def split_and_export_dataframe(df,
nrows,
sortby=None,
output_name=None,
export_csv=True):
"""SPlit dataframe with given row numbers and export it wither in csv or excel
Arguments:
df {pd.DataFrame} -- Pandas dataframe
nrows {int} -- Number of rows to split the database
Keyword Arguments:
sortby {string} -- Sort DataFrame before splitting (default: {None})
output_name {string} -- Output file name to save, input if not given (default: {None})
export_csv {bool} -- Output file format, (default: {True}) ';' seperated csv file'
"""
if sortby is not None:
df.sort_values(by=sortby, inplace=True)
n_iter = int(pd.np.ceil(len(df) / nrows))
if output_name is None:
output_name = input("Please name the output file to save: ")
if export_csv:
for i in range(n_iter):
df[nrows * i:nrows * i + nrows].to_csv(checkfile(
r'./{}.csv'.format(output_name)),
sep=';',
index=False)
else:
for i in range(n_iter):
df[nrows * i:nrows * i + nrows].to_excel(checkfile(
r'./{}.xlsx'.format(output_name)),
index=False)
def rotate(self,
pages: List[int] = None,
step: int = 1,
save_as: os.PathLike = None) -> None:
"""
Rotate given pages in steps of 90 degrees, negative sign for counter clockwise direction
Args:
pages (List[int], optional): _description_. Defaults to None.
step (int, optional): _description_. Defaults to 1.
save_as (os.PathLike, optional): _description_. Defaults to None.
"""
if pages is None or pages == []:
pages = list(range(len(self.pdf.pages)))
# page_to_rotate = [0, ]
for page in self.document:
if page.number in pages and isinstance(step, int):
# Rotate page clockwise 90 degrees and add to new pdf
page.set_rotation(step * 90)
if save_as is None:
self.document.save(checkfile(path))
else:
self.document.save(checkfile(save_as))
def merge(docpaths: List[os.PathLike, ] = None,
output: os.PathLike = None) -> None:
"""
Summary:
Merge given list of pdf paths
Args:
docpaths (List[os.PathLike, ], optional): _description_. Defaults to None.
output (os.PathLike, optional): _description_. Saves in folder of first path as Binder
"""
pdf = PDFEditor(docpaths[0])
for path in docpaths[1:]:
pdf.insert(path, merge=True)
if output is None:
pdf.document.save(
checkfile(
os.path.join(os.path.split(docpaths[0])[0], "Binder.pdf")))
else:
pdf.document.save(checkfile(output))
def insert(self,
docpath: os.PathLike,
page_range: Tuple[int, int] = None,
start_at: int = None,
rotate: int = 0,
links: bool = True,
annots: bool = True,
show_progress: int = 0,
final: int = 1,
save_as: os.PathLike = None,
merge: bool = False) -> None:
"""
Summary:
Insert pdf, pages in any position, can also be used as append,
smartly handles table of content
Args:
docpath (os.PathLike): Path to pdf
page_range (Tuple[int, int], optional): _description_. Defaults to None. (All pages)
start_at (int, optional): _description_. Defaults to None. (End of pdf aka. append)
rotate (int, optional): _description_. Defaults to 0.
links (bool, optional): _description_. Defaults to True.
annots (bool, optional): _description_. Defaults to True.
show_progress (int, optional): _description_. Defaults to 0.
final (int, optional): _description_. Defaults to 1.
save_as (os.PathLike, optional): _description_. Defaults to None.
merge (bool, optional): _description_. Defaults to False. Does not save file for merging
Returns:
_type_: _description_
"""
# Open file to insert
# Saving TOC of file to insert
doc2 = fitz.open(docpath)
toc2 = doc2.get_toc()
# In case of inserted document having no toc, insert name as toc
if len(toc2) == 0:
toc2 = [[1, os.path.splitext(os.path.basename(docpath))[0], 1]]
# Adjusting table of content aka. bookmarks in pdf however it is TOC in pymupdf context
# Saving initial TOC of original file
toc1 = self.document.get_toc()
pcount = self.document.page_count
# Extract from_page, to_page, page_count based on pages to insert for second file
if page_range is None:
start, end = (0, doc2.page_count - 1
) # Zero indexed page numbers correction
pcount2 = doc2.page_count
else:
start, end = page_range
pcount2 = abs(end - start +
1) # Zero indexed page numbers correction
# Start at the end of first file, ie. appending to the end with no argument
# In this case TOC page number of second file is increased by the page number of first file
if start_at is None or start_at >= pcount:
start_at = pcount
# Increase page numbers in doc2 toc1 with doc1 page count
for t in toc2:
t[2] += pcount
elif start_at == 0:
# Increase page numbers in doc1 toc1 with original doc2 page count
for t in toc1:
t[2] += pcount2
else:
# In case of inserting pages in between,
# Increase toc1 page number after start_at position by pcount2
# Increase toc2 page numbers by start_at
for t in toc1[start_at:]:
t[2] += pcount2
for t in toc2:
t[2] += start_at
self.document.insert_pdf(
doc2, # cannot be the same object as doc1
from_page=start, # first page to copy, default: 0
to_page=end, # last page to copy, default: last page
start_at=start_at, # target location in doc1, default: at end
rotate=rotate, # rotate copied pages
links=links, # also copy links
annots=annots, # also copy annotations
show_progress=
show_progress, # a message like Inserted 30 of 47 pages after given int
final=final # the list of already copied objects should be dropped
)
self.document.set_toc(toc1 + toc2)
doc2.close()
if not merge:
if save_as is None:
self.document.save(checkfile(path))
else:
self.document.save(checkfile(save_as))
else:
return self.document
def split(self,
page_range: Union[int, Tuple[int, int], List[int]],
single_page: bool = False,
save_as: os.PathLike = None) -> None:
"""
Summary:
Split given pdf in one of three following method based on page_range input:
If page_range is:
1. int : Split in two by given pageno, TOC is preserved only in this option
2. tuple of two int : Split page range either single pdfs or one pdf
3. list of integers : Split given arbitrary page numbers either single pdfs
or one pdf
Args:
page_range (Union[int, Tuple[int, int], List[int]]): _description_
single_page (bool, optional): _description_. Defaults to False.
"""
# Split pdf into two pdf with given page number (zero index)
if isinstance(page_range, int):
split_gr1 = (0, page_range)
split_gr2 = (page_range + 1, self.document.page_count)
toc = self.document.get_toc()
for i in [split_gr1, split_gr2]:
new_pdf = fitz.open()
new_pdf.insert_pdf(self.document, from_page=i[0], to_page=i[1])
# Adjusting page number of TOC based on split start page
cur_toc = [[t[0], t[1], t[2] - i[0]] for t in toc
if t[2] in range(i[0], i[1] + 1)]
new_pdf.set_toc(cur_toc)
if save_as is None:
new_pdf.save(checkfile(self.path))
else:
new_pdf.save(checkfile(save_as))
else:
# Split given pages with start, end number either as single page pdfs or one pdf
if isinstance(page_range, tuple) and len(page_range) == 2:
start, end = page_range
pages_to_split = range(start, end)
# Split arbitrary pages into either single page pdfs or one pdf
elif isinstance(page_range, list) and len(page_range) != 0:
pages_to_split = [i for i in page_range]
if single_page:
for pageidx in pages_to_split:
new_pdf = fitz.open()
new_pdf.insert_pdf(self.document,
from_page=pageidx,
to_page=pageidx)
if save_as is None:
new_pdf.save(checkfile(self.path))
else:
new_pdf.save(checkfile(save_as))
new_pdf.close()
else:
self.document.select(pages_to_split)
if save_as is None:
self.document.save(checkfile(self.path))
else:
self.document.save(checkfile(save_as))