def extract_block(self, block: dict, page: int) -> List[ToCEntry]:
"""Extract matching headings in a block.
Argument
block: a block dictionary
{
'bbox': (float, float, float, float),
'lines': [dict],
'type': int
}
Returns
a list of toc entries, concatenated from the result of lines
"""
if block.get('type') != 0:
# not a text block
return []
vpos = block.get('bbox', (0, 0))[1]
try:
frags = chain.from_iterable([
self._extract_line(ln) for ln in block.get('lines')
])
titles = concatFrag(frags)
return [
ToCEntry(level, title, page, vpos)
for level, title in titles.items()
]
except FoundGreedy as e:
# return the entire block as a single entry
return [ToCEntry(e.level, blk_to_str(block), page, vpos)]
def build_final_toc(path_title_level_list, toc):
idx1, idx2 = 0, 0
last_page_num = 1
final_toc = []
for idx1, (path, title, level) in enumerate(path_title_level_list):
title_match = "".join(title.split()) == "".join(
toc[idx2].title.split())
if path is None or (path is not None or title_match):
final_toc.append(ToCEntry(level, title, toc[idx2].pagenum))
if path is not None:
if title_match:
idx2 += 1
else:
print("missing chapter: ", title)
final_toc.pop() # remove missing ones
return final_toc
def parse_entry(entry: List) -> ToCEntry:
"""parse a row in csv to a toc entry"""
# a somewhat weird hack, csv reader would read spaces as an empty '', so we
# only need to count the number of '' before an entry to determined the
# heading level
indent = len(list(takewhile(lambda x: x == '', entry)))
try:
toc_entry = ToCEntry(
int(indent / 4) + 1, # 4 spaces = 1 level
entry[indent], # heading
int(entry[indent + 1]), # pagenum
*entry[indent + 2:] # vpos
)
return toc_entry
except IndexError as e:
print(f"Unable to parse toc entry {entry};",
f"Need at least {indent + 2} parts but only have {len(entry)}.",
"Make sure the page number is present.",
file=sys.stderr)
raise e
def read_toc(doc: Document) -> List[ToCEntry]:
"""Read table of contents from a document"""
return [ToCEntry(*entry) for entry in doc.get_toc()]
import toml
from mamba import description, it, before
from fitzutils import ToCEntry
from pdftocgen.tocgen import gen_toc
dirpath = os.path.dirname(os.path.abspath(__file__))
with description("gen_toc") as self:
with before.all:
self.level2 = fitz.open(os.path.join(dirpath, "files/level2.pdf"))
self.level2_recipe = toml.load(
open(os.path.join(dirpath, "files/level2_recipe.toml")))
self.level2_expect = [
ToCEntry(level=1,
title='1 Section One',
pagenum=1,
vpos=237.6484375),
ToCEntry(level=1,
title='2 Section Two',
pagenum=1,
vpos=567.3842163085938),
ToCEntry(level=2,
title='2.1 Subsection Two.One',
pagenum=2,
vpos=452.56671142578125),
ToCEntry(
level=1,
title='3 Section Three, with looong loooong looong ti- tle',
pagenum=3,
vpos=335.569580078125),
ToCEntry(
from mamba import description, it, before
from fitzutils import ToCEntry
from pdftocio.tocio import read_toc, write_toc
dirpath = os.path.dirname(os.path.abspath(__file__))
level2 = os.path.join(dirpath, "files/level2.pdf")
hastoc = os.path.join(dirpath, "files/hastoc.pdf")
with description("read_toc") as self:
with before.all:
self.doc = fitz.open(level2)
self.reference = fitz.open(hastoc)
self.expect = [
ToCEntry(level=1, title='Section One', pagenum=1),
ToCEntry(level=1, title='Section Two', pagenum=1),
ToCEntry(level=2, title='Subsection Two.One', pagenum=2),
ToCEntry(level=1,
title='Section Three, with looong loooong looong title',
pagenum=3),
ToCEntry(
level=2,
title='Subsection Three.One, '
'with even loooooooooooonger title, and probably even more',
pagenum=3),
ToCEntry(level=2, title='Subsection Three.Two', pagenum=4),
ToCEntry(level=2, title='Subsection Three.Three', pagenum=5),
ToCEntry(level=1, title='The End', pagenum=5)
]
import os
import io
from mamba import description, it, before
from fitzutils import (dump_toc, ToCEntry)
from pdftocio.tocparser import parse_toc
dirpath = os.path.dirname(os.path.abspath(__file__))
valid_file = os.path.join(dirpath, "files/level2.pdf")
invalid_file = os.path.join(dirpath, "files/nothing.pdf")
with description("parse_toc") as self:
with before.all:
self.toc = [
ToCEntry(level=1, title="title1", pagenum=1, vpos=100.0),
ToCEntry(level=2, title="title2", pagenum=1, vpos=150.0),
ToCEntry(level=3, title="title3", pagenum=2, vpos=90.0),
ToCEntry(level=2, title="title4", pagenum=2, vpos=150.0),
ToCEntry(level=2, title="title5", pagenum=3, vpos=0.0),
ToCEntry(level=1, title="title6", pagenum=5, vpos=200.0)
]
self.toc_novpos = [
ToCEntry(level=1, title="title1", pagenum=1),
ToCEntry(level=2, title="title2", pagenum=1),
ToCEntry(level=3, title="title3", pagenum=2),
ToCEntry(level=2, title="title4", pagenum=2),
ToCEntry(level=2, title="title5", pagenum=3),
ToCEntry(level=1, title="title6", pagenum=5)
]
with it("exits if pdf file is invalid and exit_on_error is true"):
try:
with open_pdf(invalid_file, True) as doc:
assert False, "should have exited"
except AssertionError as err:
raise err
except:
pass
with description("ToCEntry") as self:
with it("matches fitz's representation"):
fitz_entry = [1, "title", 2]
fitz_entry2 = [1, "title", 2, 100.0]
toc_entry = ToCEntry(level=1, title="title", pagenum=2)
toc_entry2 = ToCEntry(level=1, title="title", pagenum=2, vpos=100.0)
assert toc_entry.to_fitz_entry() == fitz_entry
assert toc_entry2.to_fitz_entry() == fitz_entry2
assert ToCEntry(*fitz_entry) == toc_entry
assert ToCEntry(*fitz_entry2) == toc_entry2
with it("is sorted correctly"):
entries = [
ToCEntry(level=1, title="title4", pagenum=2, vpos=150.0),
ToCEntry(level=1, title="title3", pagenum=2, vpos=90.0),
ToCEntry(level=1, title="title5", pagenum=3, vpos=0.0),
ToCEntry(level=1, title="title2", pagenum=1, vpos=150.0),
ToCEntry(level=1, title="title1", pagenum=1, vpos=100.0),