Extract the bounding box of the first, leftmost rectangle dividing labels and annotations in a PDF-formatted UCSC genome browser shot

ucsc_pdf_bbox_parser.py

#!/usr/bin/env python

import sys
from pdfminer.pdfparser import PDFParser
from pdfminer.pdfdocument import PDFDocument
from pdfminer.pdfpage import PDFPage
from pdfminer.pdfpage import PDFTextExtractionNotAllowed
from pdfminer.pdfinterp import PDFResourceManager
from pdfminer.pdfinterp import PDFPageInterpreter
from pdfminer.pdfdevice import PDFDevice
from pdfminer.layout import LAParams, LTFigure, LTCurve, LTRect
from pdfminer.converter import PDFPageAggregator

g_fn = None
g_bbox = None

def parse_lt_objs (lt_objs, page_number, found_rect, text=[]):
    """
    Iterate through the list of LT* objects 
    one level deep, and capture the data contained 
    in each
    """
    content = []
    for lt_obj in lt_objs:
        if isinstance(lt_obj, LTRect) and not found_rect:
            content.append(lt_obj.bbox)
            found_rect = True
        elif isinstance(lt_obj, LTFigure) and not found_rect:
            # LTFigure objects are containers for other LT* objects, so recurse through the children
            content.extend(parse_lt_objs(lt_obj, page_number, found_rect, content))
    return content

def set_bbox(b=None):
    global g_bbox
    g_bbox = b

def get_bbox():
    return g_bbox
    
def set_fn(f=None):
    global g_fn
    g_fn = sys.argv[1] if not f else f

def get_fn():
    return g_fn

def main():
    set_fn(sys.argv[1])
    parse(True)
    
def parse(debug=False):
    fp = open(get_fn(), 'rb')
    parser = PDFParser(fp)
    doc = PDFDocument(parser, None)
    if not doc.is_extractable:
        raise PDFTextExtractionNotAllowed
    rsrcmgr = PDFResourceManager()
    laparams = LAParams()
    device = PDFPageAggregator(rsrcmgr, laparams=laparams)
    interpreter = PDFPageInterpreter(rsrcmgr, device)
    
    content = []
    for i, page in enumerate(PDFPage.create_pages(doc)):
        interpreter.process_page(page)
        layout = device.get_result()
        content.extend(parse_lt_objs(layout, (i+1), False))

    set_bbox(content[0])
    
    # print first LTLine object bbox tuple
    if debug: sys.stderr.write("First LTRect bbox in PDF -> %s\n" % (get_bbox(),))

if __name__ == "__main__":
    main()