从 PDF 中提取声音注释

Question

我有一个列出 PDF 文件注释的脚本从 pdf 解析注释：import popplerqt5import argparsedef extract(fn):    doc = popplerqt5.Poppler.Document.load(fn)    annotations = []    for i in range(doc.numPages()):        page = doc.page(i)        for annot in page.annotations():            contents = annot.contents()            if contents:                annotations.append(contents)                print(f'page={i + 1} {contents}')    print(f'{len(annotations)} annotation(s) found')    return annotationsif __name__ == '__main__':    parser = argparse.ArgumentParser()    parser.add_argument('fn')    args = parser.parse_args()    extract(args.fn)但它只适用于文本注释，有很多Python库，如Poppler、PyPDF2、PyMuPDF，我一直在搜索他们的文档和源代码很多，就我而言，他们无法提取声音注释的二进制。你知道有哪个图书馆可以做到这一点吗？我需要提取这些声音注释的二进制文件并将它们转换为 MP3。

Answer 1

PyMuPDF 的下一版本将支持提取音频注释。使用此脚本使用 PyMuPDF 从 PDF 中提取音频注释，它很容易使用，只需调用该脚本并将 PDF 文件作为第一个参数传递即可：python script.py myfile.pdf

注意：仅适用于 Windows。

import fitz, sys, os, subprocess

assert len(sys.argv) == 2, "need filename as parameter"

ifile = sys.argv[1]

doc = fitz.open(ifile)

ofolder = os.path.dirname(ifile)

if ofolder == "":

    ofolder = os.getcwd()

flnm = os.path.splitext(os.path.basename(ifile))[0]

defolder = ofolder + "\\" + flnm

os.mkdir(defolder)

defolder = defolder + "\\" + flnm

for page in doc:

    print(page)

    annotNumber = 1

    for annot in page.annots(types=[fitz.PDF_ANNOT_SOUND]):  

        try: 

            sound = annot.soundGet()  

        except Exception as e:

            print(e)

            continue

        for k, v in sound.items():

            print(k, "=", v if k != "stream" else len(v))

        ofile = defolder + ".page." + str(page.number) + ".annot." + str(annotNumber) + ".raw"

        fout = open(ofile,"wb") 

        fout.write(sound["stream"])

        fout.close()

        ofileffmpeg = defolder + ".page." + str(page.number) + ".annot." + str(annotNumber) + ".mp3"

        annotNumber += 1

        if "channels" in sound:

            channels = str(sound["channels"])

        else:

            channels = "1"

        if "encoding" in sound:

            if sound["encoding"] == "Signed":

                encoding = "s"

            else:

                encoding = "u"

        else:

            encoding = "u"

        if "bps" in sound:

            fmt = encoding + str(sound["bps"]) + "be"

        else:

            fmt = encoding + "8"

        subprocess.call(['ffmpeg', '-hide_banner', '-f', fmt, '-ar', str(sound["rate"]), '-ac', channels, '-i', str(ofile), str(ofileffmpeg)], shell=True)