def get_name2code_set():
"""
name is the glyph name,
code is the unique id for the pair of glyph name & font
one glyph name could have multiple code in different fonts.
:return:
"""
from pdfxml.path_util import SHARED_FOLDER
#name2code_cache_path = "{}/tmp/name2code.pkl".format(infty_cdb_folder)
name2code_cache_path = "{}/InftyCDB-1/cache_data/name2code.json".format(
SHARED_FOLDER)
if os.path.isfile(name2code_cache_path):
name2code_list = load_general(name2code_cache_path)
name2code_set = {}
for name, code_list in name2code_list.items():
name2code_set[name] = set(code_list)
return name2code_set
cid2info = load_char_map()
name2code_set = {}
for c in cid2info.values():
if c['name'] not in name2code_set:
name2code_set[c['name']] = set()
name2code_set[c['name']].add(c['code'])
name2code_list = {}
for name, code_set in name2code_set.items():
name2code_list[name] = list(code_set)
dump_general(name2code_list, name2code_cache_path)
return name2code_set
def get_latex2adjustment_ratio_list():
from pdfxml.InftyCDB.name2latex import name2latex
cache_path = "{}/InftyCDB-1/cache_data/all_adjustment_ratio_list_latex.json".format(
SHARED_FOLDER)
if os.path.isfile(cache_path):
return load_general(cache_path)
code2name = get_code2name()
gt_dict = {
'code2upper_ratio': {},
'code2upper_ratio_hor': {},
'code2lower_ratio': {},
'code2lower_ratio_hor': {}
}
total = unify_all_adjustment_ratio()
for adjustment_type in total.keys():
c2rlist = total[adjustment_type]
for c, rlist in c2rlist.items():
latex_val = name2latex[code2name[c]]
if latex_val not in gt_dict[adjustment_type]:
gt_dict[adjustment_type][latex_val] = []
gt_dict[adjustment_type][latex_val].extend(rlist)
dump_general(gt_dict, cache_path)
return gt_dict
def unify_glyph_type_adjustment_ratio():
"""
In comparison with `create_all_adjustment_ratio`,
this is a marginal version with each code mapping to the glyph type
:return: from glyph_type to list of adjustment ratio
"""
from pdfxml.path_util import PROJECT_FOLDER
#cache_path = "{}/tmp/all_adjustment_ratio_glyph_type_level.pkl".format(infty_cdb_folder)
cache_path = "{}/InftyCDB-1/cache_data/all_adjustment_ratio_glyph_type_level.json".format(
SHARED_FOLDER)
if os.path.isfile(cache_path):
return load_general(cache_path)
code2name = get_code2name()
total = unify_all_adjustment_ratio()
# gt is short for glyph_type
# the code here is the glyph type
gt_dict = {
'code2upper_ratio': {},
'code2upper_ratio_hor': {},
'code2lower_ratio': {},
'code2lower_ratio_hor': {}
}
# NOTE: This is what I am looking for to ajdust by group
for adjustment_type in total.keys():
c2rlist = total[adjustment_type]
for c, rlist in c2rlist.items():
gt = get_glyph_type(code2name[c])
if gt not in gt_dict[adjustment_type]:
gt_dict[adjustment_type][gt] = []
gt_dict[adjustment_type][gt].extend(rlist)
dump_general(gt_dict, cache_path)
return gt_dict
def load_char_map(refresh=False):
"""
load mathematical chars from the CSV file
NOTE: Dec. 10, later use the XLSX file with less noise.
:param refresh:
:return: the dict from the char id to char info
"""
#cached_path = "{}/InftyCDB/cache_data/chars.json".format(PROJECT_FOLDER)
cached_path = "{}/tmp/chars.json".format(infty_cdb_folder)
test_folder_exist_for_file_path(cached_path)
if os.path.isfile(cached_path) and not refresh:
return load_general(cached_path)
print('rebuild cache from the xlsx file')
me_xlsx_path = "{}/InftyCDB-1/resources/me.xlsx".format(SHARED_FOLDER)
me_elems = load_me_elems_xlsx(me_xlsx_path)
cid2info = {}
for me_elem in me_elems:
cid2info[me_elem['cid']] = me_elem
dump_general(cid2info, cached_path)
return cid2info
def create_all_adjustment_ratio():
"""
This is code level adjustment, which is a very refined level.
:return:
"""
cache_path = "{}/tmp/all_adjustment_ratio_code_level.json".format(
infty_cdb_folder)
if os.path.isfile(cache_path):
return load_general(cache_path)
total = {
'code2upper_ratio': {},
'code2upper_ratio_hor': {},
'code2lower_ratio': {},
'code2lower_ratio_hor': {}
}
def update(info_dict):
for k in total.keys():
c2rlist = info_dict[k]
for c, rlist in c2rlist.items(): # code to list of ratio values
if c not in total[k]:
total[k][c] = []
total[k][c].extend(rlist)
me_idx_list = get_me_idx_list()
#me_idx_list = get_me_idx_list_with_adjustment()
for n, me_idx in enumerate(me_idx_list):
if n % 100 == 0:
print "done loading {} files".format(n)
adjust_dict = get_char2extend_for_one(me_idx)
update(adjust_dict)
dump_general(total, cache_path)
return total
def get_page_num(fpath):
""" Get the page number for the current pdf file
https://stackoverflow.com/questions/45841012/how-can-i-get-the-total-count-of-total-pages-of-a-pdf-using-pdfminer-in-python
"""
tmp_path = get_tmp_path(fpath)
cache_path = "{}.page_num.json".format(tmp_path)
if os.path.isfile(cache_path):
tmp_dict = load_general(cache_path)
return tmp_dict['page_num']
# Open a PDF file.
fp = open(fpath, 'rb')
# Create a PDF parser object associated with the file object.
parser = PDFParser(fp)
# Create a PDF document object that stores the document structure.
# Supply the password for initialization.
document = PDFDocument(parser)
c = resolve1(document.catalog['Pages'])['Count']
tmp_dict = {'page_num': c}
dump_general(tmp_dict, cache_path)
return c