeroinferior', 'oneinferior', 'twoinferior', - 'threeinferior', 'fourinferior', 'fiveinferior', 'sixinferior', - 'seveninferior', 'eightinferior', 'nineinferior', - 'centinferior', 'dollarinferior', 'periodinferior', - 'commainferior', 'Agravesmall', 'Aacutesmall', - 'Acircumflexsmall', 'Atildesmall', 'Adieresissmall', - 'Aringsmall', 'AEsmall', 'Ccedillasmall', 'Egravesmall', - 'Eacutesmall', 'Ecircumflexsmall', 'Edieresissmall', - 'Igravesmall', 'Iacutesmall', 'Icircumflexsmall', - 'Idieresissmall', 'Ethsmall', 'Ntildesmall', 'Ogravesmall', - 'Oacutesmall', 'Ocircumflexsmall', 'Otildesmall', - 'Odieresissmall', 'OEsmall', 'Oslashsmall', 'Ugravesmall', - 'Uacutesmall', 'Ucircumflexsmall', 'Udieresissmall', - 'Yacutesmall', 'Thornsmall', 'Ydieresissmall', '001.000', - '001.001', '001.002', '001.003', 'Black', 'Bold', 'Book', - 'Light', 'Medium', 'Regular', 'Roman', 'Semibold', - ) - - class INDEX(object): - - def __init__(self, fp): - self.fp = fp - self.offsets = [] - (count, offsize) = struct.unpack('>HB', self.fp.read(3)) - for _ in range(count+1): - self.offsets.append(nunpack(self.fp.read(offsize))) - self.base = self.fp.tell()-1 - self.fp.seek(self.base+self.offsets[-1]) - return - - def __repr__(self): - return '' % len(self) - - def __len__(self): # pylint: disable=E0303 - return len(self.offsets)-1 - - def __getitem__(self, i): - self.fp.seek(self.base+self.offsets[i]) - return self.fp.read(self.offsets[i+1]-self.offsets[i]) - - def __iter__(self): - return iter(self[i] for i in range(len(self))) - - def __init__(self, name, fp): - self.name = name - self.fp = fp - # Header - (_major, _minor, hdrsize, _) = struct.unpack('BBBB', self.fp.read(4)) - self.fp.read(hdrsize-4) - # Name INDEX - self.name_index = self.INDEX(self.fp) - # Top DICT INDEX - self.dict_index = self.INDEX(self.fp) - # String INDEX - self.string_index = self.INDEX(self.fp) - # Global Subr INDEX - self.subr_index = self.INDEX(self.fp) - # Top DICT DATA - self.top_dict = getdict(self.dict_index[0]) - (charset_pos,) = self.top_dict.get(15, [0]) - (encoding_pos,) = self.top_dict.get(16, [0]) - (charstring_pos,) = self.top_dict.get(17, [0]) - # CharStrings - self.fp.seek(charstring_pos) - self.charstring = self.INDEX(self.fp) - self.nglyphs = len(self.charstring) - # Encodings - self.code2gid = {} - self.gid2code = {} - self.fp.seek(encoding_pos) - format = self.fp.read(1) - if format == b'\x00': - # Format 0 - (n,) = struct.unpack('B', self.fp.read(1)) - for (code, gid) in enumerate(struct.unpack('B'*n, self.fp.read(n))): - self.code2gid[code] = gid - self.gid2code[gid] = code - elif format == b'\x01': - # Format 1 - (n,) = struct.unpack('B', self.fp.read(1)) - code = 0 - for _ in range(n): - (first, nleft) = struct.unpack('BB', self.fp.read(2)) - for gid in range(first, first+nleft+1): - self.code2gid[code] = gid - self.gid2code[gid] = code - code += 1 - else: - raise ValueError('unsupported encoding format: %r' % format) - # Charsets - self.name2gid = {} - self.gid2name = {} - self.fp.seek(charset_pos) - format = self.fp.read(1) - if format == b'\x00': - # Format 0 - n = self.nglyphs-1 - for (gid, sid) in enumerate(struct.unpack('>'+'H'*n, self.fp.read(2*n))): - gid += 1 - name = self.getstr(sid) - self.name2gid[name] = gid - self.gid2name[gid] = name - elif format == b'\x01': - # Format 1 - (n,) = struct.unpack('B', self.fp.read(1)) - sid = 0 - for _ in range(n): - (first, nleft) = struct.unpack('BB', self.fp.read(2)) - for gid in range(first, first+nleft+1): - name = self.getstr(sid) - self.name2gid[name] = gid - self.gid2name[gid] = name - sid += 1 - elif format == b'\x02': - # Format 2 - assert False, str(('Unhandled', format)) - else: - raise ValueError('unsupported charset format: %r' % format) - #print self.code2gid - #print self.name2gid - #assert 0 - return - - def getstr(self, sid): - if sid < len(self.STANDARD_STRINGS): - return self.STANDARD_STRINGS[sid] - return self.string_index[sid-len(self.STANDARD_STRINGS)] - - -## TrueTypeFont -## -class TrueTypeFont(object): - - class CMapNotFound(Exception): - pass - - def __init__(self, name, fp): - self.name = name - self.fp = fp - self.tables = {} - self.fonttype = fp.read(4) - try: - (ntables, _1, _2, _3) = struct.unpack('>HHHH', fp.read(8)) - for _ in range(ntables): - (name, _, offset, length) = struct.unpack('>4sLLL', fp.read(16)) - self.tables[name] = (offset, length) - except struct.error: - # Do not fail if there are not enough bytes to read. Even for - # corrupted PDFs we would like to get as much information as - # possible, so continue. - pass - return - - def create_unicode_map(self): - if 'cmap' not in self.tables: - raise TrueTypeFont.CMapNotFound - (base_offset, _) = self.tables['cmap'] - fp = self.fp - fp.seek(base_offset) - (_, nsubtables) = struct.unpack('>HH', fp.read(4)) - subtables = [] - for i in range(nsubtables): - subtables.append(struct.unpack('>HHL', fp.read(8))) - char2gid = {} - # Only supports subtable type 0, 2 and 4. - for (_1, _2, st_offset) in subtables: - fp.seek(base_offset+st_offset) - (fmttype, _, _) = struct.unpack('>HHH', fp.read(6)) - if fmttype == 0: - char2gid.update(enumerate(struct.unpack('>256B', fp.read(256)))) - elif fmttype == 2: - subheaderkeys = struct.unpack('>256H', fp.read(512)) - firstbytes = [0]*8192 - for (i, k) in enumerate(subheaderkeys): - firstbytes[k//8] = i - nhdrs = max(subheaderkeys)//8 + 1 - hdrs = [] - for i in range(nhdrs): - (firstcode, entcount, delta, offset) = struct.unpack('>HHhH', fp.read(8)) - hdrs.append((i, firstcode, entcount, delta, fp.tell()-2+offset)) - for (i, firstcode, entcount, delta, pos) in hdrs: - if not entcount: - continue - first = firstcode + (firstbytes[i] << 8) - fp.seek(pos) - for c in range(entcount): - gid = struct.unpack('>H', fp.read(2)) - if gid: - gid += delta - char2gid[first+c] = gid - elif fmttype == 4: - (segcount, _1, _2, _3) = struct.unpack('>HHHH', fp.read(8)) - segcount //= 2 - ecs = struct.unpack('>%dH' % segcount, fp.read(2*segcount)) - fp.read(2) - scs = struct.unpack('>%dH' % segcount, fp.read(2*segcount)) - idds = struct.unpack('>%dh' % segcount, fp.read(2*segcount)) - pos = fp.tell() - idrs = struct.unpack('>%dH' % segcount, fp.read(2*segcount)) - for (ec, sc, idd, idr) in zip(ecs, scs, idds, idrs): - if idr: - fp.seek(pos+idr) - for c in range(sc, ec+1): - char2gid[c] = (struct.unpack('>H', fp.read(2))[0] + idd) & 0xffff - else: - for c in range(sc, ec+1): - char2gid[c] = (c + idd) & 0xffff - else: - assert False, str(('Unhandled', fmttype)) - # create unicode map - unicode_map = FileUnicodeMap() - for (char, gid) in char2gid.items(): - unicode_map.add_cid2unichr(gid, char) - return unicode_map - - -## Fonts -## -class PDFFontError(PDFException): - pass - - -class PDFUnicodeNotDefined(PDFFontError): - pass - -LITERAL_STANDARD_ENCODING = LIT('StandardEncoding') -LITERAL_TYPE1C = LIT('Type1C') - - -# PDFFont -class PDFFont(object): - - def __init__(self, descriptor, widths, default_width=None): - self.descriptor = descriptor - self.widths = widths - self.fontname = resolve1(descriptor.get('FontName', 'unknown')) - if isinstance(self.fontname, PSLiteral): - self.fontname = literal_name(self.fontname) - self.flags = int_value(descriptor.get('Flags', 0)) - self.ascent = num_value(descriptor.get('Ascent', 0)) - self.descent = num_value(descriptor.get('Descent', 0)) - self.italic_angle = num_value(descriptor.get('ItalicAngle', 0)) - self.default_width = default_width or num_value(descriptor.get('MissingWidth', 0)) - self.leading = num_value(descriptor.get('Leading', 0)) - self.bbox = list_value(descriptor.get('FontBBox', (0, 0, 0, 0))) - self.hscale = self.vscale = .001 - return - - def __repr__(self): - return '' - - def is_vertical(self): - return False - - def is_multibyte(self): - return False - - def decode(self, bytes): - return bytearray(bytes) # map(ord, bytes) - - def get_ascent(self): - return self.ascent * self.vscale - - def get_descent(self): - return self.descent * self.vscale - - def get_width(self): - w = self.bbox[2]-self.bbox[0] - if w == 0: - w = -self.default_width - return w * self.hscale - - def get_height(self): - h = self.bbox[3]-self.bbox[1] - if h == 0: - h = self.ascent - self.descent - return h * self.vscale - - def char_width(self, cid): - try: - return self.widths[cid] * self.hscale - except KeyError: - try: - return self.widths[self.to_unichr(cid)] * self.hscale - except (KeyError, PDFUnicodeNotDefined): - return self.default_width * self.hscale - - def char_disp(self, cid): - return 0 - - def string_width(self, s): - return sum(self.char_width(cid) for cid in self.decode(s)) - - @abstractmethod - def to_unichr(self, cid): - raise NotImplementedError - - @property - def font_weight(self): - if 'FontWeight' in self.descriptor.keys(): - return self.descriptor['FontWeight'] - elif 'bold' in self.fontname.lower(): - return 'bold' - else: - return None - - @property - def is_bold(self): - return self.font_weight == 'bold' or self.font_weight > 549 - - -# PDFSimpleFont -class PDFSimpleFont(PDFFont): - - def __init__(self, descriptor, widths, spec): - # Font encoding is specified either by a name of - # built-in encoding or a dictionary that describes - # the differences. - if 'Encoding' in spec: - encoding = resolve1(spec['Encoding']) - else: - encoding = LITERAL_STANDARD_ENCODING - if isinstance(encoding, dict): - name = literal_name(encoding.get('BaseEncoding', LITERAL_STANDARD_ENCODING)) - diff = list_value(encoding.get('Differences', [])) - self.cid2unicode = EncodingDB.get_encoding(name, diff) - else: - self.cid2unicode = EncodingDB.get_encoding(literal_name(encoding)) - self.unicode_map = None - if 'ToUnicode' in spec: - strm = PDFStream.validated_stream(spec['ToUnicode']) - self.unicode_map = FileUnicodeMap() - CMapParser(self.unicode_map, BytesIO(strm.get_data())).run() - PDFFont.__init__(self, descriptor, widths) - return - - def to_unichr(self, cid): - if self.unicode_map: - try: - return self.unicode_map.get_unichr(cid) - except KeyError: - pass - try: - return self.cid2unicode[cid] - except KeyError: - raise PDFUnicodeNotDefined(None, cid) - - -# PDFType1Font -class PDFType1Font(PDFSimpleFont): - - def __init__(self, rsrcmgr, spec): - try: - self.basefont = literal_name(spec['BaseFont']) - except KeyError: - if settings.STRICT: - raise PDFFontError('BaseFont is missing') - self.basefont = 'unknown' - try: - (descriptor, widths) = FontMetricsDB.get_metrics(self.basefont) - except KeyError: - descriptor = dict_value(spec.get('FontDescriptor', {})) - firstchar = int_value(spec.get('FirstChar', 0)) - #lastchar = int_value(spec.get('LastChar', 255)) - widths = list_value(spec.get('Widths', [0]*256)) - widths = dict((i+firstchar, w) for (i, w) in enumerate(widths)) - PDFSimpleFont.__init__(self, descriptor, widths, spec) - if 'Encoding' not in spec and 'FontFile' in descriptor: - # try to recover the missing encoding info from the font file. - self.fontfile = PDFStream.validated_stream(descriptor.get('FontFile')) - length1 = int_value(self.fontfile['Length1']) - data = self.fontfile.get_data()[:length1] - parser = Type1FontHeaderParser(BytesIO(data)) - self.cid2unicode = parser.get_encoding() - return - - def __repr__(self): - return '' % self.basefont - - -# PDFTrueTypeFont -class PDFTrueTypeFont(PDFType1Font): - - def __repr__(self): - return '' % self.basefont - - -# PDFType3Font -class PDFType3Font(PDFSimpleFont): - - def __init__(self, rsrcmgr, spec): - firstchar = int_value(spec.get('FirstChar', 0)) - #lastchar = int_value(spec.get('LastChar', 0)) - widths = list_value(spec.get('Widths', [0]*256)) - widths = dict((i+firstchar, w) for (i, w) in enumerate(widths)) - if 'FontDescriptor' in spec: - descriptor = dict_value(spec['FontDescriptor']) - else: - descriptor = {'Ascent': 0, 'Descent': 0, - 'FontBBox': spec['FontBBox']} - PDFSimpleFont.__init__(self, descriptor, widths, spec) - self.matrix = tuple(list_value(spec.get('FontMatrix'))) - (_, self.descent, _, self.ascent) = self.bbox - (self.hscale, self.vscale) = apply_matrix_norm(self.matrix, (1, 1)) - return - - def __repr__(self): - return '' - - -# PDFCIDFont -class PDFCIDFont(PDFFont): - - def __init__(self, rsrcmgr, spec, strict=settings.STRICT): - try: - self.basefont = literal_name(spec['BaseFont']) - except KeyError: - if strict: - raise PDFFontError('BaseFont is missing') - self.basefont = 'unknown' - self.cidsysteminfo = dict_value(spec.get('CIDSystemInfo', {})) - self.cidcoding = '%s-%s' % (resolve1(self.cidsysteminfo.get('Registry', b'unknown')).decode("latin1"), - resolve1(self.cidsysteminfo.get('Ordering', b'unknown')).decode("latin1")) - try: - name = literal_name(spec['Encoding']) - except KeyError: - if strict: - raise PDFFontError('Encoding is unspecified') - name = 'unknown' - try: - self.cmap = CMapDB.get_cmap(name) - except CMapDB.CMapNotFound as e: - if strict: - raise PDFFontError(e) - self.cmap = CMap() - try: - descriptor = dict_value(spec['FontDescriptor']) - except KeyError: - if strict: - raise PDFFontError('FontDescriptor is missing') - descriptor = {} - ttf = None - if 'FontFile2' in descriptor: - self.fontfile = PDFStream.validated_stream(descriptor.get('FontFile2')) - ttf = TrueTypeFont(self.basefont, - BytesIO(self.fontfile.get_data())) - self.unicode_map = None - if 'ToUnicode' in spec: - strm = PDFStream.validated_stream(spec['ToUnicode']) - self.unicode_map = FileUnicodeMap() - CMapParser(self.unicode_map, BytesIO(strm.get_data())).run() - elif self.cidcoding in ('Adobe-Identity', 'Adobe-UCS'): - if ttf: - try: - self.unicode_map = ttf.create_unicode_map() - except TrueTypeFont.CMapNotFound: - pass - else: - try: - self.unicode_map = CMapDB.get_unicode_map(self.cidcoding, self.cmap.is_vertical()) - except CMapDB.CMapNotFound as e: - pass - - self.vertical = self.cmap.is_vertical() - if self.vertical: - # writing mode: vertical - widths = get_widths2(list_value(spec.get('W2', []))) - self.disps = dict((cid, (vx, vy)) for (cid, (_, (vx, vy))) in iter(widths.items())) - (vy, w) = spec.get('DW2', [880, -1000]) - self.default_disp = (None, vy) - widths = dict((cid, w) for (cid, (w, _)) in iter(widths.items())) - default_width = w - else: - # writing mode: horizontal - self.disps = {} - self.default_disp = 0 - widths = get_widths(list_value(spec.get('W', []))) - default_width = spec.get('DW', 1000) - PDFFont.__init__(self, descriptor, widths, default_width=default_width) - return - - def __repr__(self): - return '' % (self.basefont, self.cidcoding) - - def is_vertical(self): - return self.vertical - - def is_multibyte(self): - return True - - def decode(self, bytes): - return self.cmap.decode(bytes) - - def char_disp(self, cid): - "Returns an integer for horizontal fonts, a tuple for vertical fonts." - return self.disps.get(cid, self.default_disp) - - def to_unichr(self, cid): - try: - if not self.unicode_map: - raise KeyError(cid) - return self.unicode_map.get_unichr(cid) - except KeyError: - raise PDFUnicodeNotDefined(self.cidcoding, cid) - - -# main -def main(argv): - for fname in argv[1:]: - fp = open(fname, 'rb') - #font = TrueTypeFont(fname, fp) - font = CFFFont(fname, fp) - print (font) - fp.close() - return - -if __name__ == '__main__': - sys.exit(main(sys.argv)) diff --git a/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/CFFFont.py b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/CFFFont.py new file mode 100644 index 0000000..8b8cc14 --- /dev/null +++ b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/CFFFont.py @@ -0,0 +1,246 @@ +## CFFFont +## (Format specified in Adobe Technical Note: #5176 +## "The Compact Font Format Specification") +## +import struct +from io import BytesIO +from pdfmajor.utils import nunpack + +NIBBLES = ('0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '.', 'e', 'e-', None, '-') + +def getdict(data): + d = {} + fp = BytesIO(data) + stack = [] + while 1: + c = fp.read(1) + if not c: + break + b0 = ord(c) + if b0 <= 21: + d[b0] = stack + stack = [] + continue + if b0 == 30: + s = '' + loop = True + while loop: + b = ord(fp.read(1)) + for n in (b >> 4, b & 15): + if n == 15: + loop = False + else: + s += NIBBLES[n] + value = float(s) + elif 32 <= b0 and b0 <= 246: + value = b0-139 + else: + b1 = ord(fp.read(1)) + if 247 <= b0 and b0 <= 250: + value = ((b0-247) << 8)+b1+108 + elif 251 <= b0 and b0 <= 254: + value = -((b0-251) << 8)-b1-108 + else: + b2 = ord(fp.read(1)) + if 128 <= b1: + b1 -= 256 + if b0 == 28: + value = b1 << 8 | b2 + else: + value = b1 << 24 | b2 << 16 | struct.unpack('>H', fp.read(2))[0] + stack.append(value) + return d + + +class CFFFont(object): + + STANDARD_STRINGS = ( + '.notdef', 'space', 'exclam', 'quotedbl', 'numbersign', + 'dollar', 'percent', 'ampersand', 'quoteright', 'parenleft', + 'parenright', 'asterisk', 'plus', 'comma', 'hyphen', 'period', + 'slash', 'zero', 'one', 'two', 'three', 'four', 'five', 'six', + 'seven', 'eight', 'nine', 'colon', 'semicolon', 'less', 'equal', + 'greater', 'question', 'at', 'A', 'B', 'C', 'D', 'E', 'F', 'G', + 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', + 'U', 'V', 'W', 'X', 'Y', 'Z', 'bracketleft', 'backslash', + 'bracketright', 'asciicircum', 'underscore', 'quoteleft', 'a', + 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', + 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z', + 'braceleft', 'bar', 'braceright', 'asciitilde', 'exclamdown', + 'cent', 'sterling', 'fraction', 'yen', 'florin', 'section', + 'currency', 'quotesingle', 'quotedblleft', 'guillemotleft', + 'guilsinglleft', 'guilsinglright', 'fi', 'fl', 'endash', + 'dagger', 'daggerdbl', 'periodcentered', 'paragraph', 'bullet', + 'quotesinglbase', 'quotedblbase', 'quotedblright', + 'guillemotright', 'ellipsis', 'perthousand', 'questiondown', + 'grave', 'acute', 'circumflex', 'tilde', 'macron', 'breve', + 'dotaccent', 'dieresis', 'ring', 'cedilla', 'hungarumlaut', + 'ogonek', 'caron', 'emdash', 'AE', 'ordfeminine', 'Lslash', + 'Oslash', 'OE', 'ordmasculine', 'ae', 'dotlessi', 'lslash', + 'oslash', 'oe', 'germandbls', 'onesuperior', 'logicalnot', 'mu', + 'trademark', 'Eth', 'onehalf', 'plusminus', 'Thorn', + 'onequarter', 'divide', 'brokenbar', 'degree', 'thorn', + 'threequarters', 'twosuperior', 'registered', 'minus', 'eth', + 'multiply', 'threesuperior', 'copyright', 'Aacute', + 'Acircumflex', 'Adieresis', 'Agrave', 'Aring', 'Atilde', + 'Ccedilla', 'Eacute', 'Ecircumflex', 'Edieresis', 'Egrave', + 'Iacute', 'Icircumflex', 'Idieresis', 'Igrave', 'Ntilde', + 'Oacute', 'Ocircumflex', 'Odieresis', 'Ograve', 'Otilde', + 'Scaron', 'Uacute', 'Ucircumflex', 'Udieresis', 'Ugrave', + 'Yacute', 'Ydieresis', 'Zcaron', 'aacute', 'acircumflex', + 'adieresis', 'agrave', 'aring', 'atilde', 'ccedilla', 'eacute', + 'ecircumflex', 'edieresis', 'egrave', 'iacute', 'icircumflex', + 'idieresis', 'igrave', 'ntilde', 'oacute', 'ocircumflex', + 'odieresis', 'ograve', 'otilde', 'scaron', 'uacute', + 'ucircumflex', 'udieresis', 'ugrave', 'yacute', 'ydieresis', + 'zcaron', 'exclamsmall', 'Hungarumlautsmall', 'dollaroldstyle', + 'dollarsuperior', 'ampersandsmall', 'Acutesmall', + 'parenleftsuperior', 'parenrightsuperior', 'twodotenleader', + 'onedotenleader', 'zerooldstyle', 'oneoldstyle', 'twooldstyle', + 'threeoldstyle', 'fouroldstyle', 'fiveoldstyle', 'sixoldstyle', + 'sevenoldstyle', 'eightoldstyle', 'nineoldstyle', + 'commasuperior', 'threequartersemdash', 'periodsuperior', + 'questionsmall', 'asuperior', 'bsuperior', 'centsuperior', + 'dsuperior', 'esuperior', 'isuperior', 'lsuperior', 'msuperior', + 'nsuperior', 'osuperior', 'rsuperior', 'ssuperior', 'tsuperior', + 'ff', 'ffi', 'ffl', 'parenleftinferior', 'parenrightinferior', + 'Circumflexsmall', 'hyphensuperior', 'Gravesmall', 'Asmall', + 'Bsmall', 'Csmall', 'Dsmall', 'Esmall', 'Fsmall', 'Gsmall', + 'Hsmall', 'Ismall', 'Jsmall', 'Ksmall', 'Lsmall', 'Msmall', + 'Nsmall', 'Osmall', 'Psmall', 'Qsmall', 'Rsmall', 'Ssmall', + 'Tsmall', 'Usmall', 'Vsmall', 'Wsmall', 'Xsmall', 'Ysmall', + 'Zsmall', 'colonmonetary', 'onefitted', 'rupiah', 'Tildesmall', + 'exclamdownsmall', 'centoldstyle', 'Lslashsmall', 'Scaronsmall', + 'Zcaronsmall', 'Dieresissmall', 'Brevesmall', 'Caronsmall', + 'Dotaccentsmall', 'Macronsmall', 'figuredash', 'hypheninferior', + 'Ogoneksmall', 'Ringsmall', 'Cedillasmall', 'questiondownsmall', + 'oneeighth', 'threeeighths', 'fiveeighths', 'seveneighths', + 'onethird', 'twothirds', 'zerosuperior', 'foursuperior', + 'fivesuperior', 'sixsuperior', 'sevensuperior', 'eightsuperior', + 'ninesuperior', 'zeroinferior', 'oneinferior', 'twoinferior', + 'threeinferior', 'fourinferior', 'fiveinferior', 'sixinferior', + 'seveninferior', 'eightinferior', 'nineinferior', + 'centinferior', 'dollarinferior', 'periodinferior', + 'commainferior', 'Agravesmall', 'Aacutesmall', + 'Acircumflexsmall', 'Atildesmall', 'Adieresissmall', + 'Aringsmall', 'AEsmall', 'Ccedillasmall', 'Egravesmall', + 'Eacutesmall', 'Ecircumflexsmall', 'Edieresissmall', + 'Igravesmall', 'Iacutesmall', 'Icircumflexsmall', + 'Idieresissmall', 'Ethsmall', 'Ntildesmall', 'Ogravesmall', + 'Oacutesmall', 'Ocircumflexsmall', 'Otildesmall', + 'Odieresissmall', 'OEsmall', 'Oslashsmall', 'Ugravesmall', + 'Uacutesmall', 'Ucircumflexsmall', 'Udieresissmall', + 'Yacutesmall', 'Thornsmall', 'Ydieresissmall', '001.000', + '001.001', '001.002', '001.003', 'Black', 'Bold', 'Book', + 'Light', 'Medium', 'Regular', 'Roman', 'Semibold', + ) + + class INDEX(object): + + def __init__(self, fp): + self.fp = fp + self.offsets = [] + (count, offsize) = struct.unpack('>HB', self.fp.read(3)) + for _ in range(count+1): + self.offsets.append(nunpack(self.fp.read(offsize))) + self.base = self.fp.tell()-1 + self.fp.seek(self.base+self.offsets[-1]) + return + + def __repr__(self): + return '' % len(self) + + def __len__(self): # pylint: disable=E0303 + return len(self.offsets)-1 + + def __getitem__(self, i): + self.fp.seek(self.base+self.offsets[i]) + return self.fp.read(self.offsets[i+1]-self.offsets[i]) + + def __iter__(self): + return iter(self[i] for i in range(len(self))) + + def __init__(self, name, fp): + self.name = name + self.fp = fp + # Header + (_major, _minor, hdrsize, _) = struct.unpack('BBBB', self.fp.read(4)) + self.fp.read(hdrsize-4) + # Name INDEX + self.name_index = self.INDEX(self.fp) + # Top DICT INDEX + self.dict_index = self.INDEX(self.fp) + # String INDEX + self.string_index = self.INDEX(self.fp) + # Global Subr INDEX + self.subr_index = self.INDEX(self.fp) + # Top DICT DATA + self.top_dict = getdict(self.dict_index[0]) + (charset_pos,) = self.top_dict.get(15, [0]) + (encoding_pos,) = self.top_dict.get(16, [0]) + (charstring_pos,) = self.top_dict.get(17, [0]) + # CharStrings + self.fp.seek(charstring_pos) + self.charstring = self.INDEX(self.fp) + self.nglyphs = len(self.charstring) + # Encodings + self.code2gid = {} + self.gid2code = {} + self.fp.seek(encoding_pos) + format = self.fp.read(1) + if format == b'\x00': + # Format 0 + (n,) = struct.unpack('B', self.fp.read(1)) + for (code, gid) in enumerate(struct.unpack('B'*n, self.fp.read(n))): + self.code2gid[code] = gid + self.gid2code[gid] = code + elif format == b'\x01': + # Format 1 + (n,) = struct.unpack('B', self.fp.read(1)) + code = 0 + for _ in range(n): + (first, nleft) = struct.unpack('BB', self.fp.read(2)) + for gid in range(first, first+nleft+1): + self.code2gid[code] = gid + self.gid2code[gid] = code + code += 1 + else: + raise ValueError('unsupported encoding format: %r' % format) + # Charsets + self.name2gid = {} + self.gid2name = {} + self.fp.seek(charset_pos) + format = self.fp.read(1) + if format == b'\x00': + # Format 0 + n = self.nglyphs-1 + for (gid, sid) in enumerate(struct.unpack('>'+'H'*n, self.fp.read(2*n))): + gid += 1 + name = self.getstr(sid) + self.name2gid[name] = gid + self.gid2name[gid] = name + elif format == b'\x01': + # Format 1 + (n,) = struct.unpack('B', self.fp.read(1)) + sid = 0 + for _ in range(n): + (first, nleft) = struct.unpack('BB', self.fp.read(2)) + for gid in range(first, first+nleft+1): + name = self.getstr(sid) + self.name2gid[name] = gid + self.gid2name[gid] = name + sid += 1 + elif format == b'\x02': + # Format 2 + assert False, str(('Unhandled', format)) + else: + raise ValueError('unsupported charset format: %r' % format) + #print self.code2gid + #print self.name2gid + #assert 0 + return + + def getstr(self, sid): + if sid < len(self.STANDARD_STRINGS): + return self.STANDARD_STRINGS[sid] + return self.string_index[sid-len(self.STANDARD_STRINGS)] diff --git a/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/PDFFont.py b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/PDFFont.py new file mode 100644 index 0000000..45df98f --- /dev/null +++ b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/PDFFont.py @@ -0,0 +1,140 @@ +# PDFFont +from abc import abstractmethod +from io import BytesIO + +from pdfmajor.interpreter.EncodingDB import EncodingDB +from pdfmajor.interpreter.PSStackParser import LIT +from pdfmajor.interpreter.PSStackParser import PSLiteral +from pdfmajor.interpreter.PSStackParser import literal_name +from pdfmajor.interpreter.PDFStream import resolve1 +from pdfmajor.interpreter.PDFStream import int_value +from pdfmajor.interpreter.PDFStream import num_value +from pdfmajor.interpreter.PDFStream import list_value +from pdfmajor.interpreter.PDFStream import PDFException +from pdfmajor.interpreter.PDFStream import PDFStream +from pdfmajor.interpreter.cmapdb import FileUnicodeMap +from pdfmajor.interpreter.cmapdb import CMapParser + +## Literals +## +LITERAL_STANDARD_ENCODING = LIT('StandardEncoding') +LITERAL_TYPE1C = LIT('Type1C') + + +class PDFFont(object): + class PDFFontError(PDFException): pass + class PDFUnicodeNotDefined(PDFFontError): pass + + def __init__(self, descriptor, widths, default_width=None): + self.descriptor = descriptor + self.widths = widths + self.fontname = resolve1(descriptor.get('FontName', 'unknown')) + if isinstance(self.fontname, PSLiteral): + self.fontname = literal_name(self.fontname) + self.flags = int_value(descriptor.get('Flags', 0)) + self.ascent = num_value(descriptor.get('Ascent', 0)) + self.descent = num_value(descriptor.get('Descent', 0)) + self.italic_angle = num_value(descriptor.get('ItalicAngle', 0)) + self.default_width = default_width or num_value(descriptor.get('MissingWidth', 0)) + self.leading = num_value(descriptor.get('Leading', 0)) + self.bbox = list_value(descriptor.get('FontBBox', (0, 0, 0, 0))) + self.hscale = self.vscale = .001 + return + + def __repr__(self): + return '' + + def is_vertical(self): + return False + + def is_multibyte(self): + return False + + def decode(self, bytes): + return bytearray(bytes) # map(ord, bytes) + + def get_ascent(self): + return self.ascent * self.vscale + + def get_descent(self): + return self.descent * self.vscale + + def get_width(self): + w = self.bbox[2]-self.bbox[0] + if w == 0: + w = -self.default_width + return w * self.hscale + + def get_height(self): + h = self.bbox[3]-self.bbox[1] + if h == 0: + h = self.ascent - self.descent + return h * self.vscale + + def char_width(self, cid): + try: + return self.widths[cid] * self.hscale + except KeyError: + try: + return self.widths[self.to_unichr(cid)] * self.hscale + except (KeyError, self.PDFUnicodeNotDefined): + return self.default_width * self.hscale + + def char_disp(self, cid): + return 0 + + def string_width(self, s): + return sum(self.char_width(cid) for cid in self.decode(s)) + + @abstractmethod + def to_unichr(self, cid): + raise NotImplementedError + + @property + def font_weight(self): + if 'FontWeight' in self.descriptor.keys(): + return self.descriptor['FontWeight'] + elif 'bold' in self.fontname.lower(): + return 'bold' + else: + return None + + @property + def is_bold(self): + return self.font_weight == 'bold' or self.font_weight > 549 + +# PDFSimpleFont +class PDFSimpleFont(PDFFont): + + def __init__(self, descriptor, widths, spec): + # Font encoding is specified either by a name of + # built-in encoding or a dictionary that describes + # the differences. + if 'Encoding' in spec: + encoding = resolve1(spec['Encoding']) + else: + encoding = LITERAL_STANDARD_ENCODING + if isinstance(encoding, dict): + name = literal_name(encoding.get('BaseEncoding', LITERAL_STANDARD_ENCODING)) + diff = list_value(encoding.get('Differences', [])) + self.cid2unicode = EncodingDB.get_encoding(name, diff) + else: + self.cid2unicode = EncodingDB.get_encoding(literal_name(encoding)) + self.unicode_map = None + if 'ToUnicode' in spec: + strm = PDFStream.validated_stream(spec['ToUnicode']) + self.unicode_map = FileUnicodeMap() + CMapParser(self.unicode_map, BytesIO(strm.get_data())).run() + PDFFont.__init__(self, descriptor, widths) + return + + def to_unichr(self, cid): + if self.unicode_map: + try: + return self.unicode_map.get_unichr(cid) + except KeyError: + pass + try: + return self.cid2unicode[cid] + except KeyError: + raise self.PDFUnicodeNotDefined(None, cid) diff --git a/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/TrueTypeFont.py b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/TrueTypeFont.py new file mode 100644 index 0000000..ec05113 --- /dev/null +++ b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/TrueTypeFont.py @@ -0,0 +1,89 @@ +## TrueTypeFont +## +import struct +from pdfmajor.interpreter.cmapdb import FileUnicodeMap + +class TrueTypeFont(object): + + class CMapNotFound(Exception): + pass + + def __init__(self, name, fp): + self.name = name + self.fp = fp + self.tables = {} + self.fonttype = fp.read(4) + try: + (ntables, _1, _2, _3) = struct.unpack('>HHHH', fp.read(8)) + for _ in range(ntables): + (name, _, offset, length) = struct.unpack('>4sLLL', fp.read(16)) + self.tables[name] = (offset, length) + except struct.error: + # Do not fail if there are not enough bytes to read. Even for + # corrupted PDFs we would like to get as much information as + # possible, so continue. + pass + return + + def create_unicode_map(self): + if 'cmap' not in self.tables: + raise TrueTypeFont.CMapNotFound + (base_offset, _) = self.tables['cmap'] + fp = self.fp + fp.seek(base_offset) + (_, nsubtables) = struct.unpack('>HH', fp.read(4)) + subtables = [] + for i in range(nsubtables): + subtables.append(struct.unpack('>HHL', fp.read(8))) + char2gid = {} + # Only supports subtable type 0, 2 and 4. + for (_1, _2, st_offset) in subtables: + fp.seek(base_offset+st_offset) + (fmttype, _, _) = struct.unpack('>HHH', fp.read(6)) + if fmttype == 0: + char2gid.update(enumerate(struct.unpack('>256B', fp.read(256)))) + elif fmttype == 2: + subheaderkeys = struct.unpack('>256H', fp.read(512)) + firstbytes = [0]*8192 + for (i, k) in enumerate(subheaderkeys): + firstbytes[k//8] = i + nhdrs = max(subheaderkeys)//8 + 1 + hdrs = [] + for i in range(nhdrs): + (firstcode, entcount, delta, offset) = struct.unpack('>HHhH', fp.read(8)) + hdrs.append((i, firstcode, entcount, delta, fp.tell()-2+offset)) + for (i, firstcode, entcount, delta, pos) in hdrs: + if not entcount: + continue + first = firstcode + (firstbytes[i] << 8) + fp.seek(pos) + for c in range(entcount): + gid = struct.unpack('>H', fp.read(2)) + if gid: + gid += delta + char2gid[first+c] = gid + elif fmttype == 4: + (segcount, _1, _2, _3) = struct.unpack('>HHHH', fp.read(8)) + segcount //= 2 + ecs = struct.unpack('>%dH' % segcount, fp.read(2*segcount)) + fp.read(2) + scs = struct.unpack('>%dH' % segcount, fp.read(2*segcount)) + idds = struct.unpack('>%dh' % segcount, fp.read(2*segcount)) + pos = fp.tell() + idrs = struct.unpack('>%dH' % segcount, fp.read(2*segcount)) + for (ec, sc, idd, idr) in zip(ecs, scs, idds, idrs): + if idr: + fp.seek(pos+idr) + for c in range(sc, ec+1): + char2gid[c] = (struct.unpack('>H', fp.read(2))[0] + idd) & 0xffff + else: + for c in range(sc, ec+1): + char2gid[c] = (c + idd) & 0xffff + else: + assert False, str(('Unhandled', fmttype)) + # create unicode map + unicode_map = FileUnicodeMap() + for (char, gid) in char2gid.items(): + unicode_map.add_cid2unichr(gid, char) + return unicode_map + diff --git a/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/Type1FontHeaderParser.py b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/Type1FontHeaderParser.py new file mode 100644 index 0000000..d7ea82d --- /dev/null +++ b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/Type1FontHeaderParser.py @@ -0,0 +1,45 @@ +from pdfmajor.interpreter.PSStackParser import PSStackParser +from pdfmajor.interpreter.PSStackParser import PSEOF, KWD +from pdfmajor.interpreter.PSStackParser import PSLiteral +from pdfmajor.interpreter.PSStackParser import literal_name +from pdfmajor.interpreter.EncodingDB import name2unicode + +## Type1FontHeaderParser +## +class Type1FontHeaderParser(PSStackParser): + + KEYWORD_BEGIN = KWD(b'begin') + KEYWORD_END = KWD(b'end') + KEYWORD_DEF = KWD(b'def') + KEYWORD_PUT = KWD(b'put') + KEYWORD_DICT = KWD(b'dict') + KEYWORD_ARRAY = KWD(b'array') + KEYWORD_READONLY = KWD(b'readonly') + KEYWORD_FOR = KWD(b'for') + KEYWORD_FOR = KWD(b'for') + + def __init__(self, data): + PSStackParser.__init__(self, data) + self._cid2unicode = {} + return + + def get_encoding(self): + while 1: + try: + (cid, name) = self.nextobject() + except PSEOF: + break + try: + self._cid2unicode[cid] = name2unicode(name) + except KeyError: + pass + return self._cid2unicode + + def do_keyword(self, pos, token): + if token is self.KEYWORD_PUT: + ((_, key), (_, value)) = self.pop(2) # pylint: disable=E0632 + if (isinstance(key, int) and + isinstance(value, PSLiteral)): + self.add_results((key, literal_name(value))) + return + diff --git a/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/__init__.py b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/__init__.py new file mode 100644 index 0000000..5bcd18b --- /dev/null +++ b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/__init__.py @@ -0,0 +1,57 @@ +from pdfmajor.utils import get_logger, settings +from pdfmajor.interpreter.PSStackParser import literal_name +from pdfmajor.interpreter.PDFStream import list_value +from pdfmajor.interpreter.PDFStream import dict_value +from pdfmajor.interpreter.PDFStream import resolve1 +from pdfmajor.interpreter.constants import LITERAL_FONT + +from .fonts import PDFType1Font, PDFTrueTypeFont, PDFType3Font, PDFCIDFont, PDFFont + +log = get_logger(__file__) + +def get_font(objid: int, spec: dict, cached_fonts: dict = {}): + if objid and objid in cached_fonts: + font = cached_fonts[objid] + else: + log.debug('get_font: create: objid=%r, spec=%r', objid, spec) + if settings.STRICT: + if spec['Type'] is not LITERAL_FONT: + raise PDFFont.PDFFontError('Type is not /Font') + # Create a Font object. + if 'Subtype' in spec: + subtype = literal_name(spec['Subtype']) + else: + if settings.STRICT: + raise PDFFont.PDFFontError('Font Subtype is not specified.') + subtype = 'Type1' + if subtype in ('Type1', 'MMType1'): + # Type1 Font + font = PDFType1Font(spec) + elif subtype == 'TrueType': + # TrueType Font + font = PDFTrueTypeFont(spec) + elif subtype == 'Type3': + # Type3 Font + font = PDFType3Font(spec) + elif subtype in ('CIDFontType0', 'CIDFontType2'): + # CID Font + font = PDFCIDFont(spec) + elif subtype == 'Type0': + # Type0 Font + dfonts = list_value(spec['DescendantFonts']) + assert dfonts + subspec = dict_value(dfonts[0]).copy() + for k in ('Encoding', 'ToUnicode'): + if k in spec: + subspec[k] = resolve1(spec[k]) + font = get_font(None, subspec, cached_fonts) + else: + if settings.STRICT: + raise PDFFont.PDFFontError('Invalid Font spec: %r' % spec) + font = PDFType1Font(spec) # this is so wrong! + if objid: + cached_fonts[objid] = font + return font + + + diff --git a/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/fonts.py b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/fonts.py new file mode 100644 index 0000000..69c1ee2 --- /dev/null +++ b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/fonts.py @@ -0,0 +1,174 @@ +from io import BytesIO + +from pdfmajor.interpreter.PSStackParser import literal_name +from pdfmajor.interpreter.PDFStream import int_value +from pdfmajor.interpreter.PDFStream import num_value +from pdfmajor.interpreter.PDFStream import list_value +from pdfmajor.interpreter.PDFStream import dict_value +from pdfmajor.interpreter.PDFStream import PDFStream +from pdfmajor.interpreter.PDFStream import resolve1 +from pdfmajor.interpreter.cmapdb import CMap, CMapDB, CMapParser +from pdfmajor.interpreter.cmapdb import FileUnicodeMap +from pdfmajor.utils import settings, apply_matrix_norm + +from .PDFFont import PDFFont, PDFSimpleFont +from .util import FontMetricsDB, get_widths, get_widths2 +from .Type1FontHeaderParser import Type1FontHeaderParser +from .TrueTypeFont import TrueTypeFont + +# PDFType1Font +class PDFType1Font(PDFSimpleFont): + + def __init__(self, spec): + try: + self.basefont = literal_name(spec['BaseFont']) + except KeyError: + if settings.STRICT: + raise self.PDFFontError('BaseFont is missing') + self.basefont = 'unknown' + try: + (descriptor, widths) = FontMetricsDB.get_metrics(self.basefont) + except KeyError: + descriptor = dict_value(spec.get('FontDescriptor', {})) + firstchar = int_value(spec.get('FirstChar', 0)) + #lastchar = int_value(spec.get('LastChar', 255)) + widths = list_value(spec.get('Widths', [0]*256)) + widths = dict((i+firstchar, w) for (i, w) in enumerate(widths)) + PDFSimpleFont.__init__(self, descriptor, widths, spec) + if 'Encoding' not in spec and 'FontFile' in descriptor: + # try to recover the missing encoding info from the font file. + self.fontfile = PDFStream.validated_stream(descriptor.get('FontFile')) + length1 = int_value(self.fontfile['Length1']) + data = self.fontfile.get_data()[:length1] + parser = Type1FontHeaderParser(BytesIO(data)) + self.cid2unicode = parser.get_encoding() + return + + def __repr__(self): + return '' % self.basefont + + +# PDFTrueTypeFont +class PDFTrueTypeFont(PDFType1Font): + + def __repr__(self): + return '' % self.basefont + + +# PDFType3Font +class PDFType3Font(PDFSimpleFont): + + def __init__(self, spec): + firstchar = int_value(spec.get('FirstChar', 0)) + #lastchar = int_value(spec.get('LastChar', 0)) + widths = list_value(spec.get('Widths', [0]*256)) + widths = dict((i+firstchar, w) for (i, w) in enumerate(widths)) + if 'FontDescriptor' in spec: + descriptor = dict_value(spec['FontDescriptor']) + else: + descriptor = {'Ascent': 0, 'Descent': 0, + 'FontBBox': spec['FontBBox']} + PDFSimpleFont.__init__(self, descriptor, widths, spec) + self.matrix = tuple(list_value(spec.get('FontMatrix'))) + (_, self.descent, _, self.ascent) = self.bbox + (self.hscale, self.vscale) = apply_matrix_norm(self.matrix, (1, 1)) + return + + def __repr__(self): + return '' + + +# PDFCIDFont +class PDFCIDFont(PDFFont): + + def __init__(self, spec, strict=settings.STRICT): + try: + self.basefont = literal_name(spec['BaseFont']) + except KeyError: + if strict: + raise self.PDFFontError('BaseFont is missing') + self.basefont = 'unknown' + self.cidsysteminfo = dict_value(spec.get('CIDSystemInfo', {})) + self.cidcoding = '%s-%s' % (resolve1(self.cidsysteminfo.get('Registry', b'unknown')).decode("latin1"), + resolve1(self.cidsysteminfo.get('Ordering', b'unknown')).decode("latin1")) + try: + name = literal_name(spec['Encoding']) + except KeyError: + if strict: + raise self.PDFFontError('Encoding is unspecified') + name = 'unknown' + try: + self.cmap = CMapDB.get_cmap(name) + except CMapDB.CMapNotFound as e: + if strict: + raise self.PDFFontError(e) + self.cmap = CMap() + try: + descriptor = dict_value(spec['FontDescriptor']) + except KeyError: + if strict: + raise self.PDFFontError('FontDescriptor is missing') + descriptor = {} + ttf = None + if 'FontFile2' in descriptor: + self.fontfile = PDFStream.validated_stream(descriptor.get('FontFile2')) + ttf = TrueTypeFont(self.basefont, + BytesIO(self.fontfile.get_data())) + self.unicode_map = None + if 'ToUnicode' in spec: + strm = PDFStream.validated_stream(spec['ToUnicode']) + self.unicode_map = FileUnicodeMap() + CMapParser(self.unicode_map, BytesIO(strm.get_data())).run() + elif self.cidcoding in ('Adobe-Identity', 'Adobe-UCS'): + if ttf: + try: + self.unicode_map = ttf.create_unicode_map() + except TrueTypeFont.CMapNotFound: + pass + else: + try: + self.unicode_map = CMapDB.get_unicode_map(self.cidcoding, self.cmap.is_vertical()) + except CMapDB.CMapNotFound as e: + pass + + self.vertical = self.cmap.is_vertical() + if self.vertical: + # writing mode: vertical + widths = get_widths2(list_value(spec.get('W2', []))) + self.disps = dict((cid, (vx, vy)) for (cid, (_, (vx, vy))) in iter(widths.items())) + (vy, w) = spec.get('DW2', [880, -1000]) + self.default_disp = (None, vy) + widths = dict((cid, w) for (cid, (w, _)) in iter(widths.items())) + default_width = w + else: + # writing mode: horizontal + self.disps = {} + self.default_disp = 0 + widths = get_widths(list_value(spec.get('W', []))) + default_width = spec.get('DW', 1000) + PDFFont.__init__(self, descriptor, widths, default_width=default_width) + return + + def __repr__(self): + return '' % (self.basefont, self.cidcoding) + + def is_vertical(self): + return self.vertical + + def is_multibyte(self): + return True + + def decode(self, bytes): + return self.cmap.decode(bytes) + + def char_disp(self, cid): + "Returns an integer for horizontal fonts, a tuple for vertical fonts." + return self.disps.get(cid, self.default_disp) + + def to_unichr(self, cid): + try: + if not self.unicode_map: + raise KeyError(cid) + return self.unicode_map.get_unichr(cid) + except KeyError: + raise self.PDFUnicodeNotDefined(self.cidcoding, cid) diff --git a/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/util.py b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/util.py new file mode 100644 index 0000000..fc01caf --- /dev/null +++ b/pdfmajor/interpreter_v2/commands/state/PDFTextState/PDFFont/util.py @@ -0,0 +1,57 @@ +from pdfmajor.utils import FONT_METRICS, choplist, isnumber + + +def get_widths(seq): + widths = {} + r = [] + for v in seq: + if isinstance(v, list): + if r: + char1 = r[-1] + for (i, w) in enumerate(v): + widths[char1+i] = w + r = [] + elif isnumber(v): + r.append(v) + if len(r) == 3: + (char1, char2, w) = r + for i in range(char1, char2+1): + widths[i] = w + r = [] + return widths +assert get_widths([1]) == {} +assert get_widths([1,2,3]) == {1:3, 2:3} +assert get_widths([1,[2,3],6,[7,8]]) == {1:2,2:3, 6:7,7:8} + +def get_widths2(seq): + widths = {} + r = [] + for v in seq: + if isinstance(v, list): + if r: + char1 = r[-1] + for (i, (w, vx, vy)) in enumerate(choplist(3, v)): + widths[char1+i] = (w, (vx, vy)) + r = [] + elif isnumber(v): + r.append(v) + if len(r) == 5: + (char1, char2, w, vx, vy) = r # pylint: disable=E0632 + for i in range(char1, char2+1): + widths[i] = (w, (vx, vy)) + r = [] + return widths +assert get_widths2([1]) == {} +assert get_widths2([1,2,3,4,5]) == {1:(3, (4,5)), 2:(3, (4,5))} +assert get_widths2([1,[2,3,4,5],6,[7,8,9]]) == {1:(2, (3,4)), 6:(7, (8,9))} + + +## FontMetricsDB +## +class FontMetricsDB(object): + + @classmethod + def get_metrics(klass, fontname): + return FONT_METRICS[fontname] + + diff --git a/pdfmajor/interpreter_v2/commands/state/PDFTextState/__init__.py b/pdfmajor/interpreter_v2/commands/state/PDFTextState/__init__.py index 6c93e2a..35c1482 100644 --- a/pdfmajor/interpreter_v2/commands/state/PDFTextState/__init__.py +++ b/pdfmajor/interpreter_v2/commands/state/PDFTextState/__init__.py @@ -1,4 +1,4 @@ -from .PDFFont import PDFFont +from .PDFFont import PDFFont, get_font from pdfmajor.utils import MATRIX_IDENTITY ## PDFTextState diff --git a/pdfmajor/interpreter_v2/commands/state/__init__.py b/pdfmajor/interpreter_v2/commands/state/__init__.py index f371f30..6962cef 100644 --- a/pdfmajor/interpreter_v2/commands/state/__init__.py +++ b/pdfmajor/interpreter_v2/commands/state/__init__.py @@ -5,10 +5,17 @@ from .PDFGraphicState import PDFGraphicState from .PDFGraphicState.PDFColor import PDFColor from .PDFGraphicState.PDFColorSpace import PREDEFINED_COLORSPACE, OrderedDict, PDFColorSpace -from .PDFTextState import PDFTextState +from .PDFTextState import PDFTextState, PDFFont, get_font from .Curves import CurvePath -from .PDFItem import PDFItem, PDFImage, PDFShape, PDFXObject, PDFText + +from .layout import LTItem, LTComponent, LTContainer +from .layout import LTCharBlock, LTChar +from .layout import LTCurve, LTLine, LTHorizontalLine, LTVerticalLine, LTRect +from .layout import LTXObject +from .layout import LTImage + +from .layout import make_char_block, make_curve, make_image, make_xobject class PDFStateStack: def __init__(self): @@ -18,7 +25,7 @@ class PDFStateStack: self.graphicstack : List[PDFGraphicState] = [] self.curvestacks: List[CurvePath] = [] self.argstack: List[bytes] = [] - self.complete_items : List[PDFItem] = [] + self.complete_layout_items : List[LTItem] = [] self.colorspace_map: OrderedDict = PREDEFINED_COLORSPACE.copy() self.fontmap = {} self.xobjmap = {} diff --git a/pdfmajor/interpreter_v2/commands/state/layout/LTCurves.py b/pdfmajor/interpreter_v2/commands/state/layout/LTCurves.py index afb61f5..c35c159 100644 --- a/pdfmajor/interpreter_v2/commands/state/layout/LTCurves.py +++ b/pdfmajor/interpreter_v2/commands/state/layout/LTCurves.py @@ -2,8 +2,8 @@ from typing import List, Tuple from pdfmajor.utils import Bbox -from ..state.PDFGraphicState import PDFGraphicStateColor -from ..state.Curves import CurvePath +from ..PDFGraphicState import PDFColor +from ..Curves import CurvePath from ._base import LTComponent, Bbox @@ -13,8 +13,8 @@ class LTCurve(LTComponent): linewidth: float, paths: List[CurvePath], bbox: Bbox, - stroke: PDFGraphicStateColor, - fill: PDFGraphicStateColor, + stroke: PDFColor, + fill: PDFColor, evenodd: bool, ): LTComponent.__init__(self, bbox) diff --git a/pdfmajor/interpreter_v2/commands/state/layout/LTFigure.py b/pdfmajor/interpreter_v2/commands/state/layout/LTFigure.py deleted file mode 100644 index 974d3e5..0000000 --- a/pdfmajor/interpreter_v2/commands/state/layout/LTFigure.py +++ /dev/null @@ -1,18 +0,0 @@ -from typing import List - -from ._base import LTContainer -from pdfmajor.utils import Bbox, Point, matrix2str - -## LTFigure -## -class LTFigure(LTContainer): - - def __init__(self, name: str, pts: List[Point], matrix): - LTContainer.__init__(self, Bbox.from_points(pts)) - self.name = name - self.matrix = matrix - - def __repr__(self): - return ('<%s(%s) %s matrix=%s>' % - (self.__class__.__name__, self.name, - str(self.bbox), matrix2str(self.matrix))) diff --git a/pdfmajor/interpreter_v2/commands/state/layout/LTImage.py b/pdfmajor/interpreter_v2/commands/state/layout/LTImage.py index c7ea194..672171f 100644 --- a/pdfmajor/interpreter_v2/commands/state/layout/LTImage.py +++ b/pdfmajor/interpreter_v2/commands/state/layout/LTImage.py @@ -1,5 +1,5 @@ from pdfmajor.utils import Bbox -from ..interpreter.PDFStream import PDFStream +from pdfmajor.interpreter.PDFStream import PDFStream from ._base import LTComponent ## LTImage diff --git a/pdfmajor/interpreter_v2/commands/state/layout/LTPage.py b/pdfmajor/interpreter_v2/commands/state/layout/LTPage.py deleted file mode 100644 index 11ea8b1..0000000 --- a/pdfmajor/interpreter_v2/commands/state/layout/LTPage.py +++ /dev/null @@ -1,20 +0,0 @@ -from ._base import LTContainer -from ..utils import Bbox - -class LTPage(LTContainer): - """A class reprensting an extracted page from the pdf - Arguments: - pageid {int} -- page number - bbox {Bbox} -- Boundary Box - rotate {float} -- rotation of the page (default: {0}) - """ - - def __init__(self, pageid: int, bbox: Bbox, rotate: float=0): - LTContainer.__init__(self, bbox) - self.pageid = pageid - self.rotate = rotate - - def __repr__(self): - return ('<%s(%r) %s rotate=%r>' % - (self.__class__.__name__, self.pageid, - str(self.bbox), self.rotate)) \ No newline at end of file diff --git a/pdfmajor/interpreter_v2/commands/state/layout/LTXObject.py b/pdfmajor/interpreter_v2/commands/state/layout/LTXObject.py new file mode 100644 index 0000000..9c48059 --- /dev/null +++ b/pdfmajor/interpreter_v2/commands/state/layout/LTXObject.py @@ -0,0 +1,22 @@ +from typing import List + +from pdfmajor.utils import Bbox, Point, matrix2str +from pdfmajor.interpreter.PDFStream import PDFStream + +from ._base import LTContainer + +## LTXObject +## +class LTXObject(LTContainer): + + def __init__(self, name: str, bbox: Bbox, xobj_stream: PDFStream, resources: dict, t_matrix: List[Point] ): + LTContainer.__init__(self, bbox) + self.name = name + self.stream = xobj_stream + self.t_matrix = t_matrix + self.resources = resources + + def __repr__(self): + return ('<%s(%s) %s matrix=%s>' % + (self.__class__.__name__, self.name, + str(self.bbox), matrix2str(self.t_matrix))) diff --git a/pdfmajor/interpreter_v2/commands/state/layout/PageLayoutAnalyzer.py b/pdfmajor/interpreter_v2/commands/state/layout/PageLayoutAnalyzer.py deleted file mode 100644 index c2d0a8a..0000000 --- a/pdfmajor/interpreter_v2/commands/state/layout/PageLayoutAnalyzer.py +++ /dev/null @@ -1,178 +0,0 @@ -import logging -import re -import io - -from typing import Tuple, List - -from ._base import LTItem -from .LTPage import LTPage -from .LTCurves import LTCurve, LTLine, LTRect, LTHorizontalLine, LTVerticalLine -from .LTFigure import LTFigure -from .LTImage import LTImage -from .LTChar import LTChar -from .LTCharBlock import LTCharBlock - -from pdfmajor.utils import Point, Bbox, INF -from pdfmajor.utils import apply_matrix_pt, mult_matrix - -from ..interpreter.PDFFont import PDFUnicodeNotDefined -from ..interpreter.PDFResourceManager import PDFResourceManager -from ..interpreter.PDFGraphicState import PDFGraphicState, PDFGraphicStateColor -from ..interpreter.PDFdevice import PDFTextDevice -from ..interpreter.PDFTextState import PDFTextState -from ..interpreter.PDFColorSpace import PDFColorSpace -from ..interpreter.PDFPage import PDFPage -from ..interpreter.PDFStream import PDFStream -from ..interpreter.types import CurvePath - -log = logging.getLogger(__name__) - - -## PDFLayoutAnalyzer -## -class PDFPageLayoutAnalyzer: - - - def __init__(self, page: PDFPage, ctm: Tuple[float], pageno: int=1): - self.page = page - self.pageno = pageno - self._stack = [] - self.ctm = ctm - (x0, y0, x1, y1) = page.mediabox - (x0, y0) = apply_matrix_pt(ctm, (x0, y0)) - (x1, y1) = apply_matrix_pt(ctm, (x1, y1)) - self.cur_item = LTPage( - self.pageno, - Bbox( 0, 0, abs(x0-x1), abs(y0-y1) ) - ) - - def done(self): - assert not self._stack, str(len(self._stack)) - assert isinstance(self.cur_item, LTPage), str(type(self.cur_item)) - return self.cur_item - - def begin_figure(self, name: str, bbox: List[Point], matrix: List[Point]): - self._stack.append(self.cur_item) - - matrix = mult_matrix(matrix, self.ctm) - (x, y, w, h) = bbox - - self.cur_item = LTFigure(name, [ - apply_matrix_pt(matrix, (p, q)) - for (p, q) in ((x, y), (x+w, y), (x, y+h), (x+w, y+h)) - ], matrix) - - def end_figure(self, name: str): - fig = self.cur_item - assert isinstance(self.cur_item, LTFigure), str(type(self.cur_item)) - self.cur_item = self._stack.pop() - self.cur_item.add(fig) - return - - def render_image(self, name: str, stream: PDFStream): - assert isinstance(self.cur_item, LTFigure), str(type(self.cur_item)) - item = LTImage( - name, - stream, - Bbox( - self.cur_item.x0, self.cur_item.y0, - self.cur_item.x1, self.cur_item.y1 - ) - ) - self.cur_item.add(item) - return - - def paint_path(self, gstate: PDFGraphicState, evenodd: bool, paths: List[CurvePath]): - shape = ''.join(x.method.value for x in paths) - if shape == 'ml': - # horizontal/vertical line - pt0 = paths[0].points[0] - pt1 = paths[1].points[0] - (x0, y0) = apply_matrix_pt(self.ctm, (pt0.x, pt0.y)) - (x1, y1) = apply_matrix_pt(self.ctm, (pt1.x, pt1.y)) - if x0 == x1: - self.cur_item.add(LTVerticalLine( - linewidth=gstate.linewidth, - paths=paths, - bbox=Bbox( x0, y0, x1, y1 ), - stroke=gstate.scolor, - fill=gstate.ncolor, - evenodd=evenodd, - )) - elif y0 == y1: - self.cur_item.add(LTHorizontalLine( - linewidth=gstate.linewidth, - paths=paths, - bbox=Bbox( x0, y0, x1, y1 ), - stroke=gstate.scolor, - fill=gstate.ncolor, - evenodd=evenodd, - )) - else: - self.cur_item.add(LTLine( - linewidth=gstate.linewidth, - paths=paths, - bbox=Bbox( x0, y0, x1, y1 ), - stroke=gstate.scolor, - fill=gstate.ncolor, - evenodd=evenodd, - )) - elif shape == 'mlllh': - # rectangle - pt0 = paths[0].points[0] - pt1 = paths[1].points[0] - pt2 = paths[2].points[0] - pt3 = paths[3].points[0] - (x0, y0) = apply_matrix_pt(self.ctm, (pt0.x, pt0.y)) - (x1, y1) = apply_matrix_pt(self.ctm, (pt1.x, pt1.y)) - (x2, y2) = apply_matrix_pt(self.ctm, (pt2.x, pt2.y)) - (x3, y3) = apply_matrix_pt(self.ctm, (pt3.x, pt3.y)) - if ((x0 == x1 and y1 == y2 and x2 == x3 and y3 == y0) or - (y0 == y1 and x1 == x2 and y2 == y3 and x3 == x0)): - self.cur_item.add(LTRect( - linewidth=gstate.linewidth, - paths=paths, - bbox=Bbox( - x0=min(x0,x1,x2,x3), - x1=max(x0,x1,x2,x3), - y0=min(y0,y1,y2,y3), - y1=max(y0,y1,y2,y3), - ), - evenodd=evenodd, - stroke=gstate.scolor, - fill=gstate.ncolor, - )) - else: - (x0, y0, x1, y1) = (INF, INF, -INF, -INF) - for path in paths: - if path.method in [CurvePath.METHOD.MOVE_TO, CurvePath.METHOD.LINE_TO]: - x0 = min(x0, path.points[0].x) - y0 = min(y0, path.points[0].y) - x1 = max(x1, path.points[0].x) - y1 = max(y1, path.points[0].y) - elif path.method in [CurvePath.METHOD.CURVE_BOTH_TO]: - x0 = min(x0, path.points[2].x) - y0 = min(y0, path.points[2].y) - x1 = max(x1, path.points[2].x) - y1 = max(y1, path.points[2].y) - elif path.method in [CurvePath.METHOD.CURVE_NEXT_TO, CurvePath.METHOD.CURVE_FIRST_TO]: - x0 = min(x0, path.points[1].x) - y0 = min(y0, path.points[1].y) - x1 = max(x1, path.points[1].x) - y1 = max(y1, path.points[1].y) - self.cur_item.add(LTCurve( - linewidth=gstate.linewidth, - paths=paths, - bbox=Bbox(x0,y0,x1,y1), - evenodd=evenodd, - stroke=gstate.scolor, - fill=gstate.ncolor - )) - - def render_char_box(self, chars: list, textstate: PDFTextState, graphicstate: PDFGraphicState, ncs: PDFColorSpace): - self.cur_item.add(LTCharBlock( - chars=chars, - textstate=textstate, - ncs=ncs, - graphicstate=graphicstate - )) diff --git a/pdfmajor/interpreter_v2/commands/state/layout/__init__.py b/pdfmajor/interpreter_v2/commands/state/layout/__init__.py index 854277a..cb34455 100644 --- a/pdfmajor/interpreter_v2/commands/state/layout/__init__.py +++ b/pdfmajor/interpreter_v2/commands/state/layout/__init__.py @@ -1,5 +1,5 @@ from typing import List -from pdfmajor.utils import apply_matrix_pt, INF, Bbox +from pdfmajor.utils import apply_matrix_pt, INF, Bbox, Point, MATRIX_IDENTITY, mult_matrix from ..Curves import CurvePath from ..PDFGraphicState import PDFGraphicState, PDFColor @@ -7,7 +7,9 @@ from ..PDFTextState import PDFTextState from ._base import LTItem, LTComponent, LTContainer from .LTCharBlock import LTCharBlock -from .LTCurves import LTCurve, LTHorizontalLine, LTRect, LTVerticalLine +from .LTCurves import LTCurve, LTLine, LTHorizontalLine, LTRect, LTVerticalLine +from .LTImage import LTImage +from .LTXObject import LTXObject from .utils.textdecoder import decode_text_seq @@ -19,7 +21,7 @@ def make_char_block(seq: bytearray, ctm: tuple, textstate: PDFTextState, color: textstate=textstate ) -def make_figure(ctm: tuple, gstate: PDFGraphicState, evenodd: bool, paths: List[CurvePath]): +def make_curve(ctm: tuple, gstate: PDFGraphicState, evenodd: bool, paths: List[CurvePath]): shape = ''.join(x.method.value for x in paths) if shape == 'ml': # horizontal/vertical line @@ -105,3 +107,27 @@ def make_figure(ctm: tuple, gstate: PDFGraphicState, evenodd: bool, paths: List[ stroke=gstate.scolor, fill=gstate.ncolor ) + +def make_image(obj, ctm): + matrix = mult_matrix(MATRIX_IDENTITY, ctm) + return LTImage( + name=str(id(obj)), + stream=obj, + bbox=Bbox.from_points([ + apply_matrix_pt(matrix, (p, q)) + for (p, q) in ((0, 0), (1, 0), (0, 1), (1, 1)) + ]) + ) + +def make_xobject(obj, bbox, ctm, matrix, resources): + (x, y, w, h) = bbox + return LTXObject( + name=str(id(obj)), + bbox=Bbox.from_points([ + apply_matrix_pt(matrix, (p, q)) + for (p, q) in ((x, y), (x+w, y), (x, y+h), (x+w, y+h)) + ]), + xobj_stream=obj, + resources=resources, + t_matrix = ctm + ) \ No newline at end of file diff --git a/pdfmajor/interpreter_v2/commands/state/layout/utils/textdecoder.py b/pdfmajor/interpreter_v2/commands/state/layout/utils/textdecoder.py index 870a7c0..7f7551a 100644 --- a/pdfmajor/interpreter_v2/commands/state/layout/utils/textdecoder.py +++ b/pdfmajor/interpreter_v2/commands/state/layout/utils/textdecoder.py @@ -41,7 +41,7 @@ def __render_string_along(idx: int, seq: bytearray, textstate.linematrix[idx] += adv char_meta_datas.append( - ( text, Bbox.from_points(bbox), ) + ( text, bbox ) ) if cid == 32 and textstate.wordspace: textstate.linematrix[idx] += textstate.wordspace diff --git a/pdfmajor/interpreter_v2/utils.py b/pdfmajor/interpreter_v2/utils.py new file mode 100644 index 0000000..ba889b9 --- /dev/null +++ b/pdfmajor/interpreter_v2/utils.py @@ -0,0 +1,50 @@ +from pdfmajor.interpreter.PDFParser import PDFStream +from pdfmajor.interpreter.PDFStream import list_value, dict_value, resolve1 +from pdfmajor.interpreter.PDFStream.PDFObjRef import PDFObjRef +from pdfmajor.interpreter.PSStackParser import literal_name +from pdfmajor.interpreter.constants import LITERAL_PDF, LITERAL_TEXT + +from .commands import PDFColorSpace, PREDEFINED_COLORSPACE +from .commands import PDFStateStack, get_font + +def get_colorspace(spec): + if isinstance(spec, list): + name = literal_name(spec[0]) + else: + name = literal_name(spec) + if name == 'ICCBased' and isinstance(spec, list) and 2 <= len(spec): + return PDFColorSpace(name, PDFStream.validated_stream(spec[1])['N']) + elif name == 'DeviceN' and isinstance(spec, list) and 2 <= len(spec): + return PDFColorSpace(name, len(list_value(spec[1]))) + else: + return PREDEFINED_COLORSPACE.get(name) + +def get_procset(procs): + for proc in procs: + if proc is LITERAL_PDF: + pass + elif proc is LITERAL_TEXT: + pass + else: + #raise PDFResourceError('ProcSet %r is not supported.' % proc) + pass + return + +def init_resources(state: PDFStateStack, font_cache: dict = {}): + for (k, v) in iter(dict_value(state.resources).items()): + # log.debug('Resource: %r: %r', k, v) + if k == 'Font': + for (fontid, spec) in iter(dict_value(v).items()): + objid = None + if isinstance(spec, PDFObjRef): + objid = spec.objid + spec = dict_value(spec) + state.fontmap[fontid] = get_font(objid, spec, font_cache) + elif k == 'ColorSpace': + for (csid, spec) in iter(dict_value(v).items()): + state.colorspace_map[csid] = get_colorspace(resolve1(spec)) + elif k == 'ProcSet': + get_procset(list_value(v)) + elif k == 'XObject': + for (xobjid, xobjstrm) in iter(dict_value(v).items()): + state.xobjmap[xobjid] = xobjstrm \ No newline at end of file diff --git a/pdfmajor/layouts/PDFLayoutMaker.py b/pdfmajor/layouts/PDFLayoutMaker.py deleted file mode 100644 index 97bae9a..0000000 --- a/pdfmajor/layouts/PDFLayoutMaker.py +++ /dev/null @@ -1,178 +0,0 @@ -import logging -import re -import io - -from typing import Tuple, List - -from ._base import LTItem -from .LTPage import LTPage -from .LTCurves import LTCurve, LTLine, LTRect, LTHorizontalLine, LTVerticalLine -from .LTFigure import LTFigure -from .LTImage import LTImage -from .LTChar import LTChar -from .LTCharBlock import LTCharBlock - -from ..utils import Point, Bbox, INF -from ..utils import apply_matrix_pt, mult_matrix - -from ..interpreter.PDFFont import PDFUnicodeNotDefined -from ..interpreter.PDFResourceManager import PDFResourceManager -from ..interpreter.PDFGraphicState import PDFGraphicState, PDFGraphicStateColor -from ..interpreter.PDFdevice import PDFTextDevice -from ..interpreter.PDFTextState import PDFTextState -from ..interpreter.PDFColorSpace import PDFColorSpace -from ..interpreter.PDFPage import PDFPage -from ..interpreter.PDFStream import PDFStream -from ..interpreter.types import CurvePath - -log = logging.getLogger(__name__) - - -## PDFLayoutAnalyzer -## -class PDFPageLayoutAnalyzer: - - - def __init__(self, page: PDFPage, ctm: Tuple[float], pageno: int=1): - self.page = page - self.pageno = pageno - self._stack = [] - self.ctm = ctm - (x0, y0, x1, y1) = page.mediabox - (x0, y0) = apply_matrix_pt(ctm, (x0, y0)) - (x1, y1) = apply_matrix_pt(ctm, (x1, y1)) - self.cur_item = LTPage( - self.pageno, - Bbox( 0, 0, abs(x0-x1), abs(y0-y1) ) - ) - - def done(self): - assert not self._stack, str(len(self._stack)) - assert isinstance(self.cur_item, LTPage), str(type(self.cur_item)) - return self.cur_item - - def begin_figure(self, name: str, bbox: List[Point], matrix: List[Point]): - self._stack.append(self.cur_item) - - matrix = mult_matrix(matrix, self.ctm) - (x, y, w, h) = bbox - - self.cur_item = LTFigure(name, [ - apply_matrix_pt(matrix, (p, q)) - for (p, q) in ((x, y), (x+w, y), (x, y+h), (x+w, y+h)) - ], matrix) - - def end_figure(self, name: str): - fig = self.cur_item - assert isinstance(self.cur_item, LTFigure), str(type(self.cur_item)) - self.cur_item = self._stack.pop() - self.cur_item.add(fig) - return - - def render_image(self, name: str, stream: PDFStream): - assert isinstance(self.cur_item, LTFigure), str(type(self.cur_item)) - item = LTImage( - name, - stream, - Bbox( - self.cur_item.x0, self.cur_item.y0, - self.cur_item.x1, self.cur_item.y1 - ) - ) - self.cur_item.add(item) - return - - def paint_path(self, gstate: PDFGraphicState, evenodd: bool, paths: List[CurvePath]): - shape = ''.join(x.method.value for x in paths) - if shape == 'ml': - # horizontal/vertical line - pt0 = paths[0].points[0] - pt1 = paths[1].points[0] - (x0, y0) = apply_matrix_pt(self.ctm, (pt0.x, pt0.y)) - (x1, y1) = apply_matrix_pt(self.ctm, (pt1.x, pt1.y)) - if x0 == x1: - self.cur_item.add(LTVerticalLine( - linewidth=gstate.linewidth, - paths=paths, - bbox=Bbox( x0, y0, x1, y1 ), - stroke=gstate.scolor, - fill=gstate.ncolor, - evenodd=evenodd, - )) - elif y0 == y1: - self.cur_item.add(LTHorizontalLine( - linewidth=gstate.linewidth, - paths=paths, - bbox=Bbox( x0, y0, x1, y1 ), - stroke=gstate.scolor, - fill=gstate.ncolor, - evenodd=evenodd, - )) - else: - self.cur_item.add(LTLine( - linewidth=gstate.linewidth, - paths=paths, - bbox=Bbox( x0, y0, x1, y1 ), - stroke=gstate.scolor, - fill=gstate.ncolor, - evenodd=evenodd, - )) - elif shape == 'mlllh': - # rectangle - pt0 = paths[0].points[0] - pt1 = paths[1].points[0] - pt2 = paths[2].points[0] - pt3 = paths[3].points[0] - (x0, y0) = apply_matrix_pt(self.ctm, (pt0.x, pt0.y)) - (x1, y1) = apply_matrix_pt(self.ctm, (pt1.x, pt1.y)) - (x2, y2) = apply_matrix_pt(self.ctm, (pt2.x, pt2.y)) - (x3, y3) = apply_matrix_pt(self.ctm, (pt3.x, pt3.y)) - if ((x0 == x1 and y1 == y2 and x2 == x3 and y3 == y0) or - (y0 == y1 and x1 == x2 and y2 == y3 and x3 == x0)): - self.cur_item.add(LTRect( - linewidth=gstate.linewidth, - paths=paths, - bbox=Bbox( - x0=min(x0,x1,x2,x3), - x1=max(x0,x1,x2,x3), - y0=min(y0,y1,y2,y3), - y1=max(y0,y1,y2,y3), - ), - evenodd=evenodd, - stroke=gstate.scolor, - fill=gstate.ncolor, - )) - else: - (x0, y0, x1, y1) = (INF, INF, -INF, -INF) - for path in paths: - if path.method in [CurvePath.METHOD.MOVE_TO, CurvePath.METHOD.LINE_TO]: - x0 = min(x0, path.points[0].x) - y0 = min(y0, path.points[0].y) - x1 = max(x1, path.points[0].x) - y1 = max(y1, path.points[0].y) - elif path.method in [CurvePath.METHOD.CURVE_BOTH_TO]: - x0 = min(x0, path.points[2].x) - y0 = min(y0, path.points[2].y) - x1 = max(x1, path.points[2].x) - y1 = max(y1, path.points[2].y) - elif path.method in [CurvePath.METHOD.CURVE_NEXT_TO, CurvePath.METHOD.CURVE_FIRST_TO]: - x0 = min(x0, path.points[1].x) - y0 = min(y0, path.points[1].y) - x1 = max(x1, path.points[1].x) - y1 = max(y1, path.points[1].y) - self.cur_item.add(LTCurve( - linewidth=gstate.linewidth, - paths=paths, - bbox=Bbox(x0,y0,x1,y1), - evenodd=evenodd, - stroke=gstate.scolor, - fill=gstate.ncolor - )) - - def render_char_box(self, chars: list, textstate: PDFTextState, graphicstate: PDFGraphicState, ncs: PDFColorSpace): - self.cur_item.add(LTCharBlock( - chars=chars, - textstate=textstate, - ncs=ncs, - graphicstate=graphicstate - )) diff --git a/tests/test_extractor.py b/tests/test_extractor.py deleted file mode 100644 index a8325a4..0000000 --- a/tests/test_extractor.py +++ /dev/null @@ -1,19 +0,0 @@ -import os - -from pdfmajor.extractor import v2_extract_items_from_pdf -from pdfmajor.interpreter_v2 import PDFText - -CUR_PATH = os.path.dirname(os.path.dirname(__file__)) -CUR_PATH = os.path.dirname(os.getcwd()) - -FILE_NAME = os.path.join( - CUR_PATH, - "samples/pdf/fonts.pdf" -) - -for page in v2_extract_items_from_pdf(FILE_NAME): - for item in page: - if isinstance(item, PDFText): - print(item.chars) - else: - print(item) \ No newline at end of file diff --git a/tests/test_interpreter.py b/tests/test_interpreter.py new file mode 100644 index 0000000..9bf7e7d --- /dev/null +++ b/tests/test_interpreter.py @@ -0,0 +1,22 @@ +import os + +# from pdfmajor.extractor import extract_items_from_pdf +from pdfmajor.interpreter_v2 import LTCharBlock, PDFInterpreter, logging +from pdfmajor.interpreter_v2 import PageInterpreter +from pdfmajor.interpreter_v2.commands import LTItem +from pdfmajor.interpreter_v2.commands import LTCharBlock, LTChar + +CUR_PATH = os.path.dirname(os.path.dirname(__file__)) + +FILE_NAME = os.path.join( + CUR_PATH, + "samples/pdf/tables.pdf" +) + +for page in PDFInterpreter(FILE_NAME, debug_level=logging.INFO): + assert isinstance(page, PageInterpreter) + for item in page: + assert isinstance(item, LTItem) + if isinstance(item, LTCharBlock): + for char in item: + assert(char, LTChar) \ No newline at end of file