def reorder_pages(src, dst, new_order):
"""
new_order is a list of following format:
[
{'page_num': 2, page_order: 1},
{'page_num': 1, page_order: 2},
{'page_num': 3, page_order: 3},
{'page_num': 4, page_order: 4},
]
Example above means that in current document of 4 pages,
first page was swapped with second one.
page_num = older page order
page_order = current page order
So in human language, each hash is read:
<page_num> now should be <page_order>
"""
page_count = get_pagecount(src)
cat_ranges = cat_ranges_for_reorder(page_count=page_count,
new_order=new_order)
cmd = [settings.BINARY_PDFTK, src, "cat"]
for page in cat_ranges:
cmd.append(str(page))
cmd.append("output")
cmd.append(dst)
run(cmd)
def resize_img(page_path, media_root):
local_abspath = os.path.join(
media_root,
page_path.document_path.url()
)
logger.debug(f"Resizing image {page_path.img_url()}")
ppmroot = os.path.join(media_root, page_path.ppmroot)
ppmroot_dirname = os.path.dirname(ppmroot)
width = page_path.step.width
if not os.path.exists(ppmroot_dirname):
logger.debug(f"PPMROOT {ppmroot_dirname} does not exists. Creating.")
os.makedirs(
ppmroot_dirname, exist_ok=True
)
else:
logger.debug(f"PPMROOT {ppmroot_dirname} already exists.")
cmd = (
settings.BINARY_CONVERT,
"-resize",
f"{width}x",
local_abspath,
# output directory path, similar to ppmroot
f"{ppmroot}-1.jpg"
)
run(cmd)
def delete_pages(src, dst, page_numbers):
cmd = [settings.BINARY_STAPLER, "del", src]
for page in page_numbers:
cmd.append(str(page))
cmd.append(dst)
run(cmd)
def extract_hocr(page_url, lang, media_root):
page_abspath = os.path.join(media_root, page_url.img_url())
hocr_root, hocr_ext = os.path.splitext(
os.path.join(media_root, page_url.hocr_url()))
cmd = (settings.BINARY_OCR, "-l", lang, page_abspath, hocr_root, "hocr")
run(cmd)
logger.debug(f"OCR for {page_url.img_url()} - Complete.")
logger.debug(f"OCR Result {page_url.hocr_url()}.")
def paste_pages_into_existing_doc(src,
dst,
data_list,
after_page_number=False,
before_page_number=False):
page_count = get_pagecount(src)
list1, list2 = split_ranges(total=page_count,
after=after_page_number,
before=before_page_number)
# notice missing A
# Letter A is assignent to current folder and
# pages from list1 and list2
letters = "BCDEFGHIJKLMNOPQRSTUVWXYZ"
letters_2_doc_map = []
letters_pages = []
letters_pages_before = []
letters_pages_after = []
letters_2_doc_map.append(f"A={src}")
for idx in range(0, len(data_list)):
letter = letters[idx]
src = data_list[idx]['src']
pages = data_list[idx]['page_nums']
letters_2_doc_map.append(f"{letter}={src}")
for p in pages:
letters_pages.append(f"{letter}{p}")
for p in list1:
letters_pages_before.append(f"A{p}")
for p in list2:
letters_pages_after.append(f"A{p}")
cmd = [
settings.BINARY_PDFTK,
]
# add A=doc1_path, B=doc2_path
cmd.extend(letters_2_doc_map)
cmd.append("cat")
# existing doc pages (may be empty)
cmd.extend(letters_pages_before)
# newly inserted pages
cmd.extend(letters_pages)
# existing doc pages (may be empty)
cmd.extend(letters_pages_after)
cmd.append("output")
cmd.append(dst)
run(cmd)
def delete_pages(src, dst, page_numbers):
page_count = get_pagecount(src)
cat_ranges = cat_ranges_for_delete(page_count, page_numbers)
cmd = [settings.BINARY_PDFTK, src, "cat"]
for page in cat_ranges:
cmd.append(str(page))
cmd.append("output")
cmd.append(dst)
run(cmd)
def convert_tiff2pdf(doc_url):
logger.debug(f"convert_tiff2pdf for {doc_url}")
new_doc_url, new_filename = pdfname_from_tiffname(doc_url)
logger.debug(f"tiff2pdf source={doc_url} dest={new_doc_url}")
cmd = (
settings.BINARY_CONVERT,
doc_url,
new_doc_url,
)
run(cmd)
# returns new filename
return new_filename
def extract_txt(page_url, lang, media_root):
page_abspath = os.path.join(
media_root,
page_url.img_url()
)
txt_root, txt_ext = os.path.splitext(
os.path.join(
media_root, page_url.txt_url()
)
)
cmd = (
settings.BINARY_OCR,
"-l",
lang,
page_abspath,
txt_root
)
run(cmd)
def extract_txt(page_url, lang, media_root):
page_abspath = os.path.join(
media_root,
page_url.img_url()
)
txt_root, txt_ext = os.path.splitext(
os.path.join(
media_root, page_url.txt_url()
)
)
cmd = (
"tesseract",
"-l",
lang,
page_abspath,
txt_root
)
run(cmd)
def extract_hocr(page_url, lang, media_root):
page_abspath = os.path.join(
media_root,
page_url.img_url()
)
hocr_root, hocr_ext = os.path.splitext(
os.path.join(media_root, page_url.hocr_url())
)
cmd = (
"tesseract",
"-l",
lang,
page_abspath,
hocr_root,
"hocr"
)
run(cmd)
def convert_tiff2pdf(doc_url):
logger.debug(f"convert_tiff2pdf for {doc_url}")
# basename is filename + ext (no path)
basename = os.path.basename(doc_url)
base_root, base_ext = os.path.splitext(basename)
root, ext = os.path.splitext(doc_url)
new_doc_url = f"{root}.pdf"
logger.debug(f"tiff2pdf source={doc_url} dest={new_doc_url}")
cmd = (
settings.BINARY_CONVERT,
doc_url,
new_doc_url,
)
run(cmd)
# returns new filename
return f"{base_root}.pdf"
def extract_img(page_path, media_root):
local_abspath = os.path.join(
media_root,
page_path.document_path.url()
)
logger.debug(f"Extracing image for {page_path.img_url()}")
ppmroot = os.path.join(media_root, page_path.ppmroot)
ppmroot_dirname = os.path.dirname(ppmroot)
page_num = page_path.page_num
width = page_path.step.width
if not os.path.exists(ppmroot_dirname):
logger.debug(f"PPMROOT {ppmroot_dirname} does not exists. Creating.")
os.makedirs(
ppmroot_dirname, exist_ok=True
)
else:
logger.debug(f"PPMROOT {ppmroot_dirname} already exists.")
cmd = (
settings.BINARY_PDFTOPPM,
"-jpeg",
"-f",
str(page_num),
"-l", # generate only one page
str(page_num),
"-scale-to-x",
str(width),
"-scale-to-y",
"-1", # it will adjust height according to img ratio
local_abspath,
# output directory path,
ppmroot
)
run(cmd)
def paste_pages(src,
dst,
data_list,
dst_doc_is_new=True,
after_page_number=False,
before_page_number=False):
"""
dest_doc_ep = endpoint of the doc where newly created
file will be placed.
src_doc_ep_list is a list of following format:
[
{
'doc_ep': doc_ep,
'page_nums': [page_num_1, page_num_2, page_num_3]
},
{
'doc_ep': doc_ep,
'page_nums': [page_num_1, page_num_2, page_num_3]
},
...
]
src_doc_ep_list is a list of documents where pages
(with numbers page_num_1...) will be paste from.
dst_doc_is_new = True well.. destination document was just created,
we are pasting here cutted pages into some folder as new document.
In this case 'after' and 'before' arguments are ignored
dst_doc_is_new = False, pasting pages into exiting document.
If before_page_number > 0 - paste pages before page number
'before_page_number'
If after_page_number > 0 - paste pages after page number
'after_page_number'
before_page_number argument has priority over after_page_number.
If both before_page_number and after_page_number are < 0 - just paste
pages at the end of the document.
"""
if not dst_doc_is_new:
return paste_pages_into_existing_doc(
src=src,
dst=dst,
data_list=data_list,
after_page_number=after_page_number,
before_page_number=before_page_number)
letters = "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
letters_2_doc_map = []
letters_pages = []
for idx in range(0, len(data_list)):
letter = letters[idx]
src = data_list[idx]['src']
pages = data_list[idx]['page_nums']
letters_2_doc_map.append(f"{letter}={src}")
for p in pages:
letters_pages.append(f"{letter}{p}")
cmd = [
settings.BINARY_PDFTK,
]
# add A=doc1_path, B=doc2_path
cmd.extend(letters_2_doc_map)
cmd.append("cat")
cmd.extend(letters_pages)
cmd.append("output")
cmd.append(dst)
run(cmd)
